Forum: (Closed) StackExchange/Area51 bioinformatics is back
19
gravatar for Pierre Lindenbaum
5 months ago by
France/Nantes/Institut du Thorax - INSERM UMR1087
Pierre Lindenbaum98k wrote:

i would like to bring to your attention that the old discussion of "Biostars vs StackExchange" is back.

While I don't really care about the platform, I wouldn't like that StackExchange releases a new platform without biostars.

See also : https://groups.google.com/forum/#!msg/biostar-central/bMgmtoEcQV8/NbdzVo8eRXsJ;context-place=forum/biostar-central

area51 forum biostars • 3.1k views
ADD COMMENTlink modified 4 months ago • written 5 months ago by Pierre Lindenbaum98k
15

I just don't get the impetus behind yet another site. The only answer I've seen on SE is that "biostars accepts too many low quality questions". Right, but that's also the problem with stack exchange, namely that it's a largely toxic environment for new folks.

ADD REPLYlink written 5 months ago by Devon Ryan70k
13

Low quality questions because the majority of people who come on here to learn often have no idea what they are doing and no formal training in bioinformatics. When I joined a little over a year and a half ago, my first question (on another account who's password I have forgotten) was 'how do I intersect bed files'. Literally it. Now I know enough to help other users with their own questions.

The amount that I have learned from Biostars is just crazy, and I always recommend it to people when they ask me how to get started. "Just go read Biostars posts".

I hate SE as a medium, if you ask a simple question you get downvoted to oblivion and at best you get a condescending comment about your question without even an answer. It's a horrible teaching platform.

ADD REPLYlink modified 5 months ago • written 5 months ago by Sinji2.5k
7

I hate SE as a medium, if you ask a simple question you get downvoted to oblivion and at best you get a condescending comment about your question without even an answer. It's a horrible teaching platform.

I disagree with this. I have used both Biostars and SE while learning and both helped me immensely. What I like about SE is that it is truly a Q&A site: ask a straight question get a straight answer. Yes, one gets one's butt spanked here and there after some poorly formated/reasoned questions, but it also teaches something fundamental in the process: think clearly what the question is. I have quite often answered my own questions while writing a post there. It also very easy to find what I am looking for, which makes it less likely to ask duplicated questions. Actually overall I probably get more help from SE than Biostars.

Biostars has its merits, friendly community and all that, but the downside is that it can be very frustrating when one is trying to help but the information has to be dragged out of the OP clenched fits. But my major issue with Biostars is that frequently answers are deep in comments, and often answers are comments. In effect less of a Q&A site and more of a forum. It is not easy to parse threads to look for an answer. This is a solvable issue with stricter moderation: move comments to answers, or ask users to aggregate comments as an answer to clean up the clutter. Also, politely ask users to accept those answers that solved the issue. It will not make this site less friendly, but it will make it more helpful.

ADD REPLYlink written 5 months ago by A. Domingues1.4k
5

IIRC, in 2012/2013 when someone argued biostar comments are superior to SE, I said something like (couldn't find the thread): it is not because SE doesn't have the manpower to implement nested comments; it is because such comments are distracting. SE also puts a 512-character hard limit on comments, it hides some comments if there are too many, and it doesn't add comment upvotes towards the question. All these discourage you to put answers into comments. I still think biostar should follow SE. That is the best way to keep biostar as a Q&A site, not a forum.

ADD REPLYlink written 5 months ago by lh330k
5

You forget to mention that asking 'good questions' requires a more than basic background in whatever the OP is asking. This is my problem with the SE system. You are REQUIRED to know anything in order to ask anything. How does a person whose never been exposed to bioinformatics and has absolutely no idea how to computationally solve a problem, much less the keywords to search for in order to solve that problem, create a question that is clear enough to get answered on SE?

Seriously, it seems that half the time I post on there, half my post gets removed because it's 'not necessary'. SE is great for those of us who already understand the field, and the terminology, we get quick answers that are helpful, and correct. But will SE cater to those that are new to the field, and looking to START a Bioinformatics career? No. Definitely not.

ADD REPLYlink written 5 months ago by Sinji2.5k
6

I think many people criticizing SE are generalizing their experiences in stack overflow (SO) to SE. SO is over crowded and diversified. It is a hell to newcomers. However, many focused SE communities are fairly friendly. If you look at the biology SE, you can see questions asked by amateurs and still get answered. You also haven't used the old biostar, I guess. It did impose stricter rules, but it was still friendly overall. As long as you put thoughts and work in your questions, even very basic ones could get sometimes epic answers.

ADD REPLYlink written 5 months ago by lh330k
2

I have very negative experiences with biology SE. Moderators seem urged to comment on every question, often commenting that they think that something is not possible (without providing support). Also I made one particularly bad experience, when a difficult question of mine within my very own domain was closed with some vague accusation that OP wouldn't understand the basics. However I felt most upset, when I opened their chat channel to comment. There I would see that the moderators had publicly used very unprofessional and derogatory language to discuss my post (I suspect that it never crossed their mind that seemingly superficially sounding questions can address a difficult problem, or that such questions would be asked by someone who would know the field very well - and to a level that would exceed that of other biologists like themselves that would professionally focus on other fields).

