User talk:Brooke Vibber

Note; I don't keep track of this page as often as on my Esperanto, English, or French user talk pages. Be sure to leave a note on one of them pointing at any important updates here.

---

Hi Brion. Interested as I am in the page statistics for the International wikis, I couldn't help but notice that according to the Special:Statistics page the Chinese wiki is now by far the largest wiki in the world, with a whopping 18,446,744,073,709,551,615 valid pages! Too bad I can't read Chinese ;) -Scipius

---

Hi Brion.

I have some suggestions to the Chinese wikipedia:

1. It is preferred to maintain a single version for Chinese language. Instead of a simplified version as well as a Formal Chinese version. There are possible to maintain a Formal Chinese version, but add the necessary translation module to the system which automatically translate the Formal Chinese version to the simplified version for Readers who do not understand the formal Chinese.

Such an facility could be built in sort time with the help of the unihan database (www.unicode.org).

There are some possible way to handle this:

1.a through user selection: add to user perference a selection for simplify display.

1.b through the browser/server content negoation (zh-tw, zh-hk, zh-mo to select Formal Chinese; zh-cn, zh-sg to select simplified Chinese)

1.c add an simplified link for display the simplified version.

1.d add an selection screen or seprated url for selection.

2. If the above is accepted, we still accept simplified Chinese writings. However, efforts can be done to convert them into formal Chinese later. A small tool can be used to assist this process.

3. The Search facility also need to handle the conversion: for searching, converting both the search expression and the text into simplified form will be perferred.

4. If I understand correctly, the current wikipedia software provide search on single characters for Chinese, it will be suficient for Wikipedia's usage. But I would like to point-out that the ranking algorithm would weight the consective matching a higher score, then it will be more close to the searcher's expectation.

I would be happy to write a small script to extract the simplified mapping table from the unihan database if you like.

Kent Sin kentsin@yahoo.com 2003.1.4

My impression is that converting traditional to simplified is relatively simpler (mostly many->one) than simplified to traditional (more one->many and many->many), but that it's still not going to be perfect. So, this should be doable, particularly if we can come up with a simple syntax for specifying conversion overrides. (Vaguely similarly, I'd like to see automatic furigana on the Japanese wiki, which again is going to be perfect, and would benefit from a simple override syntax.) --Brion VIBBER 2003年1月7日 09:22 (UTC)

---

Hi Brion

I have some suggestion about the tradition/simplified chinese issue. Basically I would propose to have zh_tw.wikipedia.org and zh_cn.wikipedia.org (and maybe zh_HK,zh_SG too) and use wikipedia's interlanguage to link between zh_tw and zh_cn. The reasons to have seperate page are:

• The interface. This would allow to use tradition or simplified chinese as interface. (note that many these interface items have different chinese transltions for cn/hk/tw).
• The "encoding" conversion between tradition/simplified Chinese is not difficult. However, a traditional Chinese text which is "encoding" converted from the simplified Chinese text might be very different from a text which is written in tradition Chinese originally and vice visa. Also many translation are different. For example the term 'software'-> translated into zh_cn ->converted into zh_tw is different than term 'software'->translated into zh_tw.
• This would also help about the search problem.
• I would like to propose the following structure
  • Setup zh_cn.wikipedia.org and zh_tw.wikipedia.org
  • For each item in encyclopedia one should have
    • In zh_cn
      • Item name translated into zh_cn
      • link to the corresponding item in zh_tw
      • text in zh_cn
      • encoding conversion of about text into zh_tw.
    • In zh_tw
      • Item name translated into zh_tw
      • link to the corresponding item in zh_cn
      • text in zh_tw
      • encoding conversion of about text into zh_cn.

I think the scheme is easy to implement and can avoid many problems.

Pektiong

• Be more precise. If we try to identify language/variant/encoding/script then we are facing the possible combination of lang=zh(Mandarin), variant=cn,tw,hk,sg, encoding=utf,big5,gb(but utf is can conver big5/gb), and script=simp or trand. There are some way to deal with this problem on the server side so the end user can automatically see the page in desired lang/variant/encoding/script. I am checking what is necessary for this to be done. (check Encode-HanConvedrt,and apache-filter-hanconvert maybe?). If this is difficult to implement within wikipedia.org then mayhbe maintain seperate entry is better.

Pektiong

---

Subpages

I remember reading somewhere in the EN WP that subpages will interfere with the yet-to-come system. Did I remember this wrongly?

I ask this, because the current convention established in March in distinguishing between the traditional and simplified versions by the same title is using a slash before the identifier of the style (繁 or 简). Methods that have been tried and failed, include:

1. Attaching the identifier directly: WikiWiki -> WikiWiki繁

   It is unreadable if the title is Chinese

2. Parenthesis: Wikipedia -> Wikipedia(繁)

   It interferes with ambiguitation

4. A dot: Kristen Nygaard -> Kristen Nygaard·繁

   The dot in Chinese is used as a name division mark, so it'd imply that Nygaard's surname, or at least a part of it, is 繁. (See en:punctuation)

So currently, we use:

3. 黑客 -> 黑客/繁.

If subpage will be a problem, we could use Chinese parenthesis:（） instead of Western parenthesis as in option 2. Although it still physically resembles disambiguitation, in practice the computer can't confuse it with the real one. Any suggestion? --Menchi 2003年5月22日 16:46 (UTC)

Sorry for taking so long to respond!

There is no longer any special 'magic' for slash-subpages except in the user and talk namespaces, where it automatically shows a link back to the 'parent' page... so on 对话:黑客/繁 you'll see there will automatically be a link to 对话:黑客 at the top of the page. That's actually kind of convenient. :) Also the special user links (contributions, email) should work if there is a /繁 subpage for a user page, but I haven't tested this. Slashes are otherwise just considered part of the page title, so nothing should go horribly wrong because they're being used. --Brion VIBBER 2003年5月25日 02:12 (UTC)