El dl 07 de 01 de 2013 a les 16:47 +0100, en/na Per Tunedal va escriure:
> Hi,
> I believe that it's essential to create an easy way for laymen to
> contribute to a language pair.

I don't believe that it is essential. 

> A. What happened to the web form?

It's probably in SVN somewhere. There have also been other projects that
attempted to do this:

https://apertium.svn.sourceforge.net/svnroot/apertium/branches/gsoc2010/alessiojr

https://apertium.svn.sourceforge.net/svnroot/apertium/trunk/apertium-tools/apertium-lexical-webform

https://apertium.svn.sourceforge.net/svnroot/apertium/trunk/apertium-tools/apertium-dictionary-form

https://apertium.svn.sourceforge.net/svnroot/apertium/trunk/apertium-forms-server

> B. Why not first try to manage the easy ones? A tool that makes it
> possible to easily contribute nouns and, maybe verbs, would be welcome.
> I've tried to create a simple tool for the pair sv-da, but it might not
> work for other languages.

If you want to see how to add verbs, I suggest you check out:
apertium-af-nl, apertium-es-ca, apertium-fr-es, apertium-br-fr and
apertium-sh-mk.

All of them have slightly different ways of dealing with verb entries,
and all of them are released pairs and in trunk/ -- this isn't even
considering pairs in staging, nursery or incubator.

Summary:

* af-nl (different entries for forms with ge- prefix)
* es-ca (different entries for different accentuation dímelo vs. dime)
* fr-es (uses metadix to combine paradigms for some verb forms)
* br-fr (includes entries for common synthetic tenses)
* sh-mk (uses metadix and paradigms to deal with variation)

> http://www.tunedal.nu/download/AddToDix/ (N.B. I'm a beginner in Java.)
> And I believe others have their own tools.

The benefit of Apertium is that it is easy to write tools for small use
cases. I write scripts on a case-by-case basis.

> C. Maybe there is some fundamental design problem with the dictionaries,
> if it's that difficult to create an interface? More standardisation?

Apertium language pairs are very heterogeneous. Making something that
works with all of them or even with a worthwhile subset will be
difficult.

The dictionaries are designed to be representations of finite-state
transducers. For this they are well designed. 

I agree with standardisation, but I believe it should be an organic,
bottom-up standardisation, not a top-down one. The idea of coming up
with an "all encompassing dictionary format" would lead us to something
awful and unworkable like LMF.[1]

It's better that people who work on language pairs come up with their
own ways of doing things, taking into account how things work in other
language pairs, and then sharing scripts etc.

Fran

1. http://www.lexicalmarkupframework.org/


------------------------------------------------------------------------------
Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS,
MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current
with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft
MVPs and experts. SALE $99.99 this month only -- learn more at:
http://p.sf.net/sfu/learnmore_122412
_______________________________________________
Apertium-stuff mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/apertium-stuff

Reply via email to