ADD REPLYlink modified 4 months ago • written 5 months ago by unksci150
3

Yes, and I suffered from that at the beginning, but I never had a post removed from SE. I got some slaps for poor questions and had to do some edits but otherwise fine. In the process I learned how to learn. This also means that I spend a lot of time looking for an answer before I post. Of course YMMV.

One the other hand do we want questions that show no research from the user, contain no information or are excruciatingly hard to read? Because quite a lot of these will be left without an answer.

My comment might have been read as an harsh criticism of Biostars, but I am just advocating better curation of questions and answers. Sometimes this might be as simple as telling someone that the best place to answer the question is SO or bioconductor support, and explaining how to prepare a good question. That will go a long way on improving Biostars.

Disclaimer: when I started using both SE and Biostars was a bench biologist without any knowledge of bioinformatics or programming (undergrad Turbo Pascal excluded).

ADD REPLYlink written 5 months ago by A. Domingues1.4k
3

Just want to highlight the best bit from your comment, which BioStar lacks:

In the process I learned how to learn.

Essentially SO is about "Rubber Duck Problem Solving" , makes the user to think hard about the problem, before wasting anyone's time.

ADD REPLYlink modified 4 months ago • written 4 months ago by zx87543.7k
4

I completely agree with this. I started my phd as the only bioinformatics student in a molecular lab. Biostars was extremely helpful in not just in answering my questions, but also in expanding my knowledge. I learned tons from reading questions/answers that have no relevance to my phd. I don't see Biostars as just another question/answer site. I tell people routinely that I graduated from the university of biostars.

ADD REPLYlink modified 5 months ago • written 5 months ago by Damian Kao14k
1

This, exactly this. I'm already enough of an a$$ hole for the new folks to deal with, I would prefer not to inflict to stack exchange version of me on them :)

ADD REPLYlink written 5 months ago by Devon Ryan70k
3

Agreed. I'll take low quality questions any day over the toxic environment that runs rampant on Stack Exchange. As it is, I very much appreciate that BioStars isn't a part of SE.

ADD REPLYlink written 5 months ago by steve860
1

that it's a largely toxic environment for new folks

@Devon, true, for new folks, SE gives a feeling that they should leave the field immediately, new users (also new to field) feel terribly helpless, alone. This is completely opposite with Biostars, which makes it super nice. Albeit, sometime people say, you should've googled, what have you tried, previously answered..etc everyone helps solving the problem along with learning new stuff.

ADD REPLYlink written 5 months ago by venu4.3k
10

I don't see the point in diluting the user base to another community.

ADD REPLYlink written 5 months ago by WouterDeCoster21k
1

I agree, but whatever our opinions, a 3rd community is about to open :-)

ADD REPLYlink written 5 months ago by Pierre Lindenbaum98k
10

Can we just send everyone with a question about how to filter a fasta file based on the identifier to the new community?

ADD REPLYlink written 5 months ago by WouterDeCoster21k

@wouter, also people post already answered Biostars thread links on SE, you wrote the best example thread.

ADD REPLYlink written 5 months ago by venu4.3k

Yay, what do I win? :-P

ADD REPLYlink written 5 months ago by WouterDeCoster21k

cant we just link the Biostars answers pages in SE? :P ;)

ADD REPLYlink written 4 months ago by vchris_ngs4.0k

Just a link to Biostarts is not possible, in SE answers must be self contained. But links to other sites are not forbidden

ADD REPLYlink written 4 months ago by Lluís R.550
7

Be prepared for some more cross-posts

ADD REPLYlink modified 5 months ago • written 5 months ago by venu4.3k
3

Can you remove the annoying animated GIF? It is hurting my eyes everytime I visit this thread :)

ADD REPLYlink written 5 months ago by genomax33k
3

see, on stack exchange that would have been deleted inside 30 minutes ;)

ADD REPLYlink written 5 months ago by Neilfws47k

Done! I didn't check the messages, otherwise I would've done it already :)

ADD REPLYlink written 5 months ago by venu4.3k

@venu , I am pretty sure there will be cross posting but I will just either moderate it or link to Biostars solution. One way or the other it should be done that way to let people stop asking from redundant questions.

ADD REPLYlink written 4 months ago by vchris_ngs4.0k
7

I have to say, as someone who was a SE user before a BioStars user, I do prefer it stylistically (not that that is super important). A lot of Bioinfx questions get asked on Biology.SE anyway, but often without the support of great answers.

On the other hand, I think the 'attitude' on SE is a lot more hostile some times. There is a bit of an air of condescension that I don't see on BioStars.

Just makes me think of this:

enter image description here

ADD REPLYlink written 5 months ago by jrj.healey2.3k
6

The existence of things like Forum, Discussion, and Job posts on biostars are great and would be sad to see go. Is there anything comparable on stackexchange? The stackexchange networks are notoriously hostile to things that don't have a ridiculously narrow Q&A format

ADD REPLYlink written 5 months ago by cmdcolin670

This is a good point - Tutorials and the Job Board, in particular are really nice.

ADD REPLYlink written 5 months ago by Chris Miller18k
1

Bar for commitments appears to be pretty low so I assume that is going to be reached. People who regularly contribute here could decide to stay away from other platform. Are you planning to?

ADD REPLYlink written 5 months ago by genomax33k
1

I'm certainly not planning to start sinking time into another platform.

ADD REPLYlink written 5 months ago by Lars Juhl Jensen11k

Are you planning to?

why not ? but again, I would like to avoid the creation of a 3rd community.

