12
$\begingroup$

When you look at the statistics in Area 51 it seems as though this site cannot survive. There are other examples of sites that were closed after beta (example).

What do you think? Are there any hard criteria for closing?

And in that unfortunate case: Are there possibilities of continuing the site elsewhere?

$\endgroup$
11
$\begingroup$

Excellent question! A few stats:

(bear in mind it's only been 70 of 90 days, and we often let betas go on indefinitely if they are producing great content. Quality is, as they say, Job One.)

Also refer to Robert's post here:

https://writers.meta.stackexchange.com/questions/221/what-happens-now

But if that's TL;DR, in general help us grow your site!

  1. Share great questions and answers
    http://blog.stackoverflow.com/2010/09/announcer-booster-and-publicist-badges/

  2. Vote, vote, vote
    http://blog.stackoverflow.com/2010/10/vote-early-vote-often/
    https://quant.stackexchange.com/users?tab=voters

  3. Love and reward your new users for being awesome!
    https://quant.stackexchange.com/review
    https://quant.stackexchange.com/users

  4. Try to attract experts to the site by helping them get answers, too:
    http://blog.stackoverflow.com/2011/04/helping-the-experts-get-answers/

$\endgroup$
  • 1
    $\begingroup$ @Jeff: I didn't know that the beta phase could be extended - that is reassuring! Thank you! $\endgroup$ – vonjd Apr 12 '11 at 5:55
  • 1
    $\begingroup$ @everyone Some of our top contributors here are on the committee at the R/Finance conference coming up. We should give them all the support they need to answer questions that may crop-up during the event. $\endgroup$ – chrisaycock Apr 13 '11 at 18:55
  • $\begingroup$ @chris what can we do to support the conference? see blog.stackoverflow.com/2011/02/… and of course.. swag. $\endgroup$ – Jeff Atwood Apr 13 '11 at 19:54
  • $\begingroup$ @Jeff I've posted that as a question. $\endgroup$ – chrisaycock Apr 13 '11 at 20:10
  • 2
    $\begingroup$ @Jeff "2. Vote, vote, vote" - I think I've just upvoted all valid on topic questions so far. Wouldn't mind some help here, guys. And we still need more questions for Electorate badges to come into play. ;-) $\endgroup$ – Karol J. Piczak Apr 13 '11 at 21:21
  • $\begingroup$ @Jeff, as we're 3 days from the end of the official beta period, I think it would be reassuring to know that the site is in fact going to follow in an indefinite beta. This was assumed, but never said out loud. $\endgroup$ – Karol J. Piczak Apr 29 '11 at 9:42
  • 1
    $\begingroup$ @karol yes, betas go on indefinitely so long as the sites are producing fundamentally high quality questions and answers. $\endgroup$ – Jeff Atwood Apr 29 '11 at 9:51
10
$\begingroup$

There are no HARD criteria for closing. If you're expecting someone to say "Your time is up; You have failed!", that's not how it works. The "90 day beta" threshold is just a minimum period a site has to endure to be considered for graduation.

When Will My Site Graduate? "It takes as long as it takes."

Quantitative Finance is well within the criteria for keeping it alive and in extended beta until it reaches critical mass.

Really, what keeps a smaller site going well past that 90-day mark is that it continues to provide a high-quality experience for the users. Users should have a high expectation that they will get an expert answer to their question; That's the '% answered' statistic below. 93% of the questions get answered. "Excellent"; That's what we like to see.

enter image description here

As far as the "Worrying" statistics above, that's not really all that unexpected. Quantitative Finance is a subject that runs a bit far afield of our typical Stack Exchange site. It will just take a bit more time and effort to build an audience from the wider reaches of the Internet. But means keeping the content quality high — that's job-one right now; much more important than asking lots of questions.

We find that the best way to attract new users to the site is by highlighting your most intriguing questions. People love answering questions, and a particularly interesting and intriguing question is irresistible. It's the quickest, most definitive way to help someone realize "Yeah, this site is for me!"

Use those social bookmarks!

Social Bookmarks

So how do we know that Quant SE isn't failing?

By looking at the traffic through the site:

enter image description here

The traffic of Quant SE is characteristically flat. That's actually not all that unusual for a Stack Exchange site at this stage. It means that you are "holding your own" and not succumbing to inevitable decline.

