[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: A few typos in debian-installer (Was: Re: [D-I] Variable subsitution (level2))



On Sun, Sep 25, 2005 at 01:41:40AM +0200, Frans Pop wrote:
> On Sunday 25 September 2005 00:40, Jens Seidel wrote:
> > If the patch is accepted by most of you, I will extent it the next time
> > and will try to provide it in a more suitable form.
> 
> I have done a quick check of the patch and there are a great many good 
> corrections in it. However it also confirms my unhappiness with these 
> kind of integrated patches.

It seems you misunderstood my mail. It was never indented for you Frans
or Christian Perrier. It was intended for translators of debian
installer since I know that many read this list.

I think I will wait a week until I will contact all translators directly which
did not respond up to this time. Another week later I will commit all
obvious errors by myself.

> First of all, this patch, if submitted as one big patch, should have been 
> submitted as a minor bugreport against debian-installer, not sent to the 
> i18n list.

But the patch is similar to the one of zinosat@tiscali.it ([D-I]
Variable subsitution (level2)) and there was no remark that it should be
filled as bug.

> I am amazed (and jealous) at your knowledge of all these languages; 

Yeah, it's great, isn't it :-))
No, all issues ware collected by a script, I just selected a few of the most
obvious errors.
Since I cannot guarantee that I'm right all the time I need permission
of the translators. I would never commit a change if I'm not sure about
it.

> however personally I would not feel confortable in committing some of the 
> changes you've made for languages I don't speak as I cannot judge your 
> knowledge of the language nor, if that is what you used, the correctness 
> of your dictionary and thus, the correctness of the changes.
> Example: the multiple changes of s/dosyian/dosiyan/ in ku.po.

Of course. As I already said, the patch was indended for translators of
the corresponding language, so only the dutch part is for you :-))

> Also, the patch contains some obvious errors, or at least changes that 
> need to be checked carefully:
> - in ru.po a string is deleted;

Right, but only
#~ msgid "An error occured during the previous operation"
which is a comment and contains a typo (occured), which is useless to
fix but is always triggered by my script.

> - in pt_BR.po 'bootstrap' is wrongly corrected to 'boostrap' (line 8060).

Wrongly? Mmh, probably yes.
I assumed "boostrap" is the translation of "bootstrap" since the first
occurred 4 times but the second only a single time.

> I would probably be happy to commit all the really obvious changes like 
> s/etx2/ext2/ or devconf/debconf. However, as they are mixed with all the 
> other changes I cannot judge, I doubt I would make the effort to filter 
> them out. Maybe someone else will though...

That's really my task, it was never planned that another guy (except the
corresponding) translator does this.

> If the patch had been separated for the different languages, I probably 
> would have taken the trouble to forward them to the individual 
> translators responsible with the request to check and apply if correct 
> (and maybe commit the really obvious ones myself).

This was my aim with my mail.

> By creating one "big" patch, you're basically putting the responsibility 

Right, but it was only a very very minor one. I wasn't sure long time
whether it's worth to send it, since it only fixes a very few errors.
It should be no problem for me (except time and maybe also more false
possitives) to create a similar patch which is by a factor of 10 or 50
larger.

> to check and apply the patch on a single person. IMO this decreases the 
> chance that someone will take the trouble to look at it and thus 
> increases the chance that your work will be wasted.
> By investing a little bit extra time on your side and splitting the patch 
> into different types of fixes and separate patches for the each language, 
> this would have been avoided.
> 
> Again Jens, I very much appreciate your work (and have done in the past 
> for e.g. the Release Notes), but please spare a thought for the people 
> that have to review/apply/distribute your patches.

OK.
 
> P.S. The patch for Dutch looks fine :-)

Great. But I hope you apply at least this part.

Jens



Reply to: