To LUGNET HomepageTo LUGNET News HomepageTo LUGNET Guide Homepage
 Help on Searching
 
Post new message to lugnet.faqOpen lugnet.faq in your NNTP NewsreaderTo LUGNET News Traffic PageSign In (Members)
 FAQ / 183
182  |  184
Subject: 
Re: New LUGNET directory structre?
Newsgroups: 
lugnet.faq
Date: 
Thu, 6 May 1999 18:05:59 GMT
Viewed: 
1276 times
  
In lugnet.faq, sparre@sys-323.risoe.dk (Jacob Sparre Andersen) writes:
Todd:
This just does a flat dump of the entire tree.  Columns in the data are:
   1.  URI of the group (URL minus the http://www.lugnet.com prefix)
   2.  Category ID within its parent (null for the root)
   3.  Newsgroup name (null if no associated newsgroup)
   4.  Short category name
   5.  Long category name
   6.  Category/newsgroup charter

How do we integrate the translation of the FAQ with this?  We need to
translate (and store) column four and five to translate the FAQ.

Not sure yet...only started seriously thinking about and planning for
translation a couple months ago.


Should we generate a file similar to [1] based on your table, and then
add the translations to that file,
1) <URL:http://www.ldraw.org/www_admin/phrase_dictionary>

It couldn't hurt, but I wouldn't say that the wording is stabilized enough
to be worth investing much time in this early on.


or will you find a way to integrate the translations in your database?

This, yes definitely.  It'll come out in the wash.  But probably not anytime
super-soon.  Translation for anything on LUGNET is a relatively low priority
from my P.O.V., although planning for the data expansion is relatively
medium to high priority, so as not to get painted into any corners.

I did quite a bit of multi-language translation support mechanism
programming at my last job, so I'm painfully aware of how difficult it is to
retrofit something rather than to build it in.  (It does also frighten me a
bit because I know it can get way out of hand too.  :)  In my experience,
a couple things that are especially tricky in real-world translations are:

-  Parameterized pluralization:  Things like "%d %s" where %d is a quantity
   and %s is the word.  Not only does %s have to be looked up for the plural
   and singular forms, but the string "%d %s" has to be looked up too, since
   it might be "%s %d" in some cases.

-  Table lookups:  Displaying the nth item in a list of words or displaying
   a keyed-lookup translation; needs to be the capability for constructing
   string-tags on-the-fly with sprintf-like ease, and then feeding that to
   the lookup function.  Some of this can be hidden transparently away
   inside a local object but it still has to have some kind of a global
   namespace so that translators can access large bodies of content all at
   once (rather than a string at a time) for the most efficient use of their
   time.

--Todd



Message is in Reply To:
  Re: New LUGNET directory structre?
 
Todd: (...) How do we integrate the translation of the FAQ with this? We need to translate (and store) column four and five to translate the FAQ. Should we generate a file similar to [1] based on your table, and then add the translations to that (...) (25 years ago, 6-May-99, to lugnet.faq)

21 Messages in This Thread:








Entire Thread on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact
    

Custom Search

©2005 LUGNET. All rights reserved. - hosted by steinbruch.info GbR