Conlangs and ancient languages are usually treated similarly. The issues which are related to them are, also, our relation to non-written languages, as well as non-active Wikipedias (note that I am not talking about other projects; treat the word "project" as a synonym for the word "Wikipedia"). All of them don't have a clear future at Wikimedia.
I would like to reformulate those issues in relation to our priorities. The main goal of WMF and Wikimedia community is to spread free knowledge. According to that, we need to make our priorities and to work according to them. It is, also, important to treat this issue without personal (or whichever) POV, but as more neutral as it is possible. We should, also, treat those issues not only synchronically, but with a clear vision of some very predictable parts of our future.
So, I'll write about our priorities as I see them according to "some very predictable parts of our future" as I see them.
Before I start, I want to say my POV about all of the issues: (1) I don't think that conlangs except Esperanto and a couple of specific conlangs more are too useful. Besides that, I really don't like wannabe-world languages based on a couple of Indo-European languages, including Esperanto. (2) Artistic conlangs are, at my opinion, even lower. (3) I am not interested in developing neo-classical languages. (4) In this moment non-written languages are not a Wikimedia issue; some other institutions should take care about such languages before they become our issue. (5) I already said that if for some project may be reasonably said that it is not active ("reasonable" is a criteria about we may talk...) -- then it should be locked, but unlocking should be allowed if a new speaker of that language want to take care about that project.
But, let's see what do we have:
1. (Projects in) natural and living languages: 1.1. The biggest encyclopedia in the history of humans: English Wikipedia. 1.2. Very soon, the second biggest encyclopedia in the history of humans: German Wikipedias. 1.3. Well developed projects which are at a good path to become the biggest encyclopedias in the history of humans, too. Generally, those are projects which have more than 50,000 articles or which will have that number relatively soon. 1.4. Emerging projects: active projects with, let's say at least 5000 articles and living communities. 1.5. Projects which started to exist: projects with around 1000 articles at least and a a couple of active contributors. 1.6. Not active projects which may become active: with less than around 1000 articles and a couple of not so active contributors. 1.7. Not active projects: with less than around 1000 and without active contributors. 1.8. Hundreds of living written languages which don't have a Wikipedia. 1.9. Thousands of living non-written languages which don't have a Wikipedia.
2. (Projects in) conlangs: 2.1. Two useful projects: Esperanto (the only relevant conglang community) and Volapuk (similarity with English and a lot of data added by one person). 2.2. (Do we have any other non-artistic conlang?) 2.3. A number of potentially useful conlangs which don't have a Wikipedia because of various out-of-Wikimedia reasons, usually copyright reasons. (Slovio is an example of such language; it may be read by any educated person which native language is one of the Slavic languages.) 2.4. All other non-artistic conlangs which wouldn't get a project because of the policies. 2.5. All artistic conlangs which wouldn't get a project because of the policies.
3. (Projects in) ancient/dead languages: 3.1. Actually, some of them are not dead (Latin, even a Church Slavonic, but the later one doesn't have a project, Old Church Slavonic has). Such are definitely useful: any educated Roman Catholic (in the Roman Catholic matters) should know Latin. 3.2. Some of definitely dead languages, like Gothic, Anglo-Saxon... 3.3. A number of them which don't have projects because of our policies.
And, I'll try to put them in one priority list, with explanations.
1) 1.1. English Wikipedia is definitely our first priority. This is not because I like English, but because of the fact that it is a lingua franca of the contemporary world. If you have some knowledge written in English, you may easily have that knowledge in other languages, too. However, this project may take care about itself. 2) 1.2. German Wikipedia is at the same priority as the next group, but it share one characteristics with English one: it may take care about itself. 3) 1.3. Well developed projects are, also, often a lingua franca of some region, or even more widely. Their importance is similar to the importance of English Wikipedia in that sense. Because of those projects we need to have the Volunteer Council: to give them possibility to take care about themselves. 4) 1.4.-1.5. Emerging and starting projects are our next priority: They need a lot of technical and other help to become a stable, well developed projects. Their importance lays at the fact that a lot of people are talking those languages. 5) 1.6. Of course, our next priority should be Wikipedias which have some activity. If we see that some people are interested in Wikipedia in their language, we should encourage them to participate in the project. 6) 1.7. Not active projects are important, too. At some time someone came to us and asked for the Wikipedia in their language. We should try to find some people who are interested in writing project in that language. But, it goes out of the scope of online community and it is a matter of WMF and their contacts. 7) 1.8. The same is for the written languages which don't have projects. People who are speakers of some language and asks for the project in their language are very important: it means that they would be maybe able to go into the more stable state in the near future. At this point I really support Gerard's position that MediaWiki messages should be translated: It doesn't just allow other speakers to read MW messages, but it shows to us that a person is (or persons are) really willing to create their project. 8) 1.9. The last group, non-written languages, are, again, a matter of the WMF. It should be incorporated into the international efforts to make written forms of non-written languages. 9) 2.1.-3.1. Useful conlangs should be the next priority. At least, some number of humans are able to communicate in those languages. And we should allow them to write their encyclopedias. However, in this category are only *really* useful conglangs, like Esperanto is. However, again, Volapuk became a useful one, too -- because of its similarity with English and a work of one person. This is the category for useful ancient/dead languages, too, like Latin is. Also, if Klingon (or whatever artistic language) becomes enough widespread to be useful -- it should go into this category. 10) 3.2.-3.3. Definitely dead languages are the next. If we have resources, and there are people who are willing to do some neo-classical work -- it may be useful (somehow). 11) 2.2.-2.4. Non-artistic conlangs are the next. There are a lot of them; some may be useful for scientific purposes or even for communication ;) 12) 2.5. Then, here are artistic conlangs, too. If someone wants to enjoy while making an encyclopedia in an artistic language and we have resources -- why not to allow that. Maybe such languages would be used for real communication sometime in the future. * 2.3. (and similar) Of course, the only type of conlangs (artistic or not) which are out of the scope of our interests are copyrighted languages.
And the point is the question: Where are we now? Hm. While we are doing partially other tasks, the answer is simple: We are now in the process of making Volunteer council, which means that we are finishing the third global task out of 12.
And, what to do? Of course, we should analyze our possibilities, first. Maybe it should be one of the first tasks of the VC. I am sure that the most of use will accept to take care about projects up to the priority 7. However, WMF and VC should give to us an analysis of our possibilities. If we need to spend $10 and 10 working hours (usually, steward's working hours) per year for one new project in an artistic language (priority 12), then I think that it is reasonable. However, if we need to spend $50.000 and a lot of working hours per year for useful, but not so important Volapuk Wikipedia, instead of giving $10.000 per one African language for making five relevant encyclopedias in their languages: I am definitely for the second choice.
So, this was my contribution to relatively connected issues about we are talking a lot. I tried to move discussion from arbitrary choices to a bigger picture. Of course, I don't pretend for a perfect construction. I just hope that we may move toward more rational talks than arguing for one or another option.