zompist bboard

You may want to consider specifying Unicode collation during the creation of your database tables, and then ensuring to specify Unicode when interacting with PHP. This way you'll avoid any encoding mishaps on the server-side for other users who download your code.

Anyhow, you should consider putting up the project on a repository management website such as Bitbucket. That way others can pitch in, comment, write up issues, or even fork the code if needed.

Hakaku wrote:You may want to consider specifying Unicode collation during the creation of your database tables, and then ensuring to specify Unicode when interacting with PHP. This way you'll avoid any encoding mishaps on the server-side for other users who download your code

Oh yes, definitely. I learned this the hard way during an early version, since my Novegradian conlang uses Cyrillic. Ran into some horrible messiness when trying to create and fetch entries...

Hakaku wrote: Anyhow, you should consider putting up the project on a repository management website such as Bitbucket. That way others can pitch in, comment, write up issues, or even fork the code if needed.

Thanks, I'll look into that.

Not sure if this has been noted, but in the current version there is a bug where the type-subtype is not always recorded, even when you move on to adding the next word. Also, exactly how does the ordering work, as what I'm seeing is nothing like alphabetical order.

Excuse the noob. But how does one see what your code does? As in test it without knowing the programing language.

What code do you mean? Sko's database schema? That's just a database schema. It doesn't do anything, except create an empty database.

Alright, just an update to confirm that I have in fact been working on this. I was a little too busy to do much in the two weeks or so after I originally agreed to work on this (dealing with moving into a new apartment), but since then I've been able to put a bit of work into it most days. I've now got a mostly functioning control panel that will allow you to configure the lexicon manager from scratch, add new languages with custom fields, custom alphabetical ordering, and various other qualities, and am currently in the process of improving management of individual lexicons (adding words, editing words, etc). Once that's taken care of I'll get started on the public-facing portion, but that should be relatively quick since most of the code for that already exists in the personal lexicons I've made. A 0.1 release (fully functional, but very basic) should hopefully happen within the month, and along with that I'll release the source as well.

(Actually, can PHP even be closed-source, at least in the sense of 'you can't see my code'? You can't really hide the code if you're distributing it, though I guess copyright could restrict you from messing with it.)

Also, regarding things like that autodeclension script I had in the Novegradian lexicon... it seems like it should be fairly simple to allow for the lexicon to load plugins written in PHP and to tie those into a particular field in each lexicon entry. Later on I may adapt my current autodecliner script into a proper plugin that could be adapted for other languages. It'd still require code editing, but if most of the framework is already there, it hopefully shouldn't be too hard to do with fairly minimal programming knowledge (so long as the morphology isn't too fiendish).

Mecislau wrote:Alright, just an update to confirm that I have in fact been working on this. I was a little too busy to do much in the two weeks or so after I originally agreed to work on this (dealing with moving into a new apartment), but since then I've been able to put a bit of work into it most days. I've now got a mostly functioning control panel that will allow you to configure the lexicon manager from scratch, add new languages with custom fields, custom alphabetical ordering, and various other qualities, and am currently in the process of improving management of individual lexicons (adding words, editing words, etc). Once that's taken care of I'll get started on the public-facing portion, but that should be relatively quick since most of the code for that already exists in the personal lexicons I've made. A 0.1 release (fully functional, but very basic) should hopefully happen within the month, and along with that I'll release the source as well.

(Actually, can PHP even be closed-source, at least in the sense of 'you can't see my code'? You can't really hide the code if you're distributing it, though I guess copyright could restrict you from messing with it.)

Also, regarding things like that autodeclension script I had in the Novegradian lexicon... it seems like it should be fairly simple to allow for the lexicon to load plugins written in PHP and to tie those into a particular field in each lexicon entry. Later on I may adapt my current autodecliner script into a proper plugin that could be adapted for other languages. It'd still require code editing, but if most of the framework is already there, it hopefully shouldn't be too hard to do with fairly minimal programming knowledge (so long as the morphology isn't too fiendish).

Sounds pretty useful. I'd certainly like to see if I could build upon it to add the functionality for consonant root based languages (i.e. Semitic-style triliteral roots).

Why not release it under a licence like the GPL? That at least gives you and your code some protection and yet still allows others to make modifications. Creative Commons (Noncommercial, No derivative works, etc) is another idea, and a licence similar to Vim's is another idea (although creating your own licence is not a particularly good idea, as there are plenty out there already.)

Alright, version 0.1 is out!

http://www.veche.net/programming/lexmanager.html

(Yes, I know the name sucks; it's temporary, and I'm open to suggestions. The only one I've received so far is "Lummox", which... I have to say, it's growing on me a bit.)

Lemmator?

Hey, has anybody had any success using lex manager on a local server? I tried setting it up with WAMP and XAMPP, and I ran through the config, but LexManager cannot see the tables in the lexicons database. That is, I can create a new lexicon through the LexManager interface and see the tables that it creates in phpMyAdmin, but LexManager doesn't see those tables and says there are no lexicons. Any advice?

Mecislau wrote: The one thing there that is rather rigid (though not hardcoded) is the autodeclension tool that produces complete declension charts for every noun and adjective. What you see there is tailored specifically for Novegradian (just pulling some metadata for each noun identifying which subparadigm is needed), and off the top of my head I can't think of a good way to make an easily-customizable autodecliner. It can be done, it would just be a lot of work.

The problem is that to cover every possibility you effectively need a full programming language, which is a bit much for the non-technical user. Although it's exactly the same with SCAs... the amount of complexity required to cover rare corner cases is a lot higher than to cover the most common changes.

The other problem with declensions is... what do you consider a declension? If a language potentially inflects nouns for five different things your 2d table is going to start looking a bit unwieldy. Tables which work for IE style noun inflection work less well for languages with more agglutinating and potentially recursive noun marking.

In terms of integrated SCAs and declention tools, you also need the ability to override the results to get irregular forms.

Anyway, if you all build a good online dictionary system I 'll be interested. I considered doing it myself after I finished my SCA but I don't have a lot of time right now so I'm using a wiki instead. It's not a great way to do it, and it makes me wonder why Wiktionary chose to do it that way.

Let me say, as someone who doesn't know much about programming but has been in situations away from conlanging where this kind of project is really needed, this is great. I'll get playing with it right away.

Though I will need to figure out how to do the PHP and MySQL stuff first. Hrrm.

[razed and salted]