We in the Selenium community were trying hard to get Selenium a dedicated site at Area 51.

Of late there was discussion about merging the Selenium proposal with a software testing proposal. There was great opposition from the Selenium community to merging with the more general software testing proposal.

We got notified today that the Selenium proposal has already been merged with the software testing one. Also that the Selenium proposal has been closed, which is very very sad.

We want to know -

  1. When there was great discontent about the merge, why did the merge still happen?
  2. How could we reopen the Selenium proposal?
  3. If SE sites are all about openness then why was such an autocratic decision made?

Remember we already have Google-Groups for Selenium users which receives the highest amount of traffic for Selenium qs. The idea was to close Google-Groups once we had a dedicated Selenium area51 site.

When we don’t have a dedicated Selenium site, Google-Groups would also continue and there would be fewer questions on Selenium asked on the new merged site, or on any other site for that matter.

Refer to this discussion about the merge on Google Groups which shows how helpless we feel.

  • 2
    +1 (because I can't vote up with 15 reputation here and vote down with 51 reputation on the merge proposition) – Jérémy Hérault Apr 27 '11 at 8:57

Remember we already have Google-Groups for Selenium users which receives highest amount of traffic for Selenium qs. Idea was to close Google-Groups once we have dedicated Selenium area51 site

Yeah... I think I know what's bothering you, and I'm afraid it comes down to a misconception about what Stack Exchange can be used for.

See, we don't really like to do product-support sites. That was an idea early on - prior to and during the SE 1.0 days - that ended up just not working out well. As the platform has evolved, it's become clear that running an effective site requires some careful management: the Q&A format works very well for some topics, and extremely poorly for others.

An example of something that doesn't work well is long-format discussion: we kinda allow this on the Meta sites, but it's - at best - a poor substitute for a proper forum. We actively discourage such questions on the main sites.

So setting up a site for a particular product, and then letting users of that product talk about... whatever... is a dangerous suggestion. We could easily end up with a really crappy forum / bugs database / social network instead of an awesome Q&A repository. Instead, the idea is to put together a community of expert users in a particular field and let them answer questions on topics germane to that field. Could browser automation be considered a "field"? Could Selenium alone be? Maybe...

But it doesn't appear that this was actually anyone's intention. Reading the posts in your Google Groups, the prevailing idea seemed to be that you were trying to do just what you stated above: shut down the groups and move everything over to SE. That's a bad idea for both of us!

  • We can't import existing posts, or existing users. You just can't convert arbitrary threads into structured Q&A and expect it to work.

  • We can't operate like a mailing list - so mailing-list users like Bob McConnell would be unhappy.

  • We don't really have support for lengthy back-and-forth discussions - so it's unclear that this would even work for replacing the developer mailing list. (FWIW, it's telling that while the Area51 proposal did include "contributors" in the target audience, all the example questions are on usage.)

In short - for developer discussions, announcements, and other non-Q&A topics, you're probably best off staying in Google Groups rather than trying to hammer your square peg into the round hole that is SE. For questions on working with Selenium, we very much hope you'll find the proposed Testing and QA site a welcoming environment.

I'm sorry if you feel ignored or otherwise offended by this decision. We're trying to create the best possible site for SQA engineer and testers, a place for sharing knowledge, and that includes knowledge of web automation software - just as it does desktop automation software. Yes, you'll be a small fish in a big pond relative to both the proposed site and your current situation... But that's kinda the point: you'll have room to grow, exposure to people who aren't currently Selenium users.

