Smalltalk › Squeak › SqueakLand

Please help us for translation (An experimental version of OLPC image for non-XO environments)

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

6 messages Options

Yoshiki Ohshima-2

Please help us for translation (An experimental version of OLPC image for non-XO environments)

Hello,

We are still trying to package the Etoys for non-XO environments,
along with, of course, for XO.

Currently, end of June is the *real* target of release day of non-XO
packages. The idea is to put relatively small set of languages in it,
and it looks like English, German and French are most complete ones.
(Thank you!)

But, for the US audience, (new world's) Spanish is also important.
And Portugese (br and pt) seems to be fairly good to me, but of course
I can't tell the quality right away.

Can you help us giving the information of the translation quality
and completeness for these languages? For the non-XO environment, we
will not have good font rendering mechanism other than what we have
now, so still the main target is the Latin languages.

The completeness doesn't have to be so close to 100%, as now many of
the phrases are not for Etoys (rather for the Smalltalk development).
But we would like to hear the opinions of the native speakers.

Please visit

https://dev.laptop.org/translate/

and see how far the translation are going in terms of quality and
coverages. Bert told us that there is an old style .translation file
for Spanish. We can convert it to .po file, but not sure again if we
should merge them to the site, as it might wipe out newly translated
phrases. Does anybody has good ideas how to proceed?

Thank you!

-- Yoshiki

_______________________________________________
Squeakland mailing list
[hidden email]
http://squeakland.org/mailman/listinfo/squeakland

José Luis Redrejo

Re: Please help us for translation (An experimental version of OLPC image for non-XO environments)

2008/6/2 Yoshiki Ohshima <[hidden email]>:

Hello,

We are still trying to package the Etoys for non-XO environments,
along with, of course, for XO.

Currently, end of June is the *real* target of release day of non-XO
packages. The idea is to put relatively small set of languages in it,
and it looks like English, German and French are most complete ones.
(Thank you!)

But, for the US audience, (new world's) Spanish is also important.
And Portugese (br and pt) seems to be fairly good to me, but of course
I can't tell the quality right away.

Can you help us giving the information of the translation quality
and completeness for these languages? For the non-XO environment, we
will not have good font rendering mechanism other than what we have
now, so still the main target is the Latin languages.

The completeness doesn't have to be so close to 100%, as now many of
the phrases are not for Etoys (rather for the Smalltalk development).
But we would like to hear the opinions of the native speakers.

Please visit

https://dev.laptop.org/translate/

and see how far the translation are going in terms of quality and
coverages. Bert told us that there is an old style .translation file
for Spanish. We can convert it to .po file, but not sure again if we
should merge them to the site, as it might wipe out newly translated
phrases. Does anybody has good ideas how to proceed?

Yoshiki, as you know, in LinEx we have all the strings translated for the etoys image (plus some other strings from some packages as drgeo, skeleton, etc.). Antonio Moreno has contacted with some of the etoys people, but the answers are not really clear about the way you want us to give you the translation. Our current image is totally based on the current olpc image , with changesets added from us, but we have all the classes and objects the etoys image has, so if you give the exact way you want us to extract the translation you can have the spanish translation finished (and tested & checked) right now.
About the newly translated phrases, if our image has been used and tested for several years, with teachers fixing translations when some mistake was found, what's the problem on wiping out newly translated phrases that has not been checked for so many people?

Regards.
José L.

_______________________________________________
Squeakland mailing list
[hidden email]
http://squeakland.org/mailman/listinfo/squeakland

Korakurider-2

Re: [Localization] Please help us for translation (An experimental version of OLPC image for non-XO environments)

Hello.

On 6/3/08, José Luis Redrejo <[hidden email]> wrote:

