From karlmax at oberland.net Thu Feb 1 11:56:05 2001 From: karlmax at oberland.net (Karl-Max Wagner) Date: Fri Feb 10 19:39:08 2006 Subject: [Pipet Users] FreeSpeech Test results In-Reply-To: <3A779757.CABD82B3@gel.usherb.ca> from "Jean-Marc Valin" at Jan 30, 2001 11:40:55 PM Message-ID: <200102011656.QAA00658@schreyer.oberland.net> > My mail server blocked for a couple days, so I got all the messages in batch, in > the wrong order... Are you still having problems with Overflow or is everything I didn't work on the new snapshot any more. So the failure logging is still "state of the art". > fine now? BTW, did I say you should track and remove any Overflow library that > was previously installed elsewhere (like /usr/lib or /usr/local/lib)? If there's Is that already a problem at compile time ? I thought it is only a problem at runtime. > P.S. Another thing you need to make sure of: The Overflow ABI changes > frequently, so make sure that the Piper version you have is linked to the exact > same version of Overflow that's installed. Otherwise, it's almost guarantied not > to work. Is a cvs snapshot of the same day OK ? BTW: when not mailing me on the list you may also use french ( but no quebecois please ) as I'm in good command of the language. ======================================================================= "It was hell. They knew it. Karl-Max Wagner But they called it karlmax@oberland.net W-I-N-D-O-Z-E" ham radio: DB8CO *********Member of No Code International********* ***********Visit http://www.nocode.org*********** ********Membership Number NCI-2563-DB8CO********* _ / / (_)__ __ ____ __ / /__/ / _ \/ // /\ \/ / . . . t h e c h o i c e o f a /____/_/_//_/\_,_/ /_/\_\ G N U g e n e r a t i o n . . "Et ceterum censeo ut Microsoftem delendum esse" (Cato, adapted) "Quo usque tandem abutere nostra patientia ?" (Cicero, original) ======================================================================= From jarl at xs4all.nl Tue Feb 6 14:24:08 2001 From: jarl at xs4all.nl (jarl van katwijk) Date: Fri Feb 10 19:39:08 2006 Subject: [Pipet Users] Update References: <200101290125.BAA00866@schreyer.oberland.net> <3A74E417.CBDEF0FE@bioinformatics.org> <3A74E664.B3392DB3@bioinformatics.org> Message-ID: <3A804F58.162FB00D@xs4all.nl> Hi Pipers, I'm working on the OOP version of the phoenix BL, and this is consuming all the time my doctor allows me to spend on Piper. It looks very promising, and parallel to the oop-ing I'm implementing the security system, from bottom up. Very simplistic this will look like this: data node (nodes, containers for plugins) legacy (OS specific stuff.. I still have hopes one day this will wrap all that a 'lsof' lists) recourse (scheduling, allocation, relocation, etc. of nodes) | \ / area (security stuff) | \ / engine (runtime logic) | \ / plugins (like the DL and PL) Something else the BL needs, in my view, is a standard to communicate with other systems. For sure with neighboring instances of Piper, but also with other systems. KQML seems to be the ideal candidate: it's designed for intercommunication between agents, to share knowledge. (Knowledge Query and Manipulation Language) A nasty side effect the work on the BL core is that the efforts on implementing kqml is stalled for some time. On the development list we already had a discussion about Kapi, which is a kqml library in C. So we've 2 problems: it aint OOP and the code is licensed by something that prohibits us to merge it into Piper. The only solution left is to use Kapi as an example and to build our own kqml implementation. And yes, there doesn't seem to be an other kqml library around we even can get source code of. So you can wait for me to do all the work, but I think the growing crowd of people interested in Piper want to have something workable a bit sooner ;) Have any comments on kqml? Is there something else\better? Do you want to take up the coding of a library for piper? thnx 4 reading, jarl From Nicolas.Chauvat at logilab.fr Tue Feb 6 14:23:04 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:08 2006 Subject: [Pipet Users] Update In-Reply-To: <3A804F58.162FB00D@xs4all.nl> Message-ID: > Something else the BL needs, in my view, is a standard to communicate > with other systems. For sure with neighboring instances of Piper, but > also with other systems. KQML seems to be the ideal candidate: it's > designed for intercommunication between agents, to share knowledge. > (Knowledge Query and Manipulation Language) > > ... > > So you can wait for me to do all the work, but I think the growing crowd of > people interested in Piper want to have something workable a bit sooner ;) > Have any comments on kqml? Is there something else\better? Do you want to take up > the coding of a library for piper? I've used KQML in the past. I'm not sure it fits... or maybe it is because I have bad KQML memories! What are you looking for and what would you like to achieve? Maybe I can help in choosing the proper language/tool by discussing the topic with you? -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From jarl at xs4all.nl Tue Feb 6 14:49:31 2001 From: jarl at xs4all.nl (jarl van katwijk) Date: Fri Feb 10 19:39:08 2006 Subject: [Pipet Users] Update References: Message-ID: <3A80554B.AC720497@xs4all.nl> > I've used KQML in the past. I'm not sure it fits... or maybe it is because > I have bad KQML memories! > > What are you looking for and what would you like to achieve? Maybe I can > help in choosing the proper language/tool by discussing the topic with > you? > Currently Piper doesn't support inter instance communications, just because we never had the time for it. (Sounds like a deja-vu) So we still have the luxury of a free choose now. By inter instance communication I'm trying to say this: Piper Piper running on --?KQML?-- running on machine A machine B It's the BL that will handle this communications, so we COULD just design what we need and implement this. But I think a mayor feature would be if these communications were based on a standard so it can comm. with other legacy systems to. Kqml seems to be designed for this purpose. Why do you think it isn't? jarl From Nicolas.Chauvat at logilab.fr Tue Feb 6 14:53:21 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:08 2006 Subject: [Pipet Users] Update In-Reply-To: <3A80554B.AC720497@xs4all.nl> Message-ID: > Currently Piper doesn't support inter instance communications, just > because we never had the time for it. (Sounds like a deja-vu) So we > still have the luxury of a free choose now. > > By inter instance communication I'm trying to say this: > > Piper Piper > running on --?KQML?-- running on > machine A machine B > > It's the BL that will handle this communications, so we COULD just > design what we need and implement this. But I think a mayor feature > would be if these communications were based on a standard so it can > comm. with other legacy systems to. Kqml seems to be designed for this > purpose. Why do you think it isn't? I'm not sure I have an opinion yet. Mainly because I'm not clear what kind of messages you want these piper instances to exchange. I'm sorry if the best answer to my question is "tfm says that", I never managed to get a clear understanding of the roles of the various piper layers when reading the documentation. And I never got time to read the code... Could you explain again what those messages would be like: what is their purpose, what kind of embedded information/data do they transport, etc. As far as KQML as an intercommunication language, I've had bad experiences with it, mainly because it's not really a standard, so you end up with a lot of KQML dialects. That can be worse than having "piper's own communication language" and providing KQML translators for specific applications. -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From jarl at xs4all.nl Tue Feb 6 15:22:23 2001 From: jarl at xs4all.nl (jarl van katwijk) Date: Fri Feb 10 19:39:08 2006 Subject: [Pipet Users] Update References: Message-ID: <3A805CFF.22DC6532@xs4all.nl> > > Could you explain again what those messages would be like: what is their > purpose, what kind of embedded information/data do they transport, etc. The main task of the BL is communications between instances of Piper. Emerging tasks are resource management and security. Some issues that come to mind atm are: 1. Resource localisation (a central registration server that keeps track of Piper instances) 2. Offered data streaming (push) 3. Requested data streaming (pull) Other stuff we'll need, and that's outside the scope of kqml, are 4. Node relocation (numero uno pain in the butt) 5. Authentication (probably Akenti, but maybe Sesame) 6. Resouce management (like ram, HD space, ea) > As far as KQML as an intercommunication language, I've had bad experiences > with it, mainly because it's not really a standard, so you end up with a > lot of KQML dialects. That can be worse than having "piper's own > communication language" and providing KQML translators for specific > applications. Ic. This is bad indead. darn.. really bad. Maybe that's why all the code I could find was made 5-6 years ago. From jarl at xs4all.nl Tue Feb 6 15:32:36 2001 From: jarl at xs4all.nl (jarl van katwijk) Date: Fri Feb 10 19:39:08 2006 Subject: [Pipet Users] Documentation References: Message-ID: <3A805F64.3125B89B@xs4all.nl> > I'm not sure I have an opinion yet. Mainly because I'm not clear what kind > of messages you want these piper instances to exchange. I'm sorry if the > best answer to my question is "tfm says that", I never managed to get a > clear understanding of the roles of the various piper layers when reading > the documentation. And I never got time to read the code... Oops. Probably too obvious to us. There will be a layer document soon, I'll write a draft tonight. jarl From Nicolas.Chauvat at logilab.fr Tue Feb 6 15:38:05 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:08 2006 Subject: [Pipet Users] Update In-Reply-To: <3A805CFF.22DC6532@xs4all.nl> Message-ID: > The main task of the BL is communications between instances of Piper. > Emerging tasks are resource management and security. > > Some issues that come to mind atm are: > 1. Resource localisation (a central registration server that keeps > track of Piper instances) > 2. Offered data streaming (push) > 3. Requested data streaming (pull) > > Other stuff we'll need, and that's outside the scope of kqml, are > 4. Node relocation (numero uno pain in the butt) > 5. Authentication (probably Akenti, but maybe Sesame) > 6. Resouce management (like ram, HD space, ea) Ok, let me rephrase that: You want the BL to manage both the process distribution aspects and the node coordination aspects. * 1, 2 and 3 are directory services (who does what and where) * 4 and 5 belongs to the node coordination topic * I'm still not clear what 6 is. Two more question: * are you sure you want 4 and how good do you want it? * are you sure you want to use a centralized approach for 1? The main comment I have is that this looks like we need to do some reading and look at distributed architectures, for the above subjects have been worked on already. I can't suggest a solution as I don't know what kind of information the directory srevices would have to publish and query, nor what 4 and 6 precisely are. > Ic. This is bad indead. darn.. really bad. Maybe that's why all the > code I could find was made 5-6 years ago. Depending on what you actually need, you may get away with something like RPC-over-something-else. Ever thought of SOAP maybe? Narval is using the python implementation by pythonware.com. Slow, but standard. -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From jarl at xs4all.nl Tue Feb 6 16:11:02 2001 From: jarl at xs4all.nl (jarl van katwijk) Date: Fri Feb 10 19:39:08 2006 Subject: [Pipet Users] Update References: Message-ID: <3A806866.E6F65AAD@xs4all.nl> Hi Nicolas, I started to answer your email, but halfway I quit. Maybe you have some time to spare for an irc session? Else we'll be mailing about this subject for weeks. jarl From jeff at bioinformatics.org Tue Feb 6 17:09:08 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:08 2006 Subject: [Pipet Users] Update References: Message-ID: <3A807604.D5060A8A@bioinformatics.org> Nicolas Chauvat wrote: > > > 1. Resource localisation (a central registration server that keeps > > track of Piper instances) > * are you sure you want to use a centralized approach for 1? We (Jarl, Jean-Marc, Brad and I) discussed directory services a while back. Brad suggested that CORBA objects be registered through a single, CORBA-based directory service, perhaps located at Bioinformatics.org. But, since then, I have been leaning toward having peer-to-peer directory services, meaning each local instance of Piper would possess its own directory service, and that would register local nodes PLUS the nodes of whatever other remote Piper instance it is connected to (a la Gnutella). Since there are so many P2P systems out there, we probably don't need to invent anything of our own when it comes to directory services. But I don't know if a CORBA-based directory service is still on the slate. > > 6. Resouce management (like ram, HD space, ea) > * I'm still not clear what 6 is. I think Jarl is talking about providing resource information with each node. IOW, Piper would let a remote user know what your resources are (esp. network speed). This would be helpful if your node(s) is not the only one on the Net. Take Napster, for example. You may find 10 mp3's that are exactly the same, but you would prefer to download from the guy with a T3 over someone with a 28.8 modem. > Depending on what you actually need, you may get away with something like > RPC-over-something-else. Ever thought of SOAP maybe? Narval is using the > python implementation by pythonware.com. Slow, but standard. I know little about SOAP, but I know that CORBA is pretty much in the same class. To what extent can we make use of CORBA over KQML (or SOAP or XML-RPC)? Also, how about using only the parts of KQML that we like or need? There probably isn't much use in sticking to a standard that doesn't exist ;-) Cheers. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jarl at xs4all.nl Tue Feb 6 17:25:07 2001 From: jarl at xs4all.nl (jarl van katwijk) Date: Fri Feb 10 19:39:08 2006 Subject: [Pipet Users] Update References: <3A807604.D5060A8A@bioinformatics.org> Message-ID: <3A8079C3.9217C1DF@xs4all.nl> > > Since there are so many P2P systems out there, we probably don't need to > invent anything of our own when it comes to directory services. But I don't > know if a CORBA-based directory service is still on the slate. Ok, we wont go for the all-on-one-place design, that's for sure. The central service I have in mind is just for locating other piper systems. Nothing else, just a list of IP numbers. All other stuff is P2P. > > Depending on what you actually need, you may get away with something like > > RPC-over-something-else. Ever thought of SOAP maybe? Narval is using the > > python implementation by pythonware.com. Slow, but standard. > > I know little about SOAP, but I know that CORBA is pretty much in the same > class. To what extent can we make use of CORBA over KQML (or SOAP or > XML-RPC)? There're two thing here: transport and language. I wasn't even considering the transport until now ;) But after the bad news about the dialects of KQML I found this: http://www.fipa.org/specs/pesspecs.tar.gz. I'll try to isolate relevant documents\pages and make them subject to discussion. > Also, how about using only the parts of KQML that we like or need? There > probably isn't much use in sticking to a standard that doesn't exist ;-) No:( I find consolation in the fact that I hardly started coding on kqml. jarl From jeff at bioinformatics.org Tue Feb 6 17:42:07 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:08 2006 Subject: [Pipet Users] O'Reilly Network: The Open Lab [Feb. 01, 2001] Message-ID: <3A807DBF.FD6FF5CE@bioinformatics.org> TOL and Piper are mentioned on the O'Reilly P2P Conference website: http://www.oreillynet.com/pub/d/455 Jeff From jeff at bioinformatics.org Tue Feb 6 17:45:09 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] O'Reilly Network: Piper [Feb. 04, 2001] Message-ID: <3A807E75.8D66689D@bioinformatics.org> Piper has its own listing :-) http://www.oreillynet.com/pub/d/454 These descriptions are not my own, BTW, so don't complain to me ;-) Jeff From jeff at bioinformatics.org Tue Feb 6 18:05:50 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] Update References: <3A807604.D5060A8A@bioinformatics.org> <3A8079C3.9217C1DF@xs4all.nl> Message-ID: <3A80834E.5B6F6D99@bioinformatics.org> jarl van katwijk wrote: > > Ok, we wont go for the all-on-one-place design, that's for sure. The central > service I have in mind is just for locating other piper systems. Nothing else, > just a list of IP numbers. All other stuff is P2P. I thought about this for a while: How will a user know who they can connect to (who is out there) when there will be no central directory service? And I think the solution is simple. We can do what XChat does: include a list of addresses with the software. Perhaps we can have a modifiable list for each user, let's say an "address book". We'll simply put a few addresses in there to start them off. Actually, if we have a few "big nodes" (or instances of Piper) running (at Bioinformatics.org for one), all that the users will have to do is connect to them, and, because Piper is a P2P system, users will have access to more than what is local to the big node. Users will have access to all instances of Piper connected (all that are public anyway). The effect would be nearly the same as having a central directory service. > There're two thing here: transport and language. I wasn't even considering the > transport until now ;) But after the bad news about the dialects of KQML I found > this: http://www.fipa.org/specs/pesspecs.tar.gz. I'll try to isolate relevant > documents\pages and make them subject to discussion. If we're thinking of Piper-to-Piper communication as something like agent-to-agent communication, perhaps there is something we can share with Narval? But, I don't know, maybe we don't want to get into SOAP when we are already using CORBA. Cheers. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jarl at xs4all.nl Tue Feb 6 18:25:08 2001 From: jarl at xs4all.nl (jarl van katwijk) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] P2P References: <3A807604.D5060A8A@bioinformatics.org> <3A8079C3.9217C1DF@xs4all.nl> <3A80834E.5B6F6D99@bioinformatics.org> Message-ID: <3A8087D4.F549931E@xs4all.nl> > > I thought about this for a while: How will a user know who they can connect to > (who is out there) when there will be no central directory service? And I > think the solution is simple. We can do what XChat does: include a list of > addresses with the software. Perhaps we can have a modifiable list for each > user, let's say an "address book". We'll simply put a few addresses in there > to start them off. > > Actually, if we have a few "big nodes" (or instances of Piper) running (at > Bioinformatics.org for one), all that the users will have to do is connect to > them, and, because Piper is a P2P system, users will have access to more than > what is local to the big node. Users will have access to all instances of > Piper connected (all that are public anyway). The effect would be nearly the > same as having a central directory service. Very good thinking Jeff! I like this best of what passed until now. You just got 2 cents richer jarl From jeff at bioinformatics.org Tue Feb 6 18:42:48 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] Piper: El futuro del p2p. Message-ID: <3A808BF8.15BFF1D2@bioinformatics.org> We're the future of P2P, at least according to this great article at Hispamp3: http://www.hispamp3.com/noticias/0102/010205_5.shtml It's in Spanish, so you may want to use a translator: http://www.systransoft.com/ I'm not so sure that Piper can be used to tranfer "bellboys and eyelashes" though 8^D I'll blame that one on the translator. Jeff From Nicolas.Chauvat at logilab.fr Wed Feb 7 06:38:19 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] Update In-Reply-To: <3A807604.D5060A8A@bioinformatics.org> Message-ID: > > > 6. Resouce management (like ram, HD space, ea) > > > * I'm still not clear what 6 is. > > I think Jarl is talking about providing resource information with each node. > IOW, Piper would let a remote user know what your resources are (esp. network > speed). This would be helpful if your node(s) is not the only one on the > Net. Take Napster, for example. You may find 10 mp3's that are exactly the > same, but you would prefer to download from the guy with a T3 over someone > with a 28.8 modem. Got it. I would advocate looking at distributed computing litterature in case we'd want to implement algorithms that choose the "best node" for a certain task. Principles are not very different from deciding when to migrate processes from a CPU to another in clusters, or when having mobile code (agents?) move around host nodes. -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From Nicolas.Chauvat at logilab.fr Wed Feb 7 06:48:07 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] Update In-Reply-To: <3A8079C3.9217C1DF@xs4all.nl> Message-ID: > > > Depending on what you actually need, you may get away with something like > > > RPC-over-something-else. Ever thought of SOAP maybe? Narval is using the > > > python implementation by pythonware.com. Slow, but standard. > > > > I know little about SOAP, but I know that CORBA is pretty much in the same > > class. To what extent can we make use of CORBA over KQML (or SOAP or > > XML-RPC)? > > There're two thing here: transport and language. I wasn't even > considering the transport until now ;) I agree: CORBA, SOAP and XML-RPC are transports. KQML or FIPA are agent languages and messages can be exchanged using any of the previous tools. The two questions should be dealt with separately. FWIW, SOAP is a "better" XML-RPC. > But after the bad news about the dialects of KQML I found this: > http://www.fipa.org/specs/pesspecs.tar.gz. I'll try to isolate > relevant documents\pages and make them subject to discussion. For what I know, FIPA is more of a standard than KQML and is supposed to replace it. I've never used it, though, so I can't comment that much. But if the consensus is that an agent communication language is the right tool for piper nodes to exchange information about each other, then FIPA is probably a good choice. -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From Nicolas.Chauvat at logilab.fr Wed Feb 7 06:59:04 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] Update In-Reply-To: <3A80834E.5B6F6D99@bioinformatics.org> Message-ID: > I thought about this for a while: How will a user know who they can > ... > Actually, if we have a few "big nodes" (or instances of Piper) running > (at Bioinformatics.org for one), all that the users will have to do is > connect to them, and, because Piper is a P2P system, users will have > access to more than what is local to the big node. Users will have > access to all instances of Piper connected (all that are public > anyway). The effect would be nearly the same as having a central > directory service. This sounds good to me. > If we're thinking of Piper-to-Piper communication as something like > agent-to-agent communication, perhaps there is something we can share > with Narval? Narval is a language+interpreter to set up personal assistants (see http://www.logilab.org/narval/app.html for examples). We did not commit to any agent communication language yet, and probably won't as we'd better offer different modules that implement different languages. So I guess we will share things :-) It would also be interesting for both groups to look at each others' way of defining executable resources (or whatever we could agree on calling them). In Narval we have actions and transforms that are "executable". They all have a prototype so that we can check at run time that they have what they need as input and that their output is correct. Originally, inspiration came from pre/post conditions in Eiffel. We put a lot of work in designing this and my intuition tells me that we could probably share things there (didn't get to install and test piper to look into the matter further, so I have to fall back to intuition, sorry). > But, I don't know, maybe we don't want to get into SOAP when we are > already using CORBA. Narval is already using SOAP to connect the interpreter and the client GUI. We might end up using something else, as our current SOAP implementation is depressingly slow, but using SOAP has the benefit of not requiring an ORB. -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From jeff at bioinformatics.org Wed Feb 7 10:53:57 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] Update References: Message-ID: <3A816F95.EC4EAD9A@bioinformatics.org> Nicolas Chauvat wrote: > > Narval is already using SOAP to connect the interpreter and the client > GUI. We might end up using something else, as our current SOAP > implementation is depressingly slow, but using SOAP has the benefit of not > requiring an ORB. Way back when Brad joined the Loci project, we designed and built our own XML-based RPC, which used standard UNIX sockets. It too proved to be "depressingly slow", so we switched to CORBA. CORBA was so much faster that we trashed all of our previous work and stuck with CORBA wherever applicable. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jarl at xs4all.nl Thu Feb 8 16:14:35 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] Update References: <3A816F95.EC4EAD9A@bioinformatics.org> Message-ID: <3A830C3B.3B029F2D@xs4all.nl> > Way back when Brad joined the Loci project, we designed and built our own > XML-based RPC, which used standard UNIX sockets. It too proved to be > "depressingly slow", so we switched to CORBA. CORBA was so much faster that > we trashed all of our previous work and stuck with CORBA wherever applicable. > ;) Corba is fine with me ofcourse, but I'm not sure whether it is good over internet connections. Will need stuff like SECIOP to become usable. Had a quick look just now, couldn't find any library that is GPL'd. It can turn out a gnet based interface is better here. (glib = lib in the glib/gdk/gtk library family, special for network connections) jarl From jeff at bioinformatics.org Thu Feb 8 16:33:54 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] Update References: <3A816F95.EC4EAD9A@bioinformatics.org> <3A830C3B.3B029F2D@xs4all.nl> Message-ID: <3A8310C2.47D5A5F9@bioinformatics.org> Jarl van Katwijk wrote: > > ;) Corba is fine with me ofcourse, but I'm not sure whether it is good > over internet connections. Will need stuff like SECIOP to become usable. > Had a quick look just now, couldn't find any library that is GPL'd. It > can turn out a gnet based interface is better here. (glib = lib in the > glib/gdk/gtk library family, special for network connections) I remember when the gnet was first discussed on the Gtk or Gnome list. It uses sockets, right? Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jeff at bioinformatics.org Thu Feb 8 19:54:46 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] Intel locks up security code for P2P Message-ID: <3A833FD6.33B8B259@bioinformatics.org> http://news.cnet.com/news/0-1003-200-4754163.html Jeff From jeff at bioinformatics.org Fri Feb 9 22:13:58 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] Khoros and Cantata Message-ID: <3A84B1F6.89BC2BED@bioinformatics.org> Pipers, I don't think I mentioned Khoros directly on this list. Although I have seen a number of references to it, I never bothered to look at it :-P I turns out that Khoros, with the Cantata visual shell, is probably the closest thing to Piper I have yet seen (with DX, AVS, OpenBSA, and ISYS next in line) . Here is an excerpt from a message by Ed Hall back in 1999: ---------------------------------------- There was a system developed starting back in the mid-1980's at UNM called "Khoros" that did statistical and signal processing using a "visual shell" called "Cantata." It's an incredibly powerful environment that allowed the construction of highly complex processes via the interconnection and manipulation of block icons. It once was free software (it was developed at public expense, so it was required to be). Over the years it has been commercialized, unfortunately (a good argument for the (L)GPL, to be sure). But it's still free for academic use--check out www.khoral.com--under a highly restrictive license that somehow managed to placate the original funders. It might be worth a look. It runs under Linux... ---------------------------------------- Indeed, Khoros is being sold by Khoral, although it seems only within the last couple years: http://www.khoral.com/products/products.html You may find the white paper an amazing read, considering the concepts are those we developed independently for Piper: http://www.khoral.com/ideas/technology/cantata.pdf So, if Khoros is much older, I guess we can say that Piper is like an Open Source Khoros work-alike :-) But I think we can continue to innovate, improve upon the concepts, and add features, like contemporary P2P capabilities. Cheers. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From karlmax at oberland.net Sat Feb 10 16:52:55 2001 From: karlmax at oberland.net (Karl-Max Wagner) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] Khoros and Cantata In-Reply-To: <3A84B1F6.89BC2BED@bioinformatics.org> from "J.W. Bizzaro" at Feb 10, 2001 03:13:58 AM Message-ID: <200102102152.VAA00566@schreyer.oberland.net> > So, if Khoros is much older, I guess we can say that Piper is like an Open > Source Khoros work-alike :-) But I think we can continue to innovate, improve > upon the concepts, and add features, like contemporary P2P capabilities. In your place I'd not care much whether it is older or not or whether it exists or suchlike. In real life the question is not who was first but who was first to put a thing into common practice. Linux is the rererere.....incarnation of Unix, but nevertheless, being free it became everybody's friend and devoured pretty much all of its older non free peers. Free Software is also very much about putting technologies into common practice and as such free stuff in the long run eats up proprietary stuff. Thus work put into piper is work well spent for sure. Just my 2 cents.... ======================================================================= "It was hell. They knew it. Karl-Max Wagner But they called it karlmax@oberland.net W-I-N-D-O-Z-E" ham radio: DB8CO *********Member of No Code International********* ***********Visit http://www.nocode.org*********** ********Membership Number NCI-2563-DB8CO********* _ / / (_)__ __ ____ __ / /__/ / _ \/ // /\ \/ / . . . t h e c h o i c e o f a /____/_/_//_/\_,_/ /_/\_\ G N U g e n e r a t i o n . . "Et ceterum censeo ut Microsoftem delendum esse" (Cato, adapted) "Quo usque tandem abutere nostra patientia ?" (Cicero, original) ======================================================================= From jeff at bioinformatics.org Sat Feb 10 17:15:23 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] Khoros and Cantata References: <200102102152.VAA00566@schreyer.oberland.net> Message-ID: <3A85BD7B.7182F0B1@bioinformatics.org> I agree completely. The existence of UNIX (even BSD UNIX) did not deter Linus from developing Linux, and it is a Good Thing that it did not. Plus, I think we have many good ideas that do not exist anywhere else. Jeff Karl-Max Wagner wrote: > > In your place I'd not care much whether it is older or not or > whether it exists or suchlike. In real life the question is not > who was first but who was first to put a thing into common > practice. Linux is the rererere.....incarnation of Unix, but > nevertheless, being free it became everybody's friend and > devoured pretty much all of its older non free peers. Free > Software is also very much about putting technologies into > common practice and as such free stuff in the long run eats up > proprietary stuff. Thus work put into piper is work well spent > for sure. -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From karlmax at oberland.net Sat Feb 10 19:04:58 2001 From: karlmax at oberland.net (Karl-Max Wagner) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] Khoros and Cantata In-Reply-To: <3A85BD7B.7182F0B1@bioinformatics.org> from "J.W. Bizzaro" at Feb 10, 2001 10:15:23 PM Message-ID: <200102110004.AAA00750@schreyer.oberland.net> > I agree completely. The existence of UNIX (even BSD UNIX) did not deter Linus ....and the existence of Linux didn't deter the xBSD people from advancing their projects, so that by now we have not only one, but four production quality OS's to choose from. > from developing Linux, and it is a Good Thing that it did not. Plus, I think > we have many good ideas that do not exist anywhere else. Jeff I think the best thing with piper is the least obvious and most abstract: AFAIK piper is the first time pretty diverse projects found out that they had a lot of overlap and joined in solving the overlapping problems together. This is truly astounding, because, not using hindsight, it required thinking on a level of high abstraction from all those participating to find that out. This led to a state of hight modularization in the project which is already paying off: more projects now are joining in ( lately we got Narval on board, for an example ). Here some crazy idea from the area of mathematics: coupling some symbolic mathematics package with the piper UI engine it should be possible to draw up mathematical relationships as block diagrams and then automatically manipulate them to one's heart content using the underlying symbolic mathematics package. For lack of a better term, let's call that formula compilation. The graphical entry is a high level representation of the mathematical relationship, the actual formula something like assembly code which is automatically generated, processed and optimized to provide what is desired. Woudn't that relieve us from the drudgery of low level formula handling and lift mathematics to a new level of abstraction ? Just food for thought...... Anyway, due to our stuff being free, we are also free to do things in ways and apply our stuff in ways we see fit. This is a luxury closed solutions lack. On the long run this is evolution at work: among many contenders those with the most mileage and the best abilities to join up with others for the common advantage just make the race and leave the others in the dust on the long run. Simply biology at work.... BTW, could someone enlighten me on how all those projects found each other to constitute piper ? Just curious..... ======================================================================= "It was hell. They knew it. Karl-Max Wagner But they called it karlmax@oberland.net W-I-N-D-O-Z-E" ham radio: DB8CO *********Member of No Code International********* ***********Visit http://www.nocode.org*********** ********Membership Number NCI-2563-DB8CO********* _ / / (_)__ __ ____ __ / /__/ / _ \/ // /\ \/ / . . . t h e c h o i c e o f a /____/_/_//_/\_,_/ /_/\_\ G N U g e n e r a t i o n . . "Et ceterum censeo ut Microsoftem delendum esse" (Cato, adapted) "Quo usque tandem abutere nostra patientia ?" (Cicero, original) ======================================================================= From jeff at bioinformatics.org Sat Feb 10 19:50:35 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] Khoros and Cantata References: <200102110004.AAA00750@schreyer.oberland.net> Message-ID: <3A85E1DB.A761B2CA@bioinformatics.org> Karl-Max Wagner wrote: > > I think the best thing with piper is the least obvious and most > abstract: AFAIK piper is the first time pretty diverse projects > found out that they had a lot of overlap and joined in solving > the overlapping problems together. This is truly astounding, > because, not using hindsight, it required thinking on a level of > high abstraction from all those participating to find that out. And Free/Open Source projects are often criticized for lacking innovation. > This led to a state of hight modularization in the project which > is already paying off: more projects now are joining in ( lately > we got Narval on board, for an example ). I hope so. I think that once we get the pilot version out and posted on Freshmeat, we'll be turning up the volume quite a bit (no pun about audio processing intended). > Here some crazy idea from the area of mathematics: That reminds me of this Slashdot article I read the other day: http://slashdot.org/article.pl?sid=01/02/08/2344210&mode=flat But perhaps it has nothing to do with visual programming. It'd be nice to see someone follow up on that thought: using visual programming for mathematical formula representation. Very interesting. > BTW, could someone enlighten me on how all those projects found > each other to constitute piper ? Just curious..... It happened when Jean-Marc posted a Gnotice (Gnome News item) about Overflow. I read it, and replied (paraphrasing), "Hey, that's something like what we want to do with Loci!" Jarl then replied and said, "Hey, that's my goal for GMS too!". We then e-mailed each other and discovered that our applications each lacked (to get to our ultimate goals) something that another was quite adept at: Loci had a fancy visual scripting shell, which GMS and (to some extent) Overflow lacked GMS had sophisticated (program-to-program) communication mechanisms, which both Loci and Overflow lacked. Overflow had a system for fast and local data/workflow, which both Loci and GMS lacked. I think that we all saw our systems someday, individually becoming "Piper", but we all needed help getting there. The most practical thing for us to do was join forces and code. And, besides our same goals and existing overlaps, we were all avid GNU advocates. Who could have asked for anything more? :-) Cheers. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jarl at xs4all.nl Sun Feb 11 06:46:18 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] Khoros and Cantata References: <200102102152.VAA00566@schreyer.oberland.net> Message-ID: <3A867B8A.F2CE2E15@xs4all.nl> > In your place I'd not care much whether it is older or not or > whether it exists or suchlike. In real life the question is not > who was first but who was first to put a thing into common > practice. Linux is the rererere.....incarnation of Unix, but > nevertheless, being free it became everybody's friend and > devoured pretty much all of its older non free peers. Free > Software is also very much about putting technologies into > common practice and as such free stuff in the long run eats up > proprietary stuff. Thus work put into piper is work well spent > for sure. Right, just starting a project within a (L)GPL environment has so much advantage over a closed one. I see interaction as the energy feed for self organisation. Maybe somebody get's my point ;) jarl From jarl at xs4all.nl Sun Feb 11 07:07:41 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] Khoros and Cantata References: <200102110004.AAA00750@schreyer.oberland.net> Message-ID: <3A86808D.5A836F08@xs4all.nl> > Here some crazy idea from the area of mathematics: coupling some > symbolic mathematics package with the piper UI engine it should > be possible to draw up mathematical relationships as block > diagrams and then automatically manipulate them to one's heart > content using the underlying symbolic mathematics package. For > lack of a better term, let's call that formula compilation. The > graphical entry is a high level representation of the > mathematical relationship, the actual formula something like > assembly code which is automatically generated, processed and > optimized to provide what is desired. Woudn't that relieve us > from the drudgery of low level formula handling and lift > mathematics to a new level of abstraction ? Just food for > thought...... As I understand you, you're saying Piper embodies model I call fenotipical representation. It's a general model named by the the biological dna working: charactirstics are 'stored' in the GUI (like genes), and fenotipic features are 'born' in the backend. This can be called abstraction, although it actually is the other way around: the abstractions come first, from which the 'real thing' is gerenated. This model seems to be a requierement for an autocatalistic system, which I hope the Piper backend to become. > > On the long run this is evolution at work: among many contenders > those with the most mileage and the best abilities to join up > with others for the common advantage just make the race and > leave the others in the dust on the long run. Simply biology at > work.... I had to add 'Maybe somebody get's my point' to my previous post. It seems at least one does ;) jarl From karlmax at oberland.net Sun Feb 11 07:59:15 2001 From: karlmax at oberland.net (Karl-Max Wagner) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] Khoros and Cantata In-Reply-To: <3A85E1DB.A761B2CA@bioinformatics.org> from "J.W. Bizzaro" at Feb 11, 2001 12:50:35 AM Message-ID: <200102111259.MAA00268@schreyer.oberland.net> > And Free/Open Source projects are often criticized for lacking innovation. Somehow I wonder what these people are thinking about. Would Linux have a chance by being a clone of Windows ? Of course not ! Linux has a chance because it is technically far superior. By reiterating old technology Linux sure wouldn't beat everything else speedwise, for example. > > This led to a state of hight modularization in the project which > > is already paying off: more projects now are joining in ( lately > > we got Narval on board, for an example ). > > I hope so. I think that once we get the pilot version out and posted on > Freshmeat, we'll be turning up the volume quite a bit (no pun about audio > processing intended). You may do it - or not. I don't think that it makes a real difference. The popularity comes automatically when the bundle of projects piper comes with solve a lot of problems no other product can solve. It is much more important to finally make some sort of distribution ( maybe by bundling in the stuff it needs as prerequisites and make an overall makefile, so that the ./configure - make - make install procedure works without a hitch on just about any *nix system ). And much more and more detailed documentation is required. IMHO it is better to work at sorting out all those issues before turning up volume. And in any case: we do not have to work according to a timeline. We deliver when it's ready. Full Stop. > http://slashdot.org/article.pl?sid=01/02/08/2344210&mode=flat > > But perhaps it has nothing to do with visual programming. It'd be nice to see It has - but more with conventional visual programming. ======================================================================= "It was hell. They knew it. Karl-Max Wagner But they called it karlmax@oberland.net W-I-N-D-O-Z-E" ham radio: DB8CO *********Member of No Code International********* ***********Visit http://www.nocode.org*********** ********Membership Number NCI-2563-DB8CO********* _ / / (_)__ __ ____ __ / /__/ / _ \/ // /\ \/ / . . . t h e c h o i c e o f a /____/_/_//_/\_,_/ /_/\_\ G N U g e n e r a t i o n . . "Et ceterum censeo ut Microsoftem delendum esse" (Cato, adapted) "Quo usque tandem abutere nostra patientia ?" (Cicero, original) ======================================================================= From karlmax at oberland.net Sun Feb 11 09:16:19 2001 From: karlmax at oberland.net (Karl-Max Wagner) Date: Fri Feb 10 19:39:09 2006 Subject: [Pipet Users] Khoros and Cantata In-Reply-To: <3A86808D.5A836F08@xs4all.nl> from "Jarl van Katwijk" at Feb 11, 2001 01:07:41 PM Message-ID: <200102111416.OAA00336@schreyer.oberland.net> > > Here some crazy idea from the area of mathematics: coupling some > > symbolic mathematics package with the piper UI engine it should > > be possible to draw up mathematical relationships as block > > diagrams and then automatically manipulate them to one's heart > > content using the underlying symbolic mathematics package. For > > lack of a better term, let's call that formula compilation. The > > graphical entry is a high level representation of the > > mathematical relationship, the actual formula something like > > assembly code which is automatically generated, processed and > > optimized to provide what is desired. Woudn't that relieve us > > from the drudgery of low level formula handling and lift > > mathematics to a new level of abstraction ? Just food for > > thought...... > > As I understand you, you're saying Piper embodies model I call > fenotipical representation. It's a general model named by the the Good term. Like it :-)))). > biological dna working: charactirstics are 'stored' in the GUI (like > genes), and fenotipic features are 'born' in the backend. This can be Exactly. Actually, this has been that way since the inception of the compiler sometime in 1945 ( Konrad Zuse's "Plankalkuel" ). > called abstraction, although it actually is the other way around: the > abstractions come first, from which the 'real thing' is gerenated. Yes, exactly: in my previous example the highly abstract representation is what you enter into the GUI and the backend generates formulas from it and also manipulates them. > This model seems to be a requierement for an autocatalistic system, > which I hope the Piper backend to become. Enter Narval. The present layering model of piper looks like this: UIL -----> DL -----> BL -----> PL Adding Narval we can do this: UIL -----> DL -----> AIL -----> BL -----> PL AIL: Artificial Intelligence Layer The AIL in fact accepts an even higher abstraction representation of what the user wants to do and converts it into a representation that the PL in the end can use. In the extreme the UIL and the DL may be left away altogether and the AIL automatically generates systems to be built up, collects the results and acts accordingly. Actually, using the UIL and the DL we can do three things: 1. Design very high level representations of distributed systems. 2. We can control the actions of the AIL. 3. We can add new capabilities to the AIL. Of course, the AIL may also be in contact with remote instances of other AIL's all over the network ( using the BL as a communications controller ). ======================================================================= "It was hell. They knew it. Karl-Max Wagner But they called it karlmax@oberland.net W-I-N-D-O-Z-E" ham radio: DB8CO *********Member of No Code International********* ***********Visit http://www.nocode.org*********** ********Membership Number NCI-2563-DB8CO********* _ / / (_)__ __ ____ __ / /__/ / _ \/ // /\ \/ / . . . t h e c h o i c e o f a /____/_/_//_/\_,_/ /_/\_\ G N U g e n e r a t i o n . . "Et ceterum censeo ut Microsoftem delendum esse" (Cato, adapted) "Quo usque tandem abutere nostra patientia ?" (Cicero, original) ======================================================================= From jarl at xs4all.nl Sun Feb 11 09:57:30 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] Khoros and Cantata References: <200102111416.OAA00336@schreyer.oberland.net> Message-ID: <3A86A85A.7EAD6C1C@xs4all.nl> > 1. Design very high level representations of distributed > systems. Can you rephrase this? I dont get what you're trying to point out here. > > 2. We can control the actions of the AIL. Need manual overruling. > > 3. We can add new capabilities to the AIL. AIL training ;) Sweet. I always took a module like your AIL in mind during the design. I already have some hooks for lilgp (a GP library) in the BL. We're still not sure whether such a module is a DL or something that's extra. Note that the BL can handle multiple DL's, even when they belong to the same area (pool on nodes). This is something that's outside the pilot, we though it wise not to rush this into the Piper design. jarl From jeff at bioinformatics.org Sun Feb 11 12:41:30 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] Khoros and Cantata References: <200102111416.OAA00336@schreyer.oberland.net> <3A86A85A.7EAD6C1C@xs4all.nl> Message-ID: <3A86CECA.58536138@bioinformatics.org> Jarl van Katwijk wrote: > > > 1. Design very high level representations of distributed > > systems. > Can you rephrase this? I dont get what you're trying to point out here. For this point, Karl-Max is just stating the known capabilities of the UIL+DL: It can be used to design a "network" of distributed nodes. > Sweet. I always took a module like your AIL in mind during the design. I > already have some hooks for lilgp (a GP library) in the BL. We're still > not sure whether such a module is a DL or something that's extra. Eh, why not do it as Karl-Max suggested? Make an "AIL" a separate layer. I like the idea. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jeff at bioinformatics.org Sun Feb 11 12:49:20 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] Khoros and Cantata References: <200102110004.AAA00750@schreyer.oberland.net> <3A85E1DB.A761B2CA@bioinformatics.org> Message-ID: <3A86D0A0.DF5A51F5@bioinformatics.org> "J.W. Bizzaro" wrote: > > Loci had a fancy visual scripting shell, which GMS and > (to some extent) Overflow lacked One other thing: Regarding the separation of the UIL from the DL (via RPC), and allowing multiple UIL's per DL, that came from Loci. FYI. Brad Chapman actually developed the DL. I started Loci from the GUI end (I like developing user interfaces), and when Brad came along, he put some meat behind it with the DL. From then on, we worked on separate code bases (except Brad also did a lot to the Loci UI). So the modularization that you see is really an artifact of the different code bases and coders: UIL <------------> DL <------------> BL <-------------> PL Jeff <----------> Brad <----------> Jarl <----------> Jean-Marc Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jeff at bioinformatics.org Sun Feb 11 13:03:54 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] slideshow Message-ID: <3A86D40A.8A109BD@bioinformatics.org> I have been cleaning up the documentation page on the web. There's still more to do, but I added a link to the slide presentation that I gave at a bioinformatics conference last summer. It's the first doc listed: http://www.bioinformatics.org/piper/documentation/index.html Jeff From karlmax at oberland.net Sun Feb 11 15:46:51 2001 From: karlmax at oberland.net (Karl-Max Wagner) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] Khoros and Cantata In-Reply-To: <3A86D0A0.DF5A51F5@bioinformatics.org> from "J.W. Bizzaro" at Feb 11, 2001 05:49:20 PM Message-ID: <200102112046.UAA00830@schreyer.oberland.net> > So the modularization that you see is really an artifact of the different code > bases and coders: > > UIL <------------> DL <------------> BL <-------------> PL > > Jeff <----------> Brad <----------> Jarl <----------> Jean-Marc OK, now with AIL: UIL <-----> DL <-----> AIL <-----> BL <-----> PL Jeff <---> Brad <---> Nicolas <---> Jarl <---> Jean-Marc ;-))))). ======================================================================= "It was hell. They knew it. Karl-Max Wagner But they called it karlmax@oberland.net W-I-N-D-O-Z-E" ham radio: DB8CO *********Member of No Code International********* ***********Visit http://www.nocode.org*********** ********Membership Number NCI-2563-DB8CO********* _ / / (_)__ __ ____ __ / /__/ / _ \/ // /\ \/ / . . . t h e c h o i c e o f a /____/_/_//_/\_,_/ /_/\_\ G N U g e n e r a t i o n . . "Et ceterum censeo ut Microsoftem delendum esse" (Cato, adapted) "Quo usque tandem abutere nostra patientia ?" (Cicero, original) ======================================================================= From karlmax at oberland.net Sun Feb 11 15:30:36 2001 From: karlmax at oberland.net (Karl-Max Wagner) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] Khoros and Cantata In-Reply-To: <3A86A85A.7EAD6C1C@xs4all.nl> from "Jarl van Katwijk" at Feb 11, 2001 03:57:30 PM Message-ID: <200102112030.UAA00804@schreyer.oberland.net> > > > > 1. Design very high level representations of distributed > > systems. > Can you rephrase this? I dont get what you're trying to point out here. That means that you can design even on a higher abstraction level because the AIL does some detail work for you. > > 2. We can control the actions of the AIL. > Need manual overruling. That's exactly what is meant by the above. ======================================================================= "It was hell. They knew it. Karl-Max Wagner But they called it karlmax@oberland.net W-I-N-D-O-Z-E" ham radio: DB8CO *********Member of No Code International********* ***********Visit http://www.nocode.org*********** ********Membership Number NCI-2563-DB8CO********* _ / / (_)__ __ ____ __ / /__/ / _ \/ // /\ \/ / . . . t h e c h o i c e o f a /____/_/_//_/\_,_/ /_/\_\ G N U g e n e r a t i o n . . "Et ceterum censeo ut Microsoftem delendum esse" (Cato, adapted) "Quo usque tandem abutere nostra patientia ?" (Cicero, original) ======================================================================= From karlmax at oberland.net Sun Feb 11 15:39:01 2001 From: karlmax at oberland.net (Karl-Max Wagner) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] Khoros and Cantata In-Reply-To: <3A86CECA.58536138@bioinformatics.org> from "J.W. Bizzaro" at Feb 11, 2001 05:41:30 PM Message-ID: <200102112039.UAA00821@schreyer.oberland.net> > Jarl van Katwijk wrote: > > > > > 1. Design very high level representations of distributed > > > systems. > > Can you rephrase this? I dont get what you're trying to point out here. > > For this point, Karl-Max is just stating the known capabilities of the UIL+DL: > It can be used to design a "network" of distributed nodes. Not exactly. As the AIL can do some of the nitty gritty work you can jack it up to a higher level of abstraction. > > Sweet. I always took a module like your AIL in mind during the design. I > > already have some hooks for lilgp (a GP library) in the BL. We're still > > not sure whether such a module is a DL or something that's extra. > > Eh, why not do it as Karl-Max suggested? Make an "AIL" a separate layer. I > like the idea. Me, too, I am not in favour of integrating the AIL into the BL. Keep it as modular as possible. However, I understand Jarl's idea of integrating it into the BL somewhat: if you think of networkwide operation a AI engine can set up the topology according to the connectivity found at the moment. Which solves a lot of large scale networking headaches for sure. However, this can also be done using a separate AIL: the BL may contact the separate AIL and have it work for it. This way the design of the existing system need not be changed needlessly, and we can enjoy the advantages of an AIL as soon as it becomes available ( hi, Nicolas ! ). ======================================================================= "It was hell. They knew it. Karl-Max Wagner But they called it karlmax@oberland.net W-I-N-D-O-Z-E" ham radio: DB8CO *********Member of No Code International********* ***********Visit http://www.nocode.org*********** ********Membership Number NCI-2563-DB8CO********* _ / / (_)__ __ ____ __ / /__/ / _ \/ // /\ \/ / . . . t h e c h o i c e o f a /____/_/_//_/\_,_/ /_/\_\ G N U g e n e r a t i o n . . "Et ceterum censeo ut Microsoftem delendum esse" (Cato, adapted) "Quo usque tandem abutere nostra patientia ?" (Cicero, original) ======================================================================= From jarl at xs4all.nl Sun Feb 11 16:49:45 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] Khoros and Cantata References: <200102112046.UAA00830@schreyer.oberland.net> Message-ID: <3A8708F9.56074005@xs4all.nl> > > So the modularization that you see is really an artifact of the different code > > bases and coders: > > > > UIL <------------> DL <------------> BL <-------------> PL > > > > Jeff <----------> Brad <----------> Jarl <----------> Jean-Marc > > OK, now with AIL: > > UIL <-----> DL <-----> AIL <-----> BL <-----> PL > > Jeff <---> Brad <---> Nicolas <---> Jarl <---> Jean-Marc > Ok, have a AIL (I'll just adopt the naming) as a seperate module (ofcourse with CORBA inferface(s)) is obvious to me. But putting it inbetween the DL and BL doesn't seem a good design to me. Neither intergrating it into the BL. The way I see a GP engine fit into piper is as a general support module. Something all layers could make use of. Sort off a base class for the total monster we call Piper ;) jarl From jeff at bioinformatics.org Sun Feb 11 18:18:42 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] modular design (was: Khoros and Cantata) References: <200102112046.UAA00830@schreyer.oberland.net> <3A8708F9.56074005@xs4all.nl> Message-ID: <3A871DD2.6D74D6B0@bioinformatics.org> Jarl van Katwijk wrote: > > Ok, have a AIL (I'll just adopt the naming) as a seperate module > (ofcourse with CORBA inferface(s)) is obvious to me. But putting it > inbetween the DL and BL doesn't seem a good design to me. Neither > intergrating it into the BL. The way I see a GP engine fit into piper is > as a general support module. Something all layers could make use of. > Sort off a base class for the total monster we call Piper ;) Between the DL and the BL, even I am confused about which layer is "middle" (and "base") . I would define the base as the layer that would connect to "shared" layers like the AIL, and I'd define the middle as the layer most responsible for P2P connectivity. They don't have to be one-in-the-same, but I would make them so if it were entirely up to me. During some of the discussions about design, I had been in favor of the BL acting as the middle (and base), but some decisions by Brad and Jarl lead me to believe that the DL is, at least in part, the middle. Or maybe "there is no middle"? Jarl, I would like to know your thoughts on this. Cheers. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From karlmax at oberland.net Sun Feb 11 18:36:59 2001 From: karlmax at oberland.net (Karl-Max Wagner) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] Khoros and Cantata In-Reply-To: <3A8708F9.56074005@xs4all.nl> from "Jarl van Katwijk" at Feb 11, 2001 10:49:45 PM Message-ID: <200102112337.XAA01102@schreyer.oberland.net> > Ok, have a AIL (I'll just adopt the naming) as a seperate module > (ofcourse with CORBA inferface(s)) is obvious to me. But putting it > inbetween the DL and BL doesn't seem a good design to me. Neither > intergrating it into the BL. The way I see a GP engine fit into piper is > as a general support module. Something all layers could make use of. > Sort off a base class for the total monster we call Piper ;) Makes sense. In fact, the more I think about it it seems that the AIL could be used for a lot of things in all layers. Maybe it makes sense to integrate some sort of a switchbox function into the BL that allows the BL to chain in the AIL everywhere where it is needed. Maybe it makes even more sense to split up the AIL in a number of blocks ( rule based, genetic etc. ) to enhance its configurability and consequently its versatility. ======================================================================= "It was hell. They knew it. Karl-Max Wagner But they called it karlmax@oberland.net W-I-N-D-O-Z-E" ham radio: DB8CO *********Member of No Code International********* ***********Visit http://www.nocode.org*********** ********Membership Number NCI-2563-DB8CO********* _ / / (_)__ __ ____ __ / /__/ / _ \/ // /\ \/ / . . . t h e c h o i c e o f a /____/_/_//_/\_,_/ /_/\_\ G N U g e n e r a t i o n . . "Et ceterum censeo ut Microsoftem delendum esse" (Cato, adapted) "Quo usque tandem abutere nostra patientia ?" (Cicero, original) ======================================================================= From jeff at bioinformatics.org Sun Feb 11 20:02:25 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] Articles: Are Unix GUIs All Wrong? Message-ID: <3A873621.2C6F3C82@bioinformatics.org> Good question. I think so! http://slashdot.org/comments.pl?sid=01%2F02%2F11%2F1929220&mode=flat I submitted a short comment about Piper. Jeff From Nicolas.Chauvat at logilab.fr Mon Feb 12 05:43:23 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] Khoros and Cantata In-Reply-To: <3A86808D.5A836F08@xs4all.nl> Message-ID: > > On the long run this is evolution at work: among many contenders > > those with the most mileage and the best abilities to join up > > with others for the common advantage just make the race and > > leave the others in the dust on the long run. Simply biology at > > work.... > > I had to add 'Maybe somebody get's my point' to my previous post. It > seems at least one does ;) Don't worry Jarl, you're not alone out there :-) -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From Nicolas.Chauvat at logilab.fr Mon Feb 12 05:53:10 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] Khoros and Cantata In-Reply-To: <200102111259.MAA00268@schreyer.oberland.net> Message-ID: > ! Linux has a chance because it is technically far superior. By > reiterating old technology Linux sure wouldn't beat everything > else speedwise, for example. For what I know, Linux isn't superior to some BSD flavors speedwise... but I'm a Linux enthusiast too :-) > > Freshmeat, we'll be turning up the volume quite a bit (no pun about audio > > processing intended). > > You may do it - or not. I don't think that it makes a real difference. > The popularity comes automatically when the bundle of projects piper > comes with solve a lot of problems no other product can solve. It is > much more important to finally make some sort of distribution ( maybe > by bundling in the stuff it needs as prerequisites and make an overall > makefile, so that the ./configure - make - make install procedure > works without a hitch on just about any *nix system ). And much more > and more detailed documentation is required. We announced Narval on Freshmeat 4 months ago and got few contributions since then. Times have changed: there is tons of free software out there and not everybody hooked up to the net is a hacker anymore... when people download stuff, they most probably trash it if the install isn't smooth and the basic features don't work out of the box. And as for most free software projects, what we'll get are testers/users, not programmers... but testers/users are great, so that's fine :-) My point is that package and documentation are indispensable to get the user base to start growing and this is what Piper really lacks at this point. -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From Nicolas.Chauvat at logilab.fr Mon Feb 12 06:48:08 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] Is Narval the AIL Piper may need ? In-Reply-To: <3A873621.2C6F3C82@bioinformatics.org> Message-ID: Pipers, For once I went off-line for the week-end and when I came back my inbox was full of nice ideas about Piper layers and uses, and why Narval (or parts of it maybe) should be integrated to it as an Artificial Intelligence Layer. Short answer is: hold on. Long answer is: sure, why not! But I'd like all of us to be clear about what we're doing. As the history of the Piper project shows, you all seem to be pretty good at abstracting things out. So let's sit down a bit, forget about what we have today and what projects the current code base inherits from. Two years ago when I started Narval (before I could get the others to hop in so that we get it to work in less than 5 years), I was also thinking about a visual shell and discussing it with collegues. Visual-shellish ideas indeed sliped into Narval's design, but it's *not* a visual shell in the sense Piper appears to define it and doesn't intend to become one. On the other hand, it could very well be made to interact very nicely with a visual shell like Piper and share code and other things. To get back to my previous point, I'd suggest forgetting about the current code base and write down a rather precise description of what we want Piper to do (you know, requirements :-). The ideas of piper nodes that publish their capabilities, P2P capabilities for communication and resource discovery, extending the principle of pipes, visual programming, etc. are the real goal(s) of Piper. It could be that choices were made when first merging the intial projects because anything else would have been impracticle (I'm thinking about the mapping one person / one layer here) and that those choices are actually not so good when we look at what we want to get to. (Again, I can't say for sure, one way or the other, so if someone has a clear opinion on that, please speak up). I'm willing to help here, so I'll read anything about Piper you can point me too and I'll try to send to the list a write-up about Narval achievements, hopes and dreams (have quick look at http://www.logilab.com/press/linux-expo2001 as a starter). You'll have to wait before I can find time to try the code, and at this point, I'm not sure it's mandatory. Then I'd suggest we collaboratively write everything down and post it on the website so that any passer-by can comment. I'd say that's what we started doing this week-end on the mailing list and with the Wiki, but maybe staying too close to what exists and not what we're trying to get to. Once we get that document, I'm sure the questions about "Should Narval be Piper's AIL?", "I thought about adding a Genetic Programming module, where do you think it would best fit?" will get rather obvious answers as we'll figure the implications right away. If you guys agree with me on this need of discussing and writing down the goals of the Piper Project before continuing further with the design, please give me (or Logilab, so that Alexandre can use it too. Name it after Narval's mascot: ornicar :-) an access to the CVS and I'll check in the very first version of that document so that we can all start building on it. If we were to build a web browser or some common piece of software, the roadmap would be crystal clear. Here we're dealing with something that looks like research, or at least pretty innovative stuff. If we don't have a clear understanding of others' ideas and do not share the same vision, chances are the road will be winding. Free software or not :-) -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From jeff at bioinformatics.org Mon Feb 12 10:52:18 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] Slashdot | Pipes In GUI? (Redeux) Message-ID: <3A8806B2.84834A9E@bioinformatics.org> Hey, here we go! Slashdot just posted an article about Piper. Yes, our Piper: http://slashdot.org/article.pl?sid=01/02/11/0017256&mode=thread I don't know if this article will appear on the front page of Slashdot, though. Jeff From mkt.ecbd at acbdglobal.com Mon Feb 12 11:02:02 2001 From: mkt.ecbd at acbdglobal.com (Marketing) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] Peer-to-peer Computing Conference... Message-ID: <003b01c0950d$26b0ac40$0701a8c0@Sales2> Skipped content of type multipart/alternative-------------- next part -------------- From karlmax at oberland.net Mon Feb 12 16:03:08 2001 From: karlmax at oberland.net (Karl-Max Wagner) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] Is Narval the AIL Piper may need ? In-Reply-To: from "Nicolas Chauvat" at Feb 12, 2001 12:48:08 PM Message-ID: <200102122103.VAA01055@schreyer.oberland.net> > To get back to my previous point, I'd suggest forgetting about the current > code base and write down a rather precise description of what we want > Piper to do (you know, requirements :-). The ideas of piper nodes that > publish their capabilities, P2P capabilities for communication and > resource discovery, extending the principle of pipes, visual programming, > etc. are the real goal(s) of Piper. It could be that choices were made > when first merging the intial projects because anything else would have > been impracticle (I'm thinking about the mapping one person / one layer > here) and that those choices are actually not so good when we look at what > we want to get to. (Again, I can't say for sure, one way or the other, so > if someone has a clear opinion on that, please speak up). To begin with, piper is a consortium of several projects with pretty diverse goals and agendas that recognized that, abstractly considered, they had very similar needs in a way. Now that even more projects joined the consortium I think a good beginning would be if each of the projects would write up some presentation about its origins ( history ), goals and dreams. Once we all know what everyone is aiming at, we can design a framework that fits at least these requirements ( first draft ). Then we can go ahead and generalize it even more so that me can accommodate future projects with similar needs and goals. It might also be good if everyone writes up a bit about himself: we are a community with a rather extreme diversity of backgrounds and it helps a lot to know from which side each of us looks at things. I propose that everyone prepares this stuff in HTMLized form, bundles it up and submits it for having it put up somewhere on the piper site. This is much more comprehensive than spattering it all about the net. > If you guys agree with me on this need of discussing and writing down the > goals of the Piper Project before continuing further with the design, > please give me (or Logilab, so that Alexandre can use it too. Name it > after Narval's mascot: ornicar :-) an access to the CVS and I'll check in > the very first version of that document so that we can all start building > on it. I think this should be the next step after everyone has laid open what he is up to. > If we were to build a web browser or some common piece of software, the > roadmap would be crystal clear. Here we're dealing with something that > looks like research, or at least pretty innovative stuff. If we don't have > a clear understanding of others' ideas and do not share the same vision, > chances are the road will be winding. Free software or not :-) Chances are even that we fail. I could easily point to projects that are not getting anywhere because they apparently stopped understanding what they want to do. In any project, the concept phase is the most important one. Mistakes in the concept phase are from difficult to impossible to correct in subsequent development. I have seen that only too often. ======================================================================= "It was hell. They knew it. Karl-Max Wagner But they called it karlmax@oberland.net W-I-N-D-O-Z-E" ham radio: DB8CO *********Member of No Code International********* ***********Visit http://www.nocode.org*********** ********Membership Number NCI-2563-DB8CO********* _ / / (_)__ __ ____ __ / /__/ / _ \/ // /\ \/ / . . . t h e c h o i c e o f a /____/_/_//_/\_,_/ /_/\_\ G N U g e n e r a t i o n . . "Et ceterum censeo ut Microsoftem delendum esse" (Cato, adapted) "Quo usque tandem abutere nostra patientia ?" (Cicero, original) ======================================================================= From jeff at bioinformatics.org Mon Feb 12 17:58:42 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] Is Narval the AIL Piper may need ? References: <200102122103.VAA01055@schreyer.oberland.net> Message-ID: <3A886AA2.A010AF76@bioinformatics.org> Karl-Max Wagner wrote: > > I propose that everyone prepares this stuff in HTMLized form, > bundles it up and submits it for having it put up somewhere on > the piper site. This is much more comprehensive than spattering > it all about the net. We can put it in the docs directory of the Piper website. I actually have the whole website in CVS, so anyone with write access can contribute. Speaking of which (Nicolas also asked about CVS access), the best way to go about getting access (which means getting a shell account) is to join The Open Lab, our SourceForge-based project management system: http://bioinformatics.org Then, contact me (jeff@bioinformatics.org) and tell me that you are ready to be added to the project. When I add you, the system (admins actually) gives you a shell account. You'll then have CVS access. Right now, we have the following developers in the system (everyone should have a shell account): bizzaro bossekr chapmanb gears gmax jarl jmvalin theoriste https://bioinformatics.org/project/?group_id=19 > Chances are even that we fail. I could easily point to projects > that are not getting anywhere because they apparently stopped > understanding what they want to do. I like to think that the number one reason Free/Open projects fail is lost interest by the primary developers. Many of these guys just "move on", but it may be that they "stopped understanding what they want to do", as Karl-Max said, and that caused them to lose interest. > In any project, the concept phase is the most important one. > Mistakes in the concept phase are from difficult to impossible > to correct in subsequent development. I have seen that only too > often. Oh yes, I agree with that! Piper has spent years in the "concept phase", and it is probably still there. If we just plowed ahead from day one (assuming we didn't have to wait to find each other) and finished the project in a couple months, we would not have been able to include many of the concepts developed later. And they may not have ever been includable. But, how long is too long for the concept phase? ;-) Cheers. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jeff at bioinformatics.org Mon Feb 12 21:20:30 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse Message-ID: <3A8899EE.106DFA98@bioinformatics.org> *sigh* I finally put together some of my ideas about command-line compilation (known in Loci-speak as "compositing" or "constructing"). This all goes back to the Loci days: http://www.bioinformatics.org/piper/documentation/command-compilation.html Some of it should work later as a HOWTO. Some of it should work as a FAQ. Enjoy. Jeff From karlmax at oberland.net Mon Feb 12 21:06:17 2001 From: karlmax at oberland.net (Karl-Max Wagner) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] Is Narval the AIL Piper may need ? In-Reply-To: <3A886AA2.A010AF76@bioinformatics.org> from "J.W. Bizzaro" at Feb 12, 2001 10:58:42 PM Message-ID: <200102130206.CAA01582@schreyer.oberland.net> > Then, contact me (jeff@bioinformatics.org) and tell me that you are ready to > be added to the project. When I add you, the system (admins actually) gives > you a shell account. You'll then have CVS access. OK, then please add me. > > Chances are even that we fail. I could easily point to projects > > that are not getting anywhere because they apparently stopped > > understanding what they want to do. > > I like to think that the number one reason Free/Open projects fail is lost > interest by the primary developers. Many of these guys just "move on", but it > may be that they "stopped understanding what they want to do", as Karl-Max > said, and that caused them to lose interest. That is not the only problem. Projects may stay alive but in the end be of no real use. Here a particularly evil example: There is a large simulation project out there ( I'm not getting any more specific here as it is not my intention to bash on someone ) which started out as an interesting idea with a focus on signal processing systems simulation. They were not very well focused, but that is an asset in the beginning. Then they decided to dump the original project and convert it into something much larger, broader etc., just it was evident that they had no idea what real problems they wanted to solve. Since then they are floating along more or less as a wonderful theoretical project in search for a problem to solve. Or suchlike. They are still around, though. > Oh yes, I agree with that! Piper has spent years in the "concept phase", and ??? AFAIK Piper is at most 1 year old.... > it is probably still there. If we just plowed ahead from day one (assuming we > didn't have to wait to find each other) and finished the project in a couple > months, we would not have been able to include many of the concepts developed Exactly. Other extreme: focus too narrow. Serves only a small niche. End of project. > later. And they may not have ever been includable. But, how long is too long > for the concept phase? ;-) Piper is still very young. I'd not worry too much about that right now. And particularly, there is no abrupt end to this. Eventually parts will settle down, more will settle down, until the framework is in place. In fact it is best to keep the concept phase open as long as possible, at least as far as it does not hinder ongoing work too much. One key goal is to have a framework that is flexible enough to allow integrating new projects without any need to change anything ( this is the ideal we probably will never attain but we must try to get reasonably close to ). ======================================================================= "It was hell. They knew it. Karl-Max Wagner But they called it karlmax@oberland.net W-I-N-D-O-Z-E" ham radio: DB8CO *********Member of No Code International********* ***********Visit http://www.nocode.org*********** ********Membership Number NCI-2563-DB8CO********* _ / / (_)__ __ ____ __ / /__/ / _ \/ // /\ \/ / . . . t h e c h o i c e o f a /____/_/_//_/\_,_/ /_/\_\ G N U g e n e r a t i o n . . "Et ceterum censeo ut Microsoftem delendum esse" (Cato, adapted) "Quo usque tandem abutere nostra patientia ?" (Cicero, original) ======================================================================= From jeff at bioinformatics.org Mon Feb 12 21:50:44 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:10 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse References: <3A8899EE.106DFA98@bioinformatics.org> Message-ID: <3A88A104.20FD491C@bioinformatics.org> "J.W. Bizzaro" wrote: > > *sigh* I finally put together some of my ideas about command-line compilation > (known in Loci-speak as "compositing" or "constructing"). This all goes back > to the Loci days: > > http://www.bioinformatics.org/piper/documentation/command-compilation.html BTW, I would love to get some comments on this, especially from Jean-Marc, as it is the PL which would have to implement all of it. (I know that the PL can do much of it already). Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jeff at bioinformatics.org Mon Feb 12 22:08:59 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Is Narval the AIL Piper may need ? References: <200102130206.CAA01582@schreyer.oberland.net> Message-ID: <3A88A54B.DD8323A7@bioinformatics.org> Karl-Max Wagner wrote: > > > Then, contact me (jeff@bioinformatics.org) and tell me that you are ready to > > be added to the project. When I add you, the system (admins actually) gives > > you a shell account. You'll then have CVS access. > > OK, then please add me. You are hereby added. You now have a shell account with CVS access. Visit this page for some CVS instructions: http://bioinformatics.org/docs/cvs/ The website module is "piper-website". > > Oh yes, I agree with that! Piper has spent years in the "concept phase", and > > ??? AFAIK Piper is at most 1 year old.... "Piper" is almost 1 year old. The programs the merged to form Piper are older. Loci started in the fall of 1998. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From valj01 at gel.usherb.ca Mon Feb 12 22:25:52 2001 From: valj01 at gel.usherb.ca (Jean-Marc Valin) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse References: <3A8899EE.106DFA98@bioinformatics.org> <3A88A104.20FD491C@bioinformatics.org> Message-ID: <3A88A940.6ACC4010@gel.usherb.ca> > BTW, I would love to get some comments on this, especially from Jean-Marc, as > it is the PL which would have to implement all of it. (I know that the PL can > do much of it already). OK, these are just some quick comments I can make. Note that I read everything quickly, you guys generate a lot more stuff than I have time to read... I'd like to comment on your use of "processor", "flag" and "filemane" type of nodes. For Overflow (PL), a Node is a Node and there's no such discrimination. I'm attaching a screenshot (taken with vflow) of how I see building command lines in Piper. Here, the Constant would contain the string -I, and the INI and POV inputs would contain the two filenames (as strings). The ExecStream Node would have povray as a parameter. The Add nodes would just concatenate strings together. Now, the whole network you see could be used as a simple node that would take two strings as input. Does that make sense to you? Jean-Marc -- Jean-Marc Valin Universite de Sherbrooke - Genie Electrique valj01@gel.usherb.ca -------------- next part -------------- A non-text attachment was scrubbed... Name: pov.gif Type: image/gif Size: 25949 bytes Desc: not available Url : http://bioinformatics.org/pipermail/pipet-users/attachments/20010212/4b273cb9/pov.gif From jeff at bioinformatics.org Mon Feb 12 22:54:29 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse References: <3A8899EE.106DFA98@bioinformatics.org> <3A88A104.20FD491C@bioinformatics.org> <3A88A940.6ACC4010@gel.usherb.ca> Message-ID: <3A88AFF5.A5E5584@bioinformatics.org> Jean-Marc Valin wrote: > > I'd like to comment on your use of "processor", "flag" and "filemane" type of > nodes. For Overflow (PL), a Node is a Node and there's no such discrimination. The nodes in your screenshot are "Add" nodes, and they have a certain function: To add strings. It's pretty much what "flag" and "filename" do, except in my example the strings can be changed: For "flag", the user can change the string with a pull-down menu. For "filename", the name of the connected document is the string. Perhaps where you have set "constant", "INI", and even "POV", we would have a "variable" node, or something to hold these values. > I'm attaching a screenshot (taken with vflow) of how I see building command > lines in Piper. Here, the Constant would contain the string -I, and the INI and > POV inputs would contain the two filenames (as strings). The ExecStream Node > would have povray as a parameter. The Add nodes would just concatenate strings > together. Now, the whole network you see could be used as a simple node that > would take two strings as input. Does that make sense to you? Yeah, we're talking basically about the same kind of network, except where I have written +-----------+ +------+ +----------+ +----------+ | processor | ---> | flag | -----> | filename | -----> | filename | +-----------+ +------+ +----------+ +----------+ you have written the reverse, and the names have changed: +------------+ +------+ +-----+ +----------+ | ExecStream | <---- | Add | <----- | Add | <-------- | constant | +------------+ +------+ +-----+ +----------+ ^ ^ | | | | POV INI Yours is of course the actual way to implement this, since it has worked for you :-) I should modify my document. Cheers. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jeff at bioinformatics.org Mon Feb 12 23:00:09 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse References: <3A8899EE.106DFA98@bioinformatics.org> <3A88A104.20FD491C@bioinformatics.org> <3A88A940.6ACC4010@gel.usherb.ca> <3A88AFF5.A5E5584@bioinformatics.org> Message-ID: <3A88B149.C1DB28F7@bioinformatics.org> "J.W. Bizzaro" wrote: > > The nodes in your screenshot are "Add" nodes, and they have a certain > function: To add strings. It's pretty much what "flag" and "filename" do, > except in my example the strings can be changed: > > For "flag", the user can change the string with a pull-down menu. > > For "filename", the name of the connected document is the string. > > Perhaps where you have set "constant", "INI", and even "POV", we would have a > "variable" node, or something to hold these values. And with the ability to create and save subnets (compiling), we can have a node that is a subnet containing one "Add" node and one variable node, and we can name it "flag" or whatever. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jarl at xs4all.nl Tue Feb 13 05:12:48 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Is Narval the AIL Piper may need ? References: <200102122103.VAA01055@schreyer.oberland.net> Message-ID: <3A8908A0.5FA7068B@xs4all.nl> Hi Pipers, Sorry I didn't answer Karl-Max's lenghty post sooner, needed some time to think about the proposal. At first I've to admit Piper doesn't really have a fixed design fron which we're building the code. This is because we had this situation of 3 projects merging and we couldn't look much futher as a pilot. Current cvs code is more or less what this pilot should be. It's the right time to think about the final product now. A think we should keep in mind that Piper IS a reseach project, most of it's idears aren't done before, or they aren't done in this combination. So it's almost impossible to design Piper in theory and then let some coding monkeys finish it. I really feel the need for for a grayish design that doesn't cover everything. But I think it's a good move if we would try to make a design coming weeks, to see what we want Piper to become. So I propose everybody to make a scretch of what he thinks Piper should look like, we'll merge\discuss those views, and fill in that general design as much as we can. Does this sound like a good approarch? jarl From Nicolas.Chauvat at logilab.fr Tue Feb 13 05:55:36 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Is Narval the AIL Piper may need ? In-Reply-To: <3A8908A0.5FA7068B@xs4all.nl> Message-ID: On Tue, 13 Feb 2001, Jarl van Katwijk wrote: > At first I've to admit Piper doesn't really have a fixed design fron > ... > merge\discuss those views, and fill in that general design as much as we > can. Does this sound like a good approarch? Sounds good to me. Looks like we all agree on the need to describe what each of us had in mind when starting his project, then discuss things futher and use the current pilot to ground upcoming development. I registered user Ornicar at the Open Lab. Ornicar is the name of the Narval Project's mascot (it's a Narwhale BTW). It will be representing the Logilab Team (me, Alexandre.Fayolle@logilab.fr or others that will announce themselves on the list when necessary). Jeff: could you please add me to the project and grant me CVS access? I'll quickly add a write-up about Narval and the few ideas I have concerning Piper. I'll also read the doc Jeff sent a URL for. Feel free to send in more pointers to documentation, I like reading and there is not much on the website yet :-) -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From Nicolas.Chauvat at logilab.fr Tue Feb 13 08:56:26 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse In-Reply-To: <3A88A940.6ACC4010@gel.usherb.ca> Message-ID: > I'd like to comment on your use of "processor", "flag" and "filemane" > type of nodes. For Overflow (PL), a Node is a Node and there's no such > discrimination. I read the page about command-compilation Jeff posted. My experience (with Narval and other systems) is that "a Node is a Node" is nice in theory, but becomes a problem when you want to enhance the user interface and facilitate the writing of programs. If "processor", "flag" and "filename" are different things of different types, used for different purposes, make it so. Maybe they're all Nodes from a graph/network point of view, in which case the solution would be to make them inherit from a common base class. But if you don't put into your model the capacity to discriminate between a string that is actually a flag and a string that is actually a filename, you'll have troubles when you'll want to add things such as "put all your nodes in the workspace and I'll help you link them, or tell me what you want to do and I'll try to build a network that does it". Yes, I'm trying to bring in my AI background :-) To continue on the same topic, I've read all the docs I could find on the Piper website and Wiki and have a couple comments regarding the relations between Piper and Narval. The Piper introduction and slides say that "Everything is a component. Component have I/O. You can connect components I/O the same way you use pipes in a shell. Components have a location." For Narval I'd say "Everything is an element. Recipes are sequences of steps linked by transitions. Recipes are elements. Steps have I/O with prototypes and can be actions or transforms. Actions and transforms I/O are elements. Transitions have conditions on elements. Plans are executable instances of recipes and are elements." Similar isn't it? Yes and No. I'll focus on the differences. Narval manipulates high-level data: e-mails, calendar events, web pages, search results, tasks and projects, etc. whereas, if I understand correctly, Piper currently pipes data streams from a process to another and does not really care for the meaning of the data that's exchanged and Piper has higher speed requirements for data transfer. That's a sufficient difference to legitimate many design decision differences between the two. Narval uses typing extensively. Elements are trees currently represented in XML. XPath is used to express conditions on those trees. Those conditions are used in the step prototypes and in the transitions, as in "to run, this action needs an email Element such that the following xpath is true: email/headers[begins-with(string(subject),'[Pipet Users]'))." This provides a way to express simply a lot of constraints later used to verify things and predict problems before they happen (both as a compiler would do for source code and as a language like Eiffel does with pre/post conditions at run-time). Second, transitions let you control the data flow in your network/graph. With the added capability to map an action on a sequence of input elements, you get most of what the usual branch/loop of other languages would give you. Using trees as a basic structure and xpath to express contraints and conditions is a true benefit in our case and let us build other features (planning, etc.). We couldn't find a simpler way to do it. I've more to say about Narval, but could this experience be of any use to Piper? What do you guys think? -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From jmvalin at locusdialogue.com Tue Feb 13 09:47:30 2001 From: jmvalin at locusdialogue.com (Jean-Marc Valin) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse References: Message-ID: <3A894902.6340DEA6@locusdialogue.com> Nicolas Chauvat a ?crit : > > > I'd like to comment on your use of "processor", "flag" and "filemane" > > type of nodes. For Overflow (PL), a Node is a Node and there's no such > > discrimination. > > I read the page about command-compilation Jeff posted. My experience (with > Narval and other systems) is that "a Node is a Node" is nice in theory, > but becomes a problem when you want to enhance the user interface and > facilitate the writing of programs. If "processor", "flag" and "filename" > are different things of different types, used for different purposes, make > it so. Maybe they're all Nodes from a graph/network point of view, in > which case the solution would be to make them inherit from a common base > class. But if you don't put into your model the capacity to discriminate > between a string that is actually a flag and a string that is actually a > filename, you'll have troubles when you'll want to add things such as "put > all your nodes in the workspace and I'll help you link them, or tell me > what you want to do and I'll try to build a network that does it". Yes, > I'm trying to bring in my AI background :-) OK, there's two things here: internally, there's an abstract "Node" class, for which each new Node derives (as you see in the screenshot, "Constant", "Add" and "ExecStream" all derive from Node). Hovever, there's no need to discriminate between things that perform have the same function. A string is a string... regardless of whether you use it as a flag or a filename. If you look at C, for example, a (char *) or an int can be use for whatever you like. Jean-Marc From Alexandre.Fayolle at logilab.fr Tue Feb 13 10:12:28 2001 From: Alexandre.Fayolle at logilab.fr (Alexandre Fayolle) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse In-Reply-To: <3A894902.6340DEA6@locusdialogue.com> Message-ID: On Tue, 13 Feb 2001, Jean-Marc Valin wrote: > OK, there's two things here: internally, there's an abstract "Node" > class, for which each new Node derives (as you see in the screenshot, > "Constant", "Add" and "ExecStream" all derive from Node). Hovever, > there's no need to discriminate between things that perform have the > same function. A string is a string... regardless of whether you use it > as a flag or a filename. If you look at C, for example, a (char *) or an > int can be use for whatever you like. True, but this is not the point. You can use void* pointers in C, instead of char*. It"s just that sometimes it's nice to differentiate a char* and a FILE*, for instance: the underlying representation _is_ a pointer (an int ?), but the _semantics_ are different. The meaning of a command line switch, a filename passed on the command line, a keyword passed on the command line are different, even if they are all thrown in char** argv when the program receives them. What Nicolas meant is that if you give your program a bunch of arbitrary char*, it won't be able to do much with it. Whereas if you say that this is a switch, it should be followed by the name of an existing file, then it becomes possible to process the data. Alexandre Fayolle -- http://www.logilab.com Narval is the first software agent available as free software (GPL). LOGILAB, Paris (France). From Nicolas.Chauvat at logilab.fr Tue Feb 13 10:17:59 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse In-Reply-To: <3A894902.6340DEA6@locusdialogue.com> Message-ID: > OK, there's two things here: internally, there's an abstract "Node" > class, for which each new Node derives (as you see in the screenshot, > "Constant", "Add" and "ExecStream" all derive from Node). Hovever, > there's no need to discriminate between things that perform have the > same function. A string is a string... regardless of whether you use > it as a flag or a filename. If you look at C, for example, a (char *) > or an int can be use for whatever you like. What I was trying to point out was the exact thing you use as an example against my rationale :-) C manipulates memory addresses, but let you refer to those addresses as int, float, double, char *, etc. so that at compile time it can complain when you do (int = float) and warns you that it's an implied cast. Flags and filenames may be strings in a shell, but represent different things, the same way memory contains only bytes that represent ints and floats. We may choose not to care about the difference between flag and filename strings in Piper. What I was saying is that this will eventually prevent us to do things like adding a "help" button to inputs that would open a file browser when a filename is needed or a drop-down list of possible options when a flag is needed, or check that the filename you gave is an actual file, etc. -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From jeff at bioinformatics.org Tue Feb 13 11:17:50 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse References: <3A8899EE.106DFA98@bioinformatics.org> <3A88A104.20FD491C@bioinformatics.org> <3A88A940.6ACC4010@gel.usherb.ca> <3A88AFF5.A5E5584@bioinformatics.org> <3A88B149.C1DB28F7@bioinformatics.org> Message-ID: <3A895E2E.73145615@bioinformatics.org> There are a couple things that should be noted about the document. First, it was constructed from purely conceptual discussions we had on the Loci mailing list, before we merged to form Piper. Second, I think the major difference between the way we conceptualized command compilation working for Loci, and the way it works for Overflow, is that the Loci document refers to "string-adding terminals", while Overflow uses "string-adding nodes". I believe that they are functionally equivalent, and I need to modify the document to show how it is done (or could be improved) in Overflow. Anyway, it never ceases to amaze me how we all developed very similar concepts independently. Cheers. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From Nicolas.Chauvat at logilab.fr Tue Feb 13 11:34:38 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse In-Reply-To: <3A895E2E.73145615@bioinformatics.org> Message-ID: > Anyway, it never ceases to amaze me how we all developed very similar > concepts independently. One is never alone to get a good idea. Usually it's out there for you and everyone else to see if you look hard enough. The thing is that before the net, people that discovered the same thing or had the same good idea usually met at the patent office when registering their invention, not on the web when designing it. Anyway, it looks like the time had come for a visual shell, that's it! :-) -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From jmvalin at locusdialogue.com Tue Feb 13 11:41:56 2001 From: jmvalin at locusdialogue.com (Jean-Marc Valin) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse References: Message-ID: <3A8963D4.60FC681D@locusdialogue.com> > True, but this is not the point. You can use void* pointers in C, instead > of char*. It"s just that sometimes it's nice to differentiate a char* and > a FILE*, for instance: the underlying representation _is_ a pointer (an > int ?), but the _semantics_ are different. > > The meaning of a command line switch, a filename passed on the command > line, a keyword passed on the command line are different, even if they > are all thrown in char** argv when the program receives them. What Nicolas > meant is that if you give your program a bunch of arbitrary char*, it > won't be able to do much with it. Whereas if you say that this is a > switch, it should be followed by the name of an existing file, then it > becomes possible to process the data. Well, in my example, I still identify that "this input requires a filename" and "that input requires a config file". If you don't follw that, then too bad for you. The same thing if a C function asks for a process ID and you give it your phone number... Defining a new type always requires come code. In Overflow the implemented types are: Int, Float, Double, String, Bool, Vector, Stream, ... If we start adding types for flags, filenames, ... then it becomes endless. We'd need to divide "Vector" in "3D vector", "2D vector", "AudioFrame", "FIRFilter", "IIRFilter", .... and the same for all types. As for any language, the semantics can be taken into account with the variable name (or in this case, the input name). Jean-Marc From Nicolas.Chauvat at logilab.fr Tue Feb 13 11:55:50 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse In-Reply-To: <3A8963D4.60FC681D@locusdialogue.com> Message-ID: > Well, in my example, I still identify that "this input requires a > filename" and "that input requires a config file". If you don't follw > that, then too bad for you. The same thing if a C function asks for a > process ID and you give it your phone number... True. And that's a difference between a programming language and a person. A person would use "common sense" not to do something that's "too bad for her". Programs don't have that common sense, but some design choices in programming languages may help reducing the gap and prevent people that write code to make too many mistakes. > Defining a new type always requires come code. In Overflow the > implemented types are: Int, Float, Double, String, Bool, Vector, > Stream, ... I understood Overflow works well for its purpose. That's good and well. We're just discussing the principles, not argueing about your design. You say above that in your example you still "identify it as a filename" or "identify it as a config file". Is that implemented somehow or specified in the doc or something else ? > If we start adding types for flags, filenames, ... then it becomes > endless. We'd need to divide "Vector" in "3D vector", "2D vector", > "AudioFrame", "FIRFilter", "IIRFilter", .... and the same for all > types. As for any language, the semantics can be taken into account > with the variable name (or in this case, the input name). Not all types need be on the same level. Hierarchies are good for this. And classes of equivalent terms. And ontologies and logics. It all depends how far you want to go and how complicated you want to get. But that doesn't mean it's bad "per se" and we shouldn't mention it here, does it? -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From jeff at bioinformatics.org Tue Feb 13 15:33:52 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse References: <3A8963D4.60FC681D@locusdialogue.com> Message-ID: <3A899A30.8ECD9624@bioinformatics.org> Jean-Marc Valin wrote: > > Defining a new type always requires come code. In Overflow the > implemented types are: Int, Float, Double, String, Bool, Vector, > Stream, ... > > If we start adding types for flags, filenames, ... then it becomes > endless. We'd need to divide "Vector" in "3D vector", "2D vector", > "AudioFrame", "FIRFilter", "IIRFilter", .... and the same for all types. > As for any language, the semantics can be taken into account with the > variable name (or in this case, the input name). I agree with you that we need to have a very small number of *STANDARD* node types. For example, in most programming languages you will have types INTEGER, REAL, CHAR, STRING, etc., just a few. *BUT* (and that's a big but) you can create new types using "old" types. Take Pascal for example. You can define... TYPE MYNUMBER : INTEGER; MYARRAY : ARRAR[0..100] OF TYPE MYNUMBER; (or whatever -- I can't remember my Pascal at the moment ;-)) But, I have created my own type! :-) I'm not saying that Overflow/Piper should have types "flag" and "filename" as standard types, but that they can be created using existing types and then packaged (compiled) into a subnet. The subnet can then be called "flag" or "filename". Make sense? :-) This is the same concept as defining new methods, procedures, functions, objects, whatever. I should be doable in Piper. Cheers. Jeff From jeff at bioinformatics.org Tue Feb 13 15:51:49 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse References: <3A8963D4.60FC681D@locusdialogue.com> <3A899A30.8ECD9624@bioinformatics.org> Message-ID: <3A899E65.DE1AD1DF@bioinformatics.org> "J.W. Bizzaro" wrote: > > I'm not saying that Overflow/Piper should have types "flag" and "filename" > as standard types, but that they can be created using existing types and > then packaged (compiled) into a subnet. The subnet can then be called > "flag" or "filename". Jean-Marc, keeping track of the many *POSSIBLE* node types is indeed difficult, using the current system. That is obvious by just looking at the size of the pop-up menu! :-) However, we should keep in mind that part of the Piper project is to make nodes and networks easily accessible from anywhere in the world. I have some ideas that I'd like to try out, but for the most part, I don't want to put all of the nodes (in the world?!) on a pop-up menu. That would not be the most efficient way to do it. One use of the pop-up menu, I have been thinking, could be to show the nodes that would be *COMPATIBLE* with the terminal (I/O) the user has selected. It would narrow the field down quite a bit. Cheers. Jeff From jmvalin at locusdialogue.com Tue Feb 13 16:02:31 2001 From: jmvalin at locusdialogue.com (Jean-Marc Valin) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse References: <3A8963D4.60FC681D@locusdialogue.com> <3A899A30.8ECD9624@bioinformatics.org> Message-ID: <3A89A0E7.6993B041@locusdialogue.com> > I agree with you that we need to have a very small number of *STANDARD* node > types. For example, in most programming languages you will have types > INTEGER, REAL, CHAR, STRING, etc., just a few. *BUT* (and that's a big but) > you can create new types using "old" types. > > Take Pascal for example. You can define... > > TYPE > MYNUMBER : INTEGER; > MYARRAY : ARRAR[0..100] OF TYPE MYNUMBER; > > (or whatever -- I can't remember my Pascal at the moment ;-)) > > But, I have created my own type! :-) > > I'm not saying that Overflow/Piper should have types "flag" and "filename" > as standard types, but that they can be created using existing types and > then packaged (compiled) into a subnet. The subnet can then be called > "flag" or "filename". Make sense? :-) > > This is the same concept as defining new methods, procedures, functions, > objects, whatever. I should be doable in Piper. Wait a minute, I think there is some confusion about types here. In Overflow, there is "Data type" and "Node type". The Data types are Int, Float, Vector, ... The Node types are Constant, Add, Load, Save, ... (BTW, a Constant node is like a Node that always returns the same object/value) The data types are equivalent to the C types (int, float, ...), while the node types are equivalent to different C functions. If you want to add a new Node type, all you need to to is add the .cc (implementation) to the the Makefile, or write a new .n (subnet). You can also add new types, but it needs to be coded in C++. Right now, there's no equivalent of subnets for Data types. It's on my TODO list though (but far from the top). For everything I've written type=="Data type" (not Node type). I hope this clarifies a few things. Although you can add new data types as you wish, I really do not recommend doing so for things like "flag", "filename", as it doesn't add any functionality, but forces you to implement all the operators for those types (and things like add(flag, filename), add(string, filename)) Jean-Marc From jmvalin at locusdialogue.com Tue Feb 13 16:06:40 2001 From: jmvalin at locusdialogue.com (Jean-Marc Valin) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse References: <3A8963D4.60FC681D@locusdialogue.com> <3A899A30.8ECD9624@bioinformatics.org> <3A899E65.DE1AD1DF@bioinformatics.org> Message-ID: <3A89A1E0.AB1BE46A@locusdialogue.com> > Jean-Marc, keeping track of the many *POSSIBLE* node types is indeed > difficult, using the current system. That is obvious by just looking at the > size of the pop-up menu! :-) However, we should keep in mind that part of > the Piper project is to make nodes and networks easily accessible from > anywhere in the world. Read the note I just sent about data types vs Node types... Now about the popup menu, I know we'll need to find something else. But, remember that one way or another, you'll need to find a way to access any node you want. The way it's implemented now (which only works for local nodes) is to define a PATH for that list the toolboxes you want to see (the others don't appear in the pop-up menu). Jean-Marc From jeff at bioinformatics.org Tue Feb 13 16:25:19 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse References: <3A8963D4.60FC681D@locusdialogue.com> <3A899A30.8ECD9624@bioinformatics.org> <3A899E65.DE1AD1DF@bioinformatics.org> <3A89A1E0.AB1BE46A@locusdialogue.com> Message-ID: <3A89A63F.5A97EF65@bioinformatics.org> Jean-Marc Valin wrote: > > Read the note I just sent about data types vs Node types... I understand the difference: * Data type says what kind of data passes through a terminal * Node type says what the node will do to the data And, yes, my example was using data types (INTEGER, etc.), but the same principle applies: You can make new types ("pseudo-types") by combining several node types in a subnet, right? They are not hard-coded but built as a network. You may not have considered it to be "type defining", but it is, in effect, no? > Now about the popup menu, I know we'll need to find something else. But, > remember that one way or another, you'll need to find a way to access > any node you want. The way it's implemented now (which only works for > local nodes) is to define a PATH for that list the toolboxes you want to > see (the others don't appear in the pop-up menu). Yeah, that's just the sort of thing we need to improve with Piper. Cheers. Jeff From jmvalin at locusdialogue.com Tue Feb 13 16:36:08 2001 From: jmvalin at locusdialogue.com (Jean-Marc Valin) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse References: <3A8963D4.60FC681D@locusdialogue.com> <3A899A30.8ECD9624@bioinformatics.org> <3A899E65.DE1AD1DF@bioinformatics.org> <3A89A1E0.AB1BE46A@locusdialogue.com> <3A89A63F.5A97EF65@bioinformatics.org> Message-ID: <3A89A8C8.CB29DA7B@locusdialogue.com> > I understand the difference: > > * Data type says what kind of data passes through a terminal > * Node type says what the node will do to the data > > And, yes, my example was using data types (INTEGER, etc.), but the same > principle applies: You can make new types ("pseudo-types") by combining > several node types in a subnet, right? They are not hard-coded but built as > a network. You may not have considered it to be "type defining", but it is, > in effect, no? Well, it you combine several nodes in a subnet, you end up with a "new node" (it's in fact a subnet). You have created a new node type, but you haven't created a new data type. Right now, there's no way to create new data types without touching C++ code. C analogy: you have a couple functions (nodes) and you write a new on that calls many of the "old" functions. You end up with a new function (node) but you did not create a new type. I think I'll write an analogy between Overflow "concepts" and their Matlab (or C) equivalent. Jean-Marc From jeff at bioinformatics.org Tue Feb 13 16:35:11 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:11 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse References: <3A8963D4.60FC681D@locusdialogue.com> <3A899A30.8ECD9624@bioinformatics.org> <3A89A0E7.6993B041@locusdialogue.com> Message-ID: <3A89A88F.C5D15792@bioinformatics.org> Jean-Marc Valin wrote: > > The data types are equivalent to the C types (int, float, ...), while > the node types are equivalent to different C functions. If you want to > add a new Node type, all you need to to is add the .cc (implementation) > to the the Makefile, or write a new .n (subnet). Uh, yeah, this is confusing. When I say "make new types", I'm saying, "make a new function using existing functions". Jeff From jeff at bioinformatics.org Tue Feb 13 16:43:02 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse References: <3A8963D4.60FC681D@locusdialogue.com> <3A899A30.8ECD9624@bioinformatics.org> <3A899E65.DE1AD1DF@bioinformatics.org> <3A89A1E0.AB1BE46A@locusdialogue.com> <3A89A63F.5A97EF65@bioinformatics.org> <3A89A8C8.CB29DA7B@locusdialogue.com> Message-ID: <3A89AA66.3240CB1B@bioinformatics.org> Jean-Marc Valin wrote: > > Well, it you combine several nodes in a subnet, you end up with a "new > node" (it's in fact a subnet). You have created a new node type, but you > haven't created a new data type. Right now, there's no way to create new > data types without touching C++ code. Okay, I'm not suggesting new data types anywhere (yet) ;-) I like the way that Overflow is used to construct a command, but as outlined in the document I just posted, I would like to add some features, to make it easier for the user. Exaclty what I want to add has yet to be determined. > C analogy: you have a couple functions (nodes) and you write a new on > that calls many of the "old" functions. You end up with a new function > (node) but you did not create a new type. Got it. > I think I'll write an analogy between Overflow "concepts" and their > Matlab (or C) equivalent. That would be nice. Jeff From jeff at bioinformatics.org Tue Feb 13 17:00:50 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Command Compilation in Piper: A Discourse References: Message-ID: <3A89AE92.6B35BCF9@bioinformatics.org> Nicolas Chauvat wrote: > > We may choose not to care about the difference between flag and > filename strings in Piper. What I was saying is that this will eventually > prevent us to do things like adding a "help" button to inputs that > would open a file browser when a filename is needed or a drop-down > list of possible options when a flag is needed, or check that the > filename you gave is an actual file, etc. Right, this is exactly the difference in *FUNCTIONALITY* that I see between different "nodes used to build a command". *BUT*, after the discussions we've had, I should point out that these "nodes used to build a command", that Nicolas is referring to, are what Jean-Marc would refer to as different, individual "subnets". These nodes that Nicolas and I invision are not "basic node types". In Jean-Marc's screenshot, he built a command much more simply than I would have. He placed the "basic node type" of "Add" where I would have put some "subnets". *AND* (now this is important) THE SUBNETS WOULD BE *NAMED* "flag", "filename", ETC., AND THEY WOULD NOT SHARE THE SAME FUNCTIONALITY. I hope this cleared up the confusion between everyone. Did it? Jeff From jeff at bioinformatics.org Tue Feb 13 18:25:53 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] some gui plans Message-ID: <3A89C281.42F09827@bioinformatics.org> I just updated the screenshot page, detailing some of the plans I have for the Pied/Piper GUI: http://www.bioinformatics.org/piper/screenshots/index.html You may find that it helps you understand the meagerly documented Pied/Piper GUI. Jeff From valj01 at gel.usherb.ca Tue Feb 13 22:34:11 2001 From: valj01 at gel.usherb.ca (Jean-Marc Valin) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Overflow overview Message-ID: <3A89FCB3.9B1CE461@gel.usherb.ca> I wrote a shote overview about Overflow (PL). It's in the form of Overflow vs. C/Matlab. Comments? Jean-Marc -- Jean-Marc Valin Universite de Sherbrooke - Genie Electrique valj01@gel.usherb.ca -------------- next part -------------- An HTML attachment was scrubbed... URL: http://bioinformatics.org/pipermail/pipet-users/attachments/20010213/14b8c8ac/overview.html From jeff at bioinformatics.org Wed Feb 14 00:40:07 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Is Narval the AIL Piper may need ? References: Message-ID: <3A8A1A37.55DF5578@bioinformatics.org> Nicolas Chauvat wrote: > > I registered user Ornicar at the Open Lab. Ornicar is the name of the > Narval Project's mascot (it's a Narwhale BTW). It will be representing the > Logilab Team (me, Alexandre.Fayolle@logilab.fr or others that will > announce themselves on the list when necessary). > > Jeff: could you please add me to the project and grant me CVS access? Okay, done. You have a shell account with CVS and FTP write access. But, the account needs to be the responsibility of an individual, and that will be you :-) > I'll quickly add a write-up about Narval and the few ideas I have > concerning Piper. I'll also read the doc Jeff sent a URL for. Feel free to > send in more pointers to documentation, I like reading and there is not > much on the website yet :-) BTW, we have Wiki set up on the system, if you'd like to use that: http://www.bioinformatics.org/piperwiki/moin.cgi I haven't gotten into it myself. Cheers. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jarl at xs4all.nl Wed Feb 14 05:51:29 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Design proposal References: <3A89FCB3.9B1CE461@gel.usherb.ca> Message-ID: <3A8A6331.4C4CEA66@xs4all.nl> Hi, I've written a very basic proposal for the Piper 1.0 specs. Now I have a dillemma: should I post it and let us expand it to something good, or should I wait till others have written something so we can compare better? jarl From Nicolas.Chauvat at logilab.fr Wed Feb 14 11:06:59 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Is Narval the AIL Piper may need ? In-Reply-To: <3A8A1A37.55DF5578@bioinformatics.org> Message-ID: > > Jeff: could you please add me to the project and grant me CVS access? > > Okay, done. You have a shell account with CVS and FTP write access. > But, the account needs to be the responsibility of an individual, and > that will be you :-) That's fine, I'll be responsible for the account. Having one account will be easier for us here. Thanks. > BTW, we have Wiki set up on the system, if you'd like to use that: > > http://www.bioinformatics.org/piperwiki/moin.cgi > > I haven't gotten into it myself. I read that, but there isn't much there yet. -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From jeff at bioinformatics.org Wed Feb 14 11:25:25 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Is Narval the AIL Piper may need ? References: Message-ID: <3A8AB175.3F4ADC59@bioinformatics.org> Nicolas Chauvat wrote: > > > BTW, we have Wiki set up on the system, if you'd like to use that: > > > > http://www.bioinformatics.org/piperwiki/moin.cgi > > > > I haven't gotten into it myself. > > I read that, but there isn't much there yet. Wiki can be used to *WRITE* documentation. It is like a web-based CVS for docs. I don't know if I can recommend using it or not. Some here have used it (Jarl). What do you guys think? Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From Nicolas.Chauvat at logilab.fr Wed Feb 14 12:13:45 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Is Narval the AIL Piper may need ? In-Reply-To: <3A8AB175.3F4ADC59@bioinformatics.org> Message-ID: > > I read that, but there isn't much there yet. > > Wiki can be used to *WRITE* documentation. It is like a web-based CVS for > docs. I know that :-) I read too fast and understood you were pointing me to the doc that's in the Wiki, not that you were offering me to use the wiki for my narval write-up. > I don't know if I can recommend using it or not. Some here have used it > (Jarl). What do you guys think? Wouldn't a sgml (docbook?) document be better? It could be easily turned into HTML for web use and ps/pdf for printing. Wikis can give a hard time when it comes to assemble all the information in one single document. Unless there is a special mechanism/tool for that with MoinMoin ? -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From Nicolas.Chauvat at logilab.fr Wed Feb 14 12:20:04 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Design proposal In-Reply-To: <3A8A6331.4C4CEA66@xs4all.nl> Message-ID: > I've written a very basic proposal for the Piper 1.0 specs. Now I have a > dillemma: should I post it and let us expand it to something good, or > should I wait till others have written something so we can compare > better? +1 for posting it. -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From jeff at bioinformatics.org Wed Feb 14 12:41:42 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Is Narval the AIL Piper may need ? References: Message-ID: <3A8AC356.2615A988@bioinformatics.org> Nicolas Chauvat wrote: > > Wouldn't a sgml (docbook?) document be better? It could be easily turned > into HTML for web use and ps/pdf for printing. We've had several discussions about doc format on this list before. The suggestions have been LaTeX/LyX, DocBook, Wiki, AbiWord, RTF, HTML, etc., with some people very determined to stick to their favorite. The one suggestion that I like, is that whoever does the writing chooses the format. I know that it would give us a mess of different formats, but it should then be between the writers and the doc coordinator (Gary Van Domselaar) to put everything into a manual, when the time has come for one. Cheers. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jeff at bioinformatics.org Wed Feb 14 12:45:34 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Design proposal References: Message-ID: <3A8AC43E.4D3618AA@bioinformatics.org> > I've written a very basic proposal for the Piper 1.0 specs. Now I have a > dillemma: should I post it and let us expand it to something good, or > should I wait till others have written something so we can compare > better? Make that +2 for posting. Any set of specs will have to be changed. It would be better to spend time writing and changing one set than many. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jeff at bioinformatics.org Wed Feb 14 12:51:12 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Design proposal References: <3A8AC43E.4D3618AA@bioinformatics.org> Message-ID: <3A8AC590.97CAFD7A@bioinformatics.org> "J.W. Bizzaro" wrote: > > > I've written a very basic proposal for the Piper 1.0 specs. Now I have a > > dillemma: should I post it and let us expand it to something good, or > > should I wait till others have written something so we can compare > > better? > > Make that +2 for posting. Jarl, unless there are objections about posting, go ahead and put it in the "documentation" directory in the piper-website CVS module. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jeff at bioinformatics.org Wed Feb 14 19:51:43 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Overflow overview References: <3A89FCB3.9B1CE461@gel.usherb.ca> Message-ID: <3A8B281F.DB3E1B95@bioinformatics.org> > Jean-Marc Valin wrote: > > I wrote a shote overview about Overflow (PL). It's in the form of Overflow > vs. C/Matlab. Comments? It is now in the piper-website CVS module. Changes should probably be made to that copy, unless you are keeping it on the Overflow site somewhere, Jean-Marc. Anyway, I only made a couple layout changes. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jeff at bioinformatics.org Wed Feb 14 20:10:41 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Why Gnutella Can't Scale. No, Really. Message-ID: <3A8B2C91.C8D6CACB@bioinformatics.org> http://www.tch.org/gnutella.html Jeff From jarl at xs4all.nl Thu Feb 15 05:29:44 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Is Narval the AIL Piper may need ? References: <3A8AB175.3F4ADC59@bioinformatics.org> Message-ID: <3A8BAF98.B8C3DA5E@xs4all.nl> > Wiki can be used to *WRITE* documentation. It is like a web-based CVS for > docs. > > I don't know if I can recommend using it or not. Some here have used it > (Jarl). What do you guys think? Havn'y used it that much. At first I've the feeling Wiki's are more like a public message board, not really a cvs for documents. jarl From jarl at xs4all.nl Thu Feb 15 05:46:25 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Why Gnutella Can't Scale. No, Really. References: <3A8B2C91.C8D6CACB@bioinformatics.org> Message-ID: <3A8BB381.4FFBAE3A@xs4all.nl> "J.W. Bizzaro" wrote: > > http://www.tch.org/gnutella.html Did somebody with enough math knowledge read this and can post some comments on this whether or not this applies to Piper? jarl From jarl at xs4all.nl Thu Feb 15 05:48:02 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Design proposal References: <3A8AC43E.4D3618AA@bioinformatics.org> <3A8AC590.97CAFD7A@bioinformatics.org> Message-ID: <3A8BB3E2.2FA6AA59@xs4all.nl> > > Make that +2 for posting. > > Jarl, unless there are objections about posting, go ahead and put it in the > "documentation" directory in the piper-website CVS module. Checking in DesignProposal_PiperFinal.html; /cvsroot/piper/doc/DesignProposal_PiperFinal.html,v <-- DesignProposal_PiperFinal.html initial revision: 1.1 - please update this people. This is just a skeleton document. jarl From jeff at bioinformatics.org Thu Feb 15 06:06:52 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Why Gnutella Can't Scale. No, Really. References: <3A8B2C91.C8D6CACB@bioinformatics.org> <3A8BB381.4FFBAE3A@xs4all.nl> Message-ID: <3A8BB84B.47401E60@bioinformatics.org> Jarl van Katwijk wrote: > > Did somebody with enough math knowledge read this and can post some > comments on this whether or not this applies to Piper? I'll briefly note that Piper may avoid the "Gnutella problem" two ways: (1) Compared to Gnutella, very little information will be passed between instances of Piper. Program *SERVICES* are shared; programs themselves remain put. I think data transfers will be like serving web pages. Compare that to Gnutella transferring 2-4 MB mp3 files. (2) There will be computers on the Net, running Piper, that act more like "servers" than "clients". Some will be computers serving scientific applications. Unlike Gnutella, where people are almost purely interested in "getting" (music, pr0n), many Piper instances will be running specifically to provide a service (bioinformatics and other programs). Cheers. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jarl at xs4all.nl Thu Feb 15 06:22:45 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Why Gnutella Can't Scale. No, Really. References: <3A8B2C91.C8D6CACB@bioinformatics.org> <3A8BB381.4FFBAE3A@xs4all.nl> <3A8BB84B.47401E60@bioinformatics.org> Message-ID: <3A8BBC05.528F41C4@xs4all.nl> > I'll briefly note that Piper may avoid the "Gnutella problem" two ways: > > (1) Compared to Gnutella, very little information will be passed between > instances of Piper. Program *SERVICES* are shared; programs themselves remain > put. I think data transfers will be like serving web pages. Compare that to > Gnutella transferring 2-4 MB mp3 files. Hmm, in the conclusion of the article they say: ---- And it should also be noted that only search query and response traffic was accounted for, omitting various other types of Gnutella traffic such as PING, PONG, and most importantly, the bandwidth costs incurred by actual file transfers. 2.4GBps is just search and response traffic, but what about the obnoxiously large amount of bandwidth necessary to transfer files between clients? --- jarl From jeff at bioinformatics.org Thu Feb 15 06:37:12 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Why Gnutella Can't Scale. No, Really. References: <3A8B2C91.C8D6CACB@bioinformatics.org> <3A8BB381.4FFBAE3A@xs4all.nl> <3A8BB84B.47401E60@bioinformatics.org> <3A8BBC05.528F41C4@xs4all.nl> Message-ID: <3A8BBF68.C4FFFC4F@bioinformatics.org> Jarl van Katwijk wrote: > > ---- > And it should also be noted that only search query and response traffic > was accounted for, omitting various other types of Gnutella traffic such > as PING, PONG, and most importantly, the bandwidth costs incurred by > actual file transfers. 2.4GBps is just search and response traffic, but > what about the obnoxiously large amount of bandwidth necessary to > transfer files between clients? > --- Yeah, some of the comments that I read on Slashdot indicated that searching is a serious problem with all P2P networks. I think that is what the article addresses. Aside from searching, my two points are valid. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From valj01 at gel.usherb.ca Thu Feb 15 08:39:56 2001 From: valj01 at gel.usherb.ca (Jean-Marc Valin) Date: Fri Feb 10 19:39:12 2006 Subject: [Pipet Users] Overflow overview References: <3A89FCB3.9B1CE461@gel.usherb.ca> Message-ID: <3A8BDC2C.D7E0664E@gel.usherb.ca> I'd like to mention that the page I sent is on the Overflow website at: http://freespeech.sourceforge.net/FreeSpeech/html/Overflow/overview.html This is the page that will be updated, since it is in the Overflow CVS. Those with write access can modify it, otherwise send me corrections. Jean-Marc (Note that the version linked has ~200 less spelling mistakes than the one I send to the list :-) ) -- Jean-Marc Valin Universite de Sherbrooke - Genie Electrique valj01@gel.usherb.ca From jeff at bioinformatics.org Thu Feb 15 09:17:37 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:13 2006 Subject: [Pipet Users] Overflow overview References: <3A89FCB3.9B1CE461@gel.usherb.ca> <3A8BDC2C.D7E0664E@gel.usherb.ca> Message-ID: <3A8BE501.49DD520F@bioinformatics.org> Jean-Marc Valin wrote: > > I'd like to mention that the page I sent is on the Overflow website at: > http://freespeech.sourceforge.net/FreeSpeech/html/Overflow/overview.html > This is the page that will be updated, since it is in the Overflow CVS. Those > with write access can modify it, otherwise send me corrections. Okay, I probably then should just link to the one on your site. I'll take it out of piper-website. Jeff From karlmax at oberland.net Thu Feb 15 16:53:05 2001 From: karlmax at oberland.net (Karl-Max Wagner) Date: Fri Feb 10 19:39:13 2006 Subject: [Pipet Users] Back next week. In-Reply-To: <3A8BBF68.C4FFFC4F@bioinformatics.org> from "J.W. Bizzaro" at Feb 15, 2001 11:37:12 AM Message-ID: <200102152153.VAA00113@schreyer.oberland.net> Hi Pipers, before anyone wonders why I am not delivering anything right now - I had to go to Munich yesterday. I'll be back Sunday evening. So my report should be available coming Wednesday or so. Can't do anything right now as I have a tight schedule here. ======================================================================= "It was hell. They knew it. Karl-Max Wagner But they called it karlmax@oberland.net W-I-N-D-O-Z-E" ham radio: DB8CO *********Member of No Code International********* ***********Visit http://www.nocode.org*********** ********Membership Number NCI-2563-DB8CO********* _ / / (_)__ __ ____ __ / /__/ / _ \/ // /\ \/ / . . . t h e c h o i c e o f a /____/_/_//_/\_,_/ /_/\_\ G N U g e n e r a t i o n . . "Et ceterum censeo ut Microsoftem delendum esse" (Cato, adapted) "Quo usque tandem abutere nostra patientia ?" (Cicero, original) ======================================================================= From jarl at xs4all.nl Fri Feb 16 19:09:05 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:13 2006 Subject: [Pipet Users] Design References: <200102152153.VAA00113@schreyer.oberland.net> Message-ID: <3A8DC121.7E55D3F6@xs4all.nl> Hi Pipers, I thought to myself I rather should give the good example and wrote some more in the design proposal ;) Filled in the DL and BL alinea's. All the DL text I copied out of Brad's document @ http://www.bioinformatics.org/piper/documentation/dl_info.pdf. The BL text is new. For all convenience the doc is available in Piper's cvs repository, or for all you hard core techies it's attached. jarl -------------- next part -------------- An HTML attachment was scrubbed... URL: http://bioinformatics.org/pipermail/pipet-users/attachments/20010217/9904729c/DesignProposal_PiperFinal.html From jeff at bioinformatics.org Sat Feb 17 02:58:17 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:13 2006 Subject: [Pipet Users] Sun enlists peer-to-peer in war against Microsoft - Tech News - CNET.com Message-ID: <3A8E2F19.BB44DB5E@bioinformatics.org> ``Mike Clary, head of the Jxta project, said the software will include ways that computing tasks can be linked together in "pipelines" that span a peer-to-peer network. In addition, Jxta will offer a mechanism by which tasks can be monitored and controlled.'' http://news.cnet.com/news/0-1003-200-4836102.html Sounds like the competition keeps piling up. Jeff From Nicolas.Chauvat at logilab.fr Sat Feb 17 05:49:57 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:13 2006 Subject: [Pipet Users] Design In-Reply-To: <3A8DC121.7E55D3F6@xs4all.nl> Message-ID: > Filled in the DL and BL alinea's. All the DL text I copied out of > Brad's document @ > http://www.bioinformatics.org/piper/documentation/dl_info.pdf. The BL > text is new. > > For all convenience the doc is available in Piper's cvs repository, or > for all you hard core techies it's attached. Hi folks, I read both documents, and if I my take the place of the devil's advocate, I'd say that they describe *how* things will be done, much more than *what* things will be done. What is the intended usage? What will piper be good for? What do we agree it can be bad at? What piper capabilities are needed to get piper to be as useful and easy to use as we expect it to become? What will the user see in terms of features? How will the user(s) use it? I think that these are the questions we should write down answers for first of all. Then ask: "ok, now how do we design a system that does that?". I know there are already much work done on the design and a lot of good answers to that latter question, but I'm afraid we're not clear enough about the first set of questions (at least I'm not, please try to educate me). The documents talk about layers, interfaces, P2P, corba etc. Those are means, not ends. What are the ends of Piper? I'll write down what I think for everyone to comment on it. -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From Nicolas.Chauvat at logilab.fr Sat Feb 17 09:29:58 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:13 2006 Subject: [Pipet Users] Design In-Reply-To: Message-ID: > I'll write down what I think for everyone to comment on it. I just commited to the CVS a first draft of piper/doc/piper-ends.html. It's hardly more than a skeleton right now, but I don't have time to do more today. I attached a copy to this e-mail for those of you that would want to read it right away. Please post your comments to the list. -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) -------------- next part -------------- An HTML attachment was scrubbed... URL: http://bioinformatics.org/pipermail/pipet-users/attachments/20010217/2d4e9644/piper-ends.html From jarl at xs4all.nl Sat Feb 17 17:16:28 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:13 2006 Subject: [Pipet Users] Design References: Message-ID: <3A8EF83C.73563DBD@xs4all.nl> Nicolas Chauvat wrote: > > > I'll write down what I think for everyone to comment on it. > > I just commited to the CVS a first draft of piper/doc/piper-ends.html. > It's hardly more than a skeleton right now, but I don't have time to do > more today. > > I attached a copy to this e-mail for those of you that would want to read > it right away. Please post your comments to the list. Thnx Nicolas, nice start. Nowthe two documents need to be merged, every subsection should start by a 'what' and defined by a 'how'. Jeff, maybe you can coordinate (read 'do' ;) ) this? If not, I will. jarl From jeff at bioinformatics.org Sat Feb 17 20:15:23 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:13 2006 Subject: [Pipet Users] Design References: <3A8EF83C.73563DBD@xs4all.nl> Message-ID: <3A8F222B.CDD257BC@bioinformatics.org> Jarl van Katwijk wrote: > > Thnx Nicolas, nice start. Nowthe two documents need to be merged, every > subsection should start by a 'what' and defined by a 'how'. Jeff, maybe > you can coordinate (read 'do' ;) ) this? If not, I will. That sounds good. I'll give 'er a go tonight. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jeff at bioinformatics.org Sun Feb 18 00:48:09 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:14 2006 Subject: [Pipet Users] Design References: <3A8EF83C.73563DBD@xs4all.nl> <3A8F222B.CDD257BC@bioinformatics.org> Message-ID: <3A8F6219.54EC5160@bioinformatics.org> I combined Nicolas's and Jarl's documents into a single one. It is in the piper-website module, *NOT* the piper module. I think that we should *NOT* put docs in the piper module for now, because we don't have anything formalized, and scattered bits and pieces of documentation may be more than users can manage. Plus, with images and all, these docs can add a bit to the package size. Once we have a couple complete manuals, they can go in the piper module. I think that we should call this document the "Piper Developer's Manual" And I have split Nicolas's and Jarl's contributions into 2 parts: Goals and Design. I did not attempt to do as Jarl said: "every subsection should start by a 'what' and defined by a 'how'", because there is little overlap between what I've got (IOW, Nicolas didn't write "The purpose of the BL is..."). Oh, it is attached. Cheers. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://bioinformatics.org/pipermail/pipet-users/attachments/20010218/ff0b69b0/piper-devel-manual.html From jeff at bioinformatics.org Sun Feb 18 22:05:34 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:14 2006 Subject: [Pipet Users] Design References: <3A8EF83C.73563DBD@xs4all.nl> <3A8F222B.CDD257BC@bioinformatics.org> <3A8F6219.54EC5160@bioinformatics.org> Message-ID: <3A908D7E.274E780C@bioinformatics.org> > "J.W. Bizzaro" wrote: > > I think that we should *NOT* put docs in the piper module for now, Shall I make a "piper-doc" module? Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From Nicolas.Chauvat at logilab.fr Mon Feb 19 04:19:14 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:14 2006 Subject: [Pipet Users] Design In-Reply-To: <3A8F6219.54EC5160@bioinformatics.org> Message-ID: > I combined Nicolas's and Jarl's documents into a single one. It is in the > piper-website > module, *NOT* the piper module. I think that we should *NOT* put docs Agreed. That's my mistake, don't hit me on the head... please! -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From jarl at xs4all.nl Mon Feb 19 06:46:37 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:14 2006 Subject: [Pipet Users] Design References: <3A8EF83C.73563DBD@xs4all.nl> <3A8F222B.CDD257BC@bioinformatics.org> <3A8F6219.54EC5160@bioinformatics.org> <3A908D7E.274E780C@bioinformatics.org> Message-ID: <3A91079D.7D37532E@xs4all.nl> "J.W. Bizzaro" wrote: > > > "J.W. Bizzaro" wrote: > > > > I think that we should *NOT* put docs in the piper module for now, > > Shall I make a "piper-doc" module? > Or maybe something more general, pre-piper, or piper-waitingroom, or the better word you can think of. For all stuff that is not yet bundled with piper. jarl From Nicolas.Chauvat at logilab.fr Mon Feb 19 06:56:12 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:14 2006 Subject: [Pipet Users] Design In-Reply-To: <3A91079D.7D37532E@xs4all.nl> Message-ID: > > Shall I make a "piper-doc" module? > Or maybe something more general, pre-piper, or piper-waitingroom, or the > better word you can think of. For all stuff that is not yet bundled with > piper. As far as cvs is concerned, it does not care what the directory hierarchy is and makes not difference between "modules". So if you tell piper users to download separately piper/doc/, piper/src and piper/test-suite or piper/website, you'll get the same result, but "everything piper" will have a single root instead of having many piper children of /cvsroot. You do "cvs co piper/doc" or "cvs co piper/website" when you want part of it and do "cvs co piper" when you want everything. But it's not that important, is it? :-) -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From jeff at bioinformatics.org Mon Feb 19 23:59:33 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:14 2006 Subject: [Pipet Users] Design References: Message-ID: <3A91F9B5.124A4010@bioinformatics.org> Nicolas Chauvat wrote: > > Agreed. That's my mistake, don't hit me on the head... please! I'll let it go, this time ;-) We can put everything into the "piper" module, as you suggested. We'd need piper/doc piper/website piper/src I'd need to re-import the module. So, then we can package piper/src without the docs. Good idea. Jarl, I'm not sure what you have in mind for "all stuff that is not yet bundled with piper". It's not a language problem; I just need some examples. Are you talking about the misc docs? Anything else, like libraries? Perhaps that is what Nicolas suggested. Cheers. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jarl at xs4all.nl Tue Feb 20 07:11:50 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:14 2006 Subject: [Pipet Users] Design References: <3A91F9B5.124A4010@bioinformatics.org> Message-ID: <3A925F06.84F92FFA@xs4all.nl> > We can put everything into the "piper" module, as you suggested. We'd need > > piper/doc > piper/website > piper/src > Jarl, I'm not sure what you have in mind for "all stuff that is not yet > bundled with piper". It's not a language problem; I just need some examples. > Are you talking about the misc docs? Anything else, like libraries? Perhaps > that is what Nicolas suggested. like a subdir piper/pending. For stuff that could be put into the Wiki, but is exspected to be part of Piper after it has been finished. jarl From jeff at bioinformatics.org Tue Feb 20 08:53:32 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:14 2006 Subject: [Pipet Users] Design References: <3A91F9B5.124A4010@bioinformatics.org> <3A925F06.84F92FFA@xs4all.nl> Message-ID: <3A9276DC.3FFA35B6@bioinformatics.org> Jarl van Katwijk wrote: > > like a subdir piper/pending. For stuff that could be put into the Wiki, > but is exspected to be part of Piper after it has been finished. Sure, sounds grrrrreat! Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jeff at bioinformatics.org Tue Feb 20 17:57:07 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:14 2006 Subject: [Pipet Users] Design References: <3A91F9B5.124A4010@bioinformatics.org> <3A925F06.84F92FFA@xs4all.nl> Message-ID: <3A92F643.7305E245@bioinformatics.org> Okay, done. Jeff Jarl van Katwijk wrote: > > > We can put everything into the "piper" module, as you suggested. We'd need > > > > piper/doc > > piper/website > > piper/src > > Jarl, I'm not sure what you have in mind for "all stuff that is not yet > > bundled with piper". It's not a language problem; I just need some examples. > > Are you talking about the misc docs? Anything else, like libraries? Perhaps > > that is what Nicolas suggested. > > like a subdir piper/pending. For stuff that could be put into the Wiki, > but is exspected to be part of Piper after it has been finished. From karlmax at oberland.net Fri Feb 23 09:56:02 2001 From: karlmax at oberland.net (Karl-Max Wagner) Date: Fri Feb 10 19:39:14 2006 Subject: [Pipet Users] Preliminary HurricaneFactory docs up Message-ID: <200102231456.OAA00340@schreyer.oberland.net> Hi Pipers, I put together some ( a bit oldish ) stuff about my current project, HurricaneFactory. The docs actually are presentation sheets for a presentation done in Tokyo at the 6th IEICE softradio conference and at that time I didn't know about piper yet. I plan to write up some stuff of how to plumb Piper and HurricaneFactory together asap. In the meantime you might like to have a look at http://lea.hamradio.si/~db8co/HurricaneFactory/hf-HP.html to see what I'm up to. Tell me what you think and particularly of how you'd integrate that into piper ( as far as it makes sense ). I did not put up any code yet - the code is pretty hairy and is undergoing a major change right now anyway. ======================================================================= "It was hell. They knew it. Karl-Max Wagner But they called it karlmax@oberland.net W-I-N-D-O-Z-E" ham radio: DB8CO *********Member of No Code International********* ***********Visit http://www.nocode.org*********** ********Membership Number NCI-2563-DB8CO********* _ / / (_)__ __ ____ __ / /__/ / _ \/ // /\ \/ / . . . t h e c h o i c e o f a /____/_/_//_/\_,_/ /_/\_\ G N U g e n e r a t i o n . . "Et ceterum censeo ut Microsoftem delendum esse" (Cato, adapted) "Quo usque tandem abutere nostra patientia ?" (Cicero, original) ======================================================================= From karlmax at oberland.net Tue Feb 27 18:23:18 2001 From: karlmax at oberland.net (Karl-Max Wagner) Date: Fri Feb 10 19:39:14 2006 Subject: [Pipet Users] Piper config errors.... Message-ID: <200102272323.XAA01261@schreyer.oberland.net> Hi Pipers, just checked out piper via anon cvs and tried to autogen it. The first funny thing was that it complained about a missing AUTHORS file ( funny, because it's there ) and then, grepping through the config log I found the following: automake: Makefile.am: required file ./AUTHORS' not found automake: Makefile.am: required file ./AUTHORS' not found ./configure: GNOME_INIT: command not found ./configure: GNOME_COMPILE_WARNINGS: command not found ./configure: GNOME_INIT: command not found checking for GNOME_INIT GNOME_COMPILE_WARNINGS I found that similar stuff is defined in some .m4 files, these two, however, are not. So obviously the config procedure gets confused. I did not try a rollback to find out from when on this happened, but clearly something is missing right now. Need to run it once more to ask a few questions about the BL <> PL interface.... ======================================================================= "It was hell. They knew it. Karl-Max Wagner But they called it karlmax@oberland.net W-I-N-D-O-Z-E" ham radio: DB8CO *********Member of No Code International********* ***********Visit http://www.nocode.org*********** ********Membership Number NCI-2563-DB8CO********* _ / / (_)__ __ ____ __ / /__/ / _ \/ // /\ \/ / . . . t h e c h o i c e o f a /____/_/_//_/\_,_/ /_/\_\ G N U g e n e r a t i o n . . "Et ceterum censeo ut Microsoftem delendum esse" (Cato, adapted) "Quo usque tandem abutere nostra patientia ?" (Cicero, original) ======================================================================= From jeff at bioinformatics.org Tue Feb 27 20:11:00 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:14 2006 Subject: [Pipet Users] Piper config errors.... References: <200102272323.XAA01261@schreyer.oberland.net> Message-ID: <3A9C5024.9C4EDAA6@bioinformatics.org> Karl-Max Wagner wrote: > > automake: Makefile.am: required file ./AUTHORS' not found (processing in ./peep) > automake: Makefile.am: required file ./AUTHORS' not found (processing in ./dl) AUTHORS *WAS* not present in peep and dl. I just added them, so the warning should not appear again. > ./configure: GNOME_INIT: command not found > ./configure: GNOME_COMPILE_WARNINGS: command not found (configuring in ./dl) > ./configure: GNOME_INIT: command not found (configuring in ./bl) These are in the dl and bl configure.in files. They don't appear anywhere else. There are other GNOME_ directives, but they don't produce a warning. I'm not familiar with autoconf, so I can't be of much more help. None of these warnings prevent me from building Piper. Are you unable to continue with the build? Thanks for the bug report. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From karlmax at oberland.net Wed Feb 28 08:06:36 2001 From: karlmax at oberland.net (Karl-Max Wagner) Date: Fri Feb 10 19:39:14 2006 Subject: [Pipet Users] Piper config errors.... In-Reply-To: <3A9C5024.9C4EDAA6@bioinformatics.org> from "J.W. Bizzaro" at Feb 28, 2001 01:11:00 AM Message-ID: <200102281306.NAA00308@schreyer.oberland.net> > None of these warnings prevent me from building Piper. Are you unable to > continue with the build? Yes. This is what happens: gmake all-recursive gmake[1]: Entering directory /root/piper/src' Making all in config gmake[2]: Entering directory /root/piper/src/config' gmake[2]: Nothing to be done for `all'. gmake[2]: Leaving directory /root/piper/src/config' Making all in scripts gmake[2]: Entering directory /root/piper/src/scripts' gmake[2]: Nothing to be done for `all'. gmake[2]: Leaving directory /root/piper/src/scripts' Making all in xml gmake[2]: Entering directory /root/piper/src/xml' Making all in Receptacle gmake[3]: Entering directory /root/piper/src/xml/Receptacle' gmake[3]: Nothing to be done for `all'. gmake[3]: Leaving directory /root/piper/src/xml/Receptacle' gmake[3]: Entering directory /root/piper/src/xml' gmake[3]: Nothing to be done for `all-am'. gmake[3]: Leaving directory /root/piper/src/xml' gmake[2]: Leaving directory /root/piper/src/xml' Making all in dl gmake[2]: Entering directory /root/piper/src/dl' cd . && autoheader gmake all-recursive gmake[3]: Entering directory /root/piper/src/dl' Making all in modules gmake[4]: Entering directory /root/piper/src/dl/modules' gmake[4]: Nothing to be done for `all'. gmake[4]: Leaving directory /root/piper/src/dl/modules' Making all in scripts gmake[4]: Entering directory /root/piper/src/dl/scripts' gmake[4]: Nothing to be done for `all'. gmake[4]: Leaving directory /root/piper/src/dl/scripts' Making all in vflow gmake[4]: Entering directory /root/piper/src/dl/vflow' Making all in interfaces gmake[5]: Entering directory /root/piper/src/dl/vflow/interfaces' gmake[5]: *** No rule to make target `vflowmodule.so', needed by `all-am'. Stop. gmake[5]: Leaving directory /root/piper/src/dl/vflow/interfaces' gmake[4]: *** [all-recursive] Error 1 gmake[4]: Leaving directory /root/piper/src/dl/vflow' gmake[3]: *** [all-recursive] Error 1 gmake[3]: Leaving directory /root/piper/src/dl' gmake[2]: *** [all-recursive-am] Error 2 gmake[2]: Leaving directory /root/piper/src/dl' gmake[1]: *** [all-recursive] Error 1 gmake[1]: Leaving directory /root/piper/src' make: *** [all-recursive-am] Error 2 Actually, it reminds me of an older bug report I put into a longer posting before the big discussion. I repeat it here: ************************************************************ What I did now was to compare the makefiles between my old piper version and the new one in the directory piper/dl/vflow/interfaces. The result is interesting: --- ./piper_old/dl/vflow/interfaces/Makefile Sun Jan 7 00:12:39 2001 +++ ./piper/dl/vflow/interfaces/Makefile Sat Jan 27 18:57:50 2001 @@ -203,8 +203,8 @@ maintainer-clean-libtool: -vflowmodule$(SO): $(vflowmodule__SO__OBJECTS) $(vflowmodule__SO__DEPENDENCIES) - @rm -f vflowmodule$(SO) +vflowmodule0 11 10 6 4 3 2 1 0SO): $(vflowmodule__SO__OBJECTS) $(vflowmodule__ + @rm -f vflowmodule0 11 10 6 4 3 2 1 0SO) $(vflowmodule__SO__LINK) $(vflowmodule__SO__LDFLAGS) $(vflowmodule__SO_ tags: TAGS Actually, in the new vflowmodule line there is an opening bracket missing. The same in the @rm line. No wonder that make gets confused. ************************************************************ The offending line wasn't changed since then ( I checked it ) ======================================================================= "It was hell. They knew it. Karl-Max Wagner But they called it karlmax@oberland.net W-I-N-D-O-Z-E" ham radio: DB8CO *********Member of No Code International********* ***********Visit http://www.nocode.org*********** ********Membership Number NCI-2563-DB8CO********* _ / / (_)__ __ ____ __ / /__/ / _ \/ // /\ \/ / . . . t h e c h o i c e o f a /____/_/_//_/\_,_/ /_/\_\ G N U g e n e r a t i o n . . "Et ceterum censeo ut Microsoftem delendum esse" (Cato, adapted) "Quo usque tandem abutere nostra patientia ?" (Cicero, original) ======================================================================= From jmvalin at locusdialogue.com Wed Feb 28 12:08:45 2001 From: jmvalin at locusdialogue.com (Jean-Marc Valin) Date: Fri Feb 10 19:39:14 2006 Subject: [Pipet Users] Overflow (PL) updates Message-ID: <3A9D309D.E40DFB9E@locusdialogue.com> Hi everyone, A lot has been going on in the Overflow CVS and I'm soon planning to make a new release. Lots of the changes concern ease of use. Here's a qui list of changes: - The VFLOW_PATH environment variable is no longer required (it's determined from the --prefix configure option) - What was in libflowui is now in the same library than the one called libnetwork. The new library is now called libflow and is all that needs to be linked with Piper. - The FFTW library is no longer required to build Overflow. If you don't have it, the build will continue without FFT support. - It's now possible to have Feedback loops in a flow. This is done by using the "Feedback" node and a demo is available in the examples directory. - Overflow also supports throwing and catching exceptions within the flow using the "Throw" and "Catch" node. This is still undocumented. Some of these changes will require modifications to Piper, mostly in the makefiles/configure.in, though. I suggest someone gets the current Overflow version in CVS and tries to build Piper with it, so that the next version I release works well with Piper. Jean-Marc From jeff at bioinformatics.org Wed Feb 28 20:04:24 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:14 2006 Subject: [Pipet Users] Xirium | Tech | MUD-Shell Message-ID: <3A9DA018.4A73A70F@bioinformatics.org> A while back I mentioned having a Natural Language Interface (NLI) for the Peep (text-based) UI. The idea I had in mind was something like combining the CLI with Zork. Someone has done just that: http://www.xirium.com/tech/mud-shell/index.html although I would make it a little less like Zork than they did :-) Jeff From jeff at bioinformatics.org Wed Feb 28 21:17:58 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:14 2006 Subject: [Pipet Users] Re: Xirium | Tech | MUD-Shell Message-ID: <3A9DB156.C3DC56C@bioinformatics.org> I attached a couple messages containing early discussions about an NLI for Peep. Gary and I came up with the idea in January 2000. Narval guys, what do you think about using an NLI with your intelligent assistant? Jeff -------------- next part -------------- An embedded message was scrubbed... From: Gary Van Domselaar Subject: a couple minor brainwaves Date: Thu, 20 Jan 2000 21:03:34 -0700 Size: 4772 Url: http://bioinformatics.org/pipermail/pipet-users/attachments/20010301/5923934a/attachment.mht -------------- next part -------------- An embedded message was scrubbed... From: "J.W. Bizzaro" Subject: [loci] bioinformatics for the visually impared Date: Thu, 03 Feb 2000 21:35:06 +0000 Size: 4186 Url: http://bioinformatics.org/pipermail/pipet-users/attachments/20010301/5923934a/attachment-0001.mht