Sorry I couldn't answer this earlier.

On 30/12/2006, at 2:31 AM, Christian Lohmaier wrote:

On Mon, Dec 18, 2006 at 04:54:11PM +1030, Clytie Siddall wrote:
On 18/12/2006, at 3:33 AM, Christian Lohmaier wrote:
On Sun, Dec 17, 2006 at 10:38:18AM +0100, Johan Beckers wrote:
Clytie Siddall schreef:
On 17/12/2006, at 1:43 AM, Konrad Stobiecki wrote:
[...]

Well, theres still the image of the worldmap inside that button, so
even
when you don't know what "native language" means, you could guess from
the image.

Not necessarily. It could be "Our products are distributed worldwide"
or "Get our product from different locations" or even "The best
product in the world".

Well, wouldn't it still be a good guess if you want onother language
(besides the last one, but I don't really think anybody would connect
the map with that one)?

Do you have another suggestion for a different iconography?

A button with either background or foreground showing "OpenOffice,org" in as many different main scripts as possible. Or make it a language bar.

I don't like detecting of browser langauge or language by IP or
something for the content of a webpage. It might work well for
mirror-selection (select a mirror geographically near you), but
doesn't
work very well for website content.

I don't agree. Content negotiation is designed especially for website
content, and language choice is one of its strongest features.

That's not my point. It is not that it would not work technically, but
that it doesn't make sense in this case.

I don't agree. People recognize and read information in their known language. That language is exactly what they're looking for when they search for information. The fact that we have more than one NLP using the same language is not a disadvantage: it's a strength. We should have a language button or bar which takes the user either to the sole NLP representing their language, or to a list of links in their language describing the different NLPs available.

And in any case, content negotiation can take the user to the correct NLP. We can filter all four zh, zh_CN, zh_TW and zh_CN (Chinese, Chinese mainland, Chinese Taiwan and Chinese Hong Kong) to a single Chinese NLP, or to their own NLPs if they exist.

This way the user actually gets information s/he can understand.

the appropriate NL project. Every NL project could place a link to:
OpenOffice.org International, or OpenOffice.org Worldwide.

The big problem is that by far not every nl-project provides the info the english pages have. You can always visit your NL-project directly
using your <language-code>.openoffice.org - a scheme very widespread
across all possible websites.

But not necessarily known by the user.

Yes, but you miss the real point again it seems :-)

The point is: Native-lang projects have varying focus, you cannot serve
sites depending on language since the nl-projects don't all offer the
same content.
The only place where this could work as of today is the why.ooo pages.

NLPs may have varying focus, but we all have general information explaining what OOo is and how to get it. Without that information, the user who doesn't understand the main page can't access OpenOffice.org.

[...]
Indeed, but all of us provide a lot of OOo-specific information in
the user's language, including a general introduction, download links
and help info linked to further info in English. We are the most
effective entry point for non-English speakers.

I don't deny that, but redirecting the user to a different page than
www.ooo by any of the fancy language-detections things is not the way to
go in my opinion.

What would maybe work as is to dynamically add a link in the user's
language at the top of the native-language page. (saying approx: "Visit
the pages of the <language> native-lang project")

But the only technology we could use for this is javascript...

How about a language bar across the top of the main page and or NLP page?

If the NLP main page were in Arabic, which I can't read at all:

http://native-lang.openoffice.org/

I wouldn't get as far as the bottom of the page, which starts to list the language codes but only lists the first few.

We need something which is first-sight for users coming to OOo and takes them to information they can understand.

Otherwise, every day we are losing users who can't read English, and especially those who can't read a Roman alphabet.

The OpenOffice.org wiki does a much better job of providing information to non-English speakers: note the Languages box. It doesn't have to be that big, but it does need to contain something the user understands.

http://wiki.services.openoffice.org/wiki/Main_Page

And using content negotiation, the wiki displays to me in my own language, which empowers me in using it.

from Clytie (vi-VN, Vietnamese free-software translation team / nhóm Việt hóa phần mềm tự do)
http://groups-beta.google.com/group/vi-VN


Attachment: PGP.sig
Description: This is a digitally signed message part

Reply via email to