>
>
> 2008/6/2 Yoshiki Ohshima <[hidden email]>:
> > Hello,
> >
> > We are still trying to package the Etoys for non-XO environments,
> > along with, of course, for XO.
> >
> > Currently, end of June is the *real* target of release day of non-XO
> > packages. The idea is to put relatively small set of languages in it,
> > and it looks like English, German and French are most complete ones.
> > (Thank you!)
> >
> > But, for the US audience, (new world's) Spanish is also important.
> > And Portugese (br and pt) seems to be fairly good to me, but of course
> > I can't tell the quality right away.
> >
> > Can you help us giving the information of the translation quality
> > and completeness for these languages? For the non-XO environment, we
> > will not have good font rendering mechanism other than what we have
> > now, so still the main target is the Latin languages.
> >
> > The completeness doesn't have to be so close to 100%, as now many of
> > the phrases are not for Etoys (rather for the Smalltalk development).
> > But we would like to hear the opinions of the native speakers.
> >
> > Please visit
> >
> > https://dev.laptop.org/translate/
> >
> > and see how far the translation are going in terms of quality and
> > coverages. Bert told us that there is an old style .translation file
> > for Spanish. We can convert it to .po file, but not sure again if we
> > should merge them to the site, as it might wipe out newly translated
> > phrases. Does anybody has good ideas how to proceed?
> >
>
> Yoshiki, as you know, in LinEx we have all the strings translated for the
> etoys image (plus some other strings from some packages as drgeo, skeleton,
> etc.). Antonio Moreno has contacted with some of the etoys people, but the
> answers are not really clear about the way you want us to give you the
> translation. Our current image is totally based on the current olpc image ,
> with changesets added from us, but we have all the classes and objects the
> etoys image has, so if you give the exact way you want us to extract the
> translation you can have the spanish translation finished (and tested &
> checked) right now.

Thank you for sharing your work with us ! And sorry, I wasn't
enough clear in my recent mails in list.

For OLPC etoys we are using gettext (emulation) and that have
changed how translation dictionary is managed.
In legacy translation engine, there could be obsolete or unused
string in dictionary.
Now we extract strings that are receivers of #translated or
#translatedNoop as POT and populate translations into that. There
can't be obsolete or unused strings in dictionary, as long as
dictionary and source code are in sync . So legacy translation data
needs to be "finalized" before it will be merged into the working
version on Pootle.
I described the finalization process in
http://lists.laptop.org/pipermail/etoys/2008-June/002379.html (see
step (1) and (2))
What I attached to
http://lists.laptop.org/pipermail/etoys/2008-May/002362.html was the
results and I suggested people to try it.

Some strings are missing in extracted POT while that are successfully
translated in the legacy image. Because the strings aren't the
receiver of #translated and gettext exporter can't extract them.
There are missing strings also from migrated your translations. we
have been trying to fix the problem (see
https://dev.laptop.org/ticket/6704 for example) and we will fix also
for yours.

> About the newly translated phrases, if our image has been used and tested
> for several years, with teachers fixing translations when some mistake was
> found, what's the problem on wiping out newly translated phrases that has
> not been checked for so many people?
I (and Yoshiki, I think) couldn't discuss about quality of the
translation language not familiar to me. I could just point out the
fact that there are conflicts between yours and the one on Pootle
(attached to http://lists.laptop.org/pipermail/etoys/2008-June/002379.html
) .

I think it should be no problem to wipe out the working version on
Pootle if people like you think yours is authoritative. (but you
might want to review the conflicts anyway). We would like just to
see Spanish translation team agree on this, though I don't know who
they are.

Thank you!
/Korakurider

>
> Regards.
> José L.

_______________________________________________
Squeakland mailing list
[hidden email]
http://squeakland.org/mailman/listinfo/squeakland

Dreyfuss Pierre-André (EDUM)

RE : Please help us for translation (An experimentalversion of OLPC image for non-XO environments)

In reply to this post by Yoshiki Ohshima-2

Hi,
I'll check the french translation

-------- Message d'origine--------
De: [hidden email] de la part de Yoshiki Ohshima
Date: lun. 02/06/2008 23:15
À: [hidden email]; Etoys lists; Localization mailing list
Objet : [Squeakland] Please help us for translation (An experimentalversion of OLPC image for non-XO environments)

Hello,

We are still trying to package the Etoys for non-XO environments,
along with, of course, for XO.

Currently, end of June is the *real* target of release day of non-XO
packages. The idea is to put relatively small set of languages in it,
and it looks like English, German and French are most complete ones.
(Thank you!)

But, for the US audience, (new world's) Spanish is also important.
And Portugese (br and pt) seems to be fairly good to me, but of course
I can't tell the quality right away.

Can you help us giving the information of the translation quality
and completeness for these languages? For the non-XO environment, we
will not have good font rendering mechanism other than what we have
now, so still the main target is the Latin languages.

The completeness doesn't have to be so close to 100%, as now many of
the phrases are not for Etoys (rather for the Smalltalk development).
But we would like to hear the opinions of the native speakers.