ADD REPLYlink written 5 months ago by Pierre Lindenbaum98k

avoid the creation of a 3rd community

How can that be achieved?

ADD REPLYlink written 5 months ago by genomax33k
1

in the "old time", both communities could be merged:

https://groups.google.com/forum/#!msg/biostar-central/bMgmtoEcQV8/NbdzVo8eRXsJ;context-place=forum/biostar-central

but it remains Istvan 's choice :-)

ADD REPLYlink written 5 months ago by Pierre Lindenbaum98k
12

I have grown to deeply dislike the monoculture and centralization of modern "social networks" where facebooks, twitters and stackexchanges dominate every aspect of our lives.

The greatest differentiating factor of Biostars relative to all of the alternatives is that it is different, independent and is not integrated with the massive information tracking industry. People posting here are not tracked, their actions are not mined then resold etc.

ADD REPLYlink written 5 months ago by Istvan Albert ♦♦ 73k

Tagging: Istvan Albert

ADD REPLYlink written 5 months ago by genomax33k
1

Only 'pro' I could see to a Bioinfx.SE coming around would be if it actually replaced BioStars (not that I want that to happen ;) ). That would at least have the benefit of Istvan and co not having to run the site themselves and deal with the development. But maybe they enjoy it - I don't know :P. If BioStars merged with a new SE meaning the user base remained, that could work (all the mods etc would be transferrable and if the large, supportive, community here could dilute the tw*ts on SE sufficiently, Bioinfo.SE could be the only nice place to be on the whole domain haha

ADD REPLYlink written 5 months ago by jrj.healey2.3k

closing.

ADD REPLYlink modified 4 months ago • written 4 months ago by Pierre Lindenbaum98k
13
gravatar for Chris Miller
5 months ago by
Chris Miller18k
Washington University in St. Louis, MO
Chris Miller18k wrote:

I can see why lousy search, difficult UI, etc are frustrating to some folks. BioStars has the community and knowledgebase, but there doesn't seem to be a clear plan to improve the site, and reach feature parity with SE. Istvan, would you care to weigh in on those topics?

More than anything, I just want to avoid fragmentation of the community.

ADD COMMENTlink written 5 months ago by Chris Miller18k
1

I am no developer so I am not sure how much effort is involved but @Istvan in principle has agreed to several wishlist items (replacing the default search engine on main page, being one of the top) over the last year or so. I think he has been busy with his regular life (and the handbook).

Like some of us who contributed on the handbook perhaps developers amongst us will step forward to help improve Biostars on the whole.

Something good is bound to come out of SE development.

ADD REPLYlink modified 5 months ago • written 5 months ago by genomax33k
1

It's fair, and I don't expect Istvan to single-handedly shoulder the load. It would be great to have a development roadmap that's transparent to the community, as well as recruitment of contributors (paid or volunteer), if the platform is going to continue to grow. Right now, things seem to have stagnated: https://github.com/ialbert/biostar-central

ADD REPLYlink written 5 months ago by Chris Miller18k
1

I agree with the search being suboptimal - on the other hand, we always have Google and as a matter of fact that is how the majority of users come in from.

I found that the "Similar Posts" column in the right column is the best way to find relevant content.

One of the mistakes we initially made is that we rewrote Biostars too frequently (on a yearly basis almost). This took a lot of effort - for example, there is a fully rewritten but never released (and already obsolete) version of Biostars.

The current delay is by design - we want to let the ideas age a little bit more, to allow us to figure out what is it what we really need rather than what we think should have.

ADD REPLYlink written 5 months ago by Istvan Albert ♦♦ 73k

If the SE-like search function could be implemented on Biostars that would be great. So often google fails to turn up an answer/SE thread that is right, yet when I come to post a question, the suggested similar questions are more often than not, what I was looking for all along.

ADD REPLYlink written 5 months ago by jrj.healey2.3k

Well, I guess that's a feature we definitely can improve on, and probably isn't impossible. A problem which is harder to solve is that new people often don't even know which terms to search for.

ADD REPLYlink written 5 months ago by WouterDeCoster21k

Like you and I have asked people many times - if posters can formulate a succinct title for their question, some of those words are likely going to be appropriate search terms.

ADD REPLYlink modified 5 months ago • written 5 months ago by genomax33k

Yeah I think the latter is very much the 'nature of the beast'. That will never really go away - and probably shouldn't, because that means there are constantly new people (myself included) coming in to the field. Bioinformatics is an interesting discipline because its something that very few people ever have any formal training in, but people will find themselves needing to do increasingly.

ADD REPLYlink written 5 months ago by jrj.healey2.3k

Pattern searching and text mining do have a learning curve and it happens with time. I think we can always put up or re-publish the tutorials post in the forum or tutorial tags as top tags so that people who bumped in for the first time or even make a login id that should be the first post along with the welcome post that they should receive in their profile base. This can be done with easy push notification for every new user. Can't we? @Istvan Albert and @WouterDeCoster

ADD REPLYlink written 4 months ago by vchris_ngs4.0k
2

Sounds like a good suggestion, but it doesn't guarantee that new users will really read it. Do you accept the terms and conditions?

Whether it's practically feasible is a question for Istvan.

ADD REPLYlink written 4 months ago by WouterDeCoster21k
1

I think they should if certain terms and conditions are mentioned as per blog standards. This way we can avoid cross-posting and even help newbies to hone their mining skills and get idea of how to put a better question.

