Asking The Developers Before Asking Biostars

Forum: Asking The Developers Before Asking Biostars

5.6 years ago by

Milan, Italy

Arnaud Ceol • 840 wrote:

Dear all,

While most questions on Biostar are requests for methods and advices, many of them are also related to usage and installation of software and databases.

In the last case, the first thing to do is, to my opinion, to ask the developers/maintainers of the software/database. Most of the time, the group is more than happy to answer the questions. I'm wondering in how many cases this has been done before to post a question on Biostar. I think this could be a "must do", and when no answer has been provided, it could be mentioned in the post. In addition, if an answer has been provided, Biostar users would sure be happy to know about it, so it may be posted together with the answer.

I believe such behavior may help: 1) ensure the right people (the developers of the application/database) are asked first, 2) help the Biostar community know about the best answer, 3) keep track of the follow up provided by the databases and developers (by the way it may also be good to know for referees).

Any opinion about it?

forum • 3.5k views

ADD COMMENT • link •

modified 5.6 years ago by Mary ♦ 11k • written 5.6 years ago by Arnaud Ceol • 840

5.6 years ago by

Chris Miller ♦ 20k

Washington University in St. Louis, MO

Chris Miller ♦ 20k wrote:

We actually discussed this with Istvan about a year ago, and have been pointing users of our software here for support. We then have a tool that monitors the RSS feed for specific keywords or tags and then notifies the appropriate people to come answer it. (rss2jira) We do try to be clear that bug reports don't belong here and patches and such should go on github.

This has had a number of positive effects. The answers to commonly asked questions about our tools are now publicly accessible, indexed, and on a site where lots of people can find them (as opposed to buried away in some obscure mailing list archives). It also drives traffic to Biostar and makes lots of users aware of the community. Finally, several people who have started out asking questions about Breakdancer or Somatic Sniper here have gone on to become regular contributors to the site.

ADD COMMENT • link written 5.6 years ago by Chris Miller ♦ 20k

5.6 years ago by

Emily_Ensembl ♦ 18k

EMBL-EBI

Emily_Ensembl ♦ 18k wrote:

I man the Ensembl helpdesk and try to keep an eye on BioStar, answering any Ensembl questions. I think it's our responsibility to do so, and other developers should do the same thing. Sometimes, only we can answer the question or give the most up-to-date answer.

Personally I think that Ensembl-specific questions belong to our helpdesk and dev list, and more blue sky questions belong here. People posting to both helpdesk and here get my goat; if I'm quick I can answer both with a bit of copy/paste, but occasionally it results in duplicate work with someone else answering on here as well as me answering on helpdesk.

ADD COMMENT • link written 5.6 years ago by Emily_Ensembl ♦ 18k

I've tried to encourage other project teams to monitor this more actively, but it's not happening as much as I'd like. I think it would be good for them to be proactive like you are. A few times I've even taken questions from here and fed them to the appropriate developers but it hasn't been very satisfying.

ADD REPLY • link written 5.6 years ago by Mary ♦ 11k

Dual posting on any of StackExchange, BioStars, SEQanswers, or project specific mailing lists (unless cross referenced) annoys me for the exact same reason - duplicated effort for the community answering. I make a point of commenting about this when I notice.

ADD REPLY • link written 5.6 years ago by Peter ♦ 5.8k

5.6 years ago by

zx8754 ♦ 7.1k

London

zx8754 ♦ 7.1k wrote:

When we seek answers, we want them now - Biostar provides this by its Biostar Traffic - Just How Many Bioinformaticians Are Out There?. We could ask developers/maintainers, 1st they might have moved on, 2nd if and when they reply.

ADD COMMENT • link written 5.6 years ago by zx8754 ♦ 7.1k

I agree that small-scale developers might have moved on, or might not get around to replying to emails, but I think the large-scale well-maintained projects have a responsibility to the community. They should answer their help emails (we aim to answer initially within a working day, even if that answer is just "I'll look into it") and they should be on here looking for relevant questions about their resource. I can understand why that's not really possible for small-scale developers, but for projects like Ensembl, it's vital.

ADD REPLY • link modified 5.6 years ago • written 5.6 years ago by Emily_Ensembl ♦ 18k

5.6 years ago by

Jeremy Leipzig ♦ 18k

Philadelphia, PA

Jeremy Leipzig ♦ 18k wrote:

This is probably the right approach for bigger projects (UCSC, GATK), where they have an official support person who is paid to field these type of questions.

Otherwise, buttonholing the developers is a distraction to the very people who are producing something of value. Why not just post it here for people who have actually volunteered to solve problems?

I would rather simply post the question here, see who responds, and then as a last resort click the "via email" link you see above.

ADD COMMENT • link modified 5.6 years ago • written 5.6 years ago by Jeremy Leipzig ♦ 18k

5.6 years ago by

Arnaud Ceol • 840

Milan, Italy

Arnaud Ceol • 840 wrote:

I tend to relate it to "What have you tried before", i.e. show that you tried with the official resources before to ask the community to give you some time. I also think it is a good way to identify dead projects.

It is also really good to see that official resources (like ensembl) are aware of and following Biostar.

ADD COMMENT • link modified 5.6 years ago • written 5.6 years ago by Arnaud Ceol • 840

My experience: Someone sends an email to Ensembl helpdesk asking how to do something. I give them a couple of suggestions as to how to do it. They give reasons that they can't use my suggestions. I then say, well those were all the weapons in my arsenal and maybe they should rethink their reasons not to use my suggestions; otherwise I can't help.

They then post the same question onto BioStar, just the original question. Someone writes a response suggesting one of my original suggestions. Someone else suggests the other. OP gets back explaining why he can't use them. Everybody gets bored because they've given a decent answer to the question and they can't see why the OP can't do their suggestions, and they can't see why they bothered to write an answer that the OP already knew wasn't feasible but didn't bother to mention in the first place. Nobody responds any further and the OP doesn't get the answer he wants.

ADD REPLY • link modified 5.6 years ago • written 5.6 years ago by Emily_Ensembl ♦ 18k

5.6 years ago by

Mary ♦ 11k

Boston MA area

Mary ♦ 11k wrote:

I just saw this note on a discussion board for a tool and it made me laugh and think of this thread:

https://groups.google.com/forum/#!forum/gitools-users

Hi dear Gitools user.

Please use this mailing list to ask questions you may have specific questions regarding Gitools and it's use. For more genereal (bioinformatics) problems, consider if BioStars may be the appropriate site to ask for help.

Thanks

It all comes around. :)

ADD COMMENT • link written 5.6 years ago by Mary ♦ 11k

and sometimes, the UCSC redirects people to seqanswers or biostars: https://groups.google.com/a/soe.ucsc.edu/forum/#!msg/genome/ZhV-mjQoOYc/1ITH5yHm34AJ

ADD REPLY • link written 5.6 years ago by Pierre Lindenbaum ♦ 119k

Please log in to add an answer.

Similar posts • Search »