Please visit

https://dev.laptop.org/translate/

and see how far the translation are going in terms of quality and
coverages. Bert told us that there is an old style .translation file
for Spanish. We can convert it to .po file, but not sure again if we
should merge them to the site, as it might wipe out newly translated
phrases. Does anybody has good ideas how to proceed?

Thank you!

-- Yoshiki

_______________________________________________
Squeakland mailing list
[hidden email]
http://squeakland.org/mailman/listinfo/squeakland

_______________________________________________
Squeakland mailing list
[hidden email]
http://squeakland.org/mailman/listinfo/squeakland

Yoshiki Ohshima-2

Re: Please help us for translation (An experimental version of OLPC image for non-XO environments)

In reply to this post by José Luis Redrejo

José,

> Yoshiki, as you know, in LinEx we have all the strings translated for the etoys image (plus some other strings from some
> packages as drgeo, skeleton, etc.). Antonio Moreno has contacted with some of the etoys people, but the answers are not
> really clear about the way you want us to give you the translation. Our current image is totally based on the current
> olpc image , with changesets added from us, but we have all the classes and objects the etoys image has, so if you give
> the exact way you want us to extract the translation you can have the spanish translation finished (and tested &
> checked) right now.
> About the newly translated phrases, if our image has been used and tested for several years, with teachers fixing
> translations when some mistake was found, what's the problem on wiping out newly translated phrases that has not been
> checked for so many people?

Sorry for not being clear. Korakurider has responded in more detail
and we definitely proceed in the right way. As he wrote, if you think
it is fine (and perhaps it is more consitent to use phrases from one
team) to "wipe" the existing one (as probably most of them are
derivatives of LinEx work). Some people may get a bit of bad
feelings... So, please speak up before that.

Another thing is the difference of european Spanish and new world
Spanish. I don't know how much it really matters but the terminology
of new stuff like around computer can be quite different (ordenador
<=> computer^^;) What we can do is to save the current version just
as back up, upload the version from LinEx, and make sub language group
later.

-- Yoshiki

_______________________________________________
Squeakland mailing list
[hidden email]
http://squeakland.org/mailman/listinfo/squeakland

José Luis Redrejo

Re: Please help us for translation (An experimental version of OLPC image for non-XO environments)

2008/6/5 Yoshiki Ohshima <[hidden email]>:

José,

> Yoshiki, as you know, in LinEx we have all the strings translated for the etoys image (plus some other strings from some
> packages as drgeo, skeleton, etc.). Antonio Moreno has contacted with some of the etoys people, but the answers are not
> really clear about the way you want us to give you the translation. Our current image is totally based on the current
> olpc image , with changesets added from us, but we have all the classes and objects the etoys image has, so if you give
> the exact way you want us to extract the translation you can have the spanish translation finished (and tested &
> checked) right now.
> About the newly translated phrases, if our image has been used and tested for several years, with teachers fixing
> translations when some mistake was found, what's the problem on wiping out newly translated phrases that has not been
> checked for so many people?

Sorry for not being clear. Korakurider has responded in more detail
and we definitely proceed in the right way. As he wrote, if you think
it is fine (and perhaps it is more consitent to use phrases from one
team) to "wipe" the existing one (as probably most of them are
derivatives of LinEx work). Some people may get a bit of bad
feelings... So, please speak up before that.

Yes, I know, it's now on Antonio Moreno's shoulder. He's checking all the conflicts, I also took a look to them and some of them are clearly bad translations because of being literal dictionary translations, not using the context.

Another thing is the difference of european Spanish and new world
Spanish. I don't know how much it really matters but the terminology
of new stuff like around computer can be quite different (ordenador
<=> computer^^;)

Yes, but the LinEx translation begun in the smallland group (in 2002/03) where all these words were discussed between people from different countries, mainly Argentina and Spain, but also Peru, Mexico, etc. so we reached consensus on most of them. And, really, the doubts were very few words that (as far as I seen) are not in the conflict list that Korakurider has sent.

What we can do is to save the current version just
as back up, upload the version from LinEx, and make sub language group
later.

I hope that won't be needed, let's wait until Antonio Moreno check all the conflicts, but as I told you above, I guess there won't be much differences when the errors are fixed.

_______________________________________________
Squeakland mailing list
[hidden email]
http://squeakland.org/mailman/listinfo/squeakland