Most Stack Exchange sites typically experience this horizontal pattern of flat growth going sort of horizontally for awhile until they reach a specific "tipping point" where the traffic begins to slope upward and takes off unstoppably. That's when the site has reached critical mass. Quant SE is not at that point, yet, but when you do reach critical mass, all those "Worrying" statistics will take care of themselves.

$\endgroup$
  • 1
    $\begingroup$ Thanks for the reassuring notes, Robert. I like what Jeff said in the latest podcast, that the most important criteria is whether experts enjoy answering the questions. (His point was that if experts think the questions are stupid, then they'll leave and only non-experts will be left, which means that questions won't get good answers anymore.) $\endgroup$ – chrisaycock Apr 29 '11 at 18:09
  • $\begingroup$ Good to hear that coming from the official side. $\endgroup$ – Karol J. Piczak Apr 29 '11 at 20:37
  • $\begingroup$ Is there any way we could see occasional updates of the traffic graph for this site? $\endgroup$ – Tal Fishman Sep 12 '11 at 20:12
  • $\begingroup$ @Tal Fishman: There's really no way to do that at this time, but we're always looking for ways to provide a general health barometer for the sites. Suffice it to say, traffic experienced the generally flat slump followed more recently by an end-of-summer pickup which is pretty typical for most Stack Exchange sites (and the Internet, in general). $\endgroup$ – Robert Cartaino Sep 13 '11 at 15:17
7
$\begingroup$

I've often worried about seeing the site shutdown. And I can't imagine running it without the StackExchange software; it would be very difficult to rebuild the voting, the inline TeX, the moderator tools, etc. I've been able to learn a few things here as well as identify some talented people in our industry. I for one want to see this site continue.

With that said, I know some other sites were in beta for a long time before reaching production. So there still might be hope in the sense that it's too early to tell, especially given the niche-ness of what we do.

$\endgroup$
  • $\begingroup$ I thought the software was open source. I think I remember seeing other sites running it that were not from Area 51...?!? $\endgroup$ – vonjd Apr 12 '11 at 5:52
  • $\begingroup$ @vonjd The software was at one point licensed to interested parties in a hosting arrangement. A while back, the company realized that this wasn't making any money, so they ceased granting new licenses, but some of the original licensees still exist (Math Overflow, for example). Now days, the only way get access to the SE software is through Area 51. $\endgroup$ – chrisaycock Apr 12 '11 at 6:43
4
$\begingroup$

The normal beta period is not a strict deadline. There are sites which are in between (not ready for start and not that bad to be dumped) and they go into indefinite beta after 90 days (e.g. IT Security).

I have an impression that the visits/day statistic is important in such a situation to sustain the efforts (like in the presented case of Atheism). For proper maturation though, an established community is needed (with enough people/reputation to self-moderate the site with normal non-beta reputation requirements). For sure, in it's current state, quant.SE won't be able to graduate anytime soon. The important question is if that's gonna be enough to maintain the site whatsoever.

I hope so. So far this site has been really helpful with prompt and thorough responses on interesting topics. For sure it's a niche subject, and the entry level of discussion isn't low either, so it's no surprise that some time is required to gain adequate momentum. But my impression is that the stats get a bit better after a difficult first month. It's a slow but steady progress, and if presented in such a manner, I suppose there's a chance to incubate this site a little longer.

$\endgroup$
4
$\begingroup$

Those interested in tracking the general health of the site can check out http://www.quantcast.com/quant.stackexchange.com/traffic traffic graph

Looks like we're doing OK and showing decent growth!

Some interesting tidbits from quantcast: Compared to the average SE, we are

  • more male
  • older
  • richer
  • fewer kids
  • more educated
  • more Asian
  • growing about 40% faster!
$\endgroup$
2
$\begingroup$

From my point of view (a professional credit&IR derivatives quant), this site is too heavily slanted towards algo- and high-frequency trading. Questions about HFT get much broad and detailed answers than the questions I asked about IR or credit modelling.

$\endgroup$
  • 1
    $\begingroup$ I guess it's because people most likely to be on the beta of a website like this one will be closer to algo-trading than purely with pricing for example. One of my questions was also never answered completely. $\endgroup$ – SRKX Apr 26 '11 at 15:40
  • $\begingroup$ I've wondered about this. I doubt the reason is that we're in beta; there just aren't many well-established sites that cover valuations. I believe it's that algo-trading simply has more easy-to-answer questions. There's a straightforward solution to VWAP issues or data-feed handling (because knowledge here is experience-based and is broadly applicable) vs something that requires a formal proof. $\endgroup$ – chrisaycock Apr 26 '11 at 17:09
  • 1
    $\begingroup$ We're all specialists of one sort or another. And, the specialist questions asked here take too long to answer, unless you're currently working on a related issue. It's common to be able to answer half of the question in a reasonable period of time, but a half-fast answer isn't good enough. $\endgroup$ – bill_080 Apr 26 '11 at 17:58
  • $\begingroup$ @bill_080 That sort blows the whole idea of this site out of the water... $\endgroup$ – quant_dev Apr 26 '11 at 19:15
  • $\begingroup$ @quant, That's why I suggested earlier that we broaden the scope of the site. I don't think it's realistic to assume that more than a handful of people will spend more than 30 minutes answering a question. Time isn't free. If it looks too time consuming to me, I move on. I assume that most people do the same. That's also why I don't ask "thick" questions. $\endgroup$ – bill_080 Apr 26 '11 at 20:01
  • 2
    $\begingroup$ @bill_080 The previous proposal to expand the site was shot down because general finance / economics is very different from quantitative finance / econometrics. I spend as much time as I need to answer a question I feel I'm qualified to address. Options pricing isn't my background, so I just stick to stat arb and HFT. Presumably there are people on this site who can answer non-trading questions, since there are enough people on this very thread commiserating over that very issue. $\endgroup$ – chrisaycock Apr 27 '11 at 2:31
  • 1
    $\begingroup$ @bill_080 "I don't think it's realistic to assume that more than a handful of people will spend more than 30 minutes answering a question." -- In 30 minutes you can write a long exposition on a difficult topic, if you know it well. "That's also why I don't ask "thick" questions." - We were encouraged to ask such questions. $\endgroup$ – quant_dev Apr 27 '11 at 5:42
1
$\begingroup$

One of the keys seems to be the lack of newbie quant questions, as that was what doomed the artificial intelligence stack exchange site (which had excessive newbie questions and very little to draw experts).

$\endgroup$
  • $\begingroup$ lack of newbie quant questions... Did you mean the excessive amount of newbie questions? The most common question set is on how to enter the industry or a survey about what tools people use. I close far more questions than I answer. $\endgroup$ – chrisaycock Apr 13 '11 at 13:36
  • 1
    $\begingroup$ If I understand @Tangurena right, he meant that the general level of questions is also an important benchmark tool. AI.SE was doomed by lack of real expert questions from what I remember from the blog. So keep up the good work @chris. $\endgroup$ – Karol J. Piczak Apr 13 '11 at 19:18
  • $\begingroup$ @Karol is correct. $\endgroup$ – Tangurena Apr 13 '11 at 20:05
1
$\begingroup$

It might be useful to broaden the scope of the site. For instance, to allow economic questions to be asked. It's impossible to be a quant and not deal with supply/demand, interest rates, threat of recession, rise/fall of economic activity, inflation, etc.

I realize that this might attract homework questions, but that could also be useful (keeps us all on our toes).

From what I understand, there's an economics group starting. However, they'll probably have as much or more trouble than we've had here. If the two were combined, there may be enough traffic to keep everyone happy.

$\endgroup$
  • 3
    $\begingroup$ I'd love to see econometric questions; they'd definitely be appreciated around here. And a lot of PhD-level economics research is in that area anyway. What would definitely be out-of-scope is generic questions about supply and demand. The main thing is that we are dedicated to professionals. The general rule of thumb is that if a non-expert can answer the question, then it's out of scope. $\endgroup$ – chrisaycock Apr 13 '11 at 18:53
  • 3
    $\begingroup$ Econometrics by all means yes. $\endgroup$ – Karol J. Piczak Apr 13 '11 at 19:43
  • $\begingroup$ As to general broadening - I got the impression that the consensus here was that quant finance is a separate domain (bit.ly/hG6LMQ). Based on this I even proposed a corporate finance/IB site (bit.ly/i0WCIy), which I forgot to mention in the original thread. I know it's dividing even further (with the risk of not finding enough interest), but I assumed that an all-in-one economics site (macro/micro, quant, personal, corporate) is not the best idea, at least on SE (TBH, that's exactly what I'm doing in Polish, but it's another story - the prospective community is just too small). $\endgroup$ – Karol J. Piczak Apr 13 '11 at 19:49

You must log in to answer this question.

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