Mailing List Archive

Viktor Horvath wrote:

> Hello,
>
> I dropped some lines about this idea about two months ago. Now, I
> wrote a more detailed plan which is available from
>
> http://www.steckenpferde.de/wiki-translations/
>
> and I also made a reference on Meta's "Talk:Proposals for new
> projects" page. I'm not calling for a "proposal" as it can be
> integrated into an existing project like wikisource. But what can I do
> to promote this idea further?

Much of this fits in with my own vision of Wikisource. I have myself
raised the question of co-ordinated side-by-side boxes (or panes) to
facilitate translations, and footnoting. Two or three such panes seem
workable, more would probably result in unduly narrow columns.

Your edition management and pdf conversion ideas are interesting, but
should probably be deferred until after we have accomodated translations
and annotations.

All we need is for some developer to be sufficiently inspired by the
idea. Until then we just need to be patient, and limit ourselves to the
occasional friendly reminder to make sure that the idea does not get
forgotten.

Ec

Hello Ray,

>Your edition management and pdf conversion ideas are interesting, but
>should probably be deferred until after we have accomodated translations
>and annotations.

Of course, the two-pane layout and annotations are the most important. The
pdf conversion can wait, yes. But I think the edition management is
essential; at least, the possibility of two translations to coexist is
essential, as one cannot imagine that all the translators agree on a
specific translation.

>All we need is for some developer to be sufficiently inspired by the idea.
>Until then we just need to be patient, and limit ourselves to the
>occasional friendly reminder to make sure that the idea does not get
>forgotten.

OK, thanks! It's much more comforting not to wait alone :-)

Have a nice Christmas,
Viktor.

Hiho,

Viktor Horvath wrote:

>>Your edition management and pdf conversion ideas are interesting, but
>>should probably be deferred until after we have accomodated translations
>>and annotations.
>
> Of course, the two-pane layout and annotations are the most important. The
> pdf conversion can wait, yes. But I think the edition management is
> essential; at least, the possibility of two translations to coexist is
> essential, as one cannot imagine that all the translators agree on a
> specific translation.

As much as people can't agree on the content of _one_ encyclopedia
article? Sorry, I have to object. The genial thing about wikis is that
it allows people to work in cooperation on one translation and improve
it as best as possible.

Wikipedia is the proof that this concept works and I can't imagine why
it shouldn't work for translations, too. Implementing features to create
several translation versions means wasting the central advantage of a
wiki: collaboration.

>>All we need is for some developer to be sufficiently inspired by the idea.
>>Until then we just need to be patient, and limit ourselves to the
>>occasional friendly reminder to make sure that the idea does not get
>>forgotten.
>
> OK, thanks! It's much more comforting not to wait alone :-)

In my experience, waiting for the perfect software is the wrong way to
go. Nobody is going to program stuff for a project which might or might
not take off. There are more pressings things needed for the existing
ones...

So...start now in wikisource, with the possibilities which exist -
translations can be done with the table markup, putting text and
translation in small parts beside each other. And if there's content and
people working on translations, you will probably find an enthusiast who
improves the software according to what is needed.

greetings,
elian

Hi,

>As much as people can't agree on the content of _one_ encyclopedia article?
>Sorry, I have to object. The genial thing about wikis is that it allows
>people to work in cooperation on one translation and improve it as best as
>possible.

well, here we disagree. To me, an encyclopedia article and a translation are
inherently different. The one's a definition and rather a science, the other
one rather an art. Just because one thing works, you can of course *try* it
with other things, too, but you can't possibly say: It *will* work.

Did you ever do a translation of a *belletristic* work? You see, with
technical articles, it would be probably no problem to agree on a simple,
understandable translation. With literature and philosophy, it's impossible.
There are strict translations, free translations, congenial translations.
While on each of these fields people can collaborate and cooperate, they're
as different as rock music and classical music, or as Dutch painting of the
16th century and Picasso. And every style has its right to exist! You can't
judge which is the best one, as you can't find an objective winner between
music styles. It's just a matter of taste. That's the genial thing about
art. At least in my opinion. Whereas an encyclopedia article should contain
only checkable facts, and you can well discuss about facts.

>Implementing features to create several translation versions means wasting
>the central advantage of a wiki: collaboration.

No. There can be lots of people who collaborate for one style of
translation. If that were wasting of ressources, wouldn't maintaining Wikis
for languages which only very few people speak be a much bigger waste? The
authors could be writing for more widespread languages... but they don't,
and they're perfectly right: They can decide where to spend their time. The
same holds for translations.

