Hello Tim,

the statistics as well as the featues you described would indeed be
nice and I think it's worth implementing them.

Regards,
Aiet

On Apr 7, 2005 3:07 AM, Tim Morley <[EMAIL PROTECTED]> wrote:
> Sounds like a fine plan, whatever software is chosen.
> 
> If I could make one suggestion, based on our experience in the
> Esperanto project -- it's very useful for the translation co-ordinator
> to be able to see *who* did a particular translation or who made what
> change and when, and for some kind of "proof-read" flag to be set when
> a translation has been checked. If there were also some automatic
> system that insisted on a comment from any user who was changing an
> existing translation, that would be handy too, to encourage them to
> explain the rationale for their proposal.
> 
> Regards,
> 
> Tim
> 
> On 5 Apr 2005, at 16:52, [EMAIL PROTECTED]
> wrote:
> 
> >
> > From: Aiet Kolkhi <[EMAIL PROTECTED]>
> > Date: 4 April 2005 05:33:24 BST
> > To: [email protected]
> > Subject: Integrating OOo Online Glossary Management to
> > xy.openoffice.org
> > Reply-To: Aiet Kolkhi <[EMAIL PROTECTED]>
> >
> >
> > Hello,
> >
> > discussing this with Louis Suarez-Potts, he recommended that I raise
> > an issue at [EMAIL PROTECTED] regarding this.
> >
> > As there are many NL projects who do not have another office suite
> > localized to the local language and do not have a working and
> > widespread glossary of terms, it is important that before starting the
> > actual l10n of the modules, they agree on the glossary that will be
> > used during l10n.
> >
> > OOo already did a very good job by preparing the glossary of basic
> > terms that are used in OOo and advising that all the l10n teams
> > localize the glossary first and move to the next stage and start
> > translating the actual modules later.
> >
> > The problem is that not every is familiar with glossary
> > creation/discussion and some l10n teams may not be experienced in
> > using the Translation Memory (TM) during to aid their collaboration.
> >
> > While the actual l10n of the modules is best and most effectively
> > achieved off-line, coming up with the best glossary is mostly only
> > available by having the local IT experts, linguists product users
> > participate in the glossary localization, voting and discussion. All
> > the people participating in the process (e.g. university professors)
> > may not have the ability to download the glossary of terms, translate
> > it and send to the coordinator, most of them will ave access to
> > Internet and the ability to log on a website, comment on existing
> > glossary translations, add his/her own suggestions etc.
> >
> > To achieve this, many tools have been created by various NL and l10n
> > projects/teams.
> >
> > Alberto Escudero-Pascual has mentioned KiPot, a tool that was
> > successfully used for Swahili NL project. Pootle and Webtionary are
> > other good examples.  In Georgian NL, we are also using a tool I
> > created for Georgian glossary creation/discussion that enables
> > registered members to approve trasnlations, suggest their versions,
> > comment on other suggestions etc.
> >
> > If we could just come up with one tool that would suit the needs of
> > most NL projects and somehow have this tool integrated to
> > xy.openoffice.org (as this is the best place to have all the NL and
> > l10n related resources hosted), this would make a very big difference
> > for languages with not so much population or with little l10n history.
> >
> > Upon the completion of the glossary work, the terms could easily be
> > converted to an acceptable format and fed to any l10n software
> > supporting TM. As a result, all the localizations would be able to
> > keep same glossary during the process, as well as use Computer Aided
> > Translation (CAT) to dramatically speed up the process.
> >
> > The opinions of other NL representatives could also be very useful.
> >
> > Best regards,
> > Aiet Kolkhi
> >
> 
>

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to