[liblouis-liblouisxml] Re: Tables mess
- From: "John J. Boyer" <johnjboyer@xxxxxxxxxxxxx>
- To: liblouis-liblouisxml@xxxxxxxxxxxxx
- Date: Tue, 23 Dec 2008 22:32:39 -0600
Jamie,
Sorry for the mess. I just added all the tables that had been submitted.
I'm not entirely guiltless in the matter of table naming myself. This is
partly due to taking the easier route at the beginning. For example,
chardefs.cti really should be bana-chardefs.cti . BANA is the braille
Authority of North America.
I hope the tables aren't too broken and that we will hear from users of
various languages. I guess we need more QA to prevent new things from
breaking old ones.
I see that you fixed the inconsistent line endings with your latest
commit.
Thanks,
john
On Wed, Dec 24, 2008 at 09:34:37AM +1000, James Teh wrote:
> Hi all,
>
> As of liblouis 1.4.0, the slightly convoluted situation with tables
> became even more convoluted with the introduction of the converted
> BRLTTY text tables. There are now multiple tables for several languages
> and many do not include comments indicating the difference between them.
> Aside from making it difficult to know which table to use, this makes it
> extremely difficult for products using liblouis to provide a
> user-friendly list of tables. For example, in NVDA, we provide a list of
> tables, but rather than using file names, we provide human readable
> names; e.g. Unified English Braille Code grade 2 for UEBC-g2.ctb.
>
> Several issues need to be addressed:
> * While I realise that the filename extensions are arbitrary, as I
> understand it, .ctb is supposed to be for contracted tables and .utb for
> uncontracted. Many tables do not follow this.
> * There is no way to distinguish text tables; i.e. tables that just map
> from character to dot pattern, also known as grade 0. BRLTTY calls these
> text tables (.ttb).
> * All of the converted BRLTTY text tables are named .ctb, which is
> incorrect, as they are not contracted.
> * Many of the converted BRLTTY text tables already had equivalents in
> liblouis. These should either be removed or should supercede the old tables.
> * There is no consistent method used to describe tables in comments.
> Some tables have a comment like:
> # liblouis: Unified English Braille Grade 2 table
> This is a good start. Many tables don't have this at all or use a
> different pattern. The converted BRLTTY tables are a particularly
> problematic example.
> Better still would be the following:
> # liblouis table: Unified English Braille Code grade 2
> This would allow programmatic extraction of the table descriptions for
> use in third party software as described above.
> * There are duplicates such as fi1.ctb and fi2.ctb, and it-it-g1.utb and
> it-it-g1.utb2. The latter is especially confusing - what is utb2? If
> there is no reason for these duplicate,s they should be removed. The
> difficulty is in determining which is the correct table.
>
> First, I'd appreciate thoughts on all of the above.
>
> If this is to be solved, we are probably going to need help from users
> who use all of the available languages, as they will have a better idea
> of which tables should be kept and which should be removed.
>
> --
> James Teh
> Email/MSN Messenger/Jabber: jamie@xxxxxxxxxxx
> Web site: http://www.jantrid.net/
> For a description of the software and to download it go to
> http://www.jjb-software.com
--
My websites:
http://www.godtouches.org
http://www.jjb-software.com
Location: Madison, WI, USA
For a description of the software and to download it go to
http://www.jjb-software.com
Other related posts:
