Re: PO-based Documentation Translation

[Christian Rose <menthos gnu org>]

tor 2003-10-02 klockan 09.15 skrev Ismael Olea:
> > But I think automatically classifying anything we've done so far or
> > currently do as inferior, and everything else done by others as
> > automatically more advanced, more "proved" or superior by any other
> > non-factual means or not backed up statements serves no good purpose at
> > all.
> 
> Why do you say that?

Because I 1) don't believe that such statements have to be automatically
true and 2) I certainly don't get encouraged by them.


> I don't remember to have said anything about «inferior».

Oh, please read your mails again. You're calling us amateurs, while
volunteer translation efforts really have nothing to do about bad
quality or amateurism.

You're calling anything big companies and professional translation
businesses do as "the real world", essentially saying that we in the
free world aren't living in it, but instead living in an imaginary one.

You're saying that we should focus on proprietary, expensive tools while
you should certainly know that we cannot for very good and both
practical and fundamental reasons use those for translating free
software.

You call those tools "proved", as if gettext and po format hasn't
"proved" itself during its successful use during many years, and without
giving any details about what you mean by that and why you think so.

You're saying that we "should meet any professional translator in our
language and ask them what they use" and that it would be "as easy as
that", essentially ignoring that there are more complex issues involved
and that most professional translators aren't working with free software
translation to begin with, and hence don't need to take into account the
software used in that area. Instead they often just work with the format
that their client requires them to use and deliver in, just like we do
with po format.

You're citing professional translation congresses as a reference for how
irrelevant po format is, totally ignoring that the professional
translation industry much like the software industry in general
traditionally has been focused around proprietary software for the most
dominant proprietary platform, and not free software or the less common
platforms. We're having a hard time getting our free software
translators even to visit the GUADEC conference due to the usual high
costs of travelling to get there, so it should come as no surprise to
anyone that free software representatives are hardly represented at
translation congresses at all, and hence not their tools and methods
either.

You're claiming that we don't know anything about the tools used in the
professional translation world, while you cannot know if that is true.
As an example, in the year 2000 I invited the professional Swedish
translator Gudmund Areskoug, who at that time already had 10 years of
experience of professional software translation, to join the lists of
our Swedish team and share his experience and knowledge. He's been very
kind in doing so and given us lots of insight in many professional tools
he has used and their benefits and drawbacks. He was also one of the
people that convinced me to use TM:s on a larger scale, by the way. But
not once has he claimed that our work is inferior in quality and that
free software translations are irrelevant, or that proprietary tools are
better just because they happen to be proprietary tools.


> What I try to say is we clearly can get better. GTP has solved the
> main problem: manpower. Having GTP-compromised human resources you
> have the key part. The only essential one. And looking at the stats,
> with a great level of success.
> 
> In the other hand, I think is an exercise of modesty to recognise we
> still need to learn some things. The more I know about my mother
> language, the more I know I'm almost un illiterate person. Nothing to
> say about my stinking English O:-)

I'm certainly also open to new ideas and to see how others are doing
things (if anything my track record should show so). That's also what I
clearly stated in the very first lines of my first reply.

It's just that I don't agree with your comments about our translation
work being amateurish and not for real, and that proprietary tools are
better because they are just that; proprietary tools.


[...]
> And should I talk about our indiscutible strengths? About how we
> multiply our productivity sharing softweare, resources and experience as
> anyone can? Should I need to repeat it to you? Don't think so. You are
> who makes it real. You feel it on your own veins.

I certainly don't feel it that way. I don't translate into Spanish but
instead into Swedish, a small language just used by about 9 million
people, and not hundreds of millions.
I've spent almost every free time I've had for many years in personally
making sure the support for Swedish is on top, and yet I can still count
the number of occurances on one hand that someone from outside our small
Swedish development community has spontaneously given any appreciation
for that work. So I'm certainly not spoiled with appreciating comments,
and I certainly don't need my work to be belittled.

Also, there are a lot more languages with even less numbers of speakers
and translators doing an extraordinary amount of work with those, and I
also don't think they appreciate being told that they are amateurs not
knowing how things should be really be done.


> And finally, please don't think I want to change all the way you (we)
> work from one day to another. No. Mainly because there is no software
> available today. But maybe it'll be tomorrow. And I wish we'd be ready
> to be the best translation service of the world, without the lesser of
> the doubts.

So why are we wasting time on this thread? What needs to be urgently
done is to have a way for translators to easily translate the docs we
already have now and using the tools we have right now. For better and
for worse, that means po format for now.
We can always migrate to XLIFF at any time later (or even use it in
parallell to po) when the time is ready (and not to forget when our
developers are ready for that!), but in the mean time any extensive
debate about that like this one is just mostly hypothetical and taking
our time away from our current work and most pressing issues right now.


Christian