[BioC] Idea that needs feedback

Saurin D. Jani saurin_jani at yahoo.com
Wed Dec 22 16:20:44 CET 2010 

Hi Radhouane Aniba,

What I understand from quick look of this discussion is ..you want to create a 
methods (R scripts) library with description in terms ..lets say..

R script 1.R
input: 
ouptut:
what is used for... 
how output looks like: 

modificaions:
comments:


and so, learners can easily grab the code and put it in their script or use the 
whole script to start learning R/BioC's capabilities?

if that's the case..I like it..its good resource. 

Saurin



----- Original Message ----
From: Radhouane Aniba <aradwen at gmail.com>
To: ttriche at usc.edu
Cc: bioconductor at r-project.org
Sent: Wed, December 22, 2010 10:04:50 AM
Subject: Re: [BioC] Idea that needs feedback

Tim,

Thank you for your message.
GitHub is for managing projects as far as I know, I personnaly never been
there, but browsing a little bit their homepage R is not in their top 10
popular langages ! May be I am wrong.
My proposal is not to create another project management website platform,
other equivalent to GitHub are doing a great job.
A collaborative platform for sharing code sources could work either as a
platform for a scripts/code "show case" or a mean by which people get in
touch with each other so that they speed up their development using what
others already did (using is also modifying and enriching the scripts
database).

I need more feedback on that issue, would you please (some of you who did
not participated in this thread already) take two minutes of your time to
tell if you find this idea useful or interesting or not.


Regards

Rad

2010/12/21 Tim Triche <tim.triche at gmail.com>

> it sounds like you're proposing something along the lines of GitHub gists
> and/or StackOverflow for Bioconductor users.
>
> While I'm not against this idea, I wonder how much traction it would get,
> since the Internet is practically a testament to wheel reinvention.
>  (Reinventing wheels is not all bad -- if you're learning how to build and
> balance wheels, it's kind of mandatory)
>
> In any event, good luck.  The community could stand to have a more
> rough-and-tumble hub than the BioC website, if only for hashing out good vs.
> bad ideas.  Personally I put my code up on GitHub, communicate with the
> relevant people, and see what happens...
>
>
> On Tue, Dec 21, 2010 at 4:07 PM, Radhouane Aniba <aradwen at gmail.com>wrote:
>
>> Hi all, Hi arne,
>>
>> Well as I said in my first message, this is not intended to overlap any of
>> Bioc services (website, forums, tutorials, workflows) that I found very
>> well
>> done and useful by the way.
>> Well my point is that for new people getting involved in R/Bioc, the
>> learning curve is different depending on each one capabilities. It could
>> be
>> very helpful for everyone, even for "guru" R/Bioc developpers to see what
>> other already did that seems to be suitable for a part of a project. Just
>> a
>> word about tutorials, workflows and useful help in Bioc website; like
>> every
>> bio* project (Bioperl, Biojava, Biopython..) when a project is lunched and
>> when people start to get involved in, there is always "how to do" sections
>> and lot of tutorials that developers put in front so that people start to
>> practice and to get more familiar with the langage syntax and tips. I
>> remeber when I was developing using Biojava, with my respect to people
>> behing that do a great job, there is only a wiki page which contains
>> several
>> helpful sections but this is only the top of the iceberg, such projects
>> tends to evoluate exponentially and when a newbie start to get interested
>> in
>> using a package or langage X, it is already a mature field and people in
>> the
>> mailing lists and forum already speak a very hard t understand jargon, may
>> be I am wrong, at least that was my case when I first started to learn
>> biojava. Bioc is a relatively new, and more andmore used, and except the
>> workflows and tutorials that exist in their website, this is for sure othe
>> use cases and scenarios that some of us has already implemented in order
>> to
>> publish a paper or to develop a project, a lot of projects overlap in
>> their
>> biocomputational development parts, that's why I think personnaly that
>> sharing knowledge on "how I did that" part or what workflow i developed
>> used
>> existing package in Bioc could be very helpful in the sens that it will
>> lead
>> automatically to save hours of developments and accelerate research, this
>> is
>> the basics of knowledge sharing let's say, and this is how I came to the
>> point that sharing codes for R and BioC is a crucial and very helpful
>> topic.
>> Such platform that I have in mind has nothing to do with forums, actually
>> people will be able to develop profile pages and write posts about their
>> experience with R and Bioc, upload their presentations and posters so that
>> all of us we can found directly the information we need instead of
>> spending
>> hours on the web looking for an information, it could be a very small
>> portion of code, well commented and very clear, as it could be an entire
>> workflow showing how to connect different bioc packages. I don't see if I
>> am
>> enough clear about the idea but with all respect that i have to all
>> projects
>> related to bio, non of them(us) is able to imagine in advance all possible
>> scenarios using a specific tool. One example could be more speaking is
>> what
>> people behind Taverna package did developing myExperiment website, may be
>> we
>> can look at it and imagine features to a collaborative platform.
>>
>> Any comment is welcome and thank you for answering.
>>
>> Radhouane
>>
>> 2010/12/21 <arne.mueller at novartis.com>
>>
>> > Hi,
>> >
>> > in principle this sounds good, but it seems there's much overlap with
>> the
>> > bioconductor web-site that already offers, packages, workflows,
>> > documentation, tutorials and hands-on hints and discussion (mailing
>> list)
>> > ...
>> >
>> > In my experience many packages are as abstract as possible (IRanges and
>> > off-spring ...) whereas others are more tailored towards a specific
>> need.
>> > Most of the code that I write is just "glue" to stick packages together
>> in a
>> > workflow (or analysis), but it's not really re-usable workflow stuff,
>> since
>> > it's often project specific, etc ... . I think there's only limited
>> > abstraction beyond the package level that's really useful - otherwise
>> I'd
>> > write a package ;-) .
>> >
>> > Sorry, don't want to be negative, but maybe I didn't get your point. The
>> > only thing I'd really find useful is a web-based subscription free forum
>> (I
>> > think this topic was briefly discussed at the last bioc developers
>> > conference in Heidelberg, but I don't remember the outcome ;-).
>> >
>> > Maybe instead of setting up a new web-site you could join the
>> > bioc-developers and help improving their web-site (no offense, after
>> getting
>> > used to the new web-site I really started liking it ;-)
>> >
>> >    best,
>> >
>> >   arne
>> >
>> >
>> >
>> >  *Radhouane Aniba <aradwen at gmail.com>*
>> > Sent by: bioconductor-bounces at r-project.org
>> >
>> > 12/21/2010 08:39 PM
>> >   To
>> > bioconductor at r-project.org
>> > cc
>> >   Subject
>> > [BioC] Idea that needs feedback
>> >
>> >
>> >
>> >
>> > Hi BioC users and contributors,
>> >
>> > In order to improve collaboration between scientists, computer
>> engineers,
>> > bioinformaticians working with R and Bioconductor, I would like to see
>> how
>> > much among you here in the list, would be interested in developing or
>> > making
>> > part of a website that plays the role of collaborative platform for
>> > R/Bioconductor codes, packages, workflows and contributions, so that we
>> > develop the "re-use" and "don't reinvent the wheel" spirit in one hand,
>> and
>> > to put in practice a more collaborative work and interaction between
>> people
>> > working on R/Bioconductor related projects.
>> >
>> > I already have experience in the development of such platforms and
>> thought
>> > it could be useful and interesting to propose such initiative to people
>> > like
>> > you and I working with R/BioC
>> >
>> > This is in no way a competitive or alternative approach to Bioconductor
>> > mailing list but a complementary platform for sharing codes / workflows
>> /
>> > analysis scenarios using R/BioC
>> >
>> > Would you please take a few moment to answer to this topic so that I can
>> > have an idea on your position in regards of such initiative.
>> >
>> > Thank you
>> >
>> > Radhouane
>> >
>> > --
>> > *Radhouane Aniba*
>> > *Bioinformatics Research Associate*
>> > *Institute for Advanced Computer Studies
>> > Center for Bioinformatics and Computational Biology* *(CBCB)*
>> > *University of Maryland, College Park
>> > MD 20742*
>> >
>> >                 [[alternative HTML version deleted]]
>> >
>> > _______________________________________________
>> > Bioconductor mailing list
>> > Bioconductor at r-project.org
>> > https://stat.ethz.ch/mailman/listinfo/bioconductor
>> > Search the archives:
>> > http://news.gmane.org/gmane.science.biology.informatics.conductor
>> >
>> >
>>
>>
>> --
>> *Radhouane Aniba*
>> *Bioinformatics Research Associate*
>> *Institute for Advanced Computer Studies
>> Center for Bioinformatics and Computational Biology* *(CBCB)*
>> *University of Maryland, College Park
>> MD 20742*
>>
>>        [[alternative HTML version deleted]]
>>
>> _______________________________________________
>> Bioconductor mailing list
>> Bioconductor at r-project.org
>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>> Search the archives:
>> http://news.gmane.org/gmane.science.biology.informatics.conductor
>>
>
>
>
> --
> If people do not believe that mathematics is simple, it is only because
> they do not realize how complicated life is.
> John von 
>Neumann<http://www-groups.dcs.st-and.ac.uk/%7Ehistory/Biographies/Von_Neumann.html>
>
>
>


-- 
*Radhouane Aniba*
*Bioinformatics Research Associate*
*Institute for Advanced Computer Studies
Center for Bioinformatics and Computational Biology* *(CBCB)*
*University of Maryland, College Park
MD 20742*

    [[alternative HTML version deleted]]

_______________________________________________
Bioconductor mailing list
Bioconductor at r-project.org
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives: 
http://news.gmane.org/gmane.science.biology.informatics.conductor

More information about the Bioconductor mailing list