Wiktionary:Grease pit/2007/January

Is there an easy utility around which will give the Unicode value of a character copied/pasted from elsewhere? (I am using XP Sp2 and have MSOffice.) The particular problem is, although it would have wider application, in checking Greek letters (eg is something a u/c Greek Tau or a western T). A simple utility is what I am looking for - not a complete application). —Saltmarsh 11:48, 2 January 2007 (UTC)

• Have you seen the Windows Character Map (All Programs => Accessories => System Tools) The XP version is very good - you can select all manner of character sets. SemperBlotto 11:53, 2 January 2007 (UTC)

  I think that the Character Map is one way - see below.Saltmarsh 12:20, 2 January 2007 (UTC)

• Google it. For example, if I search for 𐀡 one of the first hits is LINEAR B SYLLABLE B011 PO ... and you can restrict your search to unicode.org. The other thing is that if it is an individual letter (not some weird symbol or something like RIGHT-TO-LEFT OVERRIDE) you can look it up here? e.g. Τ will tell you that it is Tau (although it won't give you the code.) Robert Ullmann 12:06, 2 January 2007 (UTC)

• Google seems to strip off accents - I might want to tell the difference between óό (the first being a Western O with grave acute accent - the second being a stressed omicron. Saltmarsh 12:20, 2 January 2007 (UTC)

I know that because I produced them - but can I ascertain exactly without knowing which Unicode page(?) it comes from. Saltmarsh 12:23, 2 January 2007 (UTC)

• I must learn to think twice and act once! Unicode search box seems to do what I was after. thank you! —Saltmarsh 12:31, 2 January 2007 (UTC)

By the way, Google tends not to strip off accents if you use a query like +"ó" and +"ό". – Minh Nguyễn (talk, contribs) 08:15, 5 January 2007 (UTC)

I've designed an experimental fr-noun template template (see documentation) to replace the myriad French noun templates (fr-noun-reg, fr-noun-reg-x, fr-noun-reg-aux, fr-noun-irreg, and fr-noun-mf). It allows for regular and irregular gender and plural inflections, supports both singular and plural words, and categorizes to French nouns, French nouns lacking gender, and Category:French plurals as needed. I think I ironed out all the bugs and exceptions now.

—{admin} Pathoschild 00:48, 4 January 2007 (UTC)

nice. should use {{m}} and {{f}} to display the genders, so that user preferences work Robert Ullmann 08:43, 4 January 2007 (UTC)

Looks nice. To stay in line with other noun templates, it should take pl=plural form for the plural; f=feminine or fem=feminine for the feminine; and sg=singular. I don't read coding well enough to know whether these are accepted currently by the template, but they are taken by other noun templates, and it would be nice to kow that the "usual" parameter coding is consistent so that users don't have to remember which templates do /don't use them. --EncycloPetey 13:52, 4 January 2007 (UTC)

I've tweaked the template as suggested; thanks. —{admin} Pathoschild 08:52, 5 January 2007 (UTC)

One more point occurs to me. The plural form should probably use {{p}} and display it in-line following the gender. --EncycloPetey 02:03, 6 January 2007 (UTC)

I don't agree with the reasoning for the abbreviation of 'plural' to 'pl' in the output text ("An abbreviation (rather than full spelling) is required here for purposes of differentiating grammatical comment from the base text" —template:p). The grammatical notation is differentiated enough by formatting and syntax. Particularly with non-English definitions (where English users may not be able to guess the meaning as easily), there's no reason to use abbreviations except where the full text would be inconveniently long.

I only found two relevant discussions about {{p}} in the incoming links. A Beer parlour discussion (Abbreviating "singular" and "plural"', Beer parlour, November 2005) apparently reached no consensus on using abbreviations rather than the full word. A more recent semi-public discussion ("#25. Gender templates", Connel MacKenzie's normalization discussion page, May 2006) shows that at least one additional user that didn't participate in the Beer parlour discussion favoured using the full text. I'd prefer to discuss that before adding it to the template, since I also disagree with abbreviating there. —{admin} Pathoschild 03:03, 6 January 2007 (UTC)

I also didn't participate in either discussion, but agree that "there's no reason to use abbreviations except where the full text would be inconveniently long". --Enginear 18:11, 7 January 2007 (UTC)

Such as a translation table? Even if the {{p}} template isn't used, then the word plural should still appear in-line in the output text of a plural form following the gender. --EncycloPetey 04:04, 8 January 2007 (UTC)

You mean as follows? That looks fine to me.