ADD REPLYlink written 4 months ago by vchris_ngs4.0k
3

Does anyone actually care about cross-posting?

I mean I guess it's annoying to read the same question in two places, but often when a cross-post does happen here and other places (generally Seq.Ans.), the cross-posted post doesn't have any good answers. Or worse, it's closed due to cross-posting!

So yeah it's a bummer but i'm not totally convinced closing x-posts is the right answer. I feel like it's a race between the other site and Biostars to close it, and Biostars often wins as there are many active people 24/7. Under this design, the more active community doesn't answer the question :P

Suspending the account for a day means they can't thank the people who do answer. Same goes for a user ban. I don't know. Maybe people should get awards for not cross-posting in a year, or something... an incentive over a punishment.

EDIT: Or hey, what about a "you're a sh**ty person" award :P

ADD REPLYlink modified 4 months ago • written 4 months ago by John11k
2

Electrons are rare and should be conserved where possible.

In seriousness, though, when I see complaints about double-posting, usually nine times out of ten it really ends up being rules-lawyering of the sort that makes places like Wikipedia and SO such tedious communities.

ADD REPLYlink modified 4 months ago • written 4 months ago by Alex Reynolds20k
3

I just wish people would at least cross-link their cross-posts so that readers know that there may be related posts elsewhere. Not doing so is in my opinion frankly disrespectful of other people's time.

ADD REPLYlink written 4 months ago by Lars Juhl Jensen11k

If you cross-link, then people from both communities will assume that the other community will provide the answer. At least that's a common assumption (would be nice to test how valid this claim is).

Also, you can argue cross-linking is disrespectful of other people's time as well. You expect everyone who reads your post to check all the other sites where you posted it.

ADD REPLYlink written 4 months ago by igor4.5k
2

I participate on SeqAnswers and here and generally try to cross-link posts.

Bigger problem is with the original posters not indicating a successful resolution (in all cross-posts) when an acceptable solution is posted on one of the posts.

But then we have a problem of getting original posters here to "accept" valid answers for a significant percentage of threads.

ADD REPLYlink written 4 months ago by genomax33k
1

The problem that I see with cross-posting is that you waste people's time. Since we already have a problem that too many questions go unanswered, having people waste time on answering a question that has already, unknown to them, been answered by others elsewhere is problematic.

ADD REPLYlink written 4 months ago by Lars Juhl Jensen11k
9
gravatar for Alex Reynolds
5 months ago by
Alex Reynolds20k
Seattle, WA USA
Alex Reynolds20k wrote:

One thing I like about BioStars is that the cool kids can't gang up and vote down people they don't like. Unlike official SE sites, say, with all the cool kids who get to vote down people whose views they don't like. Which really sucks. I'm glad we don't have that bullshit here. (I mean, we have a cool-kid problem, unfortunately, but at least the damage the cool kids do is limited to them just opening their big mouths.)

ADD COMMENTlink modified 5 months ago • written 5 months ago by Alex Reynolds20k
1

:D

ADD REPLYlink written 5 months ago by John11k
1

glad someone wrote this. This was always a problem with SE

ADD REPLYlink written 4 months ago by vchris_ngs4.0k
5

And if you look at the few contributions and suggestions there you can already see a closed post: on the account of it not being "constructive" and a moderator lecturing someone else as to what should go into a comment.

This is why I never feel the joy in contributing to any SE site. It is always about policing, rules penalties, getting points etc.

I see this as a testament to the success of Biostars - StackExchange has become envious of us - they want to heavily commercialize and push into the field of Bioinformatics.

ADD REPLYlink written 4 months ago by Istvan Albert ♦♦ 73k
4

they want to heavily commercialize and push into the field of Bioinformatics.

That’s nonsense. They didn’t raise a finger to aid the Bioinformatics proposal — which failed twice due to lack of participation. It was all done by outside people.

ADD REPLYlink written 4 months ago by Konrad70
2

StackExchange has become envious of us - they want to heavily commercialize and push into the field of Bioinformatics

Is there evidence for this?

ADD REPLYlink written 4 months ago by A. Domingues1.4k
3

Is there evidence for this?

Perhaps I am speculating on the motives - but it is a commercial enterprise that has to turn a profit some way.

In addition here are some thoughts:

We do live in an era where vast numbers of people vote against seemingly their own best interest. We all saw this last year. You are left scratching your head - why would anyone think that the solution is to blow things up, divide, split off and start from zero? Draw a parallel on what is happening in politics and you'll see how the phenomenon is real and is rooted in the same type of rationalization: blaming someone or something else.

This vote and the fact that so many are supporting it shows how scientists are not immune to being short sighted and making the wrong choice for what they think are the right reasons. Having a slightly different interface will not be a driver of high quality content, just as the ability to vote people down will not be a solution to any of the problems you percieve.

The real challenge is that bioinformatics questions are inherently difficult to answer - far more difficult than a classic programming or biology question. It is a field that integrates many sciences and requires interacting with people that may not be well trained in these subjects. Hence it needs a different approach, a different way to talk about these problems. That's why we allow more content in comments, and why the comments are threaded etc. Assuming that adopting a simplistic SE model will make the content better is an illusion.

The site works very well, it has experienced a steady growth over what is now its eight year - anyone stating and rationalizing that somehow the site does not serve its purpose is mistaken. It is simply that questions are becoming more difficult to answer because, unlike programming where languages don't quite change much our understanding of biology gets more complext and the questions that are posed become more difficult as well.

ADD REPLYlink modified 4 months ago • written 4 months ago by Istvan Albert ♦♦ 73k
8

it is a commercial enterprise that has to turn a profit some way

It allows users to create communities if they meet certain requirements. It's actually a very democratic and transparent process (see Area51 FAQs: http://area51.stackexchange.com/faq ). SE is not out to destroy Biostars. That's like saying Facebook is destroying Biostars because someone created a bioinformatics group.

Having a slightly different interface will not be a driver of high quality content

People said that about almost every popular website. And now they are popular. SE wasn't the first place to discuss programming questions, but it offered a better interface. SE built their whole business around a better interface.

The site works very well

There are multiple posts just in this thread alone where people say otherwise. I don't know if saying everything is fine will convince anyone that it is.

ADD REPLYlink modified 4 months ago • written 4 months ago by igor4.5k
2

In all fairness, it is always very easy to find something wrong with any existing system. There are also multiple posts in this thread where people say that SE is bad. I don't have a strong preference for one "chair" over the other, but I'm pretty sure that sitting on either is much preferable over trying to sit on both at the same time.

ADD REPLYlink written 4 months ago by Lars Juhl Jensen11k
1

There is nothing wrong with having many chairs and admitting that they all could be better and working to improve them.

This is the alternative that seems to be fairly popular here (there is even a chair for the chair analogy): enter image description here

ADD REPLYlink written 4 months ago by igor4.5k
4

I think the two of us will just have to agree to fundamentally disagree on this topic :-/ (BTW, you already posted this image once in this discussion ...)

ADD REPLYlink modified 4 months ago • written 4 months ago by Lars Juhl Jensen11k
1

It might just be that some people prefer one chair, other people the other chair, to follow in the analogy here. If both are happy with their chair, stay seated and don't complain about the other chair.

Honestly, I don't think this thread is going anywhere now. It's good that some issues about Biostars have been highlighted (mainly: answers end up in comments and a more accurate search function), and we'll try to solve those.

And if people want to create another 10 more communities, that's fine. Knock yourself out. I don't see the point but we can't stop anyone, can we?

ADD REPLYlink written 4 months ago by WouterDeCoster21k
5

I hope this post doesn't come across too aggressive. As stated before I like and use both SE (mostly SO) and Biostars, and I am participating in the discussion because I feel that as good as Biostars is, it could be even better with a few minor changes.

Perhaps I am speculating on the motives - but it is a commercial enterprise that has to turn a profit some way.

I know that Biostars will not make anyone rich, as opposed to SE, but Biostars is also a commercial enterprise to some extent:

The Biostars name and the website at https://www.biostars.org is the property of Biostar Genomics LLC a private bioinformatics consulting company registered in the state of Pennsylvania, USA. The company founder is Dr. Istvan Albert the lead developer of the the open source Biostar Q&A software.

I am not in least bothered by this btw, that is, people making a decent profit for a useful service, just wanted to point it out.

Draw a parallel on what is happening in politics and you'll see how the phenomenon is real and is rooted in the same type of rationalization: blaming someone or something else.

I do however take an issue with comparing the creation of a website, redundant as it may be, with recent political events (I presume you mean Brexit/Trump) which will have (and already had) an impact on millions of people's lifes and whose effects are not yet all visible, and will probably be felt for generations. A little perspective please.

The real challenge is that bioinformatics questions are inherently difficult to answer - far more difficult than a classic programming or biology question.

And that is fine. Maybe and SE will not be useful, and maybe it will complement Biostars. Maybe Biostars will change and become better.

ADD REPLYlink written 4 months ago by A. Domingues1.4k
1

The site works very well, it has experienced a steady growth

How you measure if the site works well? If I recall the purpose of the site is to learn to be a good bioinformatician and it is difficult to measure that By the number of open questions ~ 12k and latest questions ~47k (this counts the news, forum, tutorial sections) around 1/4 of questions are left unanswered.

Which has been that growth? Visits and questions ? Could we see some graphics about it? Like questions, answers, marked answers ratio, number of new users, and comments per year? Given the increase in the publications/people using bioinformatics is Biostars growing accordingly?

Many thanks (If I should make a new post with this questions let me known)

ADD REPLYlink written 4 months ago by Lluís R.550
3

The steady growth of visitors year over year is clear enough indication - there were 1890 people here in the last hour. This year we already have 200K more sessions than last year. You can see who is here and of what is going on at this time by clicking on the globe at the bottom of the page.

Then counting open questions is not the right measure - you should count the questions with no comments on it (I'll look into this to see if we can easily get this number). I am certain that there are a lot fewer of those.

Commenting is the way the community engages the poster, perhaps the question is ill-formed or and sometimes the question is answered in a comment. Perhaps the poster never answered the comment hence demonstrated that they are not coming back.

But I want to note how counterintuitive is what you are saying - do you think that you would get more answers if there were yet another place to ask questions?

ADD REPLYlink written 4 months ago by Istvan Albert ♦♦ 73k
4

I am not saying that there would be more answers if there is another place to ask questions neither that Biostars isn't helping newcomers to bioinformatics. I was asking about how to you measure the performance of the site because I think how much or how many people learn from the site is hard to measure. If in my blog has 100k visits and no comments, is it a good blog that help people learn?