  • 2
    I hate to throw a wrench in your huge post here, but... Drupal Answers and SharePoint are both one-product (well, more like one product series) sites. Perhaps, though, they'll be a good way for us to properly test the waters. Rather than launch a dozen product-based sites at once and try to monitor them all. – Grace Note Apr 27 '11 at 17:08
  • 7
    @Grace: don't forget Wordpress, AskUbuntu, TeX... Yes, this has been argued several times already. So I wrote, "Could browser automation be considered a "field"? Could Selenium alone be? Maybe... But it doesn't appear that this was actually anyone's intention" <-- now I might be completely off-base on this, but... I don't think so. I followed the discussion on A51 and Google Groups, and repeatedly argument against was simply, "We don't want to see questions that aren't Selenium questions". – Shog9 Apr 27 '11 at 17:28
  • Heh, point taken. I could argue that AU and TeX are a bit different, but that'd be pointless. – Grace Note Apr 27 '11 at 17:30
  • SO what would it be, tagged selenium qs on StackOverflow or Selenium questions on new merged site. Or both, if both the does it make sense? – Tarun Apr 27 '11 at 18:00
  • 1
    @Tarun: I would strongly encourage them on the new site, but they'll still be on-topic for SO in the foreseeable future, to the extent that they're valid programming questions. Currently, they're very much niche questions on SO - I'd hope they could become much more mainstream on T & SQA. – Shog9 Apr 27 '11 at 18:04
  • And how would you encourage them? Is there a mechanism to migrate SO questions to new site? – Tarun Apr 28 '11 at 3:27
  • 1
    @Tarun: Comments work pretty well... "Hey, you should check out TQA.SE! We have some crack Selenium experts over there who'd love to answer a question like this..." But yes, there's a (moderator-only) mechanism for migrating questions once the new site gets past the private beta period. – Shog9 Apr 28 '11 at 3:44
  1. You've already asked this same question here and here. It makes life a lot easier for mods if we can answer in just one place versus three.

  2. It's not like this was a surprise. Robert opened Expand Selenium to cover the field of Software Testing and QA? for discussion on 6 April.

  3. You mentioned that same Google group half a dozen times on Robert's discussion—and again, it's not really relevant. If the people in that group want to make their opinions known on Area51 proposals, they have to do it on Area51. There's no way of knowing which of them will/won't ever use an SE site. If they can't be bothered to help out to make a site happen, then their opinions simply don't matter.

  4. If the Google group members are happy being on Google groups—great! And whether they are or aren't, they're welcome to follow the selenium tag on sqa.stackexchange.com.

To answer your particular questions:

  1. When there was great discontent about the merge, why merge still happened?
    It was clear that you were not happy with the idea. You not being happy ≠ "great discontent."

    And not that it's a vote per se, but Robert's suggestion (merge the "Selenium" proposal with the less tool-specific scope of "Software Quality Assurance) received 20 upvotes to 6 downvotes. I find it difficult to see 14 votes net for the merger as "great opposition" to it.

  2. How could we reopen Selenium proposal?
    You already asked that question on Discuss.Area51, and that's probably the best place to keep it.

  3. SE sites are all about openness then how such autocratic decision was taken?
    You not being happy about the result is not the same thing as "autocratic."

    You got three weeks to make your case, and you tried. At the end, it didn't appear that continuing to postpone the decision would make any difference.

  • 4
    @Tarun - If three weeks wasn't enough for you, how long would have been? – Dori Apr 27 '11 at 5:28
  • 9
    I'm blown away that you are saying his Google Group link is irrelevant. Can you be anymore condescending? Do you really think because they currently use that site (knowing that there is not site here for them, and now never will be) that they're inferior somehow? – Lance Roberts Apr 27 '11 at 6:35
  • 4
    With all due respect your reasoning is deeply flawed. I would have downvoted the proposal but as I don't have 125 rep on Area 51 I am unable to do so (I have over 200 rep on stack overflow and skeptics so it's not like i've never used any of the sites before). About 350 of the users who have committed to the Selenium site have a rep below 125 so are unable to make thier views known, it doesn't matter if you give us 3 weeks or 3 months if we can't vote we can't vote! Secondly how can 20 upvotes verses 6 downvotes be a meaningful representation of approx 800 comitters across both proposals? – Ardesco Apr 27 '11 at 7:45
  • 2
    @Lance, if they have something to say in the discussion about what we do here, you'd expect them to do it here. You don't expect the team to listen to discussion on Hacker News or Twitter do you? You claim there is no site for them, there will be, it just won't be uniquely for them. As for users with insufficient rep to vote, you could have still left comments, answers or other regular methods of showing discontent. Furthermore, if you fail to get any rep, how would I know to trust you with making any decisions. Obviously you only care for your own site, not the network. – Ivo Flipse Apr 27 '11 at 7:58
  • 2
    @Ivo - Your remark "Obviously you only care for your own site, not the network" is very derogatory and unwarranted for. I have posted links of this thread on existing groups and I hope more Selenium Users would chime in soon. As for the n/w I am at my wits end why SE can not understand the role of distinguished Selenium proposal than a general purpose s/w testing site!!! I suppose you would suggest SQA forums also to merge all their testing forums to one group. Very sad. – Tarun Apr 27 '11 at 8:15
  • 2
    @Ivo secondly many people who committed to the Selenium site probably do only care about getting a selenium resource to begin with, once given that resource they will become exposed to the rest of the stack exchange network and realise how useful it can be and start to contribute to the network as a whole, surely this is the whole point? It may scare you to know that Stack Exchange is not the centre of the universe and not everybody is aware of the network, new sites that have a lot of interest are a good way to bring people into the fold. – Ardesco Apr 27 '11 at 9:06
  • 2
    @Ardesco, the point is: those that site have to be about Selenium or Software Q&A in general. My guess is that the latter will have more chance of actually attracting sufficient experts to answer all the questions. If everybody sees a ghost town Selenium proposal, they'll probably turn back around and go back to Google Groups. I've seen enough 'popular' proposals launch, only to end up with 5 questions/day. That's not a success in my book. – Ivo Flipse Apr 27 '11 at 9:11
  • 4
    A proposal with 423 committers is hardly a ghost town proposal and at 92% it was virtually there. Isn't the point of a beta phase a way to ensure that if it does turn out to be a ghost town it will not see the light of day. I have no issues with a general QA site, but isn't this one: testing.stackexchange.com? Is wanting a site for a specific framework really such a bad idea? If it is why do we have Drupal and Wordpress sites? – Ardesco Apr 27 '11 at 9:19
  • 3
    @Ivo so if people are not interested in what Stack Exchange currently offers you don't want them here anyway? Surely opening new avenues to people who don't currently use Stack Exchange by offering them something they want and can't get somewhere else is the best way to expand the community! (How did the gaming site come about with this attitude, that's nothing to do with coding so why would you want gamers here?) If we have some people who are only interested in answering Selenium questions is it really a problem as it will add to Stack Exchange as a whole anyway? – Ardesco Apr 27 '11 at 10:08
  • 2
    @Ardesco, that's a poor analogy, it would have meant we would have a Starcraft 2 site rather than a general Gaming site, because the Starcraft 2 people don't want to hang out with the rest. I have a serious problem with that attitude and think it shouldn't be allowed. As for users not want to be convinced, we can't convince every one even if it is better for them, that's the way the world works. But we shouldn't give in just because they stubbornly refuse to participate in anything greater than just Selenium. – Ivo Flipse Apr 27 '11 at 11:11
  • 3
    @Ardesco Absolutely, but my point is that proposal shouldn't have to be a Selenium proposal but a Software Testing & Q&A site. Outsides look at us as being yet another forum with a ton of noise, while that's simply not true. So they think they need their own site, but they don't. Either way, we're going to have to agree to disagree. – Ivo Flipse Apr 27 '11 at 11:23
  • 4
    @Ivo The problem seems to be that you don't personally think that selenium is a big enough niche area to break it out from testing as a whole, whereas the Selenium community thinks that it can prove that it is big enough to to be broken out from the testing as a whole, just like the Drupal community is. The sad thing is that the only way we would be able to prove it is by getting our commitment and going through the closed beta phase which it doesn't seem we will be able to do. – Ardesco Apr 27 '11 at 11:26
  • 6
    I haven't followed the discussion, but looking at the question started by Robert, I see vastly more opinions against a merge with Software Q & A than against. I don't understand why this merge had to happen. If there's a community that believes it can run a SE site, why not let them try, or at least merge into a "browser testing" proposal? Wordpress.SE seems to be doing pretty well. @Dori I'm not sure how credible those 14 upvotes really are in terms of measuring actual support. I bet 95% of the people who voted on this, are not going to participate in either of the sites in question. – Pekka Apr 27 '11 at 12:42
  • 7
    Area 51 needs a proper voting process for merges, or notification of all committed users if a merge is in the works that might affect their site. I'm thinking that if all committers to Selenium had known about this discussion, there would have been vastly more participation... Also, the Selenium proposal has 423 committers; Software Q&A has 544. While @Ivo makes good points - if the opposition to a merge is only about "we don't want to mingle with those Shelbyvillians", it's invalid - I don't think this should have been merged against the will of those participating. – Pekka Apr 27 '11 at 12:55
  • 3
    Also the process for merging is clearly flawed. As I understand it now the decision is based on some merge proposal which people can vote up but most can't vote down because of rep restrictions. How is this supposed to work? Should we have created a proposal to NOT merge and let selenium users vote that up? – Ivo Grootjes Apr 27 '11 at 16:13

I'm one of the people who's uncommitted to the Software Quality Assurance and Testing proposal after being forceably committed to it as a result of the merger of the proposals. I believe Selenium is broad enough to have a Q&A site of it's own, but if it's not, then I'm quite sure "browser automation tools" (including Watir, WatiN, HtmlUnit, and Selenium in the open-source space alone) is large enough to support it's own Q&A site. It is not necessarily true that "browser automation tools" == "software testing"; this is the point many in this discussion are missing, and was made far more eloquently than I ever could in this blog post. I've spent several days trying to calm down enough about the situation to provide a measured response to the (seemingly unilateral) merger. I've gotten past my anger about it; now I'm just sad. So much for, "We don't run [StackExchange], the community does."

  • 3
    The biggest point in that blog post being made is Selenium is NOT a testing tool. You might want to make that more prominent, as I think it's an excellent argument for not being merged with a "general testing tool q&a"... – fretje Apr 27 '11 at 14:00
  • 3
    @fretje: The roots go even deeper - google, "web application testing system" - the first result is selenium. Follow that link, and you'll find yourself on a site named "seleniumhq.org", where this tag line appears right at the top. It's clear to me that this misconception goes far, far beyond SE - perhaps this can serve as a wake-up call for the Selenium community: like it or not, their system has been co-opted by folks who somehow believe that it should be used for testing web apps. – Shog9 Apr 27 '11 at 14:37

Disclaimer: Just so you know, before reading question post, I didn't know what Selenium was or that it even existed.

When I first read your question, my thoughts were exactly like Ivo's (these guys want their own precious little site and don't want to share it with others). But after some more reading (the discussion on area51, the google groups, the link provided by JimEvans, ...) I think you guys actually have a point here, but it just is scattered around on all these places, and sometimes buried under the noise.

You might want to make a case here, clearly formulate your arguments against a merge on a central place (be it here or on the area51 discussion zone), so people don't have to dig for them.

For instance the argument brought up in the blog post JimEvans links to sais "Selenium is NOT a testing tool". Which I think is the best argument not to merge it, but I didn't see that argument anywhere else.

  • 1
    Maybe the community is just too small to have a site of its own, at least for now. – alex Apr 27 '11 at 14:48
  • 1
    While true @fretje, who says the scope has to be so narrow to only be testing. It could be wide enough so that it fits nicely around Selenium and similar products, but mainly be focused on Testing and Q&A. Furthermore, I think a lot of people are thinking they can use Selenium as a bug tracking system, which shouldn't be the goal of any SE-site – Ivo Flipse Apr 27 '11 at 14:48
  • 1
    @Ivo: I understand what you mean, but still, I think they deserve a chance to at least start a private beta (especially because they were already so close to it before the merge), and then we can evaluate again... if it turns out there are not enough questions, or most of the questions are only bug reports (although I don't think that should be an argument for closing a private beta), the proposal can still be closed (or merged with another). – fretje Apr 27 '11 at 14:55
  • Also: regarding "who says the scope has to be so narrow to only be testing"... I think the Selenium guys are even more narrow than "only testing", as they claim themselves "Selenium is not a test tool, it is a browser automation tool". I think "browser automation" is just (very) a small part of "testing" in general... – fretje Apr 27 '11 at 14:59
  • 2
    There was never any intention to use SE as a bug tracking system, that would be insane... The reason the community wanted a SE site was to deal with the questions we get. the vast majority of traffic through the selenium users list is questions on how to do things with selenium, have a look at groups.google.com/group/selenium-users and you will see what I mean. We have problems with people providing bad advice becuase they don't know any better and a SE site is the perfect way to vote down bad advice, vote up good advice and become the resource for anybody interested in selenium. – Ardesco Apr 27 '11 at 21:27
  • 1
    @fretje problem is, the rest of the world doesn't agree with that interpretation, so it is rather looking at the world through rose colored glasses. Try some Google searches and see what you find.. for example if you Google, "web application testing system" - the first result is selenium. – Jeff Atwood Apr 28 '11 at 0:31
  • @Jeff I disagree with your assessment. Try googling "browser automation framework" and you will see Selenium coming in at third place. Would you take the tool autoit (autoitscript.com/site/autoit) and bundle that in with a testing site? – Ardesco Apr 28 '11 at 9:35
  • Now I guess it is just not possible to reopen the proposal. is it? – Tarun Apr 29 '11 at 3:00

You must log in to answer this question.

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