>In my experience, waiting for the perfect software is the wrong way to go.
>Nobody is going to program stuff for a project which might or might not
>take off. There are more pressings things needed for the existing ones...

Well, I don't blame someone if he doesn't want to do so or has no time. I
have no time either, at least not at the moment, as I have lots to do for
other free projects... I just hoped to find someone who is also fascinated
by an idea like this. You're right, starting is better than waiting... but,
if I had to start something now and had enough time, I'd probably try to
patch the Wiki source or do the whole thing manually by coordinating people
in good old mailing lists like in the times before...

Greetings,
Viktor.

I don't think the line to draw is between translation and encyclopedia
writing. It is more between art and science.

Translation of a technical document, or legal, mathematical, or
scientific for that matter, would rather be straightforward. People
can agree on things easily. There could even be a set of glossaries
or standards they can refer to.

Translation of a literary work could be more difficult to agree on.
But this is not impossible, I think. For example, a short poetry
could be translated in many different ways and they would be equally
valid. But if people know that they have to agree on one version,
they will come to agreement somehow. Since it is under a free
license, people can publish different versions elsewhere. Or, for fun
and out of respect to each other or diverse views, they may come to agree
to make different versions, each occupying one revision in the database,
and offer a link to those revisions from the current version so that
readers can compare and enjoy.

When the work to be translated is large in volume, it needs a lot more
coordination. It is like creating a wikibook or a set of wikipedia
articles. They have to decide on certain editorial policies. There
are many different policy options, and some of them would be equally
valid. But again, because they know there is only one set of options
to be chosen, they can come to agreement.

I have translated various documents on wiki between Japanese and English,
some more like an essay others more like a set of rules. My general experience
is that wiki process is helpful. Translating something alone is not as fun.

I do think, though, that in case of literally works, offering multiple translations
is something valuable, something in line with the neutral point of view policy
that we value so much. So it is nice to have some way to help development of
multiple versions of translations. But since Mediawiki preserve
all past revisions in the database, this can be done. Additional feature is
definitely nice, though not necessary.

Regards,

Tomos

----- Original Message -----
From: Viktor Horvath <duke586@hotmail.com>
To: foundation-l@wikimedia.org
Sent: Tue, 21 Dec 2004 00:58:47 +0100
Subject: Re: [Foundation-l] Wiki Translations of Greek/Roman/XYZ classics

When this was first raised a while ago i thought it was a great idea,
however i think the approach suggested is just plain wrong. As it has
been pointed out, it is impossible to have the perfect translation of
Don Quijote de la Mancha or The Republic etc., equally it is far too
difficult and generaly unwiki to try and have multiple versions
together; which to me suggests the obvious solution that we need
several (completely seperate) different versions of each work. For
example we could have en.wikifoo.org/wiki/Don Quijote (literal),
en.wikifoo.org/wiki/Don Quijote (simple english) etc. etc. at the
begining of the translation there would be a page that outlines the
basic ideas to be followed in translating the work (ie. similar to the
translators note at the begining of most classics) and then that can
be used like the NPOV policy, as a basis for comprimise and
collaboration.

I would also suggest that until such a time as there is obvious
disagreements between those who are translating literaly, and those
who arent, there should only be one version (since it will make it
easier to write a literal, or poetic or whatever translation, if you
already have a rough translation to start with.)

I also think that this would be another place where my suggestion of a
reference tab (to go with the main article and talk tabs) for
foot/endnote would be of great use.

paz y amor,
[[User:The bellman]]

--
hit me: robin.shannon.id.au
jab me: saudade@jabber.zim.net.au

This work is licensed under the Creative Commons
Recombo Plus License. To view a copy of this license, visit
http://creativecommons.org/licenses/sampling+/1.0/

Robin Shannon wrote:

>I would also suggest that until such a time as there is obvious
>disagreements between those who are translating literaly, and those
>who arent, there should only be one version (since it will make it
>easier to write a literal, or poetic or whatever translation, if you
>already have a rough translation to start with.)
>
>I also think that this would be another place where my suggestion of a
>reference tab (to go with the main article and talk tabs) for
>foot/endnote would be of great use.
>
>
While I can appreciate the occasional need for several translation
versions, I don"t think that it is a realistic prospect in the near
future. Before there is a second translation of a work there needs to
be a first one. We can probably put that discussion on the shelf until
the circumstances present themselves.

Ec

Ray Saintonge wrote:
>While I can appreciate the occasional need for several translation
>versions, I don"t think that it is a realistic prospect in the near future.
>Before there is a second translation of a work there needs to be a first
>one.

Yes. But you won't get one complete translation if you don't make clear
which type of translation it should contain, as after the first translated
pages, at least two parties will start argumenting. I think Robin's idea is
a very good one: One could start with a first translation and some explicit
policy, begging everyone who wants to follow a completely different approach
to start another one, which is linked by the main page. This way, "edition
management" is very possible.

What's missing? Two-pane design *for translating* can be achieved by having
two browser windows side by side. You only have to use the same original
text, for example by linking to the Perseus Project. If you follow their
page numbers (or even set links to the individual Perseus pages into the
translation, if that's permitted), you can even *read* bilingual versions.
For footnotes, one can start by using the discussion page.

I agree, this way one can at least start and see where there is a real need
for software features. Now let's start a project and find some translators
:-)

Have a nice Christmas everybody,
Viktor.

Elisabeth Bauer wrote:
> As much as people can't agree on the content of _one_ encyclopedia
> article? Sorry, I have to object. The genial thing about wikis is that
> it allows people to work in cooperation on one translation and improve
> it as best as possible.
>
> Wikipedia is the proof that this concept works and I can't imagine why
> it shouldn't work for translations, too. Implementing features to create
> several translation versions means wasting the central advantage of a
> wiki: collaboration.

I agree completely. And there are easy ways to handle particularly
contentious bits if absolutely necessary in some context, but
generally I think the right way forward would be to find a way to
improve both versions until all parties are happier than with the
original versions.

> In my experience, waiting for the perfect software is the wrong way to
> go. Nobody is going to program stuff for a project which might or might
> not take off. There are more pressings things needed for the existing
> ones...

*nod* In a voluntary community, there's always a chicken-and-egg
problem, and the best way to solve it is just: move forward as best
you can with the tools available. This will itself generate the
demand and the expertise necessary to make those tools better.

--Jimbo

Viktor Horvath wrote:
> well, here we disagree. To me, an encyclopedia article and a translation
> are inherently different. The one's a definition and rather a science, the
> other one rather an art. Just because one thing works, you can of course
> *try* it with other things, too, but you can't possibly say: It *will*
> work.

I don't think the determination of what goes into an encyclopedia
article is a "science" in that sense. It's as much of an "art" as
finding a good translation.

> Did you ever do a translation of a *belletristic* work? You see, with
> technical articles, it would be probably no problem to agree on a simple,
> understandable translation. With literature and philosophy, it's
> impossible. There are strict translations, free translations, congenial
> translations. While on each of these fields people can collaborate and
> cooperate, they're as different as rock music and classical music, or as
> Dutch painting of the 16th century and Picasso. And every style has its
> right to exist! You can't judge which is the best one, as you can't find
> an objective winner between music styles. It's just a matter of taste.
> That's the genial thing about art. At least in my opinion. Whereas an
> encyclopedia article should contain only checkable facts, and you can well
> discuss about facts.

I agree with you to *some* extent, but I think the point is:
collaboration works well when people have a clear, objective and
agreed upon goal. At wikipedia, for example, we have a shared
understanding of what it means to have a neutral encyclopedia article.
That same kind of shared undersatnding is going to be necessary for a
wiki translation project to be successful.

This is not about saying that some style has no right to exist. If
someone wants to do a translation of Shakespeare into Japanese haiku,
hey, great, I hope they produce a great work of art. And if a group
of people can come to a shared understanding of what that means, such
that they can work together on it productively, that's great too.

But then they should *within that context*, work on a *single*
collaborative version, and make it conform to the stated goal as well
as they can, rather than doing 100 scattered random diffuse separate
works. Do you see what I mean? This is elian's point about what
makes wiki work.

> Well, I don't blame someone if he doesn't want to do so or has no time. I
> have no time either, at least not at the moment, as I have lots to do for
> other free projects... I just hoped to find someone who is also fascinated
> by an idea like this. You're right, starting is better than waiting...
> but, if I had to start something now and had enough time, I'd probably try
> to patch the Wiki source or do the whole thing manually by coordinating
> people in good old mailing lists like in the times before...

:-)

I think that's great.

--Jimbo