Conlang relay [relocated] (aka "The Cursed Relay")

[Corumayas]

Sounds really fun, cedh! I hope we'll get to see a video of your talk or something. Also great to hear from you again, TzirTzi!

If you'd like any input from me about what you're planning to talk about let me know. I really ought to stop by IRC again soon, anyway; maybe we should schedule a chat? Mondays and Tuesdays are my regular days off, and now that I'm done moving I have some free time again.


(Unrelatedly, it looks like there's been an enormous amount of spam posted on the wiki lately. Do we have the ability to limit new accounts or something?)

[Cedh]

I'm intending to go to the conference and would be happy to help/collaborate on an Akana talk Once I refresh my Akana knowledge after this last long absence, anyway...

If you'd like any input from me about what you're planning to talk about let me know. I really ought to stop by IRC again soon, anyway; maybe we should schedule a chat? Mondays and Tuesdays are my regular days off, and now that I'm done moving I have some free time again.

Cool! I'll try to visit the IRC channel regularly in the next weeks (usually between 22:00 and 1:00 CET) for discussion.

(Unrelatedly, it looks like there's been an enormous amount of spam posted on the wiki lately. Do we have the ability to limit new accounts or something?)

I think we could require a captcha when registering a new account, e.g. using the ConfirmEdit extension. Installing this would be TzirTzi's task; he's the one with access to the config files. @TzirTzi: Please make a full backup before changing the configuration.

[TzirTzi]

No problem my computer's at the shop at the moment (I'm posting on my phone) but I hope to get it back in the next few days, at which point I'll get on with that.

[Basilius]

(Unrelatedly, it looks like there's been an enormous amount of spam posted on the wiki lately. Do we have the ability to limit new accounts or something?)

I think we could require a captcha when registering a new account, e.g. using the ConfirmEdit extension. Installing this would be TzirTzi's task; he's the one with access to the config files. @TzirTzi: Please make a full backup before changing the configuration. :)

Watching how Cedh has to fight the spammers is not funny at all :(

(One trouble with this is that if part of that garbage is SEO spam, it may be in fact (cost-)effective. A few hours may be enough for a search engine to index the pages *and* the SEO links they contain. That is, currently our spammers have no reason to re-tune their scripts...)

I think there's a way to block spamming without installing any additional software. I'd suggest replacing the registration form with a page inviting one to introduce oneself and one's possible contribution on Akana forum - *without* a direct link. Cedh, is this technically possible?

May appear too radical, but I believe it's not so big an inconvenience for humans (for the link is available in the main page, and introducing oneself at the place where things are discussed seems the right way to join a project), while making submissions indeed not worth the effort for spambot owners. And there aren't too many people who can actually contribute something useful who we don't know yet...

Also, it can be a fallback solutions if captcha doesn't help (there are stupid-but-effective ways to crack captchas, y'know).

[Cedh]

Basilius, your suggestion sounds like a good idea, at least as a fallback strategy - but I would need access to the config files to implement that too. I may perform some admin tasks in the wiki frontend, but I'm not the backend admin.

(Also, registration on the forum is currently closed because of spam. There, I have backend access, and the forum software has a handy option to delete all posts by a user whose account you're in the process of deleting.)

[TzirTzi]

I have a (semi)working laptop again! Yay for me.

As a result, ConfirmEdit has now been installed. I've currently got it set to use QuestyCaptcha with the following questions:

Q: "What does ZBB stand for?" A: "Zompist Bulletin Board"
Q: "What is conlang short for?" A: "constructed language"
Q: "What is the relay thread relating to Akana on the ZBB described as?" A: "cursed"

QuestyCaptcha isn't included with the ConfirmEdit release intended for MediaWiki 1.12, so I added it manually... if there are any errors that I didn't find, that's probably why - let me know and I'll change it to one of the other captcha types. Or if people would prefer a different captcha type, I can change it. Any suggestions for additional questions I'll also be happy to add. The other thing I could do is change it so that it's only required on account creation rather than on every page edit.

[Cedh]

Thank you very much!

I did some quick tests, and I encountered the captcha only on 'createaccount', but not on any of 'create', 'edit', 'addurl', or 'badlogin'. Is this how it should work at the moment?

(Your post seemed to imply you had set it up so the captcha would show on any page edit. I don't think that would be necessary though; it would end up quite annoying for us, and most of the spambots seem not to have edited existing articles anyway, not even their own ones.)

[TzirTzi]

That's how it should work at the moment given that you're a sysop (which you are, right?). skipcaptcha is set to true for sysops and registered bots. For other user groups, captcha is required for all edits and account creation - I could certainly change this to just for account creation, if it'd be too annoying for everyone as is.

[4pq1injbok]

Glad to hear that there's broader enthusiasm for giving a talk on Akana at LCC4.

We're beginning to assemble the schedule now, so it'd be handy to have some sort of actual talk proposal soon, sent to conference@conlang.org.

[Cedh]

For those who haven't encountered the confirmation yet: TzirTzi and I are indeed preparing a talk for LCC4, entitled Akana: A Long-Lived Collaborative Conworld. Here's the abstract:

Collaborative conworlding projects are rarely successful. When they are proposed on a web forum or mailing list, they often receive enthusiastic responses at first, but then fade out of favour after only a few months of activity. The Akana project, however, has been running for six years so far. It features one of the largest known conlang families, and new material is still being created by a stable team of about a dozen main contributors. What is the key to its longevity?

The talk will summarize the rather unique history of this conworld, which has developed from two diachronic conlanging games, and outline strategies that have helped to keep it alive for such a long time. We will have a detailed look at how some significant pieces of content were assembled, describe how these particular methods of collaboration have shaped the world of Akana, and single out obstacles that have arisen as the result of a certain approach. Some of the conclusions drawn from our experience with Akana may be applicable to other collaborative projects as well, and we hope to provide some hints how to keep people interested in contributing.

---

Also, I have made some additions to the Tmaśareʔ grammar recently, and I have also translated a fairly long new text, which is an adaptation of Dewrad's Proto-Western creation myth, explaining the importance and rarity of *kʷacu:

• http://tzirtzi.ipage.com/akana/index.php?title=Tmaśareʔ#Creation_myth

[roninbodhisattva]

How would people feel about me setting Dơk Nhuang somewhere in Tuysafa?

[the duke of nuke]

I'm all for it - particularly the southeast is rather empty at the moment, and there's room for two large agricultural civilisations there.

[roninbodhisattva]

I'm all for it - particularly the southeast is rather empty at the moment, and there's room for two large agricultural civilisations there.

That's actually what I was thinking. Perhaps on the peninsula in the Southeast?

[Cedh]

How would people feel about me setting Dơk Nhuang somewhere in Tuysafa?

I'm all for it - particularly the southeast is rather empty at the moment, and there's room for two large agricultural civilisations there.

That's actually what I was thinking. Perhaps on the peninsula in the Southeast?

Sure, go ahead! (Well, if you like hurricanes... )

[roninbodhisattva]

How would people feel about me setting Dơk Nhuang somewhere in Tuysafa?

I'm all for it - particularly the southeast is rather empty at the moment, and there's room for two large agricultural civilisations there.

That's actually what I was thinking. Perhaps on the peninsula in the Southeast?

Sure, go ahead! (Well, if you like hurricanes... )

Actually, looking at that and then the map of southeast tuysafa on the wiki, it would probably be better me to center them in region twelve . There's a large river running through it, and I think I'lll probably put them somewhere along there.

[the duke of nuke]

My post actually lists you as having reserved that region already

(Sadly, I won't be doing much on Akana for another 7 weeks or so due to exams...)

[roninbodhisattva]

My post actually lists you as having reserved that region already

Part of the reason I decided to switch haha.

[Zhen Lin]

(Sadly, I won't be doing much on Akana for another 7 weeks or so due to exams...)

I should be preparing for exams too. And my career plans depend not insignificantly on me doing well this year. Yet I'm being distracted by many things, most recently, by thoughts of continuing my work on Vylessa (now almost 2 years old and still very incomplete), as well as designing further Peninsular languages. Not good, not good...

[Basilius]

I've logged in just to say that Cedh's new maps are cool.
(It's worth checking out e. g. via "Recent Changes".)

[the duke of nuke]

Very impressive, indeed!

One comment on the Ndak Empire map: Wasn't Mos Mumbasi the capital, rather than Ngaurlau? I ask because the former is specified in the article, but the latter is underlined on the map.

[Radius Solis]

Very impressive, indeed!

One comment on the Ndak Empire map: Wasn't Mos Mumbasi the capital, rather than Ngaurlau? I ask because the former is specified in the article, but the latter is underlined on the map.

Mos Mumbasi was indeed the capital of the Ndak empire. Although I'm not sure the notion of a capital city was well-established in that era, Tsinakan's palace had to be located somewhere, and that was Mos Mumbasi. Some of its ruins still exist in Momuva'e.

And Cedh, wow. Those are excellent and valuable maps! I'm still looking forward to seeing the Big Map someday too.

[dunomapuka]

cedh and TzirTzi - Just watched your presentation from LCC4. Good job! The video quality wasn't great though, so I can't read the text on the screen - I'd love to have a look at the file.

[Cedh]

Thanks! I'm also quite happy with how the presentation went; we were a bit rushed because the previous talk had taken extremely long and so we skipped over some language details we had planned to mention, and my spoken English could use some more practice, but well, it worked out quite fine even so. Especially considering that we had combined what each one of us had prepared only a few hours before the talk ... Some of the other attendees even expressed interest in contributing to Akana, which would be very cool. (If you read this, go ahead and join us!)

Here's the video of the talk (scroll forward to 01:15:30, although Christophe Grandsire-Koevoets' hour-long talk about surdéclinaison is also very interesting) and of the Q&A part (first five minutes or so). The slides can be found here, and two pictures of us presenting are here and here.

We also made two large introductory posters which were placed at the door of the conference room, one about Akana as a conworld, and the other about the Edastean languages. (That's what I made the new maps for.) As another piece of "previously unreleased information", the second poster includes the beginning of the Tsinakan text in Komejech, given to me by Legion on IRC.

All in all, LCC4 was a really awesome experience. You can't imagine what it's like to spend a whole weekend with ~25 other conlangers if you haven't done that yet! Read Guitarplayer's blog entry for a quick summary and a few photos.

[Corumayas]

Yeah! That was really fun to watch. It's kind of neat to hear the voices of people I've only communicated with in text; and I realized while watching that this was the first time I'd heard all those Akanaran names spoken out loud! It changes my perception of them slightly-- e.g. for some reason I was pleasantly surprised by how nice "Ndak Ta" and "Qedik" sounded.

The posters are really cool too-- nice graphics, and lots of info packed in them. (I hope we can look forward to seeing more maps in that style!) And your analysis of what we've done right and wrong with the Akana project is thought-provoking as well. Great work, guys!

[TzirTzi]

I'm glad you enjoyed it dunomapuka and Corumayas One thing that I hoped is that our discussions and thought that went towards making the presentation would also help the project directly as well. Thinking lots about what has worked and what hasn't should hopefully point us towards what we might be able to change in the way we do things in future as well. The todo list resulted from those planning conversations, so that's happened a little already!

On an unrelated note, I started work on some SCs for Nualis. They're not quite right yet - a little too innovative, as I had in mind that phonologically the Nualis-Takuna family would generally be extremely conservative. Anyway, I remembered Basilius and Qwynegold's work on an NT daughter a couple of years ago, and so went and re-checked that out. It has set me thinking - the daughterlang they discuss there is extremely innovative in various ways (/is at great time depth) and I seemed to be going in a more innovative direction with Nualis than I'd planned. Perhaps what's needed here is some intermediate proto langs for the NT family, to avoid the wild-random-jumps effect of the later Daiadak family. At its simplest it might look something like this -

Code: Select all

                               PNT
                                |
                                |
                                |
                      ---------------------
                      |                   |
                      |                   |
                      |                   |
           Proto-Southern-NT        Proto-Northern-NT
                 |                           |
                 |                           |
                 |                           |
         ----------------         Innovative Northern langs
         |              |
         |              |
         |              |
      Takuna         Nualis

But obviously things could be rather more convoluted if people wanted - really, the Southern grouping should split (before splitting into N and T) into Coastal and Continental (N and T being part of the Coastal subfamily). Also - and this is true regardless of whether we do some in-between langs like the above - it doesn't make much sense the family being called Nualis-Takuna when N and T are languages on just one branch, and there is a much higher branching subfamily. So either they should be on different branches and the third branch should contribute a name too, making it Nualis-Takuna-X, or the description of people-movements could be changed a bit and Nualis could be in the Northern subfamily, or the whole family could be renamed altogether.

Another problem that occurred to me is that the Siixtaguna culture is an immigrant culture, in historical terms fairly new to the Siixtaguna coast and nearby islands (having migrated from some unspecified place in the east, presumably Tuysafa). The Takuna are slightly farther from their original homeland than the Nualis, and so it would make sense that their language is more conservative (such is the typically observed pattern); but the proposed Northern grouping would probably be rather further from the homeland and yet linguistically is massively more innovative. One possibility here would be to change the currently Northern subfamily into being the aforementioned Continental subfamily, which was the result of an earlier wave of migration and so might well be more different to N and T than they are to each other whilst occupying a similar area. But I don't know how much planning Basilius and Qwynegold had done around the culture and history of their potential daughterlang? I'm afraid I haven't read through that entire thread yet

Anyway, I would welcome people's opinions and if anyone else is also interested in renewing work on this family, do let me know!