What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 131 Stack Exchange communities.

Edit: I no longer think this is a good idea. The comments and answers have convinced me. I'm going to leave this question here, because it will hopefully be useful someday.

I was looking through the list of StackExchange websites, and began to notice a strange phenomenon.

Over time, the SE network has accrued more and more language & usage sites. It started a while ago, but recently has gained steam. Except, in my opinion, it's getting a little ridiculous. This has even been noted as a problem since the beginning of 2011.

We currently have:

I propose a merge of these seven L&U sites into one Languages and Usage site. This may initially seem like a bad idea, but it may actually be beneficial. There are a few things to bear in mind:

  1. These sites all offer questions regarding the same context - language and usage. It's written into the title, literally. These sites are all intrinsically similar.
  2. Many (almost all of) the highly used tags are the same on each site. Each language has its specific tags, many of them are the same, e.g. reading, expression, etymology, mathematics, grammar, nuances, pronunciation, punctuation, etc. and then some. This is a strong indicator that these sites cover basically the same topics.
  3. While each of these sites may fail in its individual beta, a merged site would most definitely succeed, and, based off the involvement in each site, would be profoundly successful.
  4. Consolidation of services is often a good thing - in this case, there is no need to segregate these questions by site.

We have a unique tagging system on the various Meta sites around SE. Here on Meta, we have four "root tags" (for lack of a better term), one of which must be used for a post. This makes it easy for us to classify questions by type.

A similar system could be implemented on Languages.SE. One must use a given language tag for a question, e.g. the root tag for a question concerning the Japanese language is japanese. For German, Russian, etc. it would be german, russian, etc. This allows the users to easily find and look through a specific language, while keeping all questions about language together.

But a merge like this will be messy, won't it?

Not necessarily. There are a few steps to properly introduce this, and none of them are individually that complex.

  1. Each question on the language sites must be retagged with its appropriate root tag; e.g. put the japanese tag on each question on Japanese.SE, german on each German.SE, etc.
  2. Set up the tagging system to require one specific language tag.
  3. Set a redirect from [languagename].stackexchange.com to languages.stackexchange.com. If a person is redirected from a specific website, display a banner stating that all languages have moved.
  4. Be sure to alert users long before the change.

Aren't these sites still in beta? Why does it matter?

Like I said before, these websites are all very much open. I am confident that each will qualify for a full site, out of public beta. But, to be safe, I will cover both cases:

  1. All language sites succeed in the beta: If this is the case, then it would be laughable to make one site for each of these languages. It would be much cleaner and simpler to have one site for all of them.
  2. One or three of the websites fails the beta: If this is the case, then the reason and nature of failure need to be considered. Primarily, we need to consider whether or not these sites would have succeeded if they were merged in one languages website. If they would have passed, then it makes logical sense to add them to a full languages website.

What constitutes a good question for Languages.SE?

This obviously becomes more ambiguous once the websites have been merged. However, the FAQ of each of the language betas can be merged and resolved. This is actually a pretty easy task, since most of the FAQ from each of these websites is canned text. For instance, the entire block of What questions should I not ask here? is (quite nearly) the same across all of the language websites.

This is all I can think of at the moment. If you have further questions about how I personally see this working, or would otherwise like to contribute to this post, or just flat out disagree with me, please comment and let me know!

share|improve this question
    
@simchona Their FAQs are very similar. Check 'What kind of questions can I ask here?' Russian is more vague, since it's not fully defined, but the gist is the same. (However, I would not be opposed to leaving ELU out of it, because the userbase of the LU sites is predominantly English-speaking. Don't hold me to that statement, though. That's an empirical observation.) –  Emrakul Apr 4 '13 at 3:39
1  
I think it would be worth getting the actual pro tem mods in here. I don't know the others well enough to say, but unlike the ability to sort of understand a different programming language (and have them on one site), natural languages require a lot different of a user base. I could see grouping them into classes like Romantic Languages, but I also don't know if that would prevent one part of the user base from ever contributing outside their bubble and feeling entirely comfortable. –  simchona Apr 4 '13 at 3:42
    
@schimona The difference here is that we would have root tags. Each language would be grouped by its own tag type, and thus, no user actually has to involve themselves with languages outside of their tags. –  Emrakul Apr 4 '13 at 3:45
9  
Wouldn't it be nice if we could just merge all the languages in the real world, too? –  Cody Gray Apr 4 '13 at 3:51
    
@tal but someone can pick up the other root tags and expand participation. Learning human languages doesn't work that way. –  simchona Apr 4 '13 at 3:52
    
@Simchona I'm not totally sure what you mean? Yes, people could venture into other language tags. And, the goal of the languages website is not to teach a langauge, it's to clarify specific questions about a language. People learn as a result of those clarifications –  Emrakul Apr 4 '13 at 3:54
    
Do you actually use any of these sites? As a user of these sites do you see any utility to merging them? –  Mitch Apr 4 '13 at 14:37

2 Answers 2

up vote 13 down vote accepted

I cannot imagine how this could ever be useful.

An expert on Russian is almost certainly not going to be an expert in German. So there will be very little of the cross-talk we get on SO between users of different programming languages. Also, a Russian expert will generally have no need to learn German, while programmers often have need to pick up other languages.

The difference here is that we would have root tags. Each language would be grouped by its own tag type, and thus, no user actually has to involve themselves with languages outside of their tags.

So what's the point of merging them at all? If nobody ever does anything outside of their tag of expertise... why bother having them on the same site to begin with? Some programmers have only one specialty, but even they will have need of other areas in their careers (unless they are very sheltered).

Besides the notion of "get a site", what good does this accomplish? Even if you could convince experts from various languages to congregate into one site, you don't gain anything by doing this.

Furthermore, unless you take real steps to make them true subsites, you still know that the other questions exist. Searches would have to be tag-specific, lest they find other irrelevant content. Also, those other questions are still there on the front page. Speaking of which, if nobody would find the site's front page useful... something's terribly wrong with the site. At least in the case of Gaming.SE, people who play popular games when they first come out find it useful.

share|improve this answer
5  
This is spot on. Also, there's a side-effect of lumping subjects with distinct audiences into one site - new users are reluctant to join up because they don't feel a sense of community or belonging. This happened with poker, chess and the larger Board & Card Games site, for example. Poker SE and Chess SE are small, but they both have proven that they have a dedicated core audience that doesn't really overlap with other games even though they also involve cards or boards. Not every site will grow to be the size of Stack Overflow, and that's okay so long as they provide real value. –  Anna Lear Apr 4 '13 at 13:20

Throwing several sites together that are too small to achieve critical mass does not work if the sites have only minimal overlap. You're not creating a critical mass of experts if each expert on the site is only interested in a very small part of the site, and there is no overlap between those areas. By throwing those sites together you're just hiding the fact that you're lacking critical mass.

Each language site has to succeed on their own merit, throwing them together is helping nobody.

share|improve this answer

You must log in to answer this question.

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