We haven't decided yet exactly what to make the new www.wikipedia.org project front page look like, but it seems pretty well decided that we do want to move the English wiki to en.wikipedia.org and set up some sort of multilingual and/or browser-language-setting-sensitive intro page.
(See: http://meta.wikipedia.org/wiki/What_to_do_with_www.wikipedia.org )
If there's no objection, I'm going to switch the URL redirecting around tomorrow, so that www.wikipedia.org/* points to en.wikipedia.org/* rather than the other way 'round. A new intro page can later be set up at convenience.
Warning: login cookies won't automatically follow over, so everyone using the English wiki will have to log in afresh after the switch even if you've got "remember my password" checked.
-- brion vibber (brion @ pobox.com)
On 10/10/02 3:26 AM, "Brion VIBBER" brion@pobox.com wrote:
We haven't decided yet exactly what to make the new www.wikipedia.org project front page look like, but it seems pretty well decided that we do want to move the English wiki to en.wikipedia.org and set up some sort of multilingual and/or browser-language-setting-sensitive intro page.
(See: http://meta.wikipedia.org/wiki/What_to_do_with_www.wikipedia.org )
If there's no objection, I'm going to switch the URL redirecting around tomorrow, so that www.wikipedia.org/* points to en.wikipedia.org/* rather than the other way 'round. A new intro page can later be set up at convenience.
I object strenuously. This is not a slam-dunk good idea at all. Please hold off at least a good while before doing this.
On 10/10/02 2:15 PM, "Jimmy Wales" jwales@bomis.com wrote:
The Cunctator wrote:
I object strenuously. This is not a slam-dunk good idea at all. Please hold off at least a good while before doing this.
O.k., let's wait.
What are the objections?
The primary objection is that this hasn't really been discussed. That said:
1) There isn't a clear picture of what overall policy this fits in with;
2) Whether whatever that overall policy is well-thought out and correct;
3) What the consequences of the change from a usability perspective have not been delineated;
4) "A new intro page can later be set up at convenience" is probably not how we want to do this;
5) There are real, concrete benefits to having a default and preferred interface/language;
In other words, the pros and cons of such an act should be explicitly and clearly listed. There are many reasons that www.google.com, www.dmoz.org, etc. (which all have multilanguage settings) have English as the default, and it's not just that the servers are US-based).
Once that's done, we can weigh priorities.
I personally think a better focus right now for the developers is to work on maximum integration of the different language wikis. If the backend is better integrated, frontend issues become easier to deal with.
A big question that we are in the process of resolving now is whether we want to think of Wikipedia as a single project that has multiple translations, or as a bunch of largely independent projects specific to particular nations and language sets. I think the first conception is healthier and more productive over the long term. We really should think about this issue before we take broad actions that touch upon it.
The Cunctator wrote:
The primary objection is that this hasn't really been discussed.
It's been nothing _but_ discussed for months.
That said:
- There isn't a clear picture of what overall policy this fits in with;
Naming standards: every encyclopedia wiki in the project is at {languagecode}.wikipedia.(com|org) except the English one.
- Whether whatever that overall policy is well-thought out and correct;
Well, it sounds rather reasonable to me.
- What the consequences of the change from a usability perspective have not
been delineated;
URLs are deliberately being preserved. User interface isn't any different. One-time cookie change requires users to push the "login" button.
Front page not yet changed; new front page may make some people want to change their bookmarks if they wanted the English encyclopedic front page specifically.
- "A new intro page can later be set up at convenience" is probably not how
we want to do this;
Oh?
- There are real, concrete benefits to having a default and preferred
interface/language;
Which are?
In other words, the pros and cons of such an act should be explicitly and clearly listed. There are many reasons that www.google.com, www.dmoz.org, etc. (which all have multilanguage settings) have English as the default, and it's not just that the servers are US-based).
Once that's done, we can weigh priorities.
I personally think a better focus right now for the developers is to work on maximum integration of the different language wikis. If the backend is better integrated, frontend issues become easier to deal with.
If I'd just spent the 30 seconds to make the change without telling you, I'd have a lot more time to focus on such things. ;)
A big question that we are in the process of resolving now is whether we want to think of Wikipedia as a single project that has multiple translations, or as a bunch of largely independent projects specific to particular nations and language sets. I think the first conception is healthier and more productive over the long term. We really should think about this issue before we take broad actions that touch upon it.
"Translations" is an odd choice of words.
The *user interface of the program* is 'translated' from a central source, but the encyclopedia articles aren't. Ideally they should all contain the same (maximal quality, maximal amount of) information by cross-pollination: new material added in any one language can be taken over to the others. That's not a one-way street, and doesn't have to be achieved by "translation".
-- brion vibber (brion @ pobox.com)
On 10/10/02 4:10 PM, "Brion VIBBER" brion@pobox.com wrote:
The Cunctator wrote:
The primary objection is that this hasn't really been discussed.
It's been nothing _but_ discussed for months.
Please reference.
The Cunctator wrote:
On 10/10/02 4:10 PM, "Brion VIBBER" brion@pobox.com wrote:
The Cunctator wrote:
The primary objection is that this hasn't really been discussed.
It's been nothing _but_ discussed for months.
Please reference.
http://www.nupedia.com/pipermail/intlwiki-l/2002-March/000360.html http://www.nupedia.com/pipermail/wikipedia-l/2002-August/003979.html http://meta.wikipedia.org/wiki/Regarding_URLs http://meta.wikipedia.org/wiki/What_to_do_with_www.wikipedia.org
And it ain't been _done_ yet, just _discussed_.
-- brion vibber (brion @ pobox.com)
On 10-10-2002, Brion VIBBER wrote thusly :
The Cunctator wrote:
A big question that we are in the process of resolving now is whether we want to think of Wikipedia as a single project that has multiple translations, or as a bunch of largely independent projects specific to particular nations and language sets. I think the first conception is healthier and more productive over the long term. We really should think about this issue before we take broad actions that touch upon it.
"Translations" is an odd choice of words. The *user interface of the program* is 'translated' from a central source, but the encyclopedia articles aren't. Ideally they should all contain the same (maximal quality, maximal amount of) information by cross-pollination: new material added in any one language can be taken over to the others. That's not a one-way street, and doesn't have to be achieved by "translation".
Coordination, cooperation, confederation are terms that I think should describe the whole collective project of _Wikipedias_. You (the Cunctator) are entering quick sands area. Too much, too hasty and careless integration into "a single project" can be disastrous. The Spanish Wikipedia fork has not still been reverted...
We are looking forward to shifting to the Phase III software and working out ways to exchange content. I am sure Polish Wikipedians would regard it as a big disincentive to be reduced to mere translators of English (other languages?) articles. I think we have created a relatively small but precious Wikipedia with some remarkably good articles. We are, of course, willing to share our content with other Wikipedias.
Regards, [[Kpjas]]
On 10/10/02 4:29 PM, "Krzysztof P. Jasiutowicz" kpj@kki.net.pl wrote:
I think we have created a relatively small but precious Wikipedia with some remarkably good articles. We are, of course, willing to share our content with other Wikipedias.
It's this balkanization that I fear. I see there being one Wikipedia, which has multiple languages in it. Not separate Wikipedias run by separate groups that might or might not be "willing to share our content" with "other Wikipedias".
On 10-10-2002, The Cunctator wrote thusly :
On 10/10/02 4:29 PM, "Krzysztof P. Jasiutowicz" kpj@kki.net.pl wrote:
I think we have created a relatively small but precious Wikipedia with some remarkably good articles. We are, of course, willing to share our content with other Wikipedias.
It's this balkanization that I fear. I see there being one Wikipedia, which has multiple languages in it. Not separate Wikipedias run by separate groups that might or might not be "willing to share our content" with "other Wikipedias".
I agree with you in this that Wikipedia is one project there should be common goals and common means to achieve these goals. There should be common software base and good interlinking. But policies, in Wikipedias realm, oughtn't be enforced - they should be discussed, negotiated and agreed on, with room for minor regional differences. Maybe it is misunderstanding on my part and I don't know what you are implying but I must stress that the Polish Wikipedia and her contributors feel proud to part of the project and want to continue working for the common good of the project.
Not separate Wikipedias run by separate groups that might or might not be "willing to share our content" with "other Wikipedias".
Oh! Come on. I meant the difference being a colony and a sovereign international partner.
For the discussion to be productive please state : 1. what (other than domain names) signs of balkanization are there ? 2. what are you proposals to integrate other Wikipedias into one big Wikipedia so that avoid balkanization ?
Regards, Kpjas.
On Fri, 2002-10-11 at 04:05, Krzysztof P. Jasiutowicz wrote:
For the discussion to be productive please state :
- what (other than domain names) signs of balkanization are there ?
- what are you proposals to integrate other Wikipedias into one big
Wikipedia so that avoid balkanization ?
I've written about both of these in previous emails, so to reiterate:
1. From a procedural standpoint: separate username spaces; separate Recent Changes spaces; the wikipedia-l/intlwiki-l separation; different grades of software; imperfect or non-existent interlanguage linking. Many of these are actively being improved even as we speak, some of them are not.
On 11-10-2002, The Cunctator wrote thusly :
On Fri, 2002-10-11 at 04:05, Krzysztof P. Jasiutowicz wrote:
For the discussion to be productive please state :
- what (other than domain names) signs of balkanization are there ?
- what are you proposals to integrate other Wikipedias into one big
Wikipedia so that avoid balkanization ?
I've written about both of these in previous emails, so to reiterate:
- From a procedural standpoint: separate username spaces; separate
Recent Changes spaces; the wikipedia-l/intlwiki-l separation; different grades of software; imperfect or non-existent interlanguage linking. Many of these are actively being improved even as we speak, some of them are not.
We (other Wikipedias/Wikipedists) are not able to change them. It is therefore not a sign of balkanization but only deficiencies of design, implementation and/or execution of plans and policies.
From a conceptual standpoint: seeing Wikipedia as largely separate
projects specific to individual languages and nationalities.
- The procedural improvements I propose should be obvious from what I
consider to be problem areas.
From a conceptual standpoint, I think it's important to envision
Wikipedia as a single project that happens to have content in an admixture of languages (much as the World Wide Web is a single network that happens to have sites in an admixture of languages--the world does much better with a single geo-neutral Web instead of an American Web, Chinese Web, Polish Web. I am also troubled by the forces which are trying to balkanize the Web).
I take your point but what exactly are you proposing to do to avoid this ?
I thank you for discussing this in my native tongue. My Polish is worse than minimal.
You're welcome. I assure you my Polish is not perfect either.
Please do not give an impression there are some inner trends (in Wikipedias) to balkanize the whole project.
Regards, Kpjas.
Krzysztof P. Jasiutowicz wrote:
But policies, in Wikipedias realm, oughtn't be enforced - they should be discussed, negotiated and agreed on, with room for minor regional differences.
I can agree with this, so long as the differences really are _minor_.
One thing that's true is that people from different cultures are likely to behave differently, to an extent, in some wiki-relevant ways. Americans are more libertarian, with less respect for authority, than most other cultures. The Japanese are more hierarchical, with greater respect for authority, than most other cultures. I hesitate to characterize the differences between Americans and Europeans, but there are differences to be sure.
"Be bold in updating pages" comes naturally to Americans. To Japanese, it might seem better to extensively discuss a change on the /Talk pages first.
There's no reason for detailed policies to be decided on a top-down basis. If the Japanese prefer to extensively and politely and slowly discuss changes before they make them, then that's what they should do. I personally (as an American) think that's a slow way of getting things done, but they might consider our way reckless. (We're in the land of stereotypes here, I know, but flow with me to see my general point.)
As long as NPOV, openness to newcomers, encyclopedic writing are followed, then lesser social norms will naturally vary to an extent.
--Jimbo
Krzysztof P. Jasiutowicz wrote:
I am sure Polish Wikipedians would regard it as a big disincentive to be reduced to mere translators of English (other languages?) articles. I think we have created a relatively small but precious Wikipedia with some remarkably good articles.
I think Cunc will gladly take back the word "translation". I think he was just speaking loosely there.
The way I see it is that the Wikipedia is one project with many languages. The individual language wikipedias are not separate projects, but part of the large family. But neither are they merely _translations_.
We are united by the fundamental goals: NPOV, openness, encyclopedic writing.
I am opposed to the idea, for example, that an article in Spanish about the Spanish-American war should be in any way fundamentally different from an article in English about the same war. There's no need for them to be translations of each other, of course. But they should contain the same information, and both should be free of bias.
Multilingual people can read both and notice discrepancies and bring the articles into line with each other.
--Jimbo
At 04:16 AM 10/11/02 -0700, Jimbo wrote:
Krzysztof P. Jasiutowicz wrote:
I am sure Polish Wikipedians would regard it as a big disincentive to be reduced to mere translators of English (other languages?) articles. I think we have created a relatively small but precious Wikipedia with some remarkably good articles.
I think Cunc will gladly take back the word "translation". I think he was just speaking loosely there.
The way I see it is that the Wikipedia is one project with many languages. The individual language wikipedias are not separate projects, but part of the large family. But neither are they merely _translations_.
We are united by the fundamental goals: NPOV, openness, encyclopedic writing.
I am opposed to the idea, for example, that an article in Spanish about the Spanish-American war should be in any way fundamentally different from an article in English about the same war. There's no need for them to be translations of each other, of course. But they should contain the same information, and both should be free of bias.
An interesting example. I would expect an article in Spanish to have more to say about the effects on Cuba and Puerto Rico, and maybe less on Hearst; an article in Tagalog would probably spend more space on the results for the Philippines. Yes, in theory space is infinite. In practice, we cut articles up when they get too long, so there are choices to make on what is in the article and what's moved to an external link.
Multilingual people can read both and notice discrepancies and bring the articles into line with each other.
More to the point, given finite human energy and the severe flaws in current machine translation techniques, I don't expect anyone to translate all the stubs on US counties into Polish any time soon.
On 10/10/02 4:10 PM, "Brion VIBBER" brion@pobox.com wrote:
The Cunctator wrote:
The primary objection is that this hasn't really been discussed.
It's been nothing _but_ discussed for months.
That said:
- There isn't a clear picture of what overall policy this fits in with;
Naming standards: every encyclopedia wiki in the project is at {languagecode}.wikipedia.(com|org) except the English one.
Naming standards is not a strong enough policy to change the URLs of nearly all the pages of the Wikipedia project.
Interlanguage harmonization might be.
The Cunctator wrote:
A big question that we are in the process of resolving now is whether we want to think of Wikipedia as a single project that has multiple translations, or as a bunch of largely independent projects specific to particular nations and language sets. I think the first conception is healthier and more productive over the long term. We really should think about this issue before we take broad actions that touch upon it.
Neither.
It can't be the first: 1) true translation is impossible, anyone who speaks more than one language knows that 2) it would be quite insulting to the people who work on the other pedias to suggest that they are mere translators. The polyglots among us are the cross-pollinators. :-)
It can't be the 2nd: 1) all projects must work towards an NPOV, open, GFDL encyclopedia.
We have to move in two directions simultanously: 1.) we have to give the other language equal status and more independance 2.) we have to bring them closer in: same software, same name, same logo, same attitude
We also have to remember that most of the other pedias are way further behind in their development. -- that doesn't just mean in number of *articles*. Look behind you. Read the old english wiki debates on NPOV, on article names, on how to handle everything. Some of them are still at that phase. Give them time.
On 10/10/02 2:42 PM, "The Cunctator" cunctator@kband.com wrote:
A big question that we are in the process of resolving now is whether we want to think of Wikipedia as a single project that has multiple translations, or as a bunch of largely independent projects specific to particular nations and language sets. I think the first conception is healthier and more productive over the long term. We really should think about this issue before we take broad actions that touch upon it.
Let me clarify the above statement: by "a single project that has multiple translations" I don't mean "English + translations into other languages" I mean "a single language-neutral encyclopedia with entries in all different languages, with the goal that every entry exist in every language". That is, I think we should think of the entries in English as being translations of the French entries, and the Polish entries, and vice versa.
Ideally, if an entry exists in one language but not another, such as "racine Grecque" (http://fr.wikipedia.com/wiki.cgi?Racine_Grecque) a link to the unwritten [[Greek root]] should pull up a result that indicates that there's a version of the entry in French.
I hope this clarifies this conceptualization of Wikipedia.
Let me clarify the above statement: by "a single project that has multiple translations" I don't mean "English + translations into other languages" I mean "a single language-neutral encyclopedia with entries in all different languages, with the goal that every entry exist in every language". That is, I think we should think of the entries in English as being translations of the French entries, and the Polish entries, and vice versa.
Ideally, if an entry exists in one language but not another, such as "racine Grecque" (http://fr.wikipedia.com/wiki.cgi?Racine_Grecque) a link to the unwritten [[Greek root]] should pull up a result that indicates that there's a version of the entry in French.
ah, you ring a bell here. I several times felt the consuming desire just to create an empty page (or an *horrible* stub) in english, just to link a french page that existed.
Sorry for those who don't like stubs, but *this* issue is just a call for stubs. Or, we need to find another way to indicate links from a empty link.
Well, we have not used all the color of the rainbow yet. Say, when a page doesnot exist in english, but exist in polish, the link in the english page to the non existing page is purple instead of red, and when you click on it, you get an nearly empty page with "no article available yet" and "this page is available in [[polish]]" (in the page). + link to the polish page.
As long as only an "official" stub is available, this message could stay for any other langage available. Once the article is nice enough not to be a stub anymore, the automatic message is removed automatically, and the link is just available as a [[pl:]]
On the polish side, there could be an option somewhere allowing to propose the link to other wikis. Once the article is thought really good, polish people could propose it to intern linkage. If the page in french exist, the link is a [[pl:]] at the top, if the page in english doesnot exist, the link is in the page with the message.
Is this understandable ?
__________________________________________________ Do you Yahoo!? Faith Hill - Exclusive Performances, Videos & More http://faith.yahoo.com
Anthere wrote:
The Cunctator wrote:
Ideally, if an entry exists in one language but not another, such as "racine Grecque" (http://fr.wikipedia.com/wiki.cgi?Racine_Grecque) a link to the unwritten [[Greek root]] should pull up a result that indicates that there's a version of the entry in French.
[..]
Well, we have not used all the color of the rainbow yet. Say, when a page doesnot exist in english, but exist in polish, the link in the english page to the non existing page is purple instead of red, and when you click on it, you get an nearly empty page with "no article available yet" and "this page is available in [[polish]]" (in the page).
- link to the polish page.
Something like that sounds doable (I don't know about the specific colors, but in general). This could perhaps be related to the proposed stub link indication -- see http://www.nupedia.com/pipermail/wikipedia-l/2002-September/005510.html
-- brion vibber (brion @ pobox.com)
Brion VIBBER wrote:
This could perhaps be related to the proposed stub link indication -- see http://www.nupedia.com/pipermail/wikipedia-l/2002-September/005510.html
It is not proposed anymore, I just installed it. Check your user preferences. Personally, I set it to 500. Happy stub-hunting!
Magnus
On Fri, Oct 11, 2002 at 09:35:43AM +0200, Magnus Manske wrote:
This could perhaps be related to the proposed stub link indication -- see http://www.nupedia.com/pipermail/wikipedia-l/2002-September/005510.html
It is not proposed anymore, I just installed it. Check your user preferences. Personally, I set it to 500. Happy stub-hunting!
Excellent :-) Perhaps there should be an announcement on Wikipedia:Announcements?
On Fri, 2002-10-11 at 03:35, Magnus Manske wrote:
Brion VIBBER wrote:
This could perhaps be related to the proposed stub link indication -- see http://www.nupedia.com/pipermail/wikipedia-l/2002-September/005510.html
It is not proposed anymore, I just installed it. Check your user preferences. Personally, I set it to 500. Happy stub-hunting!
Quite nice. I assume newcomer default is at 0.
One request: for wikistandard.css could the stub! use the normal coloring scheme instead of #772233? Thanks.
BTW, how hard would it be to re-implement [Multiple word]? bracketed links?
Magnus Manske wrote:
Brion VIBBER wrote:
This could perhaps be related to the proposed stub link indication -- see http://www.nupedia.com/pipermail/wikipedia-l/2002-September/005510.html
It is not proposed anymore, I just installed it. Check your user preferences. Personally, I set it to 500. Happy stub-hunting!
And how does it work and what does it do exactly? I'm not sure what to set it to because I'm not at all sure what it actually does...
--- Brion VIBBER brion@pobox.com wrote:
Anthere wrote:
The Cunctator wrote:
Ideally, if an entry exists in one language but
not another, such as "racine
Grecque"
(http://fr.wikipedia.com/wiki.cgi?Racine_Grecque) a link to the
unwritten [[Greek root]] should pull up a result
that indicates that there's
a version of the entry in French.
[..]
Well, we have not used all the color of the
rainbow
yet. Say, when a page doesnot exist in english,
but
exist in polish, the link in the english page to
the
non existing page is purple instead of red, and
when
you click on it, you get an nearly empty page with
"no
article available yet" and "this page is available
in
[[polish]]" (in the page).
- link to the polish page.
Something like that sounds doable (I don't know about the specific colors, but in general). This could perhaps be related to the proposed stub link indication -- see
http://www.nupedia.com/pipermail/wikipedia-l/2002-September/005510.html
Colors or just the way it is implemented, is still something minor I think. The point is if we see a non-balkanized wikipedia, it is logical that a subject covered in at least one langage, is not considered not covered at all in another. Let's think global.
(avoiding the stub issue) articles right now only have a dual definition : they exist or they do not. We should go forward a tri-definition, they exist in "this" langage (and maybe others), they exist not in "this" langage but in "others" they do, they do not exist at all. In short, it is thinking more of coverage of a subject than coverage in one langage.
__________________________________________________ Do you Yahoo!? Faith Hill - Exclusive Performances, Videos & More http://faith.yahoo.com
Anthere wrote in part:
ah, you ring a bell here. I several times felt the consuming desire just to create an empty page (or an *horrible* stub) in english, just to link a french page that existed.
Sorry for those who don't like stubs, but *this* issue is just a call for stubs. Or, we need to find another way to indicate links from a empty link.
I've thought about this as well, and this has bothered me, since I'm supposed to be one of the people opposed to horrible stubs.
But I think that there is a way to avoid horrible stubs here. It doesn't avoid stubs completely -- but I'm not against non-horrible stubs.
My idea is that you should translate the first paragraph of the article. It's quite unreasonable that you should translate the whole thing, but translating the first paragraph will provide a good stub. Then the language link will show people the fuller article.
My phrase "the first paragraph" has to be interpreted appropriately. Some articles' first literal paragraphs are too long, some perhaps too short. There may even be badly written articles whose first paragraph doesn't properly introduce the subject, and the second would be better. But generally speaking, the first few sentences should be sufficient, without being a great burden on the putative translator.
And you don't have to translate it *well*; use bad grammar and spelling, it'll get fixed.
Your proposal (snipped) is not a bad idea either, but it will take some programming work to happen. Translating the first paragraph can be done now.
-- Toby
--- Toby Bartels toby+wikipedia@math.ucr.edu wrote:
Anthere wrote in part:
ah, you ring a bell here. I several times felt the consuming desire just to create an empty page (or
an
*horrible* stub) in english, just to link a french page that existed.
<cut>
My idea is that you should translate the first paragraph of the article. It's quite unreasonable that you should translate the whole thing, but translating the first paragraph will provide a good stub.
<cut>
And you don't have to translate it *well*; use bad grammar and spelling, it'll get fixed.
<cut>
True. I can do that for sure. (I will :-)) But then, if I can do that in english, I certainly cannot do it decently in any other language (it would be mostly google translation, and these are often terrible, plus they do not deal with every language). So it would only be on the en.wiki.
You would probably say that makes sense since it is the biggest; but it also make sense that some subjects will be over time better covered on the french/german wikis than on the english.
The other point is that, if I can do that, I'm not sure every non-english speaking can do it. I doubt very much english would appreciate being flooded with google translation article introductions.
So it can only be a temporary solution (solving my problem, but not the overall problem)
Your proposal (snipped) is not a bad idea either, but it will take some programming work to happen. Translating the first paragraph can be done now.
True. Of course. These were just thoughts
anth�re
__________________________________________________ Do you Yahoo!? Faith Hill - Exclusive Performances, Videos & More http://faith.yahoo.com
Flags & Maps -- both far too complicated. And they're contentious issues just waiting to explode in our faces.
Flags -- which ones to use? What to do for languages spoken in many countries? What to do for countries which speak several languages? World Map -- what about mixed areas? What about the load time for the image? It'll have to be pretty big to show the different areas of Switzerland, no?
What is the *basic, driving* idea behind these suggestions? to make it simple for people to find their language on the portal page.
The simplest thing is still to have the name of each language, in words.
tarquin tarquin@planetunreal.com writes:
Flags & Maps -- both far too complicated. And they're contentious issues just waiting to explode in our faces.
Flags -- which ones to use? What to do for languages spoken in many countries? What to do for countries which speak several languages? World Map -- what about mixed areas? What about the load time for the image? It'll have to be pretty big to show the different areas of Switzerland, no?
What is the *basic, driving* idea behind these suggestions? to make it simple for people to find their language on the portal page.
The simplest thing is still to have the name of each language, in words.
Just what I wanted to say. Let's take us an example from the success of Google in contrast to other search engines.
One search field + button, a wikipedia logo and below the different language wikipedias ("La encyclopedie libre française (5000 articles)", "Die deutsche freie Enzyklopädie (5800 Artikel)" and so on)
People don't like to read too much on a front page. They prefer a simple and fast-loading interface - then they will stay.
greetings, elian
Hi elian, hi list!
One search field + button, a wikipedia logo and below the different language wikipedias ("La encyclopedie libre française (5000
articles)",
"Die deutsche freie Enzyklopädie (5800 Artikel)" and so on)
People don't like to read too much on a front page. They prefer a
simple
and fast-loading interface - then they will stay.
I also don't see much gain in flags, maps, or any other kind of images. We have 25 Wikipedia language subprojects (or whatever they should be called, now that things are growing together), so you can find the one you want easily when you skim over the page.
Kurt
Flags -- which ones to use? What to do for languages spoken in many countries? What to do for countries which speak several languages? World Map -- what about mixed areas? What about the load time for the image? It'll have to be pretty big to show the different areas of Switzerland, no?
I was thinking of a just black and white line copy png, that would be a really small image, filesize wise, where ther is just enough distinction to click on most countries, obviously it would be hard to represent sealand, andorra, vatican luxemburg etc on this map..
OK map didn't work out.. how bout this JUST the text links and the wikipedia logo in a fast loading splash page that takes you to the appropriate language... too many people complaining about not every single person being represented in the map.. sorry
Lightning
Exactly. Why do we need anything other than the name of the language? It's the clearest, easiest way.
I hate to try to click a world map trying to get to the Basque language....
Stephen G.
--- tarquin tarquin@planetunreal.com wrote:
Flags & Maps -- both far too complicated. And they're contentious issues just waiting to explode in our faces.
Flags -- which ones to use? What to do for languages spoken in many countries? What to do for countries which speak several languages? World Map -- what about mixed areas? What about the load time for the image? It'll have to be pretty big to show the different areas of Switzerland, no?
What is the *basic, driving* idea behind these suggestions? to make it simple for people to find their language on the portal page.
The simplest thing is still to have the name of each language, in words.
[Wikipedia-l] To manage your subscription to this list, please go here: http://www.nupedia.com/mailman/listinfo/wikipedia-l
__________________________________________________ Do you Yahoo!? Faith Hill - Exclusive Performances, Videos & More http://faith.yahoo.com
Exactly. Why do we need anything other than the name of the language? It's the clearest, easiest way.
I hate to try to click a world map trying to get to the Basque language....
Stephen G.
We don't, it was just for looks, I thought it would look kinda cool, otherwise the first page would seem kind of empty. thats why i suggested replacing it with the wikipedia logo
Lightning
Anthere wrote in part:
Toby Bartels wrote:
Anthere wrote in part:
ah, you ring a bell here. I several times felt the consuming desire just to create an empty page (or an *horrible* stub) in english, just to link a french page that existed.
My idea is that you should translate the first paragraph of the article. It's quite unreasonable that you should translate the whole thing, but translating the first paragraph will provide a good stub.
True. I can do that for sure. (I will :-)) But then, if I can do that in english, I certainly cannot do it decently in any other language (it would be mostly google translation, and these are often terrible, plus they do not deal with every language). So it would only be on the en.wiki.
But [[en:]] is the only wiki besides [[fr:]] that you use, right?
You would probably say that makes sense since it is the biggest; but it also make sense that some subjects will be over time better covered on the french/german wikis than on the english.
I certainly don't mean to suggest an idea that's special to [[en:]]. But if you can't write in, say, German, then why would you want to create a page on [[de:]] anyway? I'm not sure that I understand what you want to do.
True. Of course. These were just thoughts
I think that we're brainstorming here.
-- Toby
--- The Cunctator cunctator@kband.com wrote:
On 10/10/02 2:42 PM, "The Cunctator" cunctator@kband.com wrote:
A big question that we are in the process of
resolving now is whether we
want to think of Wikipedia as a single project
that has multiple
translations, or as a bunch of largely independent
projects specific to
particular nations and language sets. I think the
first conception is
healthier and more productive over the long term.
We really should think
about this issue before we take broad actions that
touch upon it.
It is true that from a strict efficiency point of view, the goal of a comprenhensive and neutral encyclopedia would benefit from dealing with issues in only one central article with as many actors as possible debating/working together rather than several different articles with only a couple of persons in each place, even though they are updating their own articles from the other wikis.
__________________________________________________ Do you Yahoo!? Faith Hill - Exclusive Performances, Videos & More http://faith.yahoo.com
On 10-10-2002, Anthere wrote thusly :
--- The Cunctator cunctator@kband.com wrote:
On 10/10/02 2:42 PM, "The Cunctator" cunctator@kband.com wrote:
A big question that we are in the process of
resolving now is whether we
want to think of Wikipedia as a single project
that has multiple
translations, or as a bunch of largely independent
projects specific to
particular nations and language sets. I think the
first conception is
healthier and more productive over the long term.
We really should think
about this issue before we take broad actions that
touch upon it.
It is true that from a strict efficiency point of view, the goal of a comprenhensive and neutral encyclopedia would benefit from dealing with issues in only one central article with as many actors as possible debating/working together rather than several different articles with only a couple of persons in each place, even though they are updating their own articles from the other wikis.
Like there was a page with search though several Wikipedias for a term there could be a gateway page to working on several (two, theree, four) versions of the same articles in different languages.
It would be a bit more convenient than opening several windows with different language versions of the article.
1. in a frame set that can be programmed (vertical horizontal sizes etc) 2. that can "guess" the name of the article in languages we've chosen to work on - Warszawa, Warsaw, Warschau, Varsovie you type only Warsaw and it opens respective articles in frames
Regards, Kpjas.
--- The Cunctator cunctator@kband.com wrote:
Let me clarify the above statement: by "a single project that has multiple translations" I don't mean "English + translations into other languages" I mean "a single language-neutral encyclopedia with entries in all different languages, with the goal that every entry exist in every language". That is, I think we should think of the entries in English as being translations of the French entries, and the Polish entries, and vice versa.
Why use the term "translation", then? The articles, in most cases, are not translations of each other; they have been composed in the given language. There is a big difference between translation and composition.
Ideally, if an entry exists in one language but not another, such as "racine Grecque" (http://fr.wikipedia.com/wiki.cgi?Racine_Grecque) a link to the unwritten [[Greek root]] should pull up a result that indicates that there's a version of the entry in French.
Yes, I'm all for that.
Stephen G.
__________________________________________________ Do you Yahoo!? Faith Hill - Exclusive Performances, Videos & More http://faith.yahoo.com
On 10-10-2002, The Cunctator wrote thusly :
On 10/10/02 2:42 PM, "The Cunctator" cunctator@kband.com wrote:
A big question that we are in the process of resolving now is whether we want to think of Wikipedia as a single project that has multiple translations, or as a bunch of largely independent projects specific to particular nations and language sets. I think the first conception is healthier and more productive over the long term. We really should think about this issue before we take broad actions that touch upon it.
Let me clarify the above statement: by "a single project that has multiple translations" I don't mean "English + translations into other languages" I mean "a single language-neutral encyclopedia with entries in all different languages, with the goal that every entry exist in every language". That is, I think we should think of the entries in English as being translations of the French entries, and the Polish entries, and vice versa. Ideally, if an entry exists in one language but not another, such as "racine Grecque" (http://fr.wikipedia.com/wiki.cgi?Racine_Grecque) a link to the unwritten [[Greek root]] should pull up a result that indicates that there's a version of the entry in French. I hope this clarifies this conceptualization of Wikipedia.
This is my pet idea to notify every other Wikipedia of the appearance of lengthy, well fleshed out on any Wikipedia. Creating a compilation of recent changes seems to be awkward and resource consuming.
So what are other proposals to put this ideas into being ?
Regards, Kpjas.
- There are real, concrete benefits to having a default and preferred
interface/language;
In other words, the pros and cons of such an act should be explicitly and clearly listed. There are many reasons that www.google.com, www.dmoz.org, etc. (which all have multilanguage settings) have English as the default, and it's not just that the servers are US-based).
It's the default, but if your browser is set up to sent a preferred language as part of the request (German Netscape and German MSIE do so), google.com welcomes in the preferred language.
Once that's done, we can weigh priorities.
I personally think a better focus right now for the developers is to work on maximum integration of the different language wikis. If the backend is better integrated, frontend issues become easier to deal with.
A big question that we are in the process of resolving now is whether we want to think of Wikipedia as a single project that has multiple translations, or as a bunch of largely independent projects specific to particular nations and language sets. I think the first conception is healthier and more productive over the long term. We really should think about this issue before we take broad actions that touch upon it.
Since many non-english wikipedians contribute to the english and their native 'pedia, the projects are not independent but highly linked. Having a common starting page would reflect this.
I agree with you that we should first decide on the future starting page and change www.wikipedia.org to something else as soon as we know what exactly to do.
Regards,
JeLuF
--- The Cunctator cunctator@kband.com wrote:
In other words, the pros and cons of such an act should be explicitly and clearly listed. There are many reasons that www.google.com, www.dmoz.org, etc. (which all have multilanguage settings) have English as the default, and it's not just that the servers are US-based).
Google doesn't have a default English setting. If I go to Google from here, it gives me a Korean interface. If I want English, I have to set preferences to tell it so.
Just for the record...
Stephen G.
__________________________________________________ Do you Yahoo!? Faith Hill - Exclusive Performances, Videos & More http://faith.yahoo.com
The Cunctator wrote:
There are many reasons that www.google.com, www.dmoz.org, etc. (which all have multilanguage settings) have English as the default, and it's not just that the servers are US-based).
I suppose the main reason is that more people on the Internet can read English than any other language.
A big question that we are in the process of resolving now is whether we want to think of Wikipedia as a single project that has multiple translations, or as a bunch of largely independent projects specific to particular nations and language sets. I think the first conception is healthier and more productive over the long term. We really should think about this issue before we take broad actions that touch upon it.
I agree very much with the first conception, although I wouldn't necessarily use the word 'translations'.
--Jimbo
Jimmy Wales wrote:
The Cunctator wrote:
I object strenuously. This is not a slam-dunk good idea at all. Please hold off at least a good while before doing this.
O.k., let's wait.
What are the objections?
I ain't heard nobody but Cunc object so far, and he doesn't say why. I await the reasons with bated breath, if he's got something better than "the English Wikipedia needs a large pool of people working on it to achieve neutrality and perspective, but I apparently don't give a rat's ass about the contents of Wikipedia in other languages" to show for it.
-- brion vibber (brion @ pobox.com)
On 10/10/02 3:47 PM, "Brion VIBBER" brion@pobox.com wrote:
Jimmy Wales wrote:
The Cunctator wrote:
I object strenuously. This is not a slam-dunk good idea at all. Please hold off at least a good while before doing this.
O.k., let's wait.
What are the objections?
I ain't heard nobody but Cunc object so far, and he doesn't say why. I await the reasons with bated breath, if he's got something better than "the English Wikipedia needs a large pool of people working on it to achieve neutrality and perspective, but I apparently don't give a rat's ass about the contents of Wikipedia in other languages" to show for it.
I suspect a slightly less rude and sarcastic attitude on your part may help the discussion.
Changing your above statement to "Wikipedia needs a large pool of people working on it to achive neutrality and perspective" is a good starting point.
On 10/10/02 3:47 PM, "Brion VIBBER" brion@pobox.com wrote:
Jimmy Wales wrote:
The Cunctator wrote:
I object strenuously. This is not a slam-dunk good idea at all. Please hold off at least a good while before doing this.
O.k., let's wait.
What are the objections?
I ain't heard nobody but Cunc object so far, and he doesn't say why. I await the reasons with bated breath, if he's got something better than "the English Wikipedia needs a large pool of people working on it to achieve neutrality and perspective, but I apparently don't give a rat's ass about the contents of Wikipedia in other languages" to show for it.
http://www.nupedia.com/pipermail/wikipedia-l/2002-August/004031.html
The Cunctator wrote:
On 10/10/02 3:47 PM, "Brion VIBBER" brion@pobox.com wrote:
I ain't heard nobody but Cunc object so far, and he doesn't say why. I await the reasons with bated breath,
http://www.nupedia.com/pipermail/wikipedia-l/2002-August/004031.html
I already cited that where you posted it at meta. It doesn't list any reasons against using an en.* URL for English articles, which is as you say "only a minor component" of the language integration.
For someone who says that "more effort should be put into building the cross-link capabilities" and "there should be the option of having Recent Changes show changes from any array of the languages", you're putting an awful lot of effort into stalling useful work to those ends by yakking on over this tiny thing.
-- brion vibber (brion @ pobox.com)
On 10/10/02 6:43 PM, "Brion VIBBER" brion@pobox.com wrote:
The Cunctator wrote:
On 10/10/02 3:47 PM, "Brion VIBBER" brion@pobox.com wrote:
I ain't heard nobody but Cunc object so far, and he doesn't say why. I await the reasons with bated breath,
http://www.nupedia.com/pipermail/wikipedia-l/2002-August/004031.html
I already cited that where you posted it at meta. It doesn't list any reasons against using an en.* URL for English articles, which is as you say "only a minor component" of the language integration.
For someone who says that "more effort should be put into building the cross-link capabilities" and "there should be the option of having Recent Changes show changes from any array of the languages", you're putting an awful lot of effort into stalling useful work to those ends by yakking on over this tiny thing.
That's a very interesting argument. If you consider it a tiny thing, then drop it.
The Cunctator wrote:
On 10/10/02 3:26 AM, "Brion VIBBER" brion@pobox.com wrote:
We haven't decided yet exactly what to make the new www.wikipedia.org project front page look like, but it seems pretty well decided that we do want to move the English wiki to en.wikipedia.org and set up some sort of multilingual and/or browser-language-setting-sensitive intro page.
(See: http://meta.wikipedia.org/wiki/What_to_do_with_www.wikipedia.org )
If there's no objection, I'm going to switch the URL redirecting around tomorrow, so that www.wikipedia.org/* points to en.wikipedia.org/* rather than the other way 'round. A new intro page can later be set up at convenience.
I object strenuously. This is not a slam-dunk good idea at all. Please hold off at least a good while before doing this.
I notice Cunc didn't maintain the cross-most to intlwiki-L.
For those of you who may not be subscribed to wikipedia-l as well, check the thread here: http://www.nupedia.com/pipermail/wikipedia-l/2002-October/thread.html#5900
-- brion vibber (brion @ pobox.com)
did we work out a way of getting a message to an IP user without blocking?
195.93.33.13 http://www.wikipedia.org/w/wiki.phtml?title=Special:Contributions&target=195.93.33.13 is adding stubs of British actors with only the name and the DoB in an ambiguous format. Should I just ban for 30 minutes with a polite explanation?
On 12/17/02 10:31 AM, "tarquin" tarquin@planetunreal.com wrote:
did we work out a way of getting a message to an IP user without blocking?
195.93.33.13 http://www.wikipedia.org/w/wiki.phtml?title=Special:Contributions&target=195. 93.33.13 is adding stubs of British actors with only the name and the DoB in an ambiguous format. Should I just ban for 30 minutes with a polite explanation?
Banning would be entirely innapropriate, especially now that we have the stub threshhold which makes stubs much less of a problem.
The Cunctator wrote:
On 12/17/02 10:31 AM, "tarquin" tarquin@planetunreal.com wrote:
did we work out a way of getting a message to an IP user without blocking?
195.93.33.13 http://www.wikipedia.org/w/wiki.phtml?title=Special:Contributions&target=195. 93.33.13 is adding stubs of British actors with only the name and the DoB in an ambiguous format. Should I just ban for 30 minutes with a polite explanation?
Banning would be entirely innapropriate, especially now that we have the stub threshhold which makes stubs much less of a problem.
I only suggested that as a means to guarantee a message reached this person -- not as any kind of punishment. s/he doesn't seem to be reading Recent Changes, or reloading his pages: I've left messages there with no result.
On 12/17/02 11:13 AM, "tarquin" tarquin@planetunreal.com wrote:
The Cunctator wrote:
Banning would be entirely innapropriate, especially now that we have the stub threshhold which makes stubs much less of a problem.
I only suggested that as a means to guarantee a message reached this person -- not as any kind of punishment. s/he doesn't seem to be reading Recent Changes, or reloading his pages: I've left messages there with no result.
I still wouldn't ban. The harm of the stubs is minimal. But I won't cry if you do it.
The Cunctator wrote:
I still wouldn't ban. The harm of the stubs is minimal.
But in this case it was dumb legwork we could avoid. This person was adding dates as "4-5-1945", which means someone later on has to go look it up again.
But I won't cry if you do it.
I wasn't comfortable either about using the ban feature for this, which is why I brought it up here. :-) I'd rather we had a "soft" way of doing this. For example, make the contents of the page [[user:{IP address}]] appear above every edit box this IP sees.
That way: * they see it (hopefully ... I've seen people overlook what's right in front of them when it comes to computer interfaces) * we're not banning them, and they can still edit * they can reply to confirm they've read the message on the same page.
wikipedia-l@lists.wikimedia.org