19
$\begingroup$

I asked this question more than two years ago:

What are the chances that we make it out of Beta?

One of my favorite "quotes" from that question:

It is mid-September and things don't seem to have improved much. Perhaps it is time to take some action. – Daniel Standage Sep 19 '12 at 17:34

I'm asking this question because of my fear (or paranoia) of this community eventually becoming stagnant or deleted. I love Biology.SE, and want to see this site graduate very badly, because I feel as though as a whole it deserves to graduate.

Thus, I was hoping that we as a community could critically evaluate our site, identify our strengths are weakness, and give every member of Biology.SE some goals and a direction.

What do you think are some of the problems with Bio.SE? What are we doing well? What area should be actively focusing on? What would make this community better for you AND for Stack Exchange?

$\endgroup$
  • 14
    $\begingroup$ It's never bad to discuss possibilities for improvement, but I just want to provide an official note here: this community is not in danger of being shut down. Even if sites never "graduate", there is no time limit on their being in beta. As long as it continues to show growth and quality, Biology SE is a fully fledged member of the Stack Exchange network in no danger of going away. $\endgroup$ – hairboat Oct 8 '14 at 21:42
  • $\begingroup$ That's an interesting question when you take into account that the chemistry site is now graduating. The size of the site is approximately the same - although the accepted answer quote is a bit higher. This comes from a lot of answered question which (in my view) have proper answers which never get accepted. $\endgroup$ – Chris Oct 9 '14 at 7:34
  • $\begingroup$ @Chris I dont think Qs need to be accepted to count towards %answered - I think upvoted answers is enough. $\endgroup$ – fileunderwater Oct 9 '14 at 11:17
  • $\begingroup$ @fileunderwater You are right, I remembered this wrong. It doesn't look nice, too. Too much "post and run" questions are also not good. $\endgroup$ – Chris Oct 9 '14 at 11:56
  • $\begingroup$ If someone answers my questions biology.stackexchange.com/q/19887/8198 and biology.stackexchange.com/q/23080/8198, we will be a step farther :) $\endgroup$ – Honza Zidek Oct 16 '14 at 18:50
  • 2
    $\begingroup$ My two cents worth: All questions I've asked on biology.se have been left unanswered (0.00%); whilst the answer rate on all other stacks (where I've asked more than one question) has been at minimum 50% and usually 90%+. I don't know the reason for the exceptionally low answer rate on biology.se but I don't see it as being a lack of quality questions in general but rather a disinclination to answer every question regardless of difficulty or interest. After all, a stack grows from new members that see a site as responsive. $\endgroup$ – LateralFractal Oct 20 '14 at 12:38
  • 3
    $\begingroup$ @LateralFractal.. You should understand that most the questions you asked here are a little difficult to answer. They are very specific. And you have not asked many questions here (small sample bias :P ). But thanks for pointing out. Regarding "disinclination to answer every question regardless of difficulty or interest": A well posed question is always interesting. Niche questions are a tricky affair. Some of them can be answered based on general principles but if it goes too specific then one would have to put considerable effort to do the background reading. $\endgroup$ – WYSIWYG Oct 21 '14 at 9:36
  • $\begingroup$ @LateralFractal. Your questions are also very speculative e.g. this and this. It is not just to expect a 100% answer rate on questions which do not have a strong foundation. I am not saying the questions are bad per-se. They can be posed in a better way; simplified to fundamentals. $\endgroup$ – WYSIWYG Oct 21 '14 at 9:46
  • $\begingroup$ @WYSIWYG Shrug. As you please. It's no concern of mine; as this isn't an important stack site for me. Or others apparently. $\endgroup$ – LateralFractal Oct 21 '14 at 10:51
  • 5
    $\begingroup$ I wanted to mention that as a researcher, I found this site to be immensely informative both in practical research matters as well as in a food-for-thought way. Some questions have attracted insightful responses and others alerted me to the existence of important gaps in my knowledge and motivated me to fill them. It's one of the most valuable resources this side of actual peer-reviewed journals (sometimes more useful than even those). So, leaving aside how it's doing objectively according to the SE metrics, subjectively I think this site has been a considerable success. $\endgroup$ – Superbest Oct 24 '14 at 5:31
  • $\begingroup$ possible duplicate of When will Biology SE become a "full' SE site? $\endgroup$ – dustin Mar 29 '15 at 15:05
12
$\begingroup$

I don't know if this is an issue for the graduation process but, to me, the main issue for BioSE is the problem to attract questions from active researchers, and difficult well-researched questions from professionals. The balance between general curiosity questions and research questions is still very skewed towards the former. This is not to say that more general biology questions from people with a relatively weaker background in biology are uninteresting. However, when I look at StackOverflow (the obvious model) a good chunk of the questions come from working professionals, and are non-trivial for somebody working in a particular field (in their case, language). My perception is that we get far fewer of these, and this might lead to a negative feedback loop where we will have problems to attract more specialists (this has been covered in several BioSE Meta questions, e.g. here and here). The main issue is that first impressions by new users will be molded by the active/newest questions queue. If this is filled with basic questions users will get the impression that this is the wrong place for harder questions, which then leads to even fewer of these, etc. And to attract research questions is not only valuable in itself - just as important is that this will lead to a more knowledgeable user base, which would inspire better answers to all the other questions.

I'm guilty of not asking many questions myself, and so far I've only asked one single question. This is not only due to the issue I describe above, but since the ecology/conservation biology user base here is weak I would not be very hopeful to get answers to questions I encounter in my work. As much as I hate to say so, these would be more likely to get answers at ResearchGate, which has lots of active researchers answering questions (but on a vastly inferior platform).

So, what is the solution? I really do not know. Promotion and more contributions from already active users will certainly help. I do not think this will be enought though. What we lack is a critical mass of a certain type of users, even if this has improved during the last year.

As for the positives, I do think that we are getting better and better at answering a wider variety of questions, and this is mosty due to the growth of the user base. I also feel that we have gotten better at closing problematic/unproductive questions (my perception at least), but this is also by necessity since the influx of questions has increased.

$\endgroup$
  • 2
    $\begingroup$ The problem with Researchgate is true - there are more active researchers present on this site since it is primarily directed to these people. The question is if we really get a lot more of difficult questions from "professionals" since one of the things you learn during scientific studies to acquire knowledge yourself. That is what I do a lot, going into the literature with a specific question in mind. And that's also what I like about bio.sx - I can find out interesting stuff on related topics where I don't know very much about. $\endgroup$ – Chris Oct 9 '14 at 7:49
  • 3
    $\begingroup$ Yes I strongly agree with your point on the negative feedbacks. I was fed up because of it once (highly upvoted trivial questions). I am not sure if this is correct or not but askfedora site which is modeled on SE doesn't grant voting privileges to new (low rep) users. I didn't like it about askfedora because it is a practical application oriented/troubleshooting site and any newcomer should be able to appreciate a useful solution. We, on the other hand are not providing solutions and the right to judge if a question/answer is worthy or not should rest with those who are aware of the field $\endgroup$ – WYSIWYG Oct 9 '14 at 11:11
  • 1
    $\begingroup$ @Chris A also think BioSE is great for "reeducating" yourself on topics that you haven't thought about in a while, by answering Qs. You are probably correct that researchers are much more likely to go and look for answers themselves (maybe with tips from collegues at their department), which is why they don't show up here. And the thing with ResearchGate (and our "problem") is that they attract Q/A users as a byproduct of offering another service (personal page with an overview of your research outputs). $\endgroup$ – fileunderwater Oct 13 '14 at 11:06
  • 1
    $\begingroup$ When I first started using the site, I actually refrained from asking technical questions because I assumed this was for conceptual/theoretical discussion, and not nitty-gritty trivial details of my boring day to day lab experiments. Only after multiple comments on other SE sites (including Are 51 where I tried to start a lab-focused SE), did I come to realize that it's actually good to ask technical questions here. Perhaps we need to convey to our professional users that they are encouraged to ask technical questions (make it clearer in the tour/help/faq?). $\endgroup$ – Superbest Oct 24 '14 at 5:36
8
$\begingroup$

I'll address the Area 51 metrics first, the thresholds published on Area 51 are rather misleading.

We currently are at 14 questions per day, with the official threshold for excellent being 15 questions per day. This is one statistic whith a huge variation between sites, some sites like Skeptics have graduated with a question volume far below the threshold. Other sites can easily hit the threshold at the start of the public beta. Our current volume is easily enough to sustain a healthy site, this is nothing to worry about. More good questions is always nice, but simply more questions wouldn't help us graduate any sooner.

Our answer rate is at 87%, the official threshold for excellent is 90%. This is again a statistic that depends heavily on the site subject. I don't think this is a problem currently, but this is an aspect we can certainly improve on. Biology is a rather broad field, the more difficult or specialized questions can often only be answered by users in the same specialized field. Getting this rate way up is unrealistic.

We have 35 users with 3k+ reputation and 45 users with 2k+ reputation. Those are important threshold because these users will be able to close and edit after graduation. This is a sufficient number to have effective community moderation even at the higher privilege thresholds of a graduated site.

I'm not sure what to make of the answer ratio, I don't see a problem for this site here. I can see how extreme values would be problematic, but I think we're at a reasonable point.

I've put some images of our traffic in comparison to Chemistry, which is a site of a similar age and topic that is about to graduate:

enter image description here

They have a bit more traffic than us recently, but overall we're pretty similar in traffic growth. I think traffic looks pretty healthy and shouldn't be a barrier to graduation.

Herre is also a comparison with Physics, an example for a graduates site with a similar topic:

enter image description here

From the number alone, I'd say that we're ready to graduate. But the numbers alone don't decide this.