Having another place to ask questions I think is independent of how many answers or comments will be posted here. But we (you/mods) may be able to measure the impact of the new site if the new Bioinformatics in StackExchange starts. Have a nice day!

ADD REPLYlink written 4 months ago by Lluís R.550
2

Counting the number of visits alone is questionable, too. If you stopped accepting questions and answers now, the number of visits would not change too much in a year or so. I speculate most visits are contributed by good contents in the past. Actually, any single measurement is inevitably biased. SE have several ways to measure the activity of its sites. Here is the stat page of the biology SE when it was at the beta stage. You can also see the stat of all SE sites. The combination of these metrics looks quite informative to me.

ADD REPLYlink written 4 months ago by lh330k
3

This discussion is similar to those series of posts where Joachim was determined to "objectively" show that Biostar is failing:

Is The Biostar Fading? (Updated For 2014)

Since the first post was written the site gained more than 20x as many users and page views - not a single prediction turned out to be valid. Ironically the blog is not even accessible anymore, they were not able to keep it up these years ... whereas Biostar keeps on churning.

You want to make the site better - write an answer or a comment. Boom! You have helped someone today, and tomorrow and the world is a better place. The positive energy, creation, making things is what makes us all happy.

ADD REPLYlink written 4 months ago by Istvan Albert ♦♦ 73k
1

sometimes the question is answered in a comment

It would be great to give mods the ability of moving posts (to selected locations) so answers that are sometimes embedded in discussions could be moved or at least marked as "answers" (if they can't be logically moved). Same could also be extended to some comments that can be marked as supplemental questions (since many users never edit the original question to add additional information).

ADD REPLYlink written 4 months ago by genomax33k
3

I need to point out that comments are intrinsically different from answers. A proper answer should provide enough background and explanation to the answer. It educates the questioner, such that he/she can can understand how the answer works. A comment can be as simple as "yes, it works". The user solves the question at hand but will ask again when something is slightly different.

ADD REPLYlink written 4 months ago by lh330k

I think if you toggle an answer to "accepted" you can move it to the top, but I honestly haven't done it often.

ADD REPLYlink written 4 months ago by Devon Ryan70k
1

Having the ability to move certain comments into precise places in thread hierarchy would be great to have for mods. Currently the move can only be done to the top of the thread/original post which is not ideal most of the times.

ADD REPLYlink written 4 months ago by genomax33k
2

You can also move comments to an answer. Also, I am using the toggle accepted more often now on older question that get bounced automatically.

ADD REPLYlink written 4 months ago by Michael Dondrup43k
1

Absolutely. But I am referring to the old threads where people add comments as "new answers" and the only place we can move them to now is the original question. When there are pre-existing answers this is less than ideal.

ADD REPLYlink written 4 months ago by genomax33k
1

Yes, that would be neat, but if I had to choose, I'd prefer a state-of-the-art search engine. A lot of duplicates occur because it is impossible to find any old post using the site search.

ADD REPLYlink written 4 months ago by Michael Dondrup43k

There's always Google. Simply typing "biostars" and whatever you are looking for works most of the time. It's kind of like how people shopping on Amazon don't bother with Amazon's search engine, most of the time.

ADD REPLYlink written 4 months ago by Alex Reynolds20k

The problem is that google does not understand the 'votes' and 'answers' concepts. I would like the search to be sortable by most voted, answered, etc.That is the whole point of votes to me. I guess this must be possible to implement with tools like elasticsearch.

ADD REPLYlink written 4 months ago by Michael Dondrup43k
1

Answers generally have the keywords I'm looking for, so that invariably pushes up search results. In my own experience, I guess, I find votes have become overrated as a measure of quality on community sites, when people game the system or manipulate vote order, based on community criteria unrelated to actually having given a good answer. In any case, it seems like there should be a way to work towards improving search, without reinventing new sites with problematic mechanics and social dynamics.

ADD REPLYlink written 4 months ago by Alex Reynolds20k

I think that's something quite easy we can do to get more threads answered.

ADD REPLYlink written 4 months ago by WouterDeCoster21k
1

I agree to certain extent but not sure to what extent they can commercialize it.

ADD REPLYlink written 4 months ago by vchris_ngs4.0k
9
gravatar for Damian Kao
5 months ago by
Damian Kao14k
UK
Damian Kao14k wrote:

I think the main problem people have with BioStar boils down to an identity crisis. Is BioStar a Q and A site or a forum?

A Q and A type site requires the question to be technical and precise. If someone asks how to convert a format to another format, that can be easily answered.

But if someone asks what is the best method of assembling this genome given X, Y and Z, then there really are no clear answers that can be given. Cases like that requires more of a discussion which then becomes a long chain of comments.

An argument can be made that these types of more abstract and less technical questions do not belong on BioStar. However, I've personally benefitted hugely from the forum-type discussions that occurs within these questions. I think there are a lot of people out there that do not have the computational support or community in their local area to discuss these topics. And BioStars has become the bioinformatics community for them.

On the other hand, I can also see how these long messy comment chains ruins the organization of the site with redundant questions and vague answers.

Maybe the solution is to actually implement a real forum interface along side the Q and A interface. However, I can see that as being a lot of work to put on Istvan. Another solution is to just migrate to SE for Q/A and use SeqAnswers for forum type discussions.

ADD COMMENTlink written 5 months ago by Damian Kao14k
6

Yes, the most useful questions on SE tend to be off-topic and under threat by the rigid Asperger people who run those sites.

ADD REPLYlink written 5 months ago by Jeremy Leipzig17k
5

You are used to the forum-like biostar – you want to discuss a lot in comments. In a proper Q&A format, if you think you know better on this genome assembly problem, provide another answer. Explain clearly why it is better than other alternatives and let users play the judge. When someone reach the question via a google search, they read a few answers and make a choice based on the votes and the arguments you present in answer. They can also unfold comments if they have enough background to make their own judgement. I agree forum-like discussions are needed sometimes (e.g. this thread), but a lot of comments in biostar are unnecessary and distractive.

ADD REPLYlink written 5 months ago by lh330k
1

I agree with you to a degree about the Q&A format. I think a nicely structured searchable knowledge base would be very helpful and maybe SE can become that.

But I also think that the nature of the bioinformatics field probably means there are enough questions of the abstract type which warrants discussion and long messy comment chains. I have no data to conclude that though.

Developing a community where these types of questions can be actively discussed without being taken over by dominant personalities is hard and rare. And I don't think that's something SE can provide.

I do agree with many other people that perhaps the Q&A format doesn't make sense anymore. It has become a half-measure where we have some legitimate Q&A functionality, but also a lot of discussion. This results in a weak Q&A in practice and maybe also a weak discussion format as people might be hesitant to post given the explicit Q&A formatting of the site.

I am not against SE and I don't foresee a fragmentation of the community. I think SE will maybe become the top google search result when people search for precise technical questions. But BioStars will remain with warts and all.

ADD REPLYlink written 4 months ago by Damian Kao14k

Perhaps there is a happy medium? Maybe certain stellar threads could be archived in to an FAQ style separate region of the domain, and rather than rehashing the same straighforward questions, people can be directed there. Though I suppose this is sort of what the Handbook is.

For sure, I like the forum-esque nature of Biostars compared to SE, but it would be a shame to lose the Q/A which also more often than not works nicely.

ADD REPLYlink written 5 months ago by jrj.healey2.3k

Actually I don't read SeqAnswers because I miss threads there. That's the only reason. I have spent last 5 years on Biostars though.

ADD REPLYlink written 4 months ago by Biomonika (Noolean)3.0k
8
gravatar for Charles Plessy
5 months ago by
Charles Plessy2.2k
Japan
Charles Plessy2.2k wrote:

Biostars's engine is is Free software while StackExchange's engine is proprietary. Given the importance of opening source code in bioinformatics and science in general, I hope that the community will appreciate and continue to post on Biostars, that gives the example to follow.

ADD COMMENTlink written 5 months ago by Charles Plessy2.2k
6

Not sure the open source argument works. If someone takes an open source project and fixes a bug or forks the code to make an improved tool, the world is slightly better and everyone is happier. If someone makes a new Biostars, it's a big mess (see this post).

Biostars is not a software. It's a community. If anything, the software is probably the weakest part of Biostars (as highlighted by many comments in this post) and the biggest draw of StackExchange.

ADD REPLYlink modified 5 months ago • written 5 months ago by igor4.5k
3
gravatar for Lluís R.
5 months ago by
Lluís R.550
Spain, Barcelona
Lluís R.550 wrote:

When Bioconductor opened its own support.bioconductor forking Biostars I don't recall any post about diluting the user base.

That said, the reason of Biostars being separated from StackExchange was the license. And StackExchange has made efforts to provide guidance about the license of the code in their site. Which I resume as: if you copy textually from anywhere, copy also the link to the place otherwise do as you like. If pieces of your software end up having mixed license, so it is. But you can overpass the requirements and SE will not persecute you.

However in the same answer from Ivstar show that the goal of Biostars is to be "a destination where one could also learn how to be a good bioinformatician". So to this regards a Q&A site in Stack Exhange will not overlap with Biostars. Biostars is for learning to be a good bioinformatician. Bioinformatics.SE would be for learning how to do bioinformatics.

I usually go to Biostars, suppport.bioconductor and Stack Exchange sites, as long as I learn how to do what I want/need I will keep going wherever I find the answer.


I like much more the possibility to downvotes of StackExchange. The point of StackExchange is not to rule all the Q&A sites, but to provide a quality content of questions and answers. (Like The Biostar Handbook aims to provide "the most accurate and relevant information" in bioinformatics). My opinion is pretty much in the same lines as this old answer after moving out from SE1.
Here I can post a question related, without any effort and if people is willing I get free answers, which I am not sure it is on the lines of learning to be a good bioinformatician if I don't put effort solving my problems before asking... Surely the expectation in StackExchange (and support.bioconductor to some extent) is to show effort before posting the question but that makes better bioinformaticians and better professionals.

ADD COMMENTlink written 5 months ago by Lluís R.550
5

Seems to be an assumption by some that mechanisms to help people improve their questions equates with being hostile and condescending, and upsets or drives away users. In my experience, the more common result is that people do indeed edit and improve their questions.

ADD REPLYlink written 5 months ago by Neilfws47k
2

It is much more different, bioconductor support is mainly to help with the questions related to bioconductor packages and R programming in bioinformatics, while Biostars is a general bioinformatics plateform. A new SE site similar to biostars will not bring any thing new.

ADD REPLYlink written 5 months ago by Sirus770
1

Bioconductor converted its long-running mailing list to a biostars site and then closed the list. The same goes for the Galaxy biostar site.

I personally view the possibility for down votes as a strong negative. It's better to ignore crap questions than to give the submitter the perceived feedback of "your question is crap, go away."

ADD REPLYlink modified 5 months ago • written 5 months ago by Devon Ryan70k
6

Downvotes on SE also reduce the reputation points of the voter, so they can't exactly be handed out like candy.

ADD REPLYlink written 5 months ago by A. Domingues1.4k
1

Not on questions. Which … I’m not sure I like. But the fact is that the reputation penalty for downvoting questions was abolished because of an influx of many extremely low-quality questions, and downvoting questions (and commenting) actually makes people improve their questions, whereas just commenting without downvoting is less likely to achieve the same.

ADD REPLYlink written 4 months ago by Konrad70

That's a good point at least.

ADD REPLYlink written 5 months ago by Devon Ryan70k
4

I am a long time supporter of downvote and played an important role in bringing back post closing (though biostar still lacks the vote-to-close feature). Many early biostar users also favored downvote. When used properly, these harsh mechanisms help to maintain the quality of a Q&A site. The old biostar had downvote. It worked well. I did remember a few discussions about how and when to cast downvote and to close posts, but these were all minor. Generally, it is easy to find a few isolated cases where downvote hurts hard, but when you sum up the tiny benefit of downvote in most other cases, it is huge.

ADD REPLYlink modified 5 months ago • written 5 months ago by lh330k
4

Worth pointing out that voting to close does not necessarily upset or discourage users. They are often happy to discover that an answer exists already.

ADD REPLYlink modified 4 months ago • written 4 months ago by Neilfws47k
1

The down votes never say "go away" to the user but to the question. But it might discourage a bit posting. That's why I think Facebook dislike wasn't introduced until much recently.

ADD REPLYlink written 5 months ago by Lluís R.550
1

It may not be intended to mean that, but regardless that's the perception.

ADD REPLYlink written 5 months ago by Devon Ryan70k
1

It will definitely discourage posting. I have a lab member who tried to do some of his own analysis, asked a question on SE, got like 3 comments and never posted again. Just gave me the data to analyse.

ADD REPLYlink modified 5 months ago • written 5 months ago by Sinji2.5k
2
gravatar for WouterDeCoster
5 months ago by
Belgium
WouterDeCoster21k wrote:

So one of the "issues" it seems with biostars is that not always the (best) answer is posted as an answer, but rather hidden in comments. That's something we can easily improve on. Whenever a thread gets answered, we can either move the comment to an answer and toggle accepted status (mods) or ask users to summarise and repost their solution as an answer, and have that post be the accepted answer.

ADD COMMENTlink written 5 months ago by WouterDeCoster21k

That may not always be easy. One of the things that has been on my wish list (expressed before) is the ability for mods to mark (deeply embedded) comments as supplemental question(s)/answer(s). If a comment can't be moved (or deleted since that makes all children of that comment invisible) it can still be given some prominence or crosslinked in the original post automatically.

ask users to summarise and repost their solution as an answer, and have that post be the accepted answer

We know that is pretty unlikely. A high % of users do not bother to "accept" the answers (let us say unknowingly). Only way that may happen is if mods were to do it (after a waiting period of a day or so) as we come across these threads that seem to have acceptable answers.

ADD REPLYlink modified 5 months ago • written 5 months ago by genomax33k

If it's clear that this solution helped OP I don't see a problem with mods 'toggling accepted status' for answers. Waiting a few days is fair, but not necessary.

After all, the aim of this is that someone with a similar issue easily finds the solution in the thread and doesn't have to go in those deeply embedded comments...

ADD REPLYlink written 5 months ago by WouterDeCoster21k

I was not referring to toggling accept switch but rather the suggestion that

to summarise and repost their solution as an answer

if it was deeply embeded in a chat with credit being given to the person answering the question.

ADD REPLYlink modified 5 months ago • written 5 months ago by genomax33k

Oh right. Yes, that would indeed depend on the user. We can try :-)

ADD REPLYlink written 5 months ago by WouterDeCoster21k
1

Perhaps something that might help a few more threads get marked as answered would be for an OP to get a message in their inbox after a week (?) of post inactivity prompting them to revisit and accept an answer if possible? If it only makes 5% of people actually do so, that's probably still a large number of threads.

ADD REPLYlink written 5 months ago by jrj.healey2.3k

That sounds worth trying, also no real disadvantages to this.

ADD REPLYlink modified 5 months ago • written 5 months ago by WouterDeCoster21k
1
gravatar for Pierre Lindenbaum
4 months ago by
France/Nantes/Institut du Thorax - INSERM UMR1087
Pierre Lindenbaum98k wrote:

update:

This proposal is in: Commitment

A successful site needs people to commit to use it. This proposal is 100% complete.

Committed users will be invited to the private beta soon.

ADD COMMENTlink written 4 months ago by Pierre Lindenbaum98k
5

Time to close this thread then?

ADD REPLYlink written 4 months ago by genomax33k
Please log in to add an answer.
The thread is closed. No new answers may be added.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 892 users visited in the last hour