Jump to content

Wikipedia:Ignored feature requests

From Wikipedia, the free encyclopedia
(Redirected from Wikipedia:IFR)

Feature requests can be made at MediaZilla. For information on using MediaZilla, please see Wikipedia:Bug reports. You can discuss them first at m:MediaWiki feature request and bug report discussion.

No one ever reads this page. (Or at least no one who would actually be adding these features.) Don't request anything here. It will be ignored.

If you made a feature request at the village pump, it may well end up here. The village pump is crowded, and is explicitly not for the requesting of new features. Use MediaZilla or the correct meta page instead.

Feature Requests, Enhancements and Suggestions (from Village pump)

  • Time-Independent version of Recent Changes: Possibly viewed in terms of number of page views since. Alternatively, Unexamined Changes, those pages which have changed but no one has checked.
  • Consider moving from sourceforge to something such as Bugzilla or a stripped down version of the sourceforge code. Also considered moving Wikipedia to BerliOS, but it would mean moving the implied private user database to a third party.
  • Confirmation box for editing an earlier version: Disliked because it makes reverting more troublesome. There is already a bold alert which should be sufficient.
  • Add a floating quickbar on the right.
  • Add the ability to talk to non-logged in users (IPs).
  • What we need is a basic trust metric, a way to distinguish in Recent Changes etc. between an anonymous, unknown user and a well established Wikipedian with a good reputation. Also, being able to talk to anons would make it possible to explain to them why vandalism on Wikipedia is futile (w/ a "Why vandalism doesn't work" page)
Wouldn't it be better to change "Related Changes" so that it includes changes of the current page (and maybe also pages that link to the current page)? -Martin
Yes, I think that would be better. Patrick 13:26 Dec 26, 2002 (UTC)

I'm the guy who removed the self link on list of musical topics. I see why they wanted it now, and I very much agree that Related Changes should include changes made to the current page. Maybe if for some crazy reason people didn't like this, it could be an option in their preferences so that related changes doesn't include the current page. Then, surely we could just automatically strip out self links, as they would then be rendered completely useless, and there would be no possibility of a false positive. Just an idea for the automatically stripping out self-links, there may be some reason against it, but I definitely think related changes should include the current page Smelialichu 17:28 Dec 26, 2002 (UTC)

XHTML compliance

Wikipedia should be XHTML compliant so that it is easy to convert articles to other formats. See http://www.w3.org/TR/xhtml11/ --TakuyaMurata

That's probably a good idea to work towards. However, it's worth noting that it's possible to automatically convert (reasonable) HTML to XHTML, so anyone who needs XHTML can get it in a fairly straightforward way already. -- Dwheeler 09:44, 2 Aug 2003 (UTC)

It would not be so much of a problem to convert HTML to XHTML, but, to do that the HTML of Wikipedia needs to be valid HTML 2.0 at least.
I've tried with several pages, and somehow, it doesn't even pass that simple stage...
Not a simple task, I see that now...
Thorwald Peeters 81.100.246.62 04:49, 25 Nov 2003 (UTC)

The main task is closing the <p>s ;-) 'tidy -asxhtml' works without a hitch on pretty much anything htmlish i've come across so far. See meta:Skins -- Gabriel Wicke 15:01, 28 Jan 2004 (UTC)

Some progress has been achieved on this lately, at least most content views in the MonoBook skin (see http://wiki.aulinx.de/) validate as XHTML 1.0 trans. now. -- Gabriel Wicke 00:05, 7 Apr 2004 (UTC)

Message Boards/Chat

They'd prevent a lot of heated arguments and confusion. Lir 02:11 Nov 15, 2002 (UTC)

View linked status on Recent Changes

In order to make sure no new orphans get created - or at least not much - it may be useful to have a note show up in the page's entry in on the Recent Changes page. Maybe a red exlamation mark or so. As soon as the page is linked, it should disappear. Something similar could be done if the article is a stub (I think the definition is an article without a comma?). These tools would make detection of new "malicious" entries easier. June 24 2002 --- jheijmans

View votes on pages

It may be useful to be able to see in an article that somebody has voted for it, at least in the case of rewrite/wikification. This may remind an editor that it should check the remarks placed there and - if he thinks it is solved - react and/or remove the article from the queue. This may be tricky though, since we can vote to get an article on that list, but not to get it off. jheijmans

Where have the "votes for NPOV/rewrite/deletion" pages gone? Mswake

New Special Page

Probably wouldn't want it linked on the sidebars, but I think it'd be quite cool to have a Most Linked page that acted like "Most wanted" only it includes existing articles. DanKeshet, Saturday, June 22, 2002

Yeah, I think that would be cool. Displaying a link to this on the sidebar might be an option for people to select in their preferences (However, I don't see any problem with having this be displayed by default in the sidebar though). Determining popularity in this way is similar to how Google ranks pages and may also be tied into our search function (so long as this wouldn't violate Google's PageRank technology patent). This should also be easy to do: Just write a script to count the number of entries on each articles's "pages that link here" page and then rank the results at [[Special:Most linked to]]. --maveric149

Namespaces

(June 19, 2002)

Would it be possible to not regard links between the user: namespace and the articles? This would clean the "Pages that link here" list for the users that mention articles they edited on their page, and possibly reveal them as orphans. Same thing for the talk: namespace. - jheijmans

I like having these links show up. However, I think it perfectly justifiable to make this an option set in user preferences, with the default (which is what shows up to visitors) to not display these links. Alternatively, we could have things listed in different groups. That is, "The following encylopedia articles link to ..." and "The following other Wikipedia pages link to ...". As for detecting orphans, I agree with jheijmans that links from other namespaces should not count. — Toby Bartels, Wednesday, June 19, 2002

DO NOT do that, this is especially useful when you're writing a new article, check 'What links here' and find out to your pleasant surprise that some User: has taken pictures of the place in question and linked to the un-existing page.
If it's going to be an option at least follow Toby's suggestion and have a "The following other Wikipedia pages link to this page" link --Ævar Arnfjörð Bjarmason 17:32, 2004 Apr 24 (UTC)

Latex like equation writer

It should be cool to be able to put a latex environment to edit equations. It's a personal project that I'm working on (WikiLatex) but I should be great to have it here also.

Fix "Stub articles" sort order

The current sort order for stub articles is nearly useless. What's needed is to pick a threshold for what constitutes a stub article (e.g. 1000 characters), and then sort that list by how many pages link to the given stub. This would make the "Stub articles" list work similarly to the "Most wanted" list, the latter of which is very effective. RobLa -- April 13, 2002.

Printer-friendly display

Would be nice to have:

  • Serif font (?)
  • Footnoted URLs should be listed as footnotes! Currently you just see the [1], [2], and that's all... (In fact, all external links should have their URL's listed at the bottom as footnotes. AxelBoldt)
  • Some ability to link higher-resolution images for diagrams, maps, etc.

Brion VIBBER, Tuesday, April 2, 2002

You can do some tweaks with the user styles now, just place all print-specific rules in a @media block:

@media print {
    body { bla.. }
    something { dfgdg }
}

using :before and :after it's possible to add formatting using css, this here adds the full href of a link after it for example:

#content a:link:after, 
#content a:visited:after {
   content: " ( " attr(href) " ) ";
}

neat, isn't it? -- Gabriel Wicke 22:38, 29 Apr 2004 (UTC)

Statistics

At some point the number of users statistic is becoming meaningless due to accidental log ins which create multiple users or users who are just no longer active. Would it be possible to take users who have not been active for a long period of time (3 months?, 6 months?) out of the system or out of the statistical listing - then the statistic could be number of users active in the last x months? Trelvis Mar 12

The most popular page should have an explanation of what the number is. How long a time period are those numbers collected under? Could it be reset every week so it shows this weeks most popular, rather than the most popular all time? Trelvis

Could the search list be ranked either based on traffic or like Google by the number of pages linked to it, so that the most popular pages come up first. This would prevent some of the wading through a large list of obscure pages and redirects which are less popular?

When you get a search result could we add an option to do a Google search on the subject? (I found this useful in the old software)

Also when the search results come up any wiki links in the example text which have a searched phrase do not work because the html bold tags are added into the link - so far I haven't seen many accidentally formed pages with the bold tags built in, but this should be fixed. It might be on the bugs page already.

I know I have seen a request for advanced search capabilities somewhere, but I will restate that request here for compactness. Trelvis

I would like to have redirected articles and "Complete list of Encyclopedia topics" pages omitted from search results. AxelBoldt

I agree that it shouldn't search the body of a #REDIRECT, but it should still search the title; otherwise, the failed searches page become unnecessarily longer when users type in a misspelled word for which there is a redirect but which is not used in any regular article.

--Damian Yerrick

Looking at mis-spelled search requests for 'Circumsision', 'Circumsicion', 'Circumsission' and 'Lamberghini', we should probably use something like Soundex or Metaphone to search for a sound-alike article if a literal search fails. Note that recent versions of PHP have a metaphone() function built in.

See http://www.zend.com/manual/function.metaphone.php for more details.


In fact, I would go further: search engines should

  • match exact (case sensitive!) matches first, and then
  • match (case insensitive) 'exact' matches second, and then
I disagree strongly; we should never be case-sensitive. --BV
  • fall back to ignoring diacritics (as if ASCII-only), and then
  • fall back to Metaphone or other similar scheme (just replacing all vowel sequences by '*' normalises to a sufficient extent for many purposes)
Agreed. --Brion VIBBER

Just looking at the failed searches shows that about half of them would succeed given some very simple normalisation. Many of the others would work if a combination of guess-the-spaces and stemming was used. Wikipedia is small compared to the Web, and so techniques like this will improve recall without deluging the reader in dross, providing that exact matches and article-title matches take priority. The Anome


The failed search page could use some work. When you type in a single letter (for example, some poor chap who wants to find the entry on the letter J) it tells you that you probably entered a word with less than three characters, this should read less than TWO characters, as two character searches work. I also suggest adding the following text to the failed search page

If you are looking for Wikipedia articles on letters in the Latin alphabet, please use these links: A, B, C, D, E, F, G, H, I, J, K, L, M, N, O, P, Q, R, S, T, U, V, W, X, Y, Z

That would make it MUCH easier to find the letters from the search bar.

--Dante Alighieri


Namespaces

I'd like a way to list all pages in a namespace. For example, http://www.wikipedia/wiki/wikipedia%3A should list all pages in the Wikipedia namespace that the current user can access, and http://www.wikipedia/wiki/special%3A should do the same as special:Special pages. --Damian Yerrick

It would be nice to have a history table (and an auto-generated graph as well if possible), showing the trends, day by day, week by week, etc. for each of the monitored variables. This would give an instant overview of what's going on.

See MRTG for an example of this sort of thing (for network traffic in this case), or the trends graphs at seti@home for another. -- The Anome

See also the plots at http://www.distributed.net/statistics/ --Damian Yerrick

See http://wikimedia.org/stats/live/ for network-related graphs. -- Gabriel Wicke 00:08, 7 Apr 2004 (UTC)

Minor edits

A very minor issue: I don't want my minor edits showing up on my contributions page. I don't consider myself to have contributed the article on Agatha Christie, for instance, and my contribution to it (a typo correction, IIRC) was so minor as not to deserve notice. I would not, however, mind have the page list articles I instigated, e.g. Dziga Vertov and Dave Brubeck--those, in my mind, are more properly contributions. Best, Koyaanis Qatsi

It seems this has been changed, but I don't like it at all. Previously I've been able to set my preferences so that the meaningless distinction between major and minor edits effectively disappeared. No longer is this the case. --Zundark, 2002 Feb 2

Listing New Articles on User page

It would be nice if the ten newest articles that someone started would appear at the bottom of their user page. --Chuck Smith

source:namespace

A source:namespace that only can be edited by sysops or trusted longtime users and be called upon by individual lines in an article. For example, typing [source:Origin of Species/Chapter 1{1-15}] within the edit box of an article causes the display lines 1 through 15 of Chapter 1 of Charles Darwin's Origin of Species in that article (working in a similar way as placing a url of an uploaded image in the same edit box -- except the result will be selectable text). This way, one could call upon any set of lines within the Origin of Species and annotate until they are blue in the face without changing what Darwin said.

Shortcuts to Wiki Shortcuts

www.seedwiki.com has a great idea in its page editing: popup menus with all Wiki Shortcuts they use. When one Shortcut is selected it should be inserted in the text. On the bottom where Wikipedia writes DO NOT USE .... it has a nice summary of Wiki text formatting. Would be helpful. -GillianAnderson

Statistics

List the number of pages which are redirects. Calling redirects junk pages is not very accurate. --Chuck Smith

RDF editing

This is a rather esoteric feature, so bear with me. It would be very cool to be able to enter machine readable metadata about pages, and specifically, to allow RDF triples to be entered.

A crude prototype of how this might work is running on the RDF wiki. The basic concept is to have three URIs where the second URI describes the relationship between the first and the third (e.g. <http:...Mona+Lisa> <http:...Resides+In> <http:...Louvre>)

The way that I could see this working is that one namespace exists for relationships (the second URI), and that the first and the third URIs are normal Wikipedia articles. The relationship lists would reside on a special page associated with the first URI.

-- user:RobLa

Metadata layer

I think there is a need for some sort of metadata system on Wikipedia. XML/RDF data source would make it much easier to accomplish. But for the time being with are stuck with raw text. Is any metadata namespace or something along the lines the previous feature request in the works ?
August 16 2002

More Most Wanted

The most wanted feature is great, but it doesn't give enough results. Right now, most of the most wanted are things like '11th century BC' Olof

It probably would look better without that kind of time line article cluttering the listing, (Currently 26 out of 50) Most of them would only be templates anyway. Perhaps if we each took one on they could be cleared up quickly. Even so increasing the number to the 100 most wanted would make sense. That would give each contributor a greater chance to find something in his own filelds of interest. Eclecticology
21 out of 50 (April 15, 2002). But if we put stubs into the Year in Review articles, linked to other Years in Review, we would link to other Year in Review articles and bring them into WantedPages. I'd say if the first character of an article's title is a digit, demote it by 3 links in WantedPages. --Damian Yerrick

Plain text list of all articles

Hi! First off, I'd like to say that I think the new Wikipedia scripts are wicked cool. So, here's my little itty bitty feature request: you know the special page that lists all articles? There should be a way to fetch a text/plain listing. Just straight up one-article-name-per-line text. (The reason I'd like this is so I could fetch such a listing via cron, say, nightly, for tab completion in my Wikipedia emacs thingy.)

Least wanted pages

Maybe a Least wanted page, akin to the Most wanted one, is useful. It gives a list of those articles that nobody visits, and these pages may needs some visitors to edit them, or maybe to put up links at other/better places.

Since it is essentially no different than the most wanted page (just a small change in the query) I suppose this shouldn't be difficult to make. jheijmans

I agree, but let's call it the "least popular pages" or "least visited pages" instead. What should be the definition? We could sort them by

  • total number of visits ever, or by
  • total visits since a given date, or by
  • least recent visit timestamp, or
  • (my choice) use an exponentially decaying estimate of visit rate, with a decay half-life of about (say) 1 month.

Note that essentially all the pages will get visited by robots roughly daily, so pages not visited by people may get lost in the noise. Wikipedia visits for the most wanted pages seem to follow Zipfs law, and probably so do the least popular ones. The Anome

I agree, the name least wanted is very well chosen. As for the definition, I think any will do. The most wanted uses the second option. I figure that bot-visits do not really matter, assuming all pages approximately get the same number of bot-visits. One disadvantage, however, is that new pages will automatically show up there as well, so maybe the third option is better, or, your preference (though that requires more programming of course, so my 'it's easy' argument no longer holds. jheijmans

If I spontaneously link to a nonexistent page, I sometimes worry that the page already exists under another name. Well, if this bothers me, then that's what search engines are for (at least assuming that I'm looking for words with more than 4 letters in them!). But I realise that not everybody else is so thoughtful, so often I'll create orphan redirects to pages that I'm working on, using names that I think are likely spontaneous links.

What's missing is an option when both occasions arise. I make a spontaneous link to a nonexistent article, and while I know that it doesn't exist under another name, I worry that other people may spontaneously link to the same topic under another name. I would like to redirect alternative likely names to my name — I mean, to the one most in line with Wikipedia:naming conventions. Unfortunately, if I do this, then when such a spontaneous link appears, people will think (from looking at the link) that the article has already been written, when it hasn't. This is quite unsatisfactory — indeed, I've voted such redirects for deletion to avoid just this problem.

My proposed solution: When creating a page, each link is followed enough to see if it exists. I say, see if it's a redirect too, and if it is, then follow it further to see if the page that it redirects to exists. Then if it's a redirect to a nonexistent page, format the link as if it were itself a nonexistent page.

Toby Bartels, Friday, June 28, 2002

Safer edit form

I was just editing a page, and I accidentally pressed return when the scope of the cursor when somehow connected to the submit form buttons but not inside a specific text window. This caused the browser to react as if I had pressed the Save button, which I wasn't ready to do yet. No real harm done — I only wanted to Preview once to check for typos, and it turned out that there were no typos, so the only thing that was missing was that I didn't get a chance to check the minor edit box. However, I think that it would be safer to set things up so that the default button is Preview, rather than Save.

I'm not sure what in the HTML code makes the Save button the default; perhaps the browser (IE) is just guessing, and all that we have to do to make it guess correctly is to put the Preview button first??? That would be a good idea anyway, to make newbies think about Preview before they think about Save.

Also, while I'm on the subject, it would look keener if the Cancel link were a button as well, even though that's completely unnecessary, since it has more in common with the Save and Preview buttons than with the Editing Help link. That's just for looks though, not important.

Toby Bartels, Wednesday, July 3, 2002

I have to disagree about the default behavior with return/save. I for one hit enter all the time to save, and this is the most common behavior of webforms like edit windows. --maveric149

It certainly is the most common behaviour, but I think that we should encourage uncommon behaviour. Too often I see edits where somebody corrects minor typos that they just made — I did this often myself when I first started out. If people Preview when they expected to Save, this causes no real harm; if people Save when they expected to Preview, then things can get messed up, even if only temporarily. But I do agree that my feature request should be implemented only if writers agree with me in this respect rather than with you; this will merit discussion. — Toby Bartels, Wednesday, July 3, 2002


There is no good solution to such a black/white issue. Perhaps the problem is in viewing it as an either/or situation. My ideal solution would impose a "default default" behavior for newcomers (Preview by default, in this case), but can easily be customized by more advanced users to do the more 'natural' thing (Save by default). Just as the login info can be saved on the local machine, so would be such customization preferences. I'm sure you can think of other such useful preferences! David 12:38 Aug 14, 2002 (PDT)

For myself, I'd be quite happy if pressing return did nothing. It does nothing inside the big text field, after all, so why should it do something inside the summary? I've no objection to a user setting that changes this, of course, except clutter. (BTW, I doubt that very many people are reading this page. I'm trying to figure out how to bring these requests back to life. I intend to move some of them to SourceForge.) — Toby 00:16 Aug 15, 2002 (PDT)

Overlay article names

There are a lot of articles that follow a naming convention that is useful for disambiguation, but a bit too long for casual linking, for instance C programming language (which is in 90% of the places linked as "C"). What I thought was, to give each article an optional field to specify an "overlay name", that would be used instead of the "ordinary name". Although not directly intuitive, this could be quite convenient for many sorts of articles, so in my opinion the pros outweigh the cons. --Uri 02:32 Aug 16, 2002 (PDT)

To some extent, we have this. In an article named B (programming language), you can make a link [[|C]], and it will be rendered as [[C (programming language)|C]]. This is a situation where using "natural disambiguation" hinders rather than helps! — Toby 02:45 Aug 16, 2002 (PDT)

Footnotes

Here's something I've been thinking about which might be pretty easy to implement. How about a method for automatically generating nice, neat, numbered footnotes? You could embed something like [[note: the North American Land Giraffe is an example of a species so rare no members have ever existed.]] into the text of an article, and then when it gets converted to HTML for display the tag gets replaced with a number that's linked to an anchor down at the bottom of the page with the text of the note in it. Wikipedia's hyperlinks between articles make footnotes not as important as they would be for a paper encyclopedia, but there are still situations where they're very handy; annotating tables of data, for example, where there isn't room to include the text and it doesn't warrant a whole separate article. You could even get fancy and have [[note:]] tags within tables get placed immediately below the table itself, associating them more closely. Bryan 07:04 Oct 6, 2002 (UTC)

More book sources

Suggestions for additional book sources - see Wikipedia:ISBN

Searches and Redirects

moved to meta:Redirects and searching

Numbering and Referencing Equations, Figures, and Tables

It would be nice if Wikipedia could provide syntax for automatically numbering figures, tables, and equations, and referring to them in an article.

It is more professional and scalable to have an article say "This is shown in Fig.1" than "This is shown in the figure below." However, figure numbering is impractical in the Wikipedia as it stands, because someone else may add a figure before that point, and all the figures would have to be manually renumbered.

In LaTex, there is a facility to automatically generate such numbers. For example,

Equation (\ref{newton2}) is Newton's second law.
\begin{equation}
  F = ma
  \label{newton2}
\end{equation}

However, in Wikipedia we do not have specialized syntax for making equations, unlike \begin{equation} in LaTeX. An unofficial convention seems to be to put equations on a standalone line with ':'

like this

but this may be used for other purposes, such as quoted text. Similarly for figures. -- CYD

Format of moved pages in Recent Changes

When you move a page, the comment attached to the move appears to be "Moved to new_article_name", which is obvious from the name of the article. Wouldn't it be better to have the comment as "Moved from old_article_name"? -- SGBailey 00:04 Feb 7, 2003 (UTC)

It appears on the line of the old article name, and in the history of that, "moved to" makes therefore more sense than "moved from". It would be better though, if the name change were also in the history of the article with the new name (here of course "moved from"). - Patrick 12:09 Feb 7, 2003 (UTC)
Also, the "N" for "new article" appears on the line of the old article name. I think it's far more logical to have this "N" appearing together with the new article name. D.D. 11:54 Feb 7, 2003 (UTC)
But the article at the now article name isn't a new article: it's the old article, just in a new place. It's the old location that has the new content.
--Paul A 07:59 Feb 11, 2003 (UTC)
An example (open quote...
  • (diff) (hist) . .N Dependent areas; 09:57 . . Dhum Dhum (Talk) (moved to "Non-independent_areas")
  • (diff) (hist) . . Non-independent areas; 09:56 . . Dhum Dhum (Talk) (dependent areas / other non-independent areas)
... close quote) There was an article in Dependent areas(DA), there was no article called Non-independent areas (NIA). There is now an article called NIA which is new (it didn't exist before) so it needs an N whilst the old article (DA) doesbn't need an N since it was just an edit to replace the entire content by a redirect. It doesn't matter that that isn't how it is handled within the database, what matters is how it is perceived by the general punter. This should have read (open quote...
  • (diff) (hist) . . Dependent areas; 09:57 . . Dhum Dhum (Talk) (moved to "Non-independent_areas", now #REDIRECT)
  • (diff) (hist) . .N Non-independent areas; 09:56 . . Dhum Dhum (Talk) (moved from Dependent areas) (dependent areas / other non-independent areas)
... close quote). IMO anyway -- SGBailey 10:18 Feb 16, 2003 (UTC)

Proper em dashes

Is it possible to add to the Wiki Markup to HTML conversion automatic conversion double dashes to proper em-dashes? Wikipedia is now totally sprinkled by double dashes that ends up looking a bit odd (or perhaps I'm just picky?) -- Egil 11:26 Apr 10, 2003 (UTC)

The last time that this came up, there were still a lot of readers using Netscape 4, which can't handle proper em dashes. (If anybody wants to see how their own browser does it, here it is, inside some quotation marks: "—".) Also, I'd be wary of an automatic conversion until we're sure that double hyphens are never needed for anything common (<nowiki> will suffice for the very uncommon). I myself used to use em dashes until I realised that some people couldn't see them. -- Toby 02:53 Apr 13, 2003 (UTC)

I'm never comfortable with arguments of the form "We shouldn't use feature X because some people with old browsers don't see it properly", because where do you draw the line? Tables? Unicode? How about forms? Wikipedia is impossible without forms, but some browsers don't do them, and there are probably a few people still using those browsers. -- Tualha 19:48, 7 Dec 2003 (UTC)

As it is now some (e.g. ISO 8859) people write &mdash; nonetheless, although &#8212; is better supported, i.e. by NS4, if the HTTP Content-Type header's charset attribute is set to UTF-8. Furthermore the em dash in English is only correctly used with no spaces surrounding it ("foo---bar"), while other languages using Latin script use the en dash with (non breaking) spaces ("foo -- bar") and probably some do mix both concepts. However most Wikipedians are not that typographically knowledged and just type a sing hyphen-minus ("-"). A similar issue are curly quotes. Crissov 16:34, 20 Apr 2004 (UTC)

longer login

I often found my login session expired by the time I finish typing an article. The article was then listed by my IP address instead of my login name. It should be nice to have an option to set the session to last longer. -- User:Kowloonese Apr 28, 2003 (PDT)

macro include

There are many articles that share the same format with similar inclusion other than the main content. One example is the articles for the cneturies, decades and years. It would be nice to factorize the repeating parts in a form of macro that can be included to ensure uniformity across all similar page.

Old feature requests

accessories

  • A gloassry for babylon with all the entries of wikipedia.

Ordered Lists

Does anyone else think it would be a good idea to modify the default stylesheet to add space between ordered list items (see User:Merphant/sandbox)? It would make paragraphs in ordered lists much easier to read. For simple bulleted lists we don't have that problem, since we can just insert a space between bullets. -- Merphant 04:38 22 Jul 2003 (UTC)

It does look much better and more readable, certainly.
James F. 07:04 22 Jul 2003 (UTC)
I agree. I definitely think it should apply to bulleted lists also. Inserting a space (in the wiki-code) between bullet points actually causes the software to generate multiple lists of one item, which is silly. -- Wapcaplet 10:27 22 Jul 2003 (UTC)
I like the 0.4 em spacing option over 0 (hard to read) and 1 (ugly and a space hog). --mav
I've mentioned it on Wikitech-l. -- Merphant 22:34 22 Jul 2003 (UTC)

Enhance experience

Or, rather dully, I've two ideas someone could implement to spread wikipedia:

(1) A little "ask-wikip" tool/script for linux distributions, that from the unix prompt can be asked for wikipedia definitions and gives them or a no such definition if their is no page. Should be like the go-button, but a bit better in regard to uppercase/lowercase (try first the exact text, if this doesn't work, try it with all uppercase first letters, maybe even play a bit with hyphenation). Could get an X interface too, and some command line options for output (html, printable html to PS, pure ascii/using lynx as filter), and even display the wikipedia page in lynx or some other browser, maybe even including the edit functionality.


> ask-wikip --ascii-only "Wikipedia:village pump"
[http://...]
Wikipedia Village Pump
This page is for asking questions. ...
 ((gives the actual text of the Village pump))
> ask-wikip ants
Opens lynx browser pointing to the article about "Ant" 

(2) Another idea would be a "go" (or better performace given, even "search") button one could include in ones own website (like the amazon partner programme). So I could have a "Look something up at Wikipedia"-text input field + button on my website. This should be fairly easy and maybe does exist already.

What do you think about these ideas? -- till we *) 12:28, Aug 1, 2003 (UTC)

Great idea! Please see the pyWikiAPI. Once it is done I (and others I am sure) will start making programs just like this. So feel free to post your suggestion on the forums there. MB 14:43, Aug 1, 2003 (UTC)
I really like the search text-field/button idea. I know I'd include it on my web site! Another tool to think about would be an add-on toolbar for IE. One that Merriam-Webster supplies (http://www.m-w.com) I find invaluable. It docks at the top of IE. Whenever I need to look up a word or synonym, I just type it in there and *BAM* a small window pops up with the info I need. I'd love a similar tool for Wikipedia. :-) —Frecklefoot 14:53, 1 Aug 2003 (UTC)
Brilliant! Something much easier to code than an IE toolbar would be a Sherlock/Mycroft plugin to Mozilla. OTOH, only a few hundred thousand people use Mozilla, but I'm pretty sure that they're rather thickly concentrated on Wikipedia. -Smack 01:00, 2 Aug 2003 (UTC)
Well, well, well. It seems that a certain Carey Evans made just such a thing - a year and a half ago. Now, children, go download the plugin. -Smack 01:13, 2 Aug 2003 (UTC)
Did a google query, found Carey Evans homepage, but no plugin. And besides, I'd rather have something that could be included in websites, too. :-) -- till we *) 18:28, Aug 2, 2003 (UTC)
Don't go to Google; go here. -Smack 22:01, 5 Aug 2003 (UTC)
Now I've found it, thanks, coooooool. (For others as blind as I: the link above will display a search menu to download Mozilla Mycroft Search Plugins; use Site Category Reference and then choose your wikipedia addition to the search-sidebar in Mozilla-related browsers. QLookUp seems to be something similar (for context menu look-ups, configurable for wikipedia). -- till we *) 00:01, Aug 9, 2003 (UTC)
Side effect: The search sidebar (at least in Netscape 7) will open itself everytime a wikipedia article is displayes, even if it isn't a search result. It's a bit annoying ... -- till we *) 00:10, Aug 9, 2003 (UTC)
To MB: The pyWikiAPI seems to be a bit empty, doesn't it? At least I didn't find any place where I easyly could mention my idea. I see the necessarity for a python (and also PHP) API to access Wikipedia from other sides, but I really can't believe something like that doesn't exist in the moment (could be a simple PHP-variables-in-the-URL-based approach or something with POST/GET-FORMs). -- till we *) 15:12, Aug 1, 2003 (UTC)
It would be very nice if I could do the following:
  1. highlight a word on a web page and right click. (Well, anyone can do it.)
  2. I get an option of "look up in Wikipedia"
  3. Upon selecting the option, a new Window opens with the highlighted word.
I have seen similar one for Google in Japanese. Tomos 11:56, 2 Aug 2003 (UTC)
Well, if you use Mozilla, you can install the aforementioned plugin, which is almost as good. And if you don't use Mozilla, you have only yourself to blame. -Smack 18:02, 2 Aug 2003 (UTC)

second idea

For the second idea I proposed (to include a Wikipedia search or go bar into custom webpages), I found out that a quick look into the sourcecode of Wikipedia generated pages is enough. There you'll find (with the small addition of http://www.wikipedia.org done by me) all that is needed to include a wikipedia search field on custom homepages:

 <form name='search' class='inline' method=get
 action="http://www.wikipedia.org/w/wiki.phtml">
 <input type=text name="search" size=19 value="">
 <input type=submit name="go" value="Go">
 </form>

Is it okay to use this? Is there someone who could create a nice (in the result) sniplet of code out of this that could be included anywhere? -- And shouldn't the search function better be really disabled instead of only be commented out? -- till we *) 15:20, Aug 3, 2003 (UTC)

((oh, and by the way: Why does the go field (neither this nor the one in wikipedia proper not work, but brings up random pages, when the page doesn't exist ("Student union" brought me to Soviet union as well as Christian Social Union in Bavaria, Student test brought up Student nursery)? -- till we *) 15:25, Aug 3, 2003 (UTC)))


Can we have a variable name like {{TALKPAGE}} ("TALKPAGE") which expands to a link to the talk page for a page. It will make certain boilerplate text much easier to edit.

I made Template:Talk; use {{talk}} (works in main namespace only).--Patrick 12:59, 27 Jul 2004 (UTC)

move to wikipedia:ignored feature requests

Is it possible to make a link show the table of contents of a page instead of its title? This could be useful for things like archived talks or lists where you would like to see an overview but not the full content. 141.83.55.66

Oh, you should write an overview yourself. AI is just not there yet. wshun 02:38, 9 Aug 2003 (UTC)

Edit Conflicts

Would it be possible for a developer to change the edit conflict code to give the conflicter an option of overriding the previous user's edit without needing to use copy and paste? I use a text browser called links, and I cannot copy or paste, so I loose all my work if I get into an edit conflict. It would be better if I could override the other user and at least he would have his work in the history. Presently I have no way of getting the work back once I get into a conflict. Thank you very much, Greenmountainboy 21:27, 15 Dec 2003 (UTC)

If you overwrite another user during an edit conflict, you are basically reverting them for no reason, and that's very bad form. It comes down to a question of being nice to other users, or nice to yourself. I think you should explore other ways of avoiding this problem (off the top of my head: changing software; making big edits incrementally; putting a "I'm doing a big edit right now, so please wait." notice at the top of the article, making big edits to a personal copy of the article, ...) before asking a developer to make this specific feature. See also our policy on edit conflicts. -- Cyan 21:58, 15 Dec 2003 (UTC)
If it's a linux-like browser, and you are used to windows, maybe Control+Space at the beginning of the text, and Control+W at the end to cut, then Control+Y to paste. (Meta+W to copy instead of cut, if the Meta-key exists.) Or, if it's a windows-like browser, and you are used to linux, Shift+Arrows to select, Control+C to copy, Control+V to paste. Or maybe it just can't copy/paste, anyway. Κσυπ Cyp   22:06, 15 Dec 2003 (UTC)
Well, Its links, and you just cant copy/paste anyway, but I will get a "window manager" tomorrow, and then mozilla firebird, but until then I will have to wait. I just got frustrated, but I am sure that there are others who are not so fortunate as me, and will have to continue using links or even lynx into eternity. Greenmountainboy 22:14, 15 Dec 2003 (UTC)
Can't you specify using an editor for text fields? I think Lynx had an option for that - though not sure for Links... Dysprosia 22:47, 15 Dec 2003 (UTC)

Remove from watchlist coding thing

Why is it that when you click Stop watching, you are offered a link to return to Main Page when you've almost certainly come from your watchlist and want to go back there? Bmills 16:30, 10 Dec 2003 (UTC)

better stylesheet/css

I love wikipedia, but I think that the body text of articles can get a little hard to read with all the links.

Wouldn't it be a good idea to change the stylesheet a little to make the links integrate better with the text.

Not to the point where the links can't be told from text, but just enough to not make the disturb the reading.

Once users grasp the idea of wikipedia they should assume that most terms are links and a small effect when you move the mouse over a link could confirm that.

/Lasse

Not as the default stylesheet, please. I think there are two issues. One is making too many unrelated words into links. I think that definitely happens in some articles where every third word is linked (or worse). The second is that it might be worth having another style sheet that is more subtle about links. Daniel Quinlan 23:40, Dec 11, 2003 (UTC)
If you are using mozilla you can use custom styles in wikipedia by adding something like this to your userContent.css file:
    body[onload='setup("quickbar")'] <element> {
        <styles>
        ...
    }
    ...
This is a workaround because there is currently no way to make site-specific style sheets. So I took advantage of the fact that wikipedia is the only site I know of that has the onload property of its body element set to 'onload='setup("quickbar")'.
—Noldoaran (Talk) 00:07, Dec 12, 2003 (UTC)

Is there a log of "rename-events"?

(i posted this question on Wikipedia talk:How to rename (move) a page, but I don't know if anyone really frequents there)

Question: When a users renames (moves) an article, that change is not listed under the article's page history. How then are we to know who renamed (moved) the article and when it occurred? Is there a log (something akin to Wikipedia:Deletion log) that I don't know about? Kingturtle 05:14, 12 Dec 2003 (UTC)

The user automatically creates a redirect at the place from where the article was moved. Like this: [1]. --Jiang | Talk 05:20, 12 Dec 2003 (UTC)
However, if the page is moved back to its previous name for some reason, the history of the original move is erased. —Minesweeper 08:04, 12 Dec 2003 (UTC)
It would be especially helpful to have a log of all renames. An edit war of sorts could break out between two admins in regards to a name of an article. Name changes could go back and forth, and it would be very difficult to follow. Kingturtle 03:04, 13 Dec 2003 (UTC)
I agree. A move log would be a good thing. KingTurtle: why don't you post this on the requested features at sourceforge. —Noldoaran (Talk) 06:13, Dec 14, 2003 (UTC)
I agree too, but I have an idea that might be more useful than one central move log: what about a "move history" for every page, just like there's a "page history"? This way, the "move history" would be transferred along with the article's contents and editing history whenever it is renamed. As a result, the log of rename events for a particular article would be stored in one place. If it were in one central log, I fear that it would become long, spread-out, and unwieldy in keeping track of a move-war, for example. Although, maybe a centralized move log could complement an individualized move history for each article. I'll post this to SourceForge unless someone has a better idea. —Minesweeper 10:58, 14 Dec 2003 (UTC)
Why not just treat a change to the name of the article like any change in the article itself? Just include an entry in the existing change log, noting that the name was changed (the page moved) and when and by whom. -Anthropos 02:27, 16 Dec 2003 (UTC)
That would keep it one place, but it would might break up the existing list and might confuse the "cur" and "last" features. It would help in keeping track of the timeline of the page's history, though. Minesweeper 09:52, 16 Dec 2003 (UTC)
Agreed. Such a log would be useful to keep track of the insanity of those people indulged into moving wars. --Menchi (Talk)â 09:54, 16 Dec 2003 (UTC)

Rollback Edit summary

Rollback (see Wikipedia:Administrators#Reverting) currently gives a edit summary that shows up in the history as: reverted to last edit by .... Unfortunantly, this gives no indicatation as to why a edit was removed and makes no indication that a edit was reverted by an automated program. I think that one of two things should be done:

  1. Make in blindingly clear that this is an automated process. The Message could be something like: reverted to last edit by GoodUser as part of an automated rollback of all of BadUser's edits.
  2. Require rollbacks to have some kind of reason. reverted to last edit by GoodUser. Baduser has randomly deleted text in multiple articles.

This would make it clear to someone who is looking at the edit to try and figure out why on earth an edit has been reverted. It also gives a user some idea as to why their edits are being reverted (especially in cases of mistaken identity). Jrincayc 16:45, 14 Dec 2003 (UTC)

Yes, that might be a good feature (similar to what's done with deletion). Currently, I think it's done this way as a timesaver feature (one click reverting). Whenever I revert something that is not obvious I usually leave a note on the article's discussion page. Note also, if someone deletes part of an article without mentioning why, (s)he may also be reverted without comment. That's why it's always a good idea to put something in the Summary. Dori | Talk 16:49, Dec 14, 2003 (UTC)
The point of the rollback button is that it makes defending against vandalism very quick and very easy. This benefit would be destroyed if you had to type in an edit summary explaining what you were doing. Angela. 16:57, 14 Dec 2003 (UTC)
Perhaps a small set of radio buttons could be added, to help annotate that process. These would append the appropriate reason to the edit summary. Options might include "vandalism", "banned user", "see talk", producing edit summaries like "reverted to last edit by Hamish (vandalism)" -- Finlay McWalter 17:12, 14 Dec 2003 (UTC)
It would still be nice to have the option of adding an edit summary. This would be especially useful when reverting edits by a banned user; it would be very easy to write one edit summary, copy it, and then quickly paste into into the summary box with each revert. Non-obvious vandalism doesn't need to be reverted as fast and needs more annotation. --mav 18:05, 14 Dec 2003 (UTC)
Just give a place to put a summary, and if the person doing the rollback does not feel like giving a summary, than give a more detailed message like: reverted to last edit by GoodUser as part of an automated rollback of all of BadUser's edits so that it is clear that the revert is part of a rollback and not an individually checked edit. Jrincayc 20:54, 14 Dec 2003 (UTC)
Excellent idea, Jrincayc! This makes such a lot sense, gets my support anyway. Chris Jefferies 23:18, 14 Dec 2003 (UTC)
The software doesn't automatically know if the revert is part of series of them or not though. Angela.
I think it would be best to have a summary box on the user contributions page, and then use JavaScript to extract it and append it to the rollback URL on each click. -- Tim Starling 23:34, Dec 14, 2003 (UTC)
Would that work if there was more than one person doing the rollbacks though? Angela. 23:45, 14 Dec 2003 (UTC)
Yes, why wouldn't it? They wouldn't necessarily use the same edit summaries, of course. -- Tim Starling 00:08, Dec 15, 2003 (UTC)
This would be broken under Lynx (which I think would just ignore it, but that should be tested). Pakaran 00:36, 15 Dec 2003 (UTC)
This is how the simple version could be done. I think that to implement just the automatic message would just take changing the message line in Language.php to something like:
"revertpage" => "Reverted to last edit by $1 as part of an automated rollback of all of $2's edits",
and the line that uses this in rollback() in Articles.php to:
$newcomment = wfMsg( "revertpage", $s->old_user_text, $ut );
Adding a user specified message will be more complicated. Jrincayc 00:48, 15 Dec 2003 (UTC)
  1. The one-click sysop rollback exists for the sole purpose of speeding reverts of mass vandalism, which should be obvious to anyone looking at what the edit did.
  2. This is one click per page. There is no such thing as an "automated rollback of all of X's edits".
  3. If you want to put a detailed summary/reason, you can always do so with history/edit/save. --Brion 00:20, 15 Dec 2003 (UTC)
Does this mean that the revert button cannot/should not be used for isolated instances of vandalism, but only mass vandalism? -- BCorr ¤ Брайен 00:56, 15 Dec 2003 (UTC)
I hope that people are not saying it can not be used for this. It is also used to mass revert edits by banned users, which seems to be what has started this issue. Do these count as vandalism? Is it obvious enough that the revert is being made because the user is banned? I'm thinking particularly of Michael's edits, which may appear on the surface to be fine, but are very often auto-reverted. Angela. 01:01, 15 Dec 2003 (UTC)
No I think Jrincayc has misunderstood their purpose. Revert is just an easy, one-click way to revert. It is used for auto-reverting, or for reverting single contributions. I use it all the time for one-time vandalisms. If the vandal does one bad edit, followed by one good edit, I don't revert them both. Dori | Talk 01:04, Dec 15, 2003 (UTC)

I think the way it should work is that the process should be a two-step one. For this to be easier though, I think the revert should be next to a page history and there should be a diff link next to a user's contributions (see meta:MediaWiki_feature_requests_and_bug_reports#Reverting). 1) click on revert link next to a diff (either in page history, or next to contribution) 2) a new page with a text box appears that has some automatic text, and you can add your own reason, or just click OK (sort of like the delete option). Dori | Talk 01:08, Dec 15, 2003 (UTC)

I don't think you quite understand the purpose of this. This is meant to help against instances where there might be ten, twenty, fifty, junk edits that need to be reverted. Firing off a bunch of one-click reverts and then checking the results is easy. Making them require two clicks each makes the process a lot slower and more difficult. --Brion 01:27, 15 Dec 2003 (UTC)
It's just a matter of implementing it, because you could just as easily make it so that there are a bunch of checkboxes, you select all the ones you want to revert, the revert reason is the same for all of them (maybe even a select all edits still on top). You still go trhough two pages, but you have reverted more than two edits for only one extra confirm click. Dori | Talk 02:19, Dec 15, 2003 (UTC)
I just see the effect as a non-sysop user and today I saw a revert that had no visible cause and no visible reason, so I made the incorrect assumtion that the reverter was being a dink by reverting a perfectly good edit. If there was some way to make a better comment, I would have realized that the revert was part of a larger rollback of all the user's changes, which would have saved some heated words. The message as is does not give a clue as to why the edit has been reverted, which is bad (Don't bite the newcomers ...). Jrincayc 01:22, 15 Dec 2003 (UTC)

However, it is good for dissuade reputation. Which doesn't work, but still. Martin

Interlanguage Wiki concerns

Will it ever be possible to associate one user account to multiple language wikipedias? I know, I know, non-essential wishful thinking. But a couple of questions of greater immediacy:

  1. I can't seem to create an account at jp.Wikipedia.org. The page that I get sent to says "On this page currently there exists no content." 何???
  2. I have a shiny, new account at zh.Wikipedia.org, but every aspect of my user interface is in Simplified Chinese, e.g. the Quicklink bar, Preferences page, etc. Isn't there a Traditional Chinese user interface? I already tried asking at the Chinese Village Pump, and somebody replied, "What do you mean?" Not... a... good... sign. Also, they don't use the zh-cn and zh-tw tags, nor do they observe due process deletion policies. I can't get a page started over there because some guy in Beijing keeps deleting before I can get started translating in earnest! Lord of the Flies, I tell you, Lord of the Flies.

~ stardust 05:27, 16 Dec 2003 (UTC)

If you created an article with only external links or interwiki links here it would be deleted as well. I think it is unfair of you to accuse the Chinese Wikipedia of not following due deletion policy. As far as I can tell, there wasn't any actual content in that page any of the times you created it. If you need to save an article before it has any content, it is best to do this your user namespace. You can then move the page to the main namespace when it is beyond the stage at which it can be instantly deleted. Try creating it at zh:User:Stardust/Catan instead. Angela. 05:45, 16 Dec 2003 (UTC)

Accidentally posting anonymously

Suggested feature/mod: it might be useful to modify the behavior of the page-edit code a touch: I didn't realize my login cookie hadn't stuck, and did a few edits while not logged in. I'd say that it might be nice to have either the login name or anonymous:IP *right down by the submit button*... and maybe a way to catch a login and password with a combined "Log In And Submit" button as a third choice...? -Baylink 18:21, 15 Dec 2003 (UTC)

Usernames

Would it be feasible to put a 32-character limit on usernames? - Hephaestos 04:57, 18 Dec 2003 (UTC)

You read my mind. Actually I think you read the mind of many people. It might interest you to know that the longest apparently non-trolling name on the English Wikipedia is 37 characters "Johann Peter Gustav Lejeune Dirichlet". The longest name overall is our friend "What most..." at 62 chars. -- Tim Starling 06:59, Dec 18, 2003 (UTC)
I agree that such a limit would be a good idea. I think 32 or 40 characters are enough for everyone, but the limit could be just 20 or 24 too. Peace. Optim 07:28, 18 Dec 2003 (UTC)
Maybe the limit could be 3-4 characters, actually... That would probably deal with long usernames completely. Κσυπ Cyp   09:16, 18 Dec 2003 (UTC)
With a maximum of 3 characters and with the English 26-letter alphabet, we can have up to 15600 different Wikipedians. With 4 characters we can have 358800. With 5 characters 7893600. Check Permutation. how many registered users do we have right now? Optim 20:57, 18 Dec 2003 (UTC)
Just FYI, 32 bytes may be as little as 8-10 characters for some languages. --Brion 09:19, 18 Dec 2003 (UTC)
Can the limit be on characters (i.e. Unicode codepoints) rather than bytes? --Delirium 03:21, Dec 19, 2003 (UTC)
Can you guys/gals please just skip this character limit alltogather? my username is 24 characters ( my given name ), so it would not fit within Optim's latter suggestion. I could be using another family name of mine and my username would then reach 36 characters. As you get people from different nations and cultures here you get different name systems with it, please dont deny someone the use of his real name because of trolls abusing it and it not being usual in your respective countries and or naming systems. Thanks in advance, --Ævar Arnfjörð Bjarmason 18:46, 2004 Apr 26 (UTC)

Contributions count

I'd like to request a feature for generating the number of user contributions someone has made. Like ~~~~, it gets replaced with a number once the article gets saved. I would suggest a format such as {{contributions:Username}}. →Raul654 07:47, Feb 1, 2004 (UTC)

I'd suggest not. Any time the number of contributions made is easy to find, people will make contributions for the sole purpose of making that number bigger. Just look at any forum on the internet. --Carnildo 10:19, 1 Feb 2004 (UTC)
It's not like I'm asking for it to be displayed prominently on the user page. I just want something easier than going into the contribution history and counting (or using size cutoffs to count quickly, but it's still the same thing). Besides, if seeing their edit count makes them contribute more, then I'm all for it. →Raul654 10:22, Feb 1, 2004 (UTC)
Quality is more important than quantity. This is the slipperly slope to becoming another Slashdot. CGS 11:13, 1 Feb 2004 (UTC).


WHEELER's idea

Wikipedia is a great idea. The Software is awesome.

What I like is when one does = = this it bolds it and a box is created for easier viewing of sections. It is called the "Table of Contents".

Can we do something for references within an article. Some kind of keyboard movements. The reference section is automatically updated and numbered much like the Section box is. Anywhere where a person edits and puts in a reference, it is automatically numbered and put into reference section and leaves a number. And maybe if it isn't too difficult then rearranges all the numbers in the article and section. (Boy, is this going to be somebodies nightmare.) (I can hear somebody cursing now.)

Another good suggestion is that it can be seamless. Unlike the table of contents, the Reference section should be of the same color background and the same type and the reference numbers as unobtrusive as possible.

I think this would greatly improve Wikipedia and probably encourage and motivate contributors to use references!!!!

My watchlist

I propose a small modification to the header section of the Special:Watchlist page, similar to a feature already present on the Special:Recentchanges page. In Recentchanges there is an option to show/hide logged in users, bots, etc. I would like an option in "My watchlist" to show/hide my own edits so that edits by others are far more evident. - Gaz 12:32, 1 Mar 2004 (UTC)

Shouldn't this go to Wikipedia:Feature_requests? —Frecklefoot 16:44, 1 Mar 2004 (UTC)

Userpages database proposal

Summary: Proposal to have separate databases for userpages, talkpages and articles.

Hello,

I think it would be a good idea to have separate databases (or downloadable files) for userpages, talkpages and articles.

Many Wikipedia mirrors copy userpages and talkpages. I don't like that. It doesn't help to protect the privacy of Wikipedians.

Having separate databases (or downloadable files) will help the people who mirror our content to copy just what they really want (the articles) and not userpages and talkpages, which may contain information some Wikipedians (like me) would prefer be available only on servers controlled by Wikimedia Foundation.

By having separate downloadable files for articles and userpages/talkpages, we give the ability to downloaders get only the material they are really interested in: The articles. I see no reason why somebody would want to download all userpages and talkpages, except if he/she is interested to research how Wikipedians communicate with each other etc. In reality I don't see any need to have userpages and talkpages downloadable, so such downloadable files could not exist at all.

We could just have a downloadable file for Articles and another one for the whole database (Articles+Userpages+Talkpages). Actually I just want to make sure that "accidental" copying/downloading/redistribution of userpages and talkpages will not happen again.

Please comment, support, oppose, criticise and inform us about your opinion.

Thank you.

--Optim

Redirection messages

I disagree. I often make redirects for common spelling errors or alternative spellings. This may make things very confusing. For example, saying that "NATO is commonly referred to as Nato" makes little sense to me. Also, in this example there are multiple redirects (i.e. North Atlantic Treaty Organisation). Are you Anthony DiPierro? (forgot to sign)-- chris_73 00:03, 13 Apr 2004 (UTC)
Very interesting, but at the moment it would appear to be that the link to the redirect has to be the first word after "#REDIRECT". This could make for some odd grammar. Best, if possible (I've never really dealt with the Wikipedia code), to have the costume redirect message a line below the "#REDIRECT [[something]]". I gather (from Wikipedia:Redirect, obviously) that at the moment one can freely enter an explanation for the redirect at this point - never seen by the user but seen by the wiki editor - but have rarely seen this used in practice (a selection of examples for the use of this can be found at [2]). Other than that, this is an interesting idea, that can greatly contribute to the encyclopedia. Or it could make things a lot more complicated. It should be given a chance, though. (Especially as reverting to the previous version shouldn't be too complicated.) -- Itai 03:03, 13 Apr 2004 (UTC)
I agree. I like it. Something similar to this has been discussed many times, but I like this implementation than the idea as it has previously been discussed. I think the original idea was to change the syntax so one can use #mispelling or #alternative or #deprecated instead of #redirect, thus changing the end message appropriately. Anthony's system allows for more flexibility and is probably easier to use. Tuf-Kat 03:51, Apr 13, 2004 (UTC)
If the message can be adjusted flexible according to the needs, then I have no problem with it (although it is not a high priority for me). For example a syntax like Redirect[[NATO|Redirected from incorrect capitalization '''[[Nato]]''']] may be useful. Note: The comments after the redirect mentioned by Itai do not show on the target page. -- chris_73 08:00, 13 Apr 2004 (UTC)

I think this idea opens many possibilities, as I generally think redirs and disambigs are very ugly. Suppose we put an article that is today naturally at an un-paranthesed spot, like Ireland at Ireland (country) instead. This will allow us to put any disambig in the Redir message from Ireland -> Ireland (country), and articles linking to Ireland can link directly, and the user does not have to see the disambiguation. ✏ Sverdrup 10:17, 17 Apr 2004 (UTC)

Probably we have to develop something quite a bit more complicated than your proposal (but this is the trial of the idea, making it possible, gj), but I hope it will be easy to use. We'll need a way to on a redir page, (lets pick Enigma, a real example this time) put a disambig message there that will be shown instead of the current redir message. I propose putting text like
This article relates to the Enigma machine in cryptography. For other uses of Enigma, see Enigma (disambiguation).
Below the #redir instruction in Engima, and when the redirect is made, it will display much like it does now (on Enigma machine), with (redirected from Enigma or something in small where the redir message is now. When directly linked to Enigma machine, nothing will show. When a normal (non-disambig) redir is made, the standard redir message will simply show. (I propose to enclose it in < small> or something. I hope this idea is caught by the right people. ✏ Sverdrup 20:07, 18 Apr 2004 (UTC)
Just a note that this is no longer a valid example due to the move of the Enigma pages. --Dante Alighieri | Talk 20:13, Apr 20, 2004 (UTC)

What MediaWiki message where

In both of my comments above (yes, I'm mass Pumping here, sorry) I refer to system messages which are probably in the MediaWiki namespace somewhere. However, one can only guess where. Although there are lists of all messages used by the software, there doesn't seem to be any explanation of how they are used by the software. Perhaps some nice developer who knows which is which could add a note somewhere - the top of corresponding MediaWiki talk: pages seems an obvious choice, and the software could even generate these automatically for a new wiki... - IMSoP 21:06, 20 Apr 2004 (UTC)

This really belongs into the MediaWiki documentation. It is currently being worked on on Meta, but will reside in a neat help namespace in the next version of MediaWiki, and be maintained on the respective language Wikipedias. Documentation will begin in earnest as soon as MediaWiki 1.3 goes live.--Eloquence* 00:07, Apr 21, 2004 (UTC)
No - the English version will still be maintained on Meta. There will be periodic updates to the English Wikipedia, Wikibooks, Wiktionary, and Wikiquote from the Meta version. --mav 21:52, 21 Apr 2004 (UTC)


Tool bar/search bar

Does any one have the knowledge or the time to code and submit open source or donate code for a Tool bar/search bar like google.com, or dictionary.com

Do you mean a search bar for Wikipedia? Hmm. That sounds like a cool idea. I'm not a programmer myself, but it sounds like a worthwhile project. Isomorphic 20:28, 13 Apr 2004 (UTC)

If you use a modern browser like Mozilla/Firefox/anything Gecko, Safari, Konqueror or (possibly) Opera you can easily do something like this:

See http://www.mozilla.org/docs/end-user/keywords.html for more info on this. -- Gabriel Wicke 22:20, 29 Apr 2004 (UTC)


Hyperlinks between articles are a central Wikipedia feature. But it could be more useful. Take the "What's link here" functionnality: it lists many links, but it doesn't tell you why the articles are linked. I would like to be able to add an information (a predefined label) on each hyperlink. This information could then be used to comment links in the "What's link here" and even allow semantic requests like "give me the list of people born in this city". The only missing information to reach this semantic Web nirvana is hyperlink label, like that: Mozart ([[January 27:birthdate]], [[1756:birthdate]] - [[December 5:deathdate]], [[1791:deathdate]]) was born in [[Salzburg:birthplace]]. Marc Mongenet 23:26, 2004 Jun 30 (UTC)

This is one of those ideas that sounds really cool, but would be hell to implement. You'd need to develop some kind of RDF-like vocabulary and then actually get people to use it. And the wikipedia engine would need to be updated to use it. What would be cooler would be a series of templates for entering biographical data, etc, and have the engine reverse-populate links to those templated pages with metadata from them. But again, someone would need to make it happen. And in any case, this is just one guy's opinion, and you know what opinions are like. adamrice
Yes, the vocabulary would be the most controversial part. But I think it would still be easier, better and much more useful than present categories. Templates are far less general and have many constraints. They would be very useful (necessary?) to achieve a standard presentation, but not much more. Anyway, as you conclude, it is first the people coding the application, then (maybe) the people writing articles that will decide. Marc Mongenet 02:18, 2004 Jul 3 (UTC)


Arbitrary background colours in articles

So, whilst browsing random page, I can across Francisco S. Carvajal. This article has a table with background colour of #FAEBD7; problem is, my link colour is #FFFFC0 [3]. Spot the difference. I imagine this problem is duplicated in many articles containing such tables, with similar effect. Could these colours somehow be moved into CSS, perhaps with a protected MediaWiki template page containing CSS which articles use, and which could then be overridden in a user's own monobook.css or myskin.css? This problem really significantly reduces the use of allowing users to use their own skins. Lady Lysiŋe Ikiŋsile | Talk 19:55, 2004 Jul 2 (UTC)


"Preview" in "Login Successful" screen page title

The page title - I don't know if that's the correct term but I mean the wording used in a browser's title bar - for the screen confirming a successful login has the word "preview" in it. I assume this is a hangover from some development phase and should now be removed. If not, what is it previewing?? I apologise for reposting this but it's had no reply here and no reply in the 1.3 bug reports yet so I assume it's overlooked. --Nevilley 23:56, 5 Jul 2004 (UTC)

I don't know where this could be fixed, but the actual text consits of:
Login successful - Preview - Wikipedia, the free encyclopedia
Dori | Talk 15:21, Jul 6, 2004 (UTC)
Thanks for clarifying that. I suspect that it may be easily fixable, but is caught between (a) not being doable by admins but (b) being a low priority for programmers, since it is irritating but not actually broken. Oh well. --Nevilley 09:57, 12 Jul 2004 (UTC)