• chien m (plural chiens, feminine chienne)
• chiens m plural (singular chien, feminine chiennes)

—{admin} Pathoschild 04:23, 8 January 2007 (UTC)

Yes, though for nouns that have both a masculine and feminine form, I recommend:

• chien m (m plural chiens, feminine chienne, f plural chiennes)
• chiens m plural (m singular chien, feminine plural chiennes, f singular chienne)

This is the way that Spanish adjectives are handled with {{es-adj}} --EncycloPetey 04:44, 8 January 2007 (UTC)

To improve legibility, how about this? The text is shorter and abbreviations are expanded.

• chien m (plural chiens, feminine chienne and plural chiennes)
• chiens m plural (singular chien, feminine chiennes and singular chienne)

—{admin} Pathoschild 20:10, 8 January 2007 (UTC)

OK. But I recommend using a semi-colon instead of a comma if you're going to include double forms like this. Doing so will make it clearer that the second "plural" belongs only with the second gender. --EncycloPetey 00:39, 9 January 2007 (UTC)

Okay. —{admin} Pathoschild 17:58, 9 January 2007 (UTC)

I'm not sure how many notice, but Wiktionary:Grease pit archive/2006/October was recently edited by a new Wiktionarian. This perfectly matches the archive's intent - that previous conversations are maintained as a single coherent thread without being revived completely anew. But we don't seem to have a mechanism in place to detect recent edits to subpages (i.e. that need attention.)

I also noticed the curiosity, that Werdnabot seems to be archiving entries to the "correct" place - that is, the sixty days prior's archive (October 2006, in this case.) So indeed, this seems to be working much better than we had thought it would.

Anyone have brilliant ideas on how to keep track of subpage edits? --Connel MacKenzie 03:11, 7 January 2007 (UTC)

Put them on people's watchlists? --EncycloPetey 04:45, 7 January 2007 (UTC)

Sorry, my watchlist is now only for Pathoschild's experiment of the '--error: link target missing--' deletion thing. --Connel MacKenzie 05:58, 7 January 2007 (UTC)

...but as Pathoschild pointed out, you don't need to completely clear your watchlist, since new (ie previously deleted) entries are marked with N. However, I accept that they might still be hard to find if you were watching 70k words!. Also, it makes sense to prioritise watching policy archives, since they could be altered to skew (or totally delete/move/hide) past discussions to suit a POV, and this makes them unusually important. --Enginear 18:21, 7 January 2007 (UTC)

My 'experiment' involves pages that are never or virtually never edited; watching a thousand such pages is less effort than watching a single discussion page. I'm watching nearly seven thousand such pages on Wikipedia, yet it's the few discussion pages I watch that require the most effort to keep up with. Nobody forces you to watchlist those pages; don't snipe something we all agreed to in irrelevant discussions. —{admin} Pathoschild 04:58, 8 January 2007 (UTC)

It wasn't a snipe, it was a genuine concern. I'm trying to get into the habit of simply deleting anything I see when I use my watchlist (which is, as I understand it, precisely what you recommended.) I'm also considering a 15 minute cron job that checks the watchlist, and e-mails me warnings when it isn't blank. --Connel MacKenzie 09:30, 8 January 2007 (UTC)

I misunderstood, then. :) —{admin} Pathoschild 20:07, 8 January 2007 (UTC)

One way to add an archived page to everyone's watchlist is to move the page to the archive location and then move it back. This wouldn't affect Connel since he doesn't watch the page in question. DAVilla 21:15, 8 January 2007 (UTC)

I'm sorry, what does that mean? How could you move WT:GP to one of its existing archive pages, and back, without messing things up? --Connel MacKenzie 18:13, 10 January 2007 (UTC)

• On a related note, Brion reminded me of the "Related changes" button in the left column (just below "What links here".) That functions in a similar manner to Special:Recentchanges, but instead limits the results to pages that are wikified on the referent page. So, it is possible to come up with a page (e.g. WT:RFDA) that wikilinks all the redlinks we want to stay red, then have each interested sysops check that page's "Related changes" daily.

• This would mean that my removal of 70k+ items from my watchlist was not needed. Or rather, that using a watchlist for this, is not necessarily the best way to approach this. --Connel MacKenzie 18:13, 10 January 2007 (UTC)

I'd like to move Special:Prefixindex/Wiktionary:Frequency_lists to Appendix:Frequency lists Index:Frequency lists as it seems to have been lost in the shuffle. Any objections? --Connel MacKenzie 09:20, 8 January 2007 (UTC)

Wouldn't that be better placed as Index:Frequency lists? Or am I not clear on the distinction between the Index and Appendix namespaces? --EncycloPetey 18:29, 8 January 2007 (UTC)

Right you are. --Connel MacKenzie 20:10, 8 January 2007 (UTC)

I'm trying to make a hangman game which takes words from wiktionary. Firstly the game selects a random word from wiktionary. Then when the player wins or loses, it displays the meaning of that word. Is this possible to do? There are a few issues with doing this:

1. A lot of words here are very uncommon and in a language other than English. Would it be possible to only select common English words?
2. It would be wasteful to download the web page's source which the word is found, as there is a lot of formatting and other html/css code not needed. Is it possible to directly query the database to get the required information?

Note that I want to keep the game small, so I don't want to have to download a wiktionary data dump for it.

Thanks in advance! --WikiWizard 08:00, 9 January 2007 (UTC)

You could use random words from Category:English language, or you could try the MediaWiki Query Interface. —{admin} Pathoschild 18:08, 9 January 2007 (UTC)

That (MediaWiki Query Interface) was what I was looking for. Thanks. --WikiWizard 00:01, 10 January 2007 (UTC)

You're welcome. —{admin} Pathoschild 06:57, 10 January 2007 (UTC)

• I update http://tools.wikimedia.de/~cmackenzie/rnd-en-wikt.html after each XML dump. --Connel MacKenzie 18:04, 10 January 2007 (UTC)

• Moved from BP. Davilla 17:09, 2 June 2006 (UTC)

Someone (in the past couple days) changed something. ALT-A no longer functions on Wiktionary: namespace pages (like ALT-C does on NS:0 pages.) This has worked for a very long time. Anyone know what changed? --Connel MacKenzie T C 06:54, 25 May 2006 (UTC)

I think I've found it. ca-nstab-wp became ca-nstab-project by some wikimagic, and I've made ALT-C point to the new one for consistency. --Connel MacKenzie T C 17:14, 4 June 2006 (UTC)

Well done. Another browser incompatibility :-). — Vildricianus 17:16, 4 June 2006 (UTC)

Um, what? Something in the mediaWiki software changed this, not a browser thing at all. --Connel MacKenzie T C 19:16, 4 June 2006 (UTC)

• Moved from BP. Davilla 17:12, 2 June 2006 (UTC)

I just added a list of words used in Wiktionary that aren't defined in wiktionary at User:RJFJR/WTconcord. It's based on the XML dump (my program for building the list still needs some work but this list is still something to look at). RJFJR 06:46, 27 May 2006 (UTC)

Ooohh, excellent! Thanks very much. This is an extremely good idea.

The next challenge, of course -- (you were waiting for someone to bring this up, weren't you?) -- is to filter out all the simple derived words (plurals, past tenses, etc.) which do have definitions for their base words. (But having written code that tries to do this myself, I know it's not quite as simple as it sounds.) –14:28, 27 May 2006 (UTC)

This link may be helpful then: http://www.tartarus.org/~martin/PorterStemmer/ --Connel MacKenzie T C 15:19, 27 May 2006 (UTC)

The Porter Stemmer may be a bit too powerful for the purpose of inflections only. I would hope that derivations like poorly and weakness do get filled out. Davilla 17:04, 27 May 2006 (UTC)

I was about to say, "Thanks, Connel, for that Porter stemmer link", but I just tried the program, and at first blush I'm not too impressed. this → thi, apples → appl, people → peopl. Hmm. –scs 21:07, 31 May 2006 (UTC)

Also, it would probably be good to filter out headings (the XML dump marks those as such, right?), because the stylized terms used there (e.g. "romanizations") are swamping the rest. –Scs 14:54, 27 May 2006 (UTC)

Filtering out headings would be a bad idea, IMHO, as we do want those entered as well. --Connel MacKenzie T C 15:19, 27 May 2006 (UTC)

Doh! Of course. (What was I thinking?) Scs 16:39, 27 May 2006 (UTC)

Excellently done, RJFJR! I often filter out entry names that contain a colon in the headword, when trying to limit things to NS:0. This has the side benefit of excluding pseudo-namespaces as well. --Connel MacKenzie T C 15:22, 27 May 2006 (UTC)

I've been making modifications. To get around the problem with not knowing if a word is supposed to be capitalized and scanned against titles case-insensitive. And I modified it to ignore template invocations to cleanup some of the entries.

But what's really cool (at least to me): Somebody linkspammed us and they put it in so many times that part of the url (between periods) floated to near the top of the list of words that are used but not defined and when I searched for it I found and removed it. (And one of the words this happened with was in links that were made non displayign so we wouldn't even see it if we looked at that word.). RJFJR 02:10, 30 May 2006 (UTC)

Wow! That is cool. (What was it?)

Two more suggestions:

1. Can you work on filtering out inflected forms, pronunciation fragments, and the other biggest "noise" sources? Ideally, someone looking for a project, who checks this list for something to do, would find mostly words that need new entries, not a bunch of false positives to be thought about (or tagged) before finding the words that need new entries.
2. It would be nice to autogenerate, next to each word, a link to http://www.google.com/search?q=that_word+site%3Aen.wiktionary.org , to make it easier to find out how a missing word is being used.

–Scs 19:55, 30 May 2006 (UTC)

The words were allsex and virginporn (check my contribution list to find where they were). I'm trying to improve the filtering, and I've made some advances. Inflected forms are going to be a problem, they can be handled by manually adding them to my exclude list but I don't have a general way yet. The link will go into the next batch. RJFJR 13:49, 31 May 2006 (UTC)

I've been working on the program. The latest version is at User:RJFJR/WTconcord2. RJFJR 11:54, 3 June 2006 (UTC)

I've written a script (based on the "PorterStemmer" program Connel mentioned) for culling derived words. See User_talk:RJFJR/WTconcord for details. –scs 18:29, 3 June 2006 (UTC)

• cross-posted from WT:BP

• Some discusstion below merged from BP. Davilla 05:31, 3 June 2006 (UTC)

I just entered "prescriptivistic" in the search box and pressed . I was presented with a page that had "create an entry with that title" in BLUE. Clicking the link, I arrived at $1. Anyone know what the @$#% is going on? What changed in the last day? --Connel MacKenzie T C 04:34, 28 May 2006 (UTC)

Eek. No idea. "$1" should be replaced by the "prescriptivistic" argument you passed in, but the substitution is failing. Sure looks like a MediaWiki bug. Rod (A. Smith) 04:48, 28 May 2006 (UTC)

Looks like it was having trouble carrying the parameter through the nogomatch template. Strange. Rod (A. Smith) 05:07, 28 May 2006 (UTC)

Temporarily solved: Will file a bugzilla, then give a description here. --Connel MacKenzie T C 05:20, 28 May 2006 (UTC)

Subst'ing it back into the MediWiki namespace did the trick...I'm on #wikimedia-tech discussing it now. I'll have to file a bugzilla for it, then report back later. --Connel MacKenzie T C 05:21, 28 May 2006 (UTC)

See: http://en.wikipedia.orghttps://dictious.com/en/Wikipedia:Village_pump_%28technical%29#nowiki_in_links

See my changes to MediaWiki:Pagemovedtext from two days ago, same problem. Nogomotach was still working back then, however. —Vildricianus 06:42, 28 May 2006 (UTC)

I'd rather not repeat what I said at WT:BP here. I've learned my lesson about cross-posting though. Ugh.

Anyway, according to bugzilla:6115, ALL references in Special:SystemMessages that contain "<nowiki>" are now likely to break - badly. If someone feels like helping me plow through these 50 items (or so) I'd appreciate it. --Connel MacKenzie T C 06:16, 28 May 2006 (UTC)

bugzilla:6115. A change made today to the MediaWiki software allows for ridiculously punctuated characters (such as $1) to be passed through in a more consistent manner. Unfortunately, that means that "$1" is no longer expanded when referencing the "template" Wiktionary:Project-Nogomatch from MediaWiki:Nogomatch. A suggestion from Splarka was to subst: the "template" back into the MediaWiki: namespace, which did the trick. So that page is back to being sysop-edit only again. If needed, we could go back to the extenseive Javascript manipulation style, to allow it to still be a template, but I am quite leery of doing so.

Some talk-page moving around (and other assorted cleanup) is now needed for the residual "template", but otherwise, things seem to be back to normal. I don't know of any other MediaWiki: system messages that are similarly affected at this time, but I have not done any extensive searches. Yet. --Connel MacKenzie T C 06:06, 28 May 2006 (UTC)

Ah, that explains why $1 got "vandalized" last night - I had better unblock the innocent party. SemperBlotto 07:16, 28 May 2006 (UTC)

Looking in the comments, in history, the article $1 was created because it had a lot of wanted links to it, but I don't think it should have an article. The wanted links probably came out of entries intending to call the $1 variable marker in a Mediawiki message, not because anybody needs to know that it's a symbol for one dollar or one peso. When we're done with the technical corrections, let's delete the article and perhaps correct some "what links here"s in the process. --Dvortygirl 08:45, 28 May 2006 (UTC)

I think it is much better off 1) Existing, 2) remaining semi-protected. Copying this discussion on its talk page might be helpful too. --Connel MacKenzie T C 00:55, 31 May 2006 (UTC)

Hi guys. The {}rfvpassed}} template seems to be working well, but I should like to have a box similar to the "edit summary" box incorporated into the template to put in comments. For example necklace was referred for verification of the murder sense. On the talk page I used the rfvpassed template, but cannot say "Comment: Sense 2 has been verified" within the box to enable people to see what was referred in the first place and why it has been approved. Can anyone put in such a comment box into the template? Many Thanks. Andrew massyn 13:17, 28 May 2006 (UTC)

Can someone come up with a 'bot to move all rhymes: entries back to Rhymes: ? I spotted a few, moved them manually, but then checked Special:Allpages and saw there are hundreds of them. If we ever want Rhymes: to be a real namespace, these will need to be re-capitalized. Great! I would have posted at Connel's talk page but this is the place to do so, actually. —Vildricianus 13:23, 28 May 2006 (UTC)

I think there are cases like these when we can ask a dev to do a server-side script for us. This is how they did the uppercase/lowercase split for instance. It would also be a lot easier if one of us learned how to make such scripts and then we would only have to request they run it, rather than requesting they develop it too. — Hippietrail 19:15, 28 May 2006 (UTC)

I'm wondering what happens when a new namespace is made. Perhaps it's not necessary at all to move the rhymes: pages around, because they may be automatically capitalized (template:see = Template:see). —Vildricianus 13:11, 29 May 2006 (UTC)

Right, it could be rHyMeS: and still find it, once it is a namespace. Perhaps we should revive the "wanted" namespaces discussion from years gone by? --Connel MacKenzie T C 18:55, 29 May 2006 (UTC)

Absolutely. Wikisaurus to begin with, once we settle on the name (discussion still ongoing at BP). Then, Index:, Appendix:, Rhymes: and Concordance:. (Transwiki:? WT:?) —Vildricianus 19:54, 29 May 2006 (UTC)

I'm not sure that "WT:" would have as much support as the others. Transwiki:, absolutely! In fact, Transwiki: is possibly the highest priority. --Connel MacKenzie T C 01:09, 30 May 2006 (UTC)

The benefits of having a true namespace have come slowly to me, and I think they've not come at all to most users. Only yesterday did I discover the option to have random page generation for every true namespace. I'm sure there are still advantages I've not discovered. Can anyone sum them up? We'll surely need them if we want to convince people of the necessity to have these namespaces. It looks like the WikiSaurus debate is kind of stuck. —Vildricianus 20:27, 31 May 2006 (UTC)

Serious? That one looks to me to have resolved to Thesaurus: with only one objection. Davilla 14:06, 3 June 2006 (UTC)

On first sight, yes. Looking twice, one sees the main Wikisaurus contributors giving a different and quite valid opinion. I don't mind, I think Wikisaurus is equally fine. It's the technical aspect that bothers me now, as looks like it will take some time ere it is solved.

On the same topic: does someone have the technical know-how to read and more or less understand Help:Custom namespaces? One thing that bothers me is this note:

Any existing pages whose titles start with the letters "Foo:" or "Foo talk:" will become unavailable, so you'd better rename them first. (Where Foo: is the desired custom namespace).

This sounds like we can't make a namespace where there are already pages that exist in it. In other words, will we need to move, for instance, all Appendix: entries temporarily to another space, then set the namespace, and move them back? If that is the case, we'll certainly need developer help for this, if not already. — Vildricianus 14:18, 3 June 2006 (UTC)

If that's the case, it sounds like we should certainly leave them at "rhymes:" for now (unless the caselessness of namespaces means that the existence of a true "Rhymes:" namespace will make any existing pages whose titles start with the letters "Rhymes:" or "rhymes:" unavailable, in which case we'll have to temporarily rename all of them anyway.) —scs 12:17, 16 June 2006 (UTC)

Wiktionary:Discussion rooms

Any comments? I don't know where it should go, though. Top of all four pages? Community portal? The first table is the main thing I'm talking about; I've created the second to give some more overview. Ideas? Shall I include Wiktionary talk:Entry layout explained as the main "forum" for discussing layout?

Also, it's clear that I still want to start using the "Information desk" (whatever its name should be). —Vildricianus 14:05, 28 May 2006 (UTC)

I like it. Its very clear. Andrew massyn 14:12, 28 May 2006 (UTC)

Wow, that is is bigger than I'd hoped it would be...and I hoped it would be smaller, because I think it belongs near the top (2nd section?) of each affected page. But perhaps the extended-dance-remix version of the template is what is appropriate? After all, the question "where do I ask this" is what trips most people up. Novices and seasoned veterans, alike! --Connel MacKenzie T C 14:20, 28 May 2006 (UTC)

I suggest swapping Beer Parlour and Tea Room, to make the order across the top "Information desk / Tea room / Beer parlour / Grease pit" (which is more the continuum as I see it). –Scs 14:24, 28 May 2006 (UTC)

With the Beer parlour's header and archive timeline revamped, I see the following: the top box (with the 4 rooms) goes at the top of each room. It's quite small and should be no prob. The other box with the other 8 things could go then on the Community portal, together with the top box there. —Vildricianus 14:46, 28 May 2006 (UTC)

Works for me. Isn't WT:ID also for general brain-farts? (i.e. "Crap, I forgot how to foo".) --Connel MacKenzie T C 15:12, 28 May 2006 (UTC)

Feel free to change the wording. It's up to us how to define it. —Vildricianus 15:17, 28 May 2006 (UTC)

Installed. See Wiktionary:Discussion rooms (can't include here anymore, it has noinclude tags). —Vildricianus 15:51, 28 May 2006 (UTC)

I think this is awesome! It looks great and the new rooms are a brilliant idea. Widsith 16:10, 28 May 2006 (UTC) PS I moved the apostrophe in Newcomers’.

Gah, s'top pointing me to my erroneou's apo'strophes' ! —Vildricianus 16:19, 28 May 2006 (UTC)

Great !--Richardb 11:45, 29 May 2006 (UTC)

I could use a symbol for a triple bond in chemical formulae. I use the equal sign (=) for double bonds, but need one like it that has three lines instead of two. Any ideas?

'≡'? Unicode 2261, alt+240 - TheDaveRoss 07:42, 29 May 2006 (UTC)

Thanks, that works fine (after a bit of a panic - I pressed Alt Gr 240 by mistake (I think) and my screen turned sideways - took ages to get it back!) SemperBlotto 08:11, 29 May 2006 (UTC)

FYI, if you use Latex (see Help:Formula) to show your chemicals, the equivalent of the unicode character is "\equiv", e.g.:

${\displaystyle atom_{1}\equiv atom_{2}}$

Rod (A. Smith) 16:45, 29 May 2006 (UTC)

I would like to know what our actual users feel about Wiktionary. Can anyone think of an easy way to obtain feedback from them? Some sort of form-based question and answer system would be best. The questions that I would like answered include . . .

1. did you find the word you were looking for?
2. if you found it
   1. was the entry useful?
   2. how did you find it?
      1. did you use the "Go" or "Search" facility
      2. did you follow a blue link from another word
      3. did you use an Appendix or List
      4. did you use a Category
   3. what language was it
   4. what was the word (optional)
3. if you didn't find it
   1. did you then make a request for it?

Any system must not be too intrusive, and should not attract vandalism. SemperBlotto 11:33, 29 May 2006 (UTC)

We certainly could use Javascript to determine if they are logged in or not. Adding a form to the bottom of the screen would have to go to a toolserver page or something. Scs? Patrick? --Connel MacKenzie T C 18:33, 29 May 2006 (UTC)

Yes but what SB said about attracting vandals is true. Perhaps we can do something via e-mail? —Vildricianus 18:42, 29 May 2006 (UTC)

I don't get that. At all. --Connel MacKenzie T C 06:12, 9 June 2006 (UTC)

Sorry. Me neither. — Vildricianus 23:12, 9 June 2006 (UTC)

There seems to be some fighting over whether this template should produce s or sg. I think some of us here can come up with a solution which satisfies both parties and probably also those who might want sing or even the full singular. That last option would be easy to do across m, f, n, and c too but what would be a good way to do both that and allow the extra possibilities for sg? — Hippietrail 21:04, 29 May 2006 (UTC)

The ideal solution depends on easing user stylesheet replacement. Organizing our css class "namespace" hierarchy, though, is an important component to all such personalizations. (BTW, "personalization" is an industry term for user-specific customization.) So, to personalize display of "s|sg|..." for {{s}}, how about css classes .lbl-s-s, .lbl-s-sg", ...? Thus, <span class="lbl-s-s">s</span><span class="lbl-s-sg">sg</span>... in {{s}} can hide all but the user's selection in Special:Mypage/whatever.css. To support arbitrary text for {{s}}, our monobook.css solution extends to monobook.js, which can write the arbitrary label into each span of class lbl-s-s. Thoughts? Rod (A. Smith) 06:39, 31 May 2006 (UTC)

As a user I would want a single option for all of these foreign-language labels as a group, to display the single letter, two letters at most, a short abbreviation (with a period I would think), or the full term. I'm just not sure about the "two letters at most" option since the debate about s vs. sg might be a different issue, having to do with s as it stands for "noun". But when you craft a solution for that problem, remember the broader goal is simplicity.

By the way, I saw this used in a definition once, for an English word even, and just wanted to point out that because we are not paper these should be spelled out in full, just as esp. and usu. ar spelled out, in other situations where they might not be understood. Davilla 15:54, 31 May 2006 (UTC)

• Perhaps we need a set of "abbreviation" templates/CSS classes. In this case I would favour the prefix abbr-. I've thought about the "we are not paper" thing with regard to gender abbreviations. It's just two damn difficult to read a crowded translation section with everything spelled out. Only users completely unfamiliar with gender languages won't prefer the abbreviation. This doesn't mean I won't support an option to expand all abbreviations, just that plenty of us want these abbreviated even if we want other abbreviations in full. — Hippietrail 17:15, 31 May 2006 (UTC)
  • Weren't you going to start a thread here about naming conventions? I think a more generic "user customizable templates" template subnamespace is needed, right? --Connel MacKenzie T C 06:02, 9 June 2006 (UTC)
    • My two cents: The abbreviation 's' is ambiguous in a multilingual dictionary, since it stands for "substantive" (i.e. "noun") in many dictionaries (English and otherwise). I have no preference among the other possibilities, but "s" should be avoided. --EncycloPetey 02:27, 16 June 2006 (UTC)
      • Yup, fair enough. However, we only agree that we use n for nouns in Wiktionary. That said, I have seen "s. m." and the like given in some translations. These are just incorrectly formatted, and I reduce these to "m" when I see them. I've never been a fan of "s" and "p" - they don't immediately say "singular" and "plural" to me. But if people want to be able to set their options to show "s" and "p", I say let 'em. They will then know that "s" means "singular" and not "substantive", and if they see an "s" for "substantive", hopefully they will either query it or change it themselves. — Paul G 07:08, 18 June 2006 (UTC)

Here's an idea for somebody ambitious using toolserver. The scripts that some languages use don't yet have good support in all or some Operating systems. Burmese, Sinhala, and Khmer all spring to mind. Tibetan is also pretty poor in many cases.

What we could do is to provide graphical images for these languages. At least if pango and at least one good font can be found for each case and toolserver can run pango. A request could be sent off to the toolserver to generate an image that we then display.

A non-toolserver solution would be that we include images of such words just like normal images, but each one would have to be contributed by somebody capable of making them. — Hippietrail 22:06, 29 May 2006 (UTC)

Exotic scripts like IPA you mean? Davilla 15:57, 31 May 2006 (UTC)

Exactly! It's surely the one we need most, and wouldn't need a pango back-end, just 30 or so .png files. — Hippietrail 17:09, 31 May 2006 (UTC)

So, you could handwrite them on a piece of paper, scan it, then post the 30 images to commons, right? When better ones come along, they replace them... Or did I just completely miss what you were getting at? --Connel MacKenzie T C 05:59, 9 June 2006 (UTC)

There's a new parserfunction, #ifexist:, which will let us simplify the implementation of the "did you mean" blurb in MediaWiki:Noarticletext (as implemented in Template:didyoumean). I'm not sure what the current state of that code is or who's in the best position to test this change out on all the crazy cases we were worried about, but you might want to play with this. (It might let us remove the gory special cases for talk pages, too.) See m:ParserFunctions for info. –Scs 17:49, 30 May 2006 (UTC)

I meant to revisit this last week when they fixed NAMESPACE vs. NAMESPACEE (two "E"s) but haven't gotten around to it. So now, the talk pages probably don't need to be excluded from the special handling anymore. It is good to hear they added $ifexist:...that should greatly simplify it. Given the spate of recent software changes that relate to this, we probably should not push policy changes regarding #redirects too hard, for a few more weeks. --Connel MacKenzie T C 18:14, 30 May 2006 (UTC)

I'm too chicken to mess with it right now. I think I should play with #ifexist: in the sandbox for a while first. --Connel MacKenzie T C 18:19, 30 May 2006 (UTC)

I have put a policy proposal for verification in the Beer parlour. Your comments would be helpful. Andrew massyn 21:16, 31 May 2006 (UTC)

Darn. Could you leave a link behind when you move things please? I wanted to note that although the RFV process worked moderately well when it began, the nine months of books.google.com bombarding their site with questionable texts has resulted in the "three minimum" criteