search

Omikhleia / sindict

Sindarin dictionary in TEI format
Other
32 stars 6 forks source link

Automated workflow to add Quenya, Telerin (etc.) cognates to existing entries? #51

Open L-ND-L opened 3 years ago

L-ND-L commented 3 years ago

Here in English, as it's a more generic issue:

Wouldn't it be possible to develop a workflow allowing easy, structured referencing of cognate words in Quenya and Telerin (and maybe a few more languages: Nandorin, Danian, Ilkorin...)?

This would then allow for simple acceptance / rejection (maybe with comment) from the project administrator(s).

Once developed (if possible / realistic / useful), this would be easier and faster to manage the addition of cognates than the generic issue reporting, which requires further manual processing to add the cognates to the database.

Omikhleia commented 3 years ago

On one hand, adding cognates for all languages would be a monumental task - of a kind this project never touched since its inception more than 20 years ago... So a first question could be: is it worth involving time in this direction now? If we start with cognates, we'll even want roots, which changed over Tolkien's lifetime - that's a lot of things to properly encode...

I only know one project that sort of achieved it: Eldamo... So a second question could be: what would we solve here differently, for the effort to be worth of it?

As for developing things, let's face it, there was always a lack of long-term contributors, not to say developers.

Regarding developers,

Regarding contributors... I am afraid I don't think it's realistic, unless you have a lot (and a whole lot) of free time. Those who could contribute, in a way, preferred doing their own things and even to sell them (I'm sure you know this one and this other one for an example. All derivatives, in a way or another, of this very project. All failing short, if I may say...)

This dictionary was once one of the best of its kind - if not the best -, I don't think saying that is so pretentious. Nowadays, though, one cannot do as if Eldamo didn't existed ;-) Yet, this dictionary can still count, despite its age. But asking for more than that... I don't know.

I'm sorry to sound quite negative here. Find developers (and for such a workflow, you mean coders(2), right?), and I'll gladly reconsider my position.

(1) It's partly lost now. Bits were recovered, though I have no idea if it could be revived without lots of tweaks. But the video link above was cool...

(2) Just in case I am not clear here: I don't consider the lexicon to be "code". It's just tagged information (ok, in XML, but heh). A big text file with mark-up. The code, for me, is what transforms it to HTML, PDF, or whatever.

Omikhleia commented 3 years ago

I had even forgotten the other Ladon experiment video. Showing the main dictionary tool (not the editor). Makes me feel even older :stuck_out_tongue:

L-ND-L commented 3 years ago

On one hand, adding cognates for all languages would be a monumental task - of a kind this project never touched since its inception more than 20 years ago... So a first question could be: is it worth involving time in this direction now? If we start with cognates, we'll even want roots, which changed over Tolkien's lifetime - that's a lot of things to properly encode...

You're right. On the other hand, there are already some entries with Q(u)enya and Telerin cognates, and it would be sad to remove them. So for the sake of consistency, one may want to continue adding cognates to existing entries that don't have them yet. Cognates for other languages would definitely be an additional task, albeit a limited one (basically covering only The Etymologies).

As for P.Q. / C.E. words, and roots, that's even a further step. It would definitely be nice to be able to encode them properly (even if we don't start adding them), but I agree this goes well beyond the initial scope of the Dictionary, and would probably overlap more and more with Eldamo.

I only know one project that sort of achieved it: Eldamo... So a second question could be: what would we solve here differently, for the effort to be worth of it?

This brings us to your second question, which is a very good one, obviously. I do have three arguments in favor of doing things differently, and one against. Pros: (1) there are a fair number of cases where I have good reasons to think that Paul got some interpretations wrong, and it takes way too much efforts to convince him, even with a lot of evidence; (2) there is no French version of Eldamo; (3) developing one would soon create additional issues, as Paul doesn't provide a way for others to track his ongoing changes. Cons: I don't have the time to do it now, and won't for the short to medium terms.

Regarding contributors... I am afraid I don't think it's realistic, unless you have a lot (and a whole lot) of free time. Those who could contribute, in a way, preferred doing their own things and even to sell them (I'm sure you know this one and this other one for an example. All derivatives, in a way or another, of this very project. All failing short, if I may say...)

You may indeed. :)

I'm sorry to sound quite negative here. Find developers (and for such a workflow, you mean coders(2), right?), and I'll gladly reconsider my position.

I do mean developers, and of course I don't have names right now. If I did, there is a large Tolkienian French website whose overhaul is now long overdue, and I would direct them there as a priority. ;)

(2) Just in case I am not clear here: I don't consider the lexicon to be "code". It's just tagged information (ok, in XML, but heh). A big text file with mark-up. The code, for me, is what transforms it to HTML, PDF, or whatever.

I actually tried to open the xml file last week, but my computer refused to comply... I'll have to check why when I have time to do so.