A severe quality problem would also prevent graduation, but I don't think we have that here. We could certainly do better, but I think overall quality is good on the site. I'm not entirely happy with the ratio of popular biology questions to more advanced questions, but that is not a barrier to graduation.

One aspect that I think could improve is the community engagement on meta. Our meta was pretty silent most of the time, this has improved significantly recently.

My guess would be that we're ready for graduation, but that either our evalution wasn't scheduled yet or the design bottleneck is the limiting factor. But this is entirely my own speculation, I have no inside information here. It is almost impossible to get a straight answer out of an SE employee on when a site graduates, so I don't actually know what SE thinks in this case.

$\endgroup$
  • $\begingroup$ From what you write, our numbers are basically the same as chemistry (only slightly lower). Is there some qualitative difference that you can point to, that is not captured in these statistics, which can explain why they are about to graduate and BioSE is not? $\endgroup$ – fileunderwater Nov 21 '14 at 19:13
  • $\begingroup$ @fileunderwater Sites aren't evaluated at the same time, from what I could gather we simply weren't evaluated for graduation since Chemistry graduated. $\endgroup$ – Mad Scientist Nov 21 '14 at 19:16
6
$\begingroup$

My three cents: I agree with fileunderwater's answer. I am active on Gardening & Landscaping beta, which is actually 189 days older than Biology. It is a much slower site question-wise, with all-around lower stats (See the GL's Area51 stats, compared with Bio's), but what I've found is that when you want something to happen, you have to take initiative and do it yourself.

Lead by Example

What I recommend doing in this case, is: if you have complaints with the question quality (I do), ask your own, as examples. Even if you know the answers, this really helps set the theme for what good questions look like. If you let new users ask all the questions, then you will be able to predict exactly what you have now: Lots of easy, low research/effort questions flooding the site, and acting as examples for more new users to see.

On GL, I'm not sure there's anything I can ask that I couldn't answer, but that didn't stop me from asking nearly twice as many questions as the next highest asker. Do what you want others to do, to the best of your ability. I was kinda active on GL when it started, and went until the end of June '12. I came back on May 16, '14, and have been on since. I just got a graph of GL site activity, and you will be amazed at how much of a jump 1 user can make, if he/she tries it. Look at the jump in may '14, when I rejoined, highlighted in the second pic:

I think that on a bigger site, such as biology, it may take more than one person (aka teamwork), but you can get a much higher quality activity if you just put it in yourself. Good luck, folks!

$\endgroup$
6
$\begingroup$

Things that need to improve according to the Bio SE stats in order of urgency (in my opinion)...

Multiple answers is a good thing, not a bad one

Often I find questions with an accepted and up-voted answer, I used to not bother adding another answer unless I felt I could add a lot more to the matter, partly because I didn't think it would be too well received by the first answerer. Now though if I've got something small that I can add to an answer, or I feel I can offer a different interpretation or perspective on the problem, I will add an additional answer.

We currently have 1.5 answers per question and it needs to go to 2.5

We need more questions

More questions = more potential for answers = more information on the site = more users = more potential for more questions and new answers!

The statistic only needs to improve a little so it won't take much. I occasionally chip in a question I know the answer too (or know how to get to the answer) just to get more questions on to the site. If just all of the top 40 users (those with rep >~2500) contributed an additional question per month (on top of what they already do) we'd have enough questions being asked. Furthermore, once there is more questions on the site it is more likely people will find the site, and then be more likely to use it in the future with their own questions.

It's ok to not know

Often questions go unanswered because they are too specific or niche so there is very little research out there. We need to try and make "best effort" type answers, these would be more speculative and attempt to join the dots more than most "good quality" answers would. I think answers should generally good quality, but if the answer to the question is "we just don't know yet" then we can't change that. In that case we should be more tolerant of speculative answers which attempt to elucidate a likely result from other similar problems.

$\endgroup$
  • $\begingroup$ on a positive side the number of users and daily visits has really increased lately $\endgroup$ – rg255 Oct 22 '14 at 10:30
  • 3
    $\begingroup$ We shouldn't optimize for those stats, the Area 51 thresholds are very misleading. These remaining "okay" values are not a problem, question and answer quality is pretty much the only thing that matters once we have a sufficient size (which we have IMHO). $\endgroup$ – Mad Scientist Oct 22 '14 at 14:23
  • $\begingroup$ So how do we graduate? If they don't matter then there is no clear target @MadScientist $\endgroup$ – rg255 Oct 23 '14 at 7:48
  • 2
    $\begingroup$ I've posted my own answer that also addresses the statistics in more detail. $\endgroup$ – Mad Scientist Oct 23 '14 at 9:40
  • $\begingroup$ I don't think we have much of a with the quality of the answers. The idea of asking questions where you know the answer to raise the quality is interesting. I have to think about this, since this feels quite awkward. A bit like fishing for compliments. $\endgroup$ – Chris Oct 23 '14 at 10:38

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .