Wikipedia:Village pump (proposals)

The proposals section of the village pump is used to discuss new ideas and proposal that are not policy related (see Wikipedia:Village pump (policy) for that).

Please sign and date your post (by typing ~~~~ or clicking the signature icon in the edit toolbar).

Start a new discussion in the proposals section

 Policy Technical Proposals Idea lab WMF Miscellaneous 

• Table of contents
• First discussion
• End of page
• New post

« Archives, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218

Proposals archive

Discussions older than 7 days (date of last made comment) are moved here. These dicussions will be kept archived for 7 more days. During this period the discussion can be moved to a relevant talk page if appropriate. After 7 days the discussion will be permanently removed.

Remind people to cite sources while editing

I suggest adding underneath the edit box a short, simple message:

Please cite your sources so others can check your work.

If some people think that's a good idea, and others don't, perhaps we could try it out for a few days as an experiment and see if there's any harm.

There's an infinite number of style suggestions, but for other stuff, it's easy to fix things later. Formatting weird? You can fix it. Bad grammar? You can fix it. POV? Usually you can fix that too, IF you can find out where the data came from. But if you've no idea where the info came from, it's really difficult to do fact-checking later. Thus, I think citing sources is unique - it's more difficult to fix later. I'm not perfect, but I try to do this. Reminding people -- and especially telling newcomers -- would make it more likely to happen.

In particular, I think this would help deal with one of the main complaints about Wikipedia: "Can I trust it?" With sources cited, you can check things yourself, and many people have greater trust in an article when the citations are included. It also makes the fact-checking folk's work easier/possible. -- Dwheeler 23:33, 2004 Dec 17 (UTC)

I've seen absolutely zero protests about this, after 4 days, and this can easily be undone. People who only read Wikipedia won't even notice! So, I've given this change a whirl -- hopefully others will like it! Dwheeler 23:52, 2004 Dec 21 (UTC)

Actually, I don't like it. It adds even more information to the bottom, which even less people read. And, frankly, citations are in my opinion only a small priority for the user to understand. I personally prefer the old version, it puts much less load on the user. -- Chris 73 Talk 02:15, Dec 22, 2004 (UTC)

Sorry you don't like it. But I think citations are absolutely vital; as I noted above, everything else can typically fixed by at least some other user (NPOV, etc.), but without a citation, it's really hard to make progress. This simply reminds people of a policy that's already in place, but all too often unintentionally forgotten. And it's not much text, so it shouldn't have too much load on the user. Anyway, this is obviously trivial to reverse. Let's see what others think about it, and go from there. -- Dwheeler 02:53, 2004 Dec 22 (UTC)

I agree completely with Dwheeler, citing sources is enormously important for fact checking articles and helping the reader know that what they are reading is reputable. It also allows readers to find more detail on the article if they are trying to research it. Adding a one line note to cite your sources would take very little space, it'd be easy to implement, and it would remind everyone that when they add factual information to an article, they should cite their sources. --Soren9580 04:01, 22 Dec 2004 (UTC)

I tend to agree with Dwheeler. When I first found this place, it was over a nutrition subject in which I learned that a certain nutrient isn't contained in a type of food a very ill family member was eating regularly to receive that nutrient from. Without source checking however, I have no idea if the information I read here is correct or just someone's opinion. This didn't occur to me till recently because I originally blindly accepted the info I read in the article as fact and made dietary health purchase decisions based on this. So I think we should cite sources where we can, such as when making statements like "spinach has oxalic acid in it which blocks the absorption of iron." The statement is true, I'm sure, but people need to know where it came from. -Emerman, December 28, 2004, 11:40 a.m. Eastern.

We should realize that many people write articles from their own experience or knowledge, and don't necessarily have a source to cite. I don't think we should unnecessarily hamstring article development by making this an absolutely hard requuirement. Keeping it as a recommendation is fair however. --Stevietheman 16:46, 28 Dec 2004 (UTC)

This doesn't hamstring article development. It's just a recommendation/plea. The text says "Please cite your sources so others can check your work." All it says is "please", it certainly doesn't say "you must". It doesn't stop you from adding pages with references! But a lot of people actually know where their info came from, if only we reminded them to tell us. -- Dwheeler

Point taken, but if some people are discouraged from contributing because they think that have to provide a source, based on the "Please", then that's something to think about. --Stevietheman 00:57, 29 Dec 2004 (UTC)

I understand. Personally, I doubt that will become an issue. -- Dwheeler 17:05, 2004 Dec 30 (UTC)

I am definitely not a Wikipedia authority. However, I've scanned the VfD pages, and those guys are forced to Google every single time because no one cited any source whatsoever. Citation simply makes an article credible. Does it take extra work? Oh, yeah. Does it make a difference? Rhetorical question.

All for it! allie 18:50, 8 Jan 2005 (UTC)

Sources are needed, of course, but as a minimum i would make the comment below the edit box not bold, since this distracts from the DO NOT SUBMIT COPYRIGHTED WORK WITHOUT PERMISSION!, which I think is way more important. -- Chris 73 Talk 00:34, Jan 12, 2005 (UTC)

Article popularity

Chris Anderson (Wired Magazine editor) has a new blog called The Long Tail [1] which discusses the nature of Amazon and Wikipedia and Netflix (and many others) that provide access to content that ranks low on the popularity scale (obscure books, obscure encyclopedia articles, obscure movies) that traditional channels have ignored (compare the number of articles in WP to EB or titles at Blockbuster to Netflix). These obscure titles are the "long tail" on the popularity chart. His most recent entry [2] discusses the nature of "choice" and how we decide which of these obscure titles to read/watch, and the overwhelming variety of choice. He says that we depend on metadata and the wisdom of the crowds-- for example, Google rankings are based on what other people are linking too. Or on Amazon there are item popularity rankings, the circles feature, and other ways.

Wikipedia addresses this problem two ways. First it has Featured Articles to showcase the best articles, and second it relies on Google ranking for the project as a whole. However I contend that these are not enough. For one, Featured Articles is not wisdom of the crowds (more like 10 or 20 people) and is akin to the traditional method of gate-keeping recommendations such as journal editors. Second the google ranking breaks down on obscure titles that may have 1 to none external links to it.

Therefore I propose that a new system of popularity ranking be examined. Wikipedia has close to half a million articles, no one really knows whats here, if you read every article it would take a lifetime. We absolutely must have a system of recommendation.

Because this would involve resources (programming and otherwise) I'm not sure how fruitfull a discussion it would be without input from the powers that be. However, I will say that wisdom of the crowds is the only way to scale such a system. Perhaps one of the easiest and most obvious suggestions would be a page hit counter for each article. --Stbalbach 16:46, 6 Jan 2005 (UTC)

There is a page hit counter, but it has been turned off for a long time because it overloads the servers. 68.237.137.57 03:25, 7 Jan 2005 (UTC)

Until recently there also used to be a monthly analysis of the server logs which could be used to count hits, but these have been halted — also on performance grounds, I believe. — Matt Crypto 09:36, 7 Jan 2005 (UTC)

My experience running a core Usenet server is that logs take a lot of resources, and eventually it become fiscally prudent to send the logs to /dev/null, so I totally understand with the number of hits Wikipedia get this may not be possible at this time. I do wonder though if there are thoughts or ideas on how to determine an articles popularity? It should of course be possible to look at article ranking at Google, and then add that information to the bottom of the article (all automated of course). This could be done on a monthly or less to keep resources at a minimum, and it could be very un-intrusive on the page. I've done something manually along those lines in my personal articles page User:Stbalbach/Articles to get an idea of what it is. --Stbalbach 17:17, 8 Jan 2005 (UTC)

No need for a seperate system. I would recommend reading featured articles... Mgm|^(talk) 12:45, Jan 20, 2005 (UTC)

Featured Articles is not wisdom of the crowds (more like 10 or 20 people) and is akin to the traditional method of gate-keeping recommendations such as journal editors. .. also many of the featured articles are, IMO, not very good compared to many articles where no one has taken the effort to nominate it. I mean, out of half a million plus articles, there are more than 900 good articles. Indeed many of the featured articles have, uh, put on some weight and no longer in shape as they were when they were nominated (and, nomination standards keep improving). --Stbalbach 16:45, 20 Jan 2005 (UTC)

I heard somewhere that a review system is being developed for Wikipedia. That would be very useful in this regard. Couple this with a rating system (and page hit counters when resources allow) and you're home and hosed. Not sure what code they use but lovefilm.com have an excellent rating system (you'll have to register to try it unfortunately). I'd love to see a rating system that attempted to take into account people's political outlook. For instance you could get members to take a political compass test like that on politicalcompass.org before they can rate articles. This would then allow readers to categorise ratings based on where raters see themselves on the political spectrum. —Christiaan - ☎ 16:29, 20 Jan 2005

Those are all great ideas. How do we account for articles changing in quality, for better or worse, over time using a rating system? --Stbalbach 16:45, 20 Jan 2005 (UTC)

That's a really good question. Maybe ratings could expire when a certain percentage (rater determined?) of the article has been modified, at which point raters would be invited to confirm or modify their rating? —Christiaan - ☎ 19:05, 20 Jan 2005

Is a rating system even practical at all? It can't be complicated. Since the articles are in constant flux and change, one can't give an article 7 stars in any meaningful way because in the future, it may be an entirely different article. So for things that are constantly changing, how does one rate it? This is a lot like the problem of ranking stocks, except we have almost no data to work with. The Google rank is sort of the markets vote on an articles popularity. Although I noticed when using the "link:" command in Google, most of the sites linking to a Wikipedia article are from within Wikipedia, a sort of insider trading pyramid scheme. --Stbalbach 19:04, 20 Jan 2005 (UTC)

Um, I don't understand your question, "So for things that are constantly changing, how does one rate it?" Why would you ask that in response to an answer? As for Google rank, that's only a measure of popularity, it's not neccessarily a measure of the quality of an article (especially obscure ones). —Christiaan - ☎ 20:25, 20 Jan 2005

It was rhetorical question. I agree about Google, although popularity is often a guage for relevance, which I think is what most people are really looking for (the reason Google is so popular). If it's relevant to other people, good chance it will be useful to yourself as well. --Stbalbach 04:47, 22 Jan 2005 (UTC)

Category counting

Not sure if this belongs here or in technical... Since the 1.4 upgrade, big categories no longer list all items on the first page, but break the category into 200-article chunks. All well and good, but is there any way that the total number of items could also be displayed (sort of like the Google-styled "displaying items 201-400 out of 2371")? A similar thing would be very useful on the search results, too. Grutness|^hello? 00:31, 7 Jan 2005 (UTC)

It would also be useful to 1-count the number of images in a category, period, and 2-have sub-categories listed independently of the no-more-than-200-per-page system. It's very difficult to track down all of the sub-categories of a large category as things currently stand. (This makes sub-categories for those categories effectively useless.) -Aranel ("Sarah") 18:33, 13 Jan 2005 (UTC)

Login bars

On this website there is a login bar at the top of the page with a link to register. Would that be possible on WP, it would make things so much easier. Please reply here rather than on my talkpage, see my talkpage for reason.--Gabriel (internal ID number: 118170) 17:49, 9 Jan 2005 (UTC)

While it makes it simple for those few who log in and don't choose to remain logged in by checking the box, it clutters the screen in a distracting way for those who don't edit and simply use Wikipedia for the content. 68.112.220.182 20:00, 9 Jan 2005 (UTC)

-- to —

I don't see a real use in allowing -- to stand for two hyphens. Most people use it incorrectly, meaning a dash (—). Correcting that takes time and energy and clogs up the source of articles with unnecessary HTML entities. It would be great if the wiki software automatically changed -- to —. 68.112.220.182 19:55, 9 Jan 2005 (UTC)

Hell yes. Since -- is intended to be an em dash —, it should be rendered as such — EXCEPT in signatures. User:Anárion/sig 20:01, 9 Jan 2005 (UTC)

I agree, this would be a useful feature to add to the auto-wiki markup. However, it should correctly differentiate between em and en dashes. Please correct me if I'm wrong, but I believe that em dashes do not have spaces on either side (e.g. blah blah blah—blah blah blah), whereas en dashes do have a space (e.g. blah blah blah – blah blah blah). —thames 03:43, 12 Jan 2005 (UTC)

Do people use two hyphens for an en dash? I only use one, as in '1934-1945'. (That is, when I forget to code them.) —Mike 03:58, Jan 12, 2005 (UTC)

Usage differs, but essentially that's correct. It's either word—word, or word – word. However, many now seem to also use word — word. "Correct" use calls for hair spaces surrounding the em dash, but virtually no fonts support this character, so it will result in an ugly box: word — word. User:Anárion/sig 08:20, 12 Jan 2005 (UTC)

Archive proposal

I see some pages have a lot of archives already, I would like to suggest if we could make a page dedicate to archive content. It's like the content table in every article, but it has several content tables(according to the number of archive pages), rather than one. Roscoe x 20:26, 9 Jan 2005 (UTC)

Your proposal is very unclear. What do you have in mind? -- Jmabel | Talk 23:36, Jan 9, 2005 (UTC)

Hm... a page with something like this.

Archive 1 Contents [hide] 1 content1 2 content2 n content...

Archive 2 Contents [hide] 1 content1 2 content2 n content...

Archive 3 Contents [hide] 1 content1 2 content2 n content...

With 'content?' were linked to the specific topic, so we could have fast access to old archive data. I hope that would clarify what I thought. Roscoe x 12:38, 10 Jan 2005 (UTC)

I suspect that any proposal to automate the generation of these will not get a high enough priority to make it to the top of the developers' list, but for pages where you think this would be useful, you can create a template like this and insclude it in the relevant main page and its archives. It would take a little bit of maintenance work, but since archives are usually systematically numbered, you can probably include the next, say, half dozen archive pages as red links, so you wouldn't have to revisit them very often. -- Jmabel | Talk 20:05, Jan 10, 2005 (UTC)

Or you could just make an archive directory as you go along, without needing templates or tables. Maurreen 03:47, 11 Jan 2005 (UTC)

Contributions count

How about a counter at "User contributions" page that shows how many contributions a particular user has made. This can be useful when that user made A LOT of contributions. --SunTzu2 09:08, 11 Jan 2005 (UTC)

Yeh or even just a number next to each contribution instead of the bullet so you can easily count them. There must be some way though already, because I've seen a lot of user pages saying exactly what their 5000th edit was, or exactly how many contributions they've made... AlbinoMonkey ^(Talk) 12:39, 11 Jan 2005 (UTC)

Well for a while, User:Kate had developed a tool that counted them automatically, but that tool is no longer online. Before that, and now again, people just counted manually. BLANKFAZE | (что??) 18:54, 11 Jan 2005 (UTC)

The tool is currently online at [3]. Aenar 02:44, 18 Jan 2005 (UTC)

I think we'd all be better off if people stopped trying to keep score. -- Cyrius|✎ 17:39, 11 Jan 2005 (UTC)

Actually, vote count is very useful when considering WP:RFA. My way is to copy the final page of listings into Excel and let it count the lines. There could be another way, increasing the offset in the URL, but the Ecxel method is quickest, IMO. Smoddy | Talk 23:05, 11 Jan 2005 (UTC)

Plus, some people are just curious about how much they've done for wikipedia. SunTzu2 03:58, 16 Jan 2005 (UTC)

There is that. Smoddy | Talk 16:47, 16 Jan 2005 (UTC)

a great idea. .

FILTERS FOR 'RANDOM PAGE' BY SUBJECT(broad and specific)!!

Anonymous request, but wasn't this already proposed? Peter O. (Talk) 19:08, Jan 11, 2005 (UTC)

pointless. we have categories for browsing. the only 'serious' use for the 'random' button is doing surveys of 'average WP'. If you filter it, it's not random anymore. dab (ᛏ) 19:43, 14 Jan 2005 (UTC)

Auto Translation

What if Wiki pages were automatically translated and placed on other-language Wiki's with a notice such as 'No matching article was found, however the below article was found in the English wiki and automatically translated. Please click here if you would like to correct the translation.'?

This may only work between certain languages, but is it worth pursuing? I tried translating a few random Spanish Wiki pages to English with Babel Fish and the results were better than I expected, certainly better than nothing. --Jperlin 07:36, Jan 14, 2005 (UTC)

Sounds ok to me, as long as the translated article still retains it's content and nothing is mistranslated. In other words, articles translated this way need double checking. SunTzu2 08:26, 14 Jan 2005 (UTC)

Doesn't sound OK to me. There are all sorts of reasons this is not a good idea. Please see Wikipedia:Translation into English and Wikipedia:Pages needing translation into English for some further discussion of this. The short of it is that if someone wants to use Babelfish etc. to read a foreign-language article, they can do that themselves. Auto-translated articles really bring down the tone of the site; auto-translated versions of articles from foreign Wikipedias are, in fact subject to deletion.

Automatic translation is a good tool for translators. It's a perfectly good way to get a first draft of an article that you will promptly edit closely, as it can "batch translate" a lot of the content of an article. However, automatic translations need to be checked closely: besides often atrocious grammar, they also introduce errors such as "translating" names (e.g. "Giuseppe Verdi" can become "Giuseppe Green") or picking the most common meaning of a word that, in fact, is being used differently in the present context. -- Jmabel | Talk 18:50, Jan 14, 2005 (UTC)

forget it. at most we could have bots dumping auto-translated articles to subpages of existing, and interwikied, stubs, and drop a note on the talkpage "if anyone wants to look at this, I dumped an autotranslation /there." But it is much preferable to have humans do this, i.e. find an excellent foreign language article to a stub in their own language, auto-translate it, and see what they can make of it. dab (ᛏ) 19:39, 14 Jan 2005 (UTC)

WikiRecipes

I am sure that recipes have been suggested, but I would like to see recipes with the following information.

• Title
• Origins
• Ingredients : what is in the recipe
• What is not in the recipe (e.g. meat, fish, eggs, milk).
• Method
• Serving suggestions
• Good with...
• Optional images

The categories "What is not in the recipe" could include "yeast free", "dairy free", "wheat free", "no nuts", "citrus fruit free", etc and these terms should be standardized.

This would be an asset to those who suffer food intolerance since you could search the database for those items you cannot eat, and then choose set of recipes.

LoopZilla 10:24, 14 Jan 2005 (UTC)

• Recipes are in the wikibooks - Skysmith 11:23, 14 Jan 2005 (UTC)

• Also, there is a similar website called recipesource.com where you can submit recipes and browse through the collection. --Munchkinguy 20:02, 14 Jan 2005 (UTC)

• You are welcome to add recipes to Wikipedia, but please do make the article encyclopedic, so stick to recipes of cultural value, and add encyclopedic information with the recipe. SweetLittleFluffyThing

Wikicalendar

I have just come up with a fantastic new idea for a "Wiki" website: Wikicalendar! I got the idea from typing years (like 2005) into the Wikipedia, and learning different facts (like that 2005 is the World Year of Physics). It could either be part of Wikipedia, or on its own. The idea is an open source calendar where people can add holidays and onto the calendar. Then, you coul click on the hyperlinked holidays, and it would take you to a page with an article about that holiday. Another cool feature might be that one could view different formats of calendars (Gregorian, Jewish, Muslim, etc.) --Munchkinguy 20:03, 14 Jan 2005 (UTC)

Hmm...nice.....and don't forget the Chinese calendar. However, it does sound like a lot of coding involved, especially all the tables (I hate making these things on Wikipedia). It can be done by regular wikipedians like us though...just need to be really hard working, even just to get things started. One question: how do want this to look like, especially the part where we put several calendars togather? Not all calendars have 365 days a year you know. SunTzu2 03:36, 16 Jan 2005 (UTC)

I also belive that tables are evil. I was thinking that the current day would be highlited on the calendar, and clicking on the link to other calendars would take you to the current day on that other calendar, with the day highlited on that calendar. I was hoping that somebody more technologically inclined than myself would come up with the coding. --Munchkinguy 20:58, 17 Jan 2005 (UTC)

Well...guess I have to learn to make these tables sooner or later...... I got 5 1/2 months holiday ahead of me anyway. I can help with the 2005 Muslim calendar, the Chinese calendar and the Gregorian calendar (lol, who can't?). Before I get things started, a few issues to settle:

1. Coding the thing to show the highlight date automatically. I got no idea how to do this.
2. Crossing the language barrier. How do we present other calendars, which may be weird to just romanize them, especailly names of months, or in some cases (like the Chinese calendar), years as well.
3. Dealing with existing articles. What do we with it? Especailly these: 2005, 2004, 2003...ect. Without being rediculous for reinventing the wheel. SunTzu2 05:05, 18 Jan 2005 (UTC)

I suppose that somebody will have to change the highlited date every day. To colour in the backround of a cell...

hello

... you type this... <table border=black> <td bgcolor=yellow>hello</td> </table>. I think that the calendar should be created under a new article called "Wikicalendar", and then put a link to it on the "Calendar" article and articles for the years (2004, 2005, etc). As for the cross-language stuff, it is probably good to transliterate, except for the years, in which case they should be translated. I hope that made sense. --Munchkinguy 21:20, 18 Jan 2005 (UTC)

lol, manual labour it is then. As a standard, the dates should be in UTC. I don't know about other calendars, but the chinese year might be tricky to translate. Half of it (see Heavenly Stems) has no meaning, the characters used there is almost exclusively for calendars only. No problems with second half though, just names of animals. btw, I get an eerie feeling we're alone in this, lol. SunTzu2 12:04, 19 Jan 2005 (UTC)

Tada~~ here comes Wikicalendar: 2005. Needs a lot of work though, so I listed it for expansion. What kind of holidays should we list there? National holidays? State holidays? Religious holidays? Cultural holidays? School holidays? (just kidding on the last one). Do we put all the holidays in every calendar or do we put them in the calendar that marks the occasion only? SunTzu2 15:37, 19 Jan 2005 (UTC)

Wonderful! My brainchild has been born. Bwahaha! I think that any important holiday from anywhere in the world can be added. That way, each day will have a holiday. Just one thing... (please don't get angry) ...I think it might be better if the weekdays were at the top of the calendar. --Munchkinguy 20:52, 19 Jan 2005 (UTC)

You got to be kidding me......*faints*. Don't forget that's my sweat we're talking about, I've got a huge headache trying to get the tables to the right spot. And you haven't answer me about what to do with the Heavenly Stems. SunTzu2 02:47, 20 Jan 2005 (UTC)

Okay, here's the deal: I deleted my "Edgar Allan Poe Birthday". I don't really know what you should do with the Chineese Calendar beacuse I don't know anything about it. You should probably write the months phonetically. Just go with your intuition. I will try to rotate the calendar so the weekdays are at the top, but I won't make any changes until it looks perfect in preview so don't worry. --Munchkinguy 18:49, 20 Jan 2005 (UTC)

Can I just ask why you haven't done the tables in Wiki markup, because the other usages are deprecated? (See m:Help:Table.) Noisy | Talk 19:11, Jan 20, 2005 (UTC)

Sure you can! It's because I can get most of the work done in HTML with MsFrontPage. You've seen how much coding it involves right? And after I'm done with the Gregorian Calendar, I still have the Islamic calendar and the Chinese calendar. (Nice thing about living in Malaysia is that you get 4 in 1 calendars, the other one is a Hindu calendar, not sure which variant it is though.)SunTzu2 07:06, 21 Jan 2005 (UTC)

50% of my work is done now. I found a HTML=>Wiki converter and cut down it's size, but it's still almost 40kb of codes, hope it's worth all the space it's taking. Meanwhile, I need help adding all the holidays, linking pages to and from Wikicalendar and Wikicalendar: 2005, updating the dates daily, and discuss which other holidays to add. (See dicussion page for more info) I'll finish the rest of the Chinese calendar and Muslim calendar myself, in stages. --SunTzu2 11:04, 25 Jan 2005 (UTC)

Polygon articles

Many of Wikipedia's polygon articles are simply stubs that talk about simply the polygon name and the number of sides it has, the measure of the angle in degrees, and, perhaps, a picture, and don't get updated on a regular enough basis. I propose the following:

All polygons past the octagon should be merged and re-directed to polygon. Add a degree column to the polygon table telling the angular measure of a regular polygon in degrees. (Note: this was brought to my attention when I saw a preparation of a page move of Enneagon.) As for the usage of enneagon/nonagon, it is already mentioned at the bottom of the Greek numerical prefixes article. Any objections to this merge?? Georgia guy 00:20, 15 Jan 2005 (UTC)

Linking users among different language wiki

Why is it that I can't use the same user name across wiki in different languages? For those of us that are multilingual, it would be nice to be able to log in and use one username across the different wikipedia to which we contribute.

Can it be done?

Because people keep asking for it but nobody steps up to the task of writing the code necessary. -- Cyrius|✎ 01:41, 15 Jan 2005 (UTC)

Improving the backup system

I am concerned about how easy is for people to download and save wkipedia database, witch i find difficult now. i think that if we save less.... we can save more, because more people will download the database. At the same time the servers are getting overloaded and from what i saw there is a nead for multiple hard disks inside the database servers.

This is more related to madiawiki, but anyway, we should be able to delete archives older than 3 months maybe, after a vote. I dont know if this is possilbe alredy but thanks to tell me that. This may alrady be implemented since this page does not have very old archives...

seccond proposal: We nead also a system that will show us how offten the backups were made. The most simple is a nomal wiki table where each time someone makes a backup he writes a line in the table (maybe his country, etc...). Of course we cannot check the backup was real, we must trust people... Anyway it should be a good begeining in knowing how many people did the backup.

bittorrent is a good idea if you want more people to download the database. 'because a project as big as wikipedia and wiki* should be backuped in most big countries of the earth at least by one or two people in each country (at least every 6 months); this is somethink around more than 60 backups every 6 months, about 10 every month, this is one backup every 3 days. Of course to save bandwith we can use bittorrent with no problem :)

Finaly the backups are generated every week, but once every month is enaught. Hey, we are speaking here about backups in the case somethink very bad happends with the main servers, and this does not happened every week :) Moa3333 01:47, 15 Jan 2005 (UTC)

There's a difference between backup, which should be happening on a continuous basis for a site like Wikipedia, and the ability of users to make their own copies of the database. If the servers are destroyed in a fire at the co-location place which houses them, then I would expect all but the last few edits to be restored when the servers are replaced. If the whole of southern Florida is wiped out, I still would expect at the very most a day's worth of edits to be lost, irrespective of any user backups made.

The problem with deleting archives older than a certain date is that the terms of the GFDL under which all content on Wikipedia is licensed (more or less) requires attribution to be kept for all authors. We keep this attribution in the article histories.-gadfium 02:05, 15 Jan 2005 (UTC)

I rethinked my proposal, and if somethink bad happends we only nead the current edit. On this page, the current edit, with the free photos will take about 1.5Go for all the wikimedia projects. (comparig with 100Go for the entire database). We shoud think about two points now: First, you pointed out the problem related to the gfdl, well if the current edit is in free download, and some day we will use one to restore wikimedia, how legal would that be? Maybe we can list people without including al the edits... Seccond, we still nead also a system that will show us how offten the backups were made, especialy because it is not hard to do, i already did one for this little project two days ago: here

wikiannotations

How about putting up a couple of rich interesting literary works in full text - making it possible for any old reader to go to a point in the text and attach a note/footnote etc that could :define an unusual work; identify a location or building or historical or mythical character or any other reference; identify a source; suggest an interpretation; make a connection with other point in the text, etc. Over time we would have an extremely interesting thoroughly annotated text. [by anonymous]

You might want to try Wikibooks or Wikisource. RickK 00:58, Jan 16, 2005 (UTC)

wikitranslations

I suggest putting up a couple of nonEnglish texts - novels or shortstories might be good - and inviting the public to translate them into English. Any one person need only translate a line or two. If the idea took off, we could have complete translations that could be continully improved. Of course, this could work just as well from English to other languages but that would be best done in the otherlanguage Wiki web sites.

I'd be careful with that idea. Each person may have different translations for the same text. Even now, there are many interpretions of the works of Confucius, Buddha, the Bible, and many more. I assure you there will be even MORE chaos when people try to translate it. btw, I don't think an encyclopedia is a good place for this. Try Wikibooks or Wikisource instead. SunTzu2 03:54, 16 Jan 2005 (UTC)

We can always use help at Wikipedia:Translation into English. -- Jmabel | Talk 23:16, Jan 16, 2005 (UTC)

That is kind of what Wikisource is about. ~ mlk ^✉♬ 03:06, 21 Jan 2005 (UTC) ~

Random article

Random page is a fun feature, but it seems like a good many of them are stubs and/or very short articles. I would like to see a "random article" link, as well, that would either a) only return pages with more than one section or b) only return pages over a certain length (1000 characters seems reasonable.) Perhaps simply appending "(article)" as a separate link to the right of "Random page" could facilitate this. - Chardish 19:37, 17 Jan 2005 (UTC)

But then it wouldn't be random... Part of the beauty of the random page function is to find stub articles for an editor to expand. Smoddy | ειπετε 23:13, 19 Jan 2005 (UTC)

I wasn't proposing a change in how Random Page works, but simply an additional functionality for the feature. - Chardish 03:50, 20 Jan 2005 (UTC)

Change category link at bottom of articles

Currently, at the bottom of many articles, there is a list of categories that the article belongs in. The first link is one to the general Categories page. That page is just an unformatted list of every category in alphabetical order. A better page for it to link to would be Wikipedia:Browse. What do you think of this proposal? --kenb215 02:01, Jan 18, 2005 (UTC)

Practically anything would be better than linking to Special:Categories, which is practically useless. Wikipedia:Browse would be an improvement. (Perhaps Wikipedia:Category?) -Aranel ("Sarah") 01:07, 19 Jan 2005 (UTC)

I filed an enhancement request a few months ago. —AlanBarrett 08:19, 22 Jan 2005 (UTC)

New project proposal - Wikibuilder

Wikibuilder - a knowledge base covering the design and construction of the built environment, in its entirety, in all languages. See meta:Wikibuilder for more information, add your name to meta:Proposals for new projects#Wikibuilder if you're interested in joining, and be sure to leave your comments on meta:Talk:Wikibuilder —Christiaan - ☎ 20:28, 18 Jan 2005

Redirect Alert

I propose that there should be a "redirect alert".

This could simply be a weekly check of page access counts to identify any redirect-page that counts for nearly 50% (say 35% +) of the total accesses to the article being re-directed to.

This would allow a automated means to identify any article that should be listed for consideration of a title change. This would be especially usefull for obtuse subjects whose titles have become subjected to edit wars; as it would clarify which English name the Internet community most commonly knows the subject as in accord with existing Wikipedia naming policy.

I'm not opposed to this existing, but I'm not sure how accurate it might be. I type the title I want straight into the URL and frequently use shorter redirects when I want to go to articles with long titles. Tuf-Kat 03:47, Jan 19, 2005 (UTC)

That's why it's an 'alert' and not an automated edit; it merely highlights where someone should have a look at the name. In fact, it need not be done even on a weekly basis, a review every few months should be enough to catch any titles that are grossly mis-labeled.

I seem to recall that Mediawiki can already count accesses to individual pages, but that it's explicitly turned off in the English Wikipedia because of the additional server load. —Korath (Talk) 03:59, Jan 19, 2005 (UTC)

So perhaps just some kind of Perl script or the like to check for any titles that have an unusually high re-direct count.?

Generic image names (i.e., no file extension)

I have spent a good amount of time during my Wikipedia career finding PNG alternatives to bad JPEG images (for example, the CIA World Factbook maps). It seems to me that the current scheme of naming images has a major flaw: namely, the inclusion of the file type as part of the file name itself. It would make it far easier to simply upload a PNG with the same name as the previous JPEG, rather than uploading a new file, indicating its copyright status, and changing image names on all the pages that use it.

Wikipedia articles don't indicate any specific file type. It would be beneficial to treat images the same way; instead of Image:St Patrick's saltire.png, for example, it would be nice to have simply Image:St Patrick's saltire. Has this been discussed before? —Bkell 07:07, 19 Jan 2005 (UTC)

Nice idea, though I'm sure it would require a software update. —Mike 07:33, Jan 19, 2005 (UTC)

Yeah, almost certainly. Should I try to bring this up at the MediaWiki site? —Bkell 07:47, 19 Jan 2005 (UTC)

it may lead to some collisions between existing images, but nothing that couldn't be sorted out. The images could then also be converted internally to jpeg or png (e.g., make all gifs into pngs, and make all images jpegs for which that results in a considerably smaller filesize), without interfering with article contents. dab (ᛏ) 15:42, 19 Jan 2005 (UTC)

Article structuring

I tried to find some good advice on structuring articles in wikipedia namespaces. All I found were some pages in the Style and How-to Directory, some pages in the Wikipedia:Manual of Style, and some links on template:FAPath. They mostly deal with the issue from a general stylistic and aesthetic viewpoint, and only touch on the underlying reasons for giving a logical and fairly standardized structure to encyclopedia articles.

I've therefore written up a draft guideline for writing articles in a "pyramid structure". Being mostly based on common sense, it is in part a description of what we already do, but its goal is also to explain why structuring articles in this way is good. It's at User:Zocky/Pyramid structure. Please feel free to improve and comment.

I'm not sure how to proceed. Proper structure should obviously be a FA requirement, and this should probably also be in the how-to series. But since structure is a fundamental editorial issue which heavily affects both quality and NPOV, it could also be a part of the Manual of Style. Zocky 15:54, 19 Jan 2005 (UTC)

I'd suggest also posting a notice at Wikipedia talk:Manual of Style if you haven't already done so. -- Jmabel | Talk 18:55, Jan 19, 2005 (UTC)

Stopping Link Spam / Comment Spam

Link spam or Comment Spam is the practice of someone putting links to their own pages in Wikis, Blogs and other places in order to raise their page's ranking in search engines by creating bogus links.

Google has devised a method to make this undesirable practice ineffective. I recommend that Wikipedia place this feature on the list of things to be implemented.

How it is done is, for external links, automatically insert rel="nofollow" in the generated html. Here is a quote from the example:

Q: How does a link change?
A: Any link that a user can create on your site automatically gets a new "nofollow" attribute. So if a blog spammer previously added a comment like
Visit my <a href="http://www.example.com/">discount pharmaceuticals</a> site.
That comment would be transformed to
Visit my <a href="http://www.example.com/" rel="nofollow">discount pharmaceuticals</a> site.

I think this will be a useful feature to make link spam not worth bothering to do or less significant.
Paul Robinson Rfc1394 18:37, 19 Jan 2005 (UTC)

Yes, it would do that, but it would also mean that we don't contribute to the Google ranks of the good articles we link to, which is just plain unfair. I think that outweighs the advantage of the proposed change. -- Jmabel | Talk 19:10, Jan 19, 2005 (UTC)

I think this is a great idea. There are is a lot of paranoia here (some of it justified, some of it not) by visitors who like to add links. If the links don't count for search engine rankings, there is no point in adding spam links. However, good links that help the article would still be added. As time went by, this community would become less paranoid about external links. As to this comment, "Yes, it would do that, but it would also mean that we don't contribute to the Google ranks of the good articles we link to, which is just plain unfair. I think that outweighs the advantage of the proposed change." Why should any site get a bump for being in Wikipedia? No one should be making money off of the efforts of the Wikipedia community. Some do but we don't have to condone it or make it easier. 141.209.33.148 21:10, 19 Jan 2005 (UTC)

It's not about "making money":

1. If we use "nofollow", Wikipedia itself will not be known to Google a "backward link" from any of these articles.
2. Presuming most links in Wikipedia are valid, and point to useful, scholarly resources, we want to boost their rankings. This is the same means by which Wikipedia itself becomes visible on Google. Our valid links help good, scholarly resources compete in the Google rankings.

Jmabel | Talk 22:10, Jan 19, 2005 (UTC)

How is it a good thing to be less paranoid about external links? Advertisements and other spam, as well as links to sites that don't contain information, should be removed regardless of whether the links are followed by search engines. Fredrik | talk 07:31, 21 Jan 2005 (UTC)

I agree with Jmabel. We're always going to have vandalism. Link spam is just another form of vandalism, and it can always be dealt with by reverting. It's not worth taking away the benefit to the quality resources (most of which are noncommercial) that are listed for good reason. – flamurai^TM 22:27, Jan 19, 2005 (UTC)

In addition, link spamming is not a problem we encounter every day -- it's very infrequent. Peter O. (Talk) 01:07, Jan 20, 2005 (UTC)

• Especially since the spam blacklist was added, and is maintained. I too think adding nofollow would be a bad idea. —Ben Brockert (42) 04:20, Jan 20, 2005 (UTC)

I agree with Jmabel and others in opposition to this proposal. Link spam is infrequent and reverted in a timely manner, whereas in blogs, they usually don't have the same editing standards and Wikipedia's "watchful eyes." --Stevietheman 06:27, 20 Jan 2005 (UTC)

Personally, I just figure that someone trying to insert link spam here is probably not sufficiently aware of how Wikipedia works to be deterred by this. (If they were that aware of how Wikipedia works, they would know better than to try in the first place.) -Aranel ("Sarah") 17:30, 20 Jan 2005 (UTC)

I think the nofollow tags are a great idea, and should be used whenever an anonymous user adds an external link. Dealing with each vandalism manually when we have technical tools to help lessen the load is a waste of our valuable time. - Omegatron 00:47, Jan 21, 2005 (UTC)

It does sound good at first, but adding nofollow to all links is a bit like dropping out of the Web altogether. By its nature, Wikipedia's community self-moderates spam links, and this is meet and good. Gives us something to do on a slow day. —Michael Z. 08:14, 2005 Jan 21 (UTC)

• It would only be applied to external links, so it wouldn't make any difference as far as wikipedia articles coming up in google searches.
• Although wikipedia is a good source for highly relevant, quality external links, I'm sure it doesn't make a huge difference in the overall popularity of such links.
• It could be applied only to anonymous editors, making the effect negligible. - Omegatron 16:30, Jan 21, 2005 (UTC)

This has been turned on! I am very unhappy that it has, who do we need to ask to turn it back off? —Ben Brockert (42) 02:30, Jan 22, 2005 (UTC)

Is it time to vote? - Omegatron 02:44, Jan 22, 2005 (UTC)

I hope so, though I would have thought that the time to vote was before a change in longstanding policy. -- Jmabel | Talk 03:43, Jan 22, 2005 (UTC)

As a public and appreciative wikipedia user, I heard about the change and stopped by to see how the community had handled it so quickly, given the democratic nature of wikipedia. Wow... tis is amazing that it was turned on without discussion? And with all of the negative opinions expressed? What's with that?

Since this affects every Wikipedia, the proper place for discussion is meta. I put up a page at m:Meta:nofollow, I hope you'll all go there to enhance it and express your opinion. —Ben Brockert (42) 05:35, Jan 22, 2005 (UTC)

But it isn't that hard to spam Wikipedia if you know what you are doing. Always create a login name, only add one or two links per user name, use different IP addresses by logging in at your local public library and having subs to services like Earthlink and AOL that give you a different IP address every time, only add one or two links a day, etc. There are some successful spam sites in Wikipedia. The serious ones learn from their mistakes and figure out how to edit here and get away with it. Further, I am aware of several sites that are on the spam filter because a competitor spammed the hell out of them to get them banned from Wikipedia. They know about the spam filter and are using this method to get their competition out. The actual site owners are innocent but even legit adds from these sites are filtered out. Having a no follow tag on links will end most spamming. Most good sites don't need the help of the Wikis to rank well... 172.135.76.241 01:38, 23 Jan 2005 (UTC)

Verbal explanations and pronunciations

At the recently reopened Museum of Modern Art the audio guide has ad-lib discussions of notable pieces by current "experts" in the field. They aren't scripted - it is just someone who knows something about a subject talking about it and putting it in context with history, other pieces, artists, modernity, etc. Further, M-w.com has verbal pronunciations of many words, indicated to be available by a small speaker symbol (example).

I am proposing a somewhat mixed integration of these two ideas, and the Ogg Vorbis audio codec would probably suffice, although I admit to having not used it myself. We would openly invite those who feel comfortable with speaking, with a certain subject, and also with licensing a snippet of their voice under the GFDL, to simply speak about a subject and upload the file. This could be concerning a work of art being displayed inside an article, an entire article, a clarification of some point, or anything the speaker wishes. Because some people would find the small speaker symbols annoying, their could be an option in the user preferences to turn them off, or they could hide them with CSS as is often done here. A limitation would be that they could not introduce facts not already indicated in writing somewhere in the encyclopedia.

This is further expanded to Wiktionary, where it would be highly useful to have open source pronunciations of words in many regional accents of english speakers, and appropriately for other languages as well. --Alterego 00:26, Jan 20, 2005 (UTC)

I think the concept of having sound is a good one for pronunciations as you mentioned and pages on music. – flamurai^TM 04:29, Jan 20, 2005 (UTC)

MediaWiki:Sharedupload

Hi all, what do you think about copying the german style of this important message, compare de:MediaWiki:Sharedupload? --217.255.66.146 04:24, 20 Jan 2005 (UTC)

I like the German style better. It doesn't blend in as much as plain text does, but it is not gaudy either. —Mike 07:42, Jan 20, 2005 (UTC)

External table namespace

Could tables be created as separate "articles" in a Table: namespace? Then you could link to them like images, with right and left floating tags, for instance, and edit them separately from the article. Also a single table could be used the same way as a template directly.

This would be beneficial because:

1. Large tables are annoying to work with while editing text. They clutter up the wiki markup. You have to scroll around them and it gets confusing and easy to get lost. They are definitely confusing for some newcomers. With taxoboxes they are often the first thing you see when editing an article! Bad for newcomers.
2. Editing tables is difficult in regular text editing mode. Especially large ones. The table editing page could have non-wrapping text, for instance. It could have a list of the table elements displayed at the top for reference. Or even clickable like the ÁáÉé at the bottom of this edit window.
3. Even better, it would allow an alternative paradigm of editing tables that is more in the spirit of easy-to-edit wikis. The current table markup is harder to use than it needs to be. I am imagining a table of form cells that can each be edited individually, and the total number of cells chosen in another form box, a clickable box for spanning rows, etc. (Maybe this is related to some of the ideas that people had for m:SVG image support, of providing an edit window for each image that would allow people to change the XML directly and edit annotations and whatever.) I just like the idea of each type of unit (image, table, template) having its own efficient editor functions and abilities.

Examples:

• "Basic mode" example here.
• "Advanced mode" example here. It gets confusing fast, but with colors and more creative borders I think it could be made much more manageable than markup right in the article source. The checkboxes are for spanning cells, and could also be updated when you click the update button, refreshing the page with the affected cells merged into one. - Omegatron 19:13, Jan 21, 2005 (UTC)

More automated vandalism

I don't mean the vandalism is automated; I mean dealing with the vandalism is automated. Currently it takes several page navigations and long waiting for each one (especially lately grumble grumble) to revert something. I think there should be a single revert button (maybe only for administrators or power users or whatever they're called?) that automatically reverts to the version before the last editor edited (to handle several tests in a row by some anon), and adds a vandalism or test template to their user talk. One click; no wait. - Omegatron 03:02, Jan 21, 2005 (UTC)

See Wikipedia:How to revert a page to an earlier version#Admin-only "rollback" link. —Korath (Talk) 03:24, Jan 21, 2005 (UTC)

Hmm.. I want that ability, but I don't want any responsibility.  :-) And I don't want to delete things by accident... - Omegatron 05:17, Jan 21, 2005 (UTC)

It's very difficult to outright delete things by accident. -- Cyrius|✎ 06:16, 21 Jan 2005 (UTC)

A "TOP" link

I think it would be nice for some long pages to have a shortcut to the top of the page.
As all the pages have an id="top", why not use it?
Could there be an automated process do do that?
Cy21 16:03, 21 Jan 2005 (UTC)

Conceptually, I like this idea. But alas, the devil is in the details. Perhaps there would need to be a user setting that determined where "top" links are placed, like after every n paragraphs or something like that. --Stevietheman 07:39, 22 Jan 2005 (UTC)

Most browsers have built-in commands to go to the top of the page (e.g. the "Home" key). Given that, why would you want a link in the article? —AlanBarrett 08:26, 22 Jan 2005 (UTC)

Maybe it could go next to the "About Wikipedia" and "Disclaimers" links at the bottom.

But I've always wondered why some people insist that web designers need to spend time duplicating the most basic operating system functions. If someone managed to navigate to the web page, and even scroll to the bottom!, don't you think they can scroll back to the top without even having to give it any thought? Why only on "long" pages; hitting the page up key or dragging the thumb to the top of the scrollbar doesn't get harder if you have farther to go!

For me the top link is useless, so I would just consider it page clutter. —Michael Z. 2005-01-22 17:16 Z

It is not very useful. Anyone can just press the "home" key to go to the top. Also, you can add edit links to the bottom (Help:User_style#Duplicate_edit_links_at_the_bottom_of_the_content_area). (I really wish wikibooks had "next" "previous" and "up" links, though.) - Omegatron 18:00, Jan 22, 2005 (UTC)

To answer some concerns here... I think Top links are useful for the person keeping their hand on their mouse while casually reading articles. To have someone take their hand off their mouse to hit the Home key is a bit awkward. Of course, one could program a button on their mouse to do a "Home" but most people wouldn't bother. Top links are a regular appearance on web sites and they need not cause any clutter if designed correctly. Further, going back to the top fast is sometimes a necessity if 1) one wants to re-read the basic summary in the article a second time, or 2) one wants to select a task from the left-hand Wikipedia menu. --Stevietheman 19:14, 22 Jan 2005 (UTC)

You can also make the sidebar float in css, though I seem to be the only person who likes this feature, and my rendition is kludgy. See here for the css: meta:Help_talk:User_style#Floating_sidebar. If someone knows css and can help me finish it, I would appreciate it. We could also consider adding the bottom links css and floating sidebar css to the universal skin instead of a top link. Now that I think of it, an unobtrusive up arrow under the edit link for each section does sound kind of nice, actually. - Omegatron 21:39, Jan 22, 2005 (UTC)

Re: the "unobtrusive arrow", that's fairly close to what I was just thinking today. Put it next to the edit link. And for those who would argue a little up arrow to be clutter, then let's make it a user setting that's defaulted to Off. --Stevietheman 00:14, 23 Jan 2005 (UTC)

Proposal: Autocomplain if user doesn't enter an edit summary

It's a good deal easier to track edits if the editor uses the "Edit summary" box. Wikipedia doesn't force editors to enter something in the box, which is a good idea. It could, however, do what many email clients do if you fail to enter a subject: ask the user if they really want to do that. This could gently encourage people to enter an edit summary, without the risk of forcing people to use it (in that case, people might just type jibberish). Thoughts? jdb ❋ 01:45, 22 Jan 2005 (UTC)

Hmmm... I'm bad about this myself, and it would be good for me to have a box pop up if I didn't type anything, making it more efficient to just type something. On the other hand, that is just me. Others would be annoyed. But maybe the non-annoyance of the many outweighs the non-annoyance of the lazy? Complaining is fun. - Omegatron 02:01, Jan 22, 2005 (UTC)

Support. Compare submitting a mail with no subject in Mozilla Mail. Although I hate it there, on Wikipedia an empty summary field is more often a mistake. Deco 02:14, 22 Jan 2005 (UTC)

Support as well - it could be configured that by simply clicking enter at the prompt, which also had a field for an edit summary and "cancel" - eg no mouse clicks - it went ahead and saved. This is a minor inconvenience, and I think we would rather see the person submitting the edit inconvenienced than the multitude of people throughout history from now until the end of time who will have to check their edit. To be honest, I think edit summaries should be mandatory in the main namespace.

That's one way: you could do that with a few lines of javascript. "Press OK if you really want to submit this edit without a summary, press Cancel to go back and add one." Or you could do what some web forms do -- bring the user to a new page with a giant honking exclamation point and a "are you sure you want to do that?" admonition. Given that the latter would involve a roundtrip to the (often painfully slow) wikipedia servers and the former wouldn't, I'd favor the former. jdb ❋ 02:41, 22 Jan 2005 (UTC)

A popup javascript dialog is the way to go, with a form in it so you don't have to push cancel; you can fill in the summary right then and there and push enter. It will certainly annoy me, but that is for the better good.  :-) - Omegatron 02:45, Jan 22, 2005 (UTC)

No offence, but I think this is a bad idea. Just my 2¢. BLANKFAZE | (что??) 03:23, 22 Jan 2005 (UTC)

Oppose. Neutrality^talk 03:24, Jan 22, 2005 (UTC)

Oppose. Nagware is bad, and tracking vandalism on RC is much easier when most vandalism is by IPs who leave no edit summary. —Ben Brockert (42) 03:35, Jan 22, 2005 (UTC)

The proposal is not to force an edit summary, but suggest one. I doubt that vandals will stop leaving empty edit summaries even if WP reminds them to do so. The proposal may, however, make finding genuine vandals easier, by filtering out regular users who are in the habit of not entering summaries. jdb ❋ 19:45, 23 Jan 2005 (UTC)

I also oppose. Peter O. (Talk) 03:49, Jan 22, 2005 (UTC)

My 2ψ oppose: not everyone browses with javascript on, or allow it to pop up windows, and having to go through another page reload is sometimes tortuous. —Korath (Talk) 05:51, Jan 22, 2005 (UTC)

If someone doesn't have javascript on, this wouldn't have any effect -- it easily could be designed to fail gracefully. Also, there's no need for a page reload (see above) -- it could be implemented entirely in client-side javascript. jdb ❋ 19:08, 22 Jan 2005 (UTC)

Neutral. See my comments above. It would be somewhat annoying, but would make the entire experience less annoying in general. It would not be a pop-up window; just a dialog box. It would only take one click more than it currently does. It would save bandwidth as edit summaries would be more likely to be filled in and people wouldn't need to check edits by loading the page. A less forceful method would be preferred, though. - Omegatron 06:14, Jan 22, 2005 (UTC)

Oppose. This idea would slow down the "data entry flow" and I'm not sure the benefit is worth it. Perhaps a better approach would be to track what users do and automatically append a friendly informational message to their talk page if they submit so many edits w/o a summary. --Stevietheman 07:52, 22 Jan 2005 (UTC)

I do like that idea, but it only works for logged-on users. Some combination of the nagging dialog for not-logged-in users and a nagging message on the user's talk page would be good. jdb ❋ 19:08, 22 Jan 2005 (UTC)

oppose. It will not work. I write summaries when I feel tension, or potential of disagreement. I often don't when I think my edits are totally uncontroversial. I would just switch off Javascript, or else I'd just type something like 'asdf': nonsensical edit summaries are no more helpful than no edit summaries. dab (ᛏ) 09:14, 23 Jan 2005 (UTC)

Not to tell you how to edit, but an edit summary is more than a defense or justification. It's also informational, so people can look at the history or Recent Changes and see what changed, briefly, without interpreting a diff. An RC patroller will often interpret a missing edit summary as a potential attack ("what do they have to hide?"). Deco 10:23, 23 Jan 2005 (UTC)

I agree - edits summeries makes it possible to see what is in an edit without checking the diff. As Wikipedia:Edit_summary says, Always fill the summary field. Thue | talk 22:31, 23 Jan 2005 (UTC)

support. As a Wiki-newbie, I have repeatedly found myself hitting "submit" and then realizing that I did not enter a summary -- even though I wanted to. The solution may be as intrusive as nagware, or as simple as placing the summary box above the wiki editing area. Allowing users to go back an add a summary to an existing change may also solve this one specific issue. --Zenkat 06:00, 24 Jan 2005 (UTC)

As a side note, has anyone considered performing controlled experiments on Wikipedia using server stats as the experimental evidence? It could be interesting to try a two-week nagware experiment that tracked the percentage of edits that included a comment, as well as other statitics about edit frequency, length, secondary alterations, etc. --Zenkat 06:00, 24 Jan 2005 (UTC)

Conditionals in templates

It would be great if there were a simple way to do conditionals in templates. It would be great even to just check if a variable existed. For example,

|-
|| Release date:
|| {{{release_date}}}
{{{{if awards}}}}
|-
|| Awards:
|| {{{awards}}}
{{{{endif}}}}
|-
...

I don't know what the best syntax would be, but this would really be helpful.

– flamurai^TM 21:31, Jan 22, 2005 (UTC)

Direct link from "watch list" to article's history

When reviewing my watch list, I almost always want to go to the history of the article, to see who has been active and when, so I can compare to a version I have seen before. Currently this take two clicks and wait times (and system load), first to the article, then through to the history. A direct link on the "watch list" page would help. --Woodstone 00:16, 2005 Jan 23 (UTC)

You must have a different watch list that I do; mine has a link to the hist and diff of everything on the watchlist. What skin are you using? —Ben Brockert (42) 04:33, Jan 23, 2005 (UTC)

You are absolutely right, I must be blind. It's right there, but I never saw it. Thanks for pointing it out and request withdrawn. --Woodstone 12:29, 2005 Jan 23 (UTC)

Rating buttons

Although I'm sure this idea has come up before, I was thinking how useful it might be if there were easily available "thumbs-up" and "thumbs-down" buttons on each article page. Then anonymous readers, who are often too scared or too busy to make any real edits, could at least leave their valuable opinion on the quality of an article. These statistics would not be displayed on the article, but would be available in some manner for improving the encyclopedia. For example, we might have a list of worst-rated pages, for cleanup or deletion, and a list of best-rated pages, for consideration as featured articles. It would also be a great way for editors to get more feedback from readers and to feel more appreciated. Comments for and against appreciated. Deco 05:27, 23 Jan 2005 (UTC)

I like the idea, as long as the information is only used for internal purposes. jdb ❋ 19:40, 23 Jan 2005 (UTC)

I believe something like this was already proposed. There were many oppositions, one of which was the worry that some articles would receive higher ratings than other, equally satisfactory articles. [Was about to post here, but had a conflict.] Peter O. (Talk) 19:44, Jan 23, 2005 (UTC)

This is certainly true, I would expect a considerable amount of error and anomalies in the data, but I think it would still be useful information, and I don't intend it to be a way of proving which article is better. Deco 20:26, 23 Jan 2005 (UTC)

Protecting Wikipedia:About

I have re-opened the debate on whether Wikipedia:About should be protected. Visit the Talk page to discuss. --Slowking Man 06:45, Jan 23, 2005 (UTC)

"Hide logged-in users" in Special:Newpages

I would very much like to see a "hide logged-in users" feature in Special:Newpages, just like in Special:Recentchanges. Most of the vanity/advertisement/etc pages are created by anonymous users. That way it would be easier to spot them quickly. What do you think?User:Pt/sig 14:46, 23 Jan 2005 (UTC)

Sounds useful to me. --Stevietheman 21:59, 23 Jan 2005 (UTC)

Support. Deco 22:42, 23 Jan 2005 (UTC)

Ditto. jdb ❋ 00:55, 24 Jan 2005 (UTC)

This would be one of the most useful features ever. Neutrality^talk 06:31, Jan 24, 2005 (UTC)

Of course, as soon as any vandal got their head around it, they'd simply log in to reduce the likelihood of their vandalism bening noted. Filiocht 08:55, Jan 24, 2005 (UTC)

Right. And of course they don't do that now for regular edits to pages, which are already hideable this way. Anyone who cares about insidious vandals will simply leave these options off; people who prefer dealing only with egregious vandalism or newbie tests can turn them on. JRM 10:37, 2005 Jan 24 (UTC)

I have wanted this too for some time. I made a patch at [4]. If somebody else reads it/tests it and say it works then I think it will have a better chance of being applied. Thue | talk 20:09, 24 Jan 2005 (UTC)

It would also be nice to be able to mark pages as checked on special:newpages, to help yourself and other editors, like a software-assisted wikipedia:new pages patrol. I made a patch for that too: [5]. Note that you can vote for bugs in the bug tracking system :). Thue | talk 16:00, 25 Jan 2005 (UTC)

The Names of People

Articles concerning people seem to be titled as "A.A.Milne". It would be better to title them "Milne,A.A" or, "Milne, Alan Alexander", or preferably "Milne, Alan Alexander (1882-1956)".

If this is desirable I would be willing to assist in renaming the current articles.

Why do you consider putting an article at a title no one would ever use to be better? -- Cyrius|✎ 14:54, 24 Jan 2005 (UTC)

Why would this be better? It isn't a library card catalog where you have to browse through by last name; it's the internet. I think the names should stay in normal order. - Omegatron 14:57, Jan 24, 2005 (UTC)

The library is orgainised by last name, normal encyclopedias also, the Who's Who is by last names as well. I would assume that there is a reason for this! Why should wikipedia be different?

Because the only way to find an entry in any of those is to browse through in alphabetical order. While this is a factor for Special:Allpages, I expect that the number of people finding an article through that is vanishingly small compared to those who search or follow a link (which would either have to be piped all the time, or have a redirect for every single entry). —Korath (Talk) 03:13, Jan 25, 2005 (UTC)

Let me break this down.

1) Is it desirable to have a policy for the naming of pages concerning people ? 2) If it is, what should this policy be ? 3) If it is, is it desirable to change the existing pages ?

No, no, no, no, no. Do you know how many thousands of pages and links there are in Wikipedia? We CAN'T change this now, even if we want to, and you still haven't presented a case for why it's needed when we have links. RickK 05:37, Jan 25, 2005 (UTC)

There is not an official policy, but the de facto policy emerging from many past discussions like this one is that names should be in the normal order (first first, last last). Deco 05:44, 25 Jan 2005 (UTC)

Why not just add redirects for all the names of people in the form Milne, A. A.? They will appear in the desired place in the list, and they will get you to the article. Nothing that exists now has to change; this just would just add the desired functionality.

Well, get to work! —Michael Z. 2005-01-22 16:54 Z

The only reason other sources use names backwards is because you have to browse through them in alphabetical order to find things. Thanks to hyperlinks, much of that sort of tedious browsing has been eliminated. Likewise with backwards names. We just link to Thomas Edison like we would if we were just speaking about him. It makes much more sense, don't you think? And for things like categories of people, they are registered in backwards, so they show up in alphabetical order. For example, Category:Deceased musicians. I see no reason for this backwards naming at all. It's a step into the past. But if you want to add a bunch of redirects that will never be used, go right ahead. We'd rather see you use your editing time for something constructive, though.  :-) - Omegatron 06:26, Jan 25, 2005 (UTC)

add a new option

hello, your website is very interesting but I want to make a possibility to remove the pictures when I don't want them.

Turn off "auto load images" in your web browser. jdb ❋ 17:31, 24 Jan 2005 (UTC)

Edit Display Mode for links

A user 'preference' editing option that tags wikilinks with status info.

An example of this tag might look like stones(disamb) or perhaps Earthworks(stub).

This markup would be invisible to regular users, but anyone who likes to ramble around doing small edit work could set the preference to display the markup. This would open up the ability to work on a number of links on a single page by having a heads-up about the current status of all the links on that page. (Really the idea is just a further extension of the dead link color code that already exists.)

I know this would be a significant coding issue (of which I know nothing), but a general solution might be to append non-visible characters directly to the article title (or header?). This would negate any new sorting requirement on the wiki system by having the title altered each time a 'stub' or 'disambig' or 'cleanup' or similar template is added/removed from the article. Then it is just (just?) a matter for the end user's system to recognize the coding and display it.--Bookandcoffee 23:47, 24 Jan 2005 (UTC)

Or put it in the title field of the link! That is what I thought you meant when you said it. So when you hover over a link it shows you in the tooltip. - Omegatron 03:07, Jan 25, 2005 (UTC)

Hmm. Then you'd have to hover over each link to see what its status was. That might defeat the purpose? I was hoping to see it appended within an article itself, not the artcle holding the link. ie. in the above example it would be the title of the 'Stone' article that holds the template information. That way any link to 'stone' would be able to exploit the info, and any update would be universal.--Bookandcoffee 09:04, 25 Jan 2005 (UTC)

Good point. That would be mystery meat navigation. Perhaps color coding? That could get too elaborate, though. If we broke it down into just 4 colors or so, that would be viable. Nonexistent articles, regular articles, stubs, and disambigs? Putting words next to each link is kinda ugly. But maybe. - Omegatron 15:39, Jan 25, 2005 (UTC)

Ahh... See here: Wikipedia:Village_pump_(technical)#Stub_threshold. - Omegatron 15:47, Jan 25, 2005 (UTC)

For modern browsers (i.e. not IE): :link:after {content: " (" attr(title) ")"} User:Anárion/sig 15:48, 25 Jan 2005 (UTC)

"Stable", protected, democratically elected version of each article?

How about having two versions of every article; One, which is protected, democratically deemed accurate and NPOV enough (for the time being)etcetera and another, which is the regular kind, edited by anyone,anytime. The version under construction would then be voted to be better than the "Stable" version (or not), and take it's place (or not)? During voting there'd be a third, temporarily protected version so that the anyone,anytime editing would not suffer ... --Mikko

OH, God, no. This is the antithesis of the Wiki way. RickK 07:26, Jan 25, 2005 (UTC)

Lots of people have made lots of variants of this proposal. The devil is in the details, and the details are hard. So far, I haven't seen a detailed proposal of this sort that I'd support. jdb ❋ 07:28, 25 Jan 2005 (UTC)

Can you point me to some discussion of this or give a short summary of the devilish details? I realize that this Wiki way of evolving texts is the best atleast in a world with people committed to the truth, but I'd like to be convinced it's the best even in a world with vandals and advertisers and whatnot... --Mikko

It's been discussed in a lot of places. I think the most discussion is at Wikipedia:Approval mechanism; another place you might look is Wikipedia talk:Forum for Encyclopedic Standards (the relevant discussion is probably mostly in the archives), but someone else might have additional suggestions. -- Jmabel | Talk 09:02, Jan 25, 2005 (UTC)

Actually, this sounds just like Wikipedia 1.0. apart maybe from the 'democratically elected' bit. Filiocht 11:49, Jan 25, 2005 (UTC)

Hong Kong notice board

Hi all. I am thinking of starting a notice board on Hong Kong related materials. Any bright ideas or people wanting to join? Pls drop me a line on my talk page. Thanks. --JuntungWu 12:16, 25 Jan 2005 (UTC)

Power Users

This may have been proposed in the past but I believe it's a useful idea. I propose that a new permissions level of "Power User" be created. These users would have normal permissions except that they would have access to the rollback link. They would not have the admin powers of blocking users or deleting pages. A user would become a power user after hitting a threshold of 500 edits (or 750, 1000, the exact number would need to be agreed upon). If the Power User violates the 3RR or abuses their rollback power, they would lose their Power User status.

I believe establishing "Power Users" would allow users to revert vandalism more quickly and easily. It would also allow users to prove themselves ready for adminship by providing them with a relatively harmless admin power. Carrp 16:06, 25 Jan 2005 (UTC)

Sounds like a good idea, but I wouldn't base it on number of edits, as a number like that has little to do with the quality of edits. I would favor an application/nomination process, however. --Stevietheman 18:02, 25 Jan 2005 (UTC)

I completely agree that a set number of edits isn't an ideal system. The edit threshold was an attempt to come up with an automated system that wouldn't put additional responsibilities on admins or bureaucrats. Since adminship is supposed to be "no big deal", my opinion is that becoming a Power User should be almost automatic. The reason I don't think all users should have the rollback function is because it's too likely to be abused by new users. Perhaps the threshold should be based on time (1 or 2 months?) instead of edits. The specifics would certainly need to be discussed. Right now I'm just trying to get support for the idea of Power Users. Carrp 18:37, 25 Jan 2005 (UTC)