[BiO BB] Please guide me
dmb at mrc-dunn.cam.ac.uk
Fri Jan 30 15:28:26 EST 2004
> Michael Gruenberger wrote:
>>On Fri, 2004-01-30 at 15:20, Andrius wrote:
>>>i'm sure real world will demand it. it's kind of a feeling when you doing
>>> research with such tools..it
>>>reminds me open source climate a lot. the thing is... sometimes even
>>> programmers do not think in a real way.
>>>they tend to follow 'practices' and there's little few who follows their brain.
>>> you may disagree and of course i can ask:
>>>does our today 'market' demand thinking? same for research i think. as long as
>>> researchers demand tools which
>>>allow them to think free ( and gain better results this way ) there's a market
>>> for such tools. of course it may not be mainstream,
>>>but there's nothing wrong with that. i'm not much familiar with research
>>> routines (i'm still graduating for bachelor) and people in it, but i have few
>>> friends there. they tend to say that even in academic world there's a lot of
>>> serial writing (i mean writing papers just to increase curriculum) and there's
>>> no surprise such people do not demand open thinking tools. and there's nothing
>>> wrong with that again:) as long as there's bunch of people who care about a
>>> quality of their work (research).
>>Yes, you are right, researchers need (computer) tools, but they need to be very
>> easy to use and a new programming language would also have to be very easy to
>> use. What would your bio-language look like? What would make it easier to use
>> than existing BioJava or BioPerl? Bioinformatics is a very wide field (I work on
>> two projects at the moment, one is mainly database stuff, the other one 3D
>> images). What would your
>>language focus on?
> it may be pluggable, but i don't want to speculate. i just have an idea for a
> future. in short
> we can build language for JVM for example which uses all what Java ( notice 3d
> here too) can provide but completely hiddens it from
> bioinformatics programmer. i'm sure i'll find problems with existing solutions
> (bioperl and biojava) when i put my hands on it. and yes, ease of use is a main
> reason for such custom language. and i guess it means how language is presented
> for a researcher. i doubt
> plain java is applicable.
sounds like the tower of babel gone wrong - java, perl, c, all are good for
bioinformatics, laying the concepts on top is the tricky part, and the more you ask
a language to do for you, the more 'locked in' to a particular mind set you get.
Traditional programing languages are good because of the freedom of they give -
beyond that it is a matter of taste. I do not think each dicipline of science should
have its own language. subroutines yes, dialect no.
> BiO_Bulletin_Board maillist - BiO_Bulletin_Board at bioinformatics.org
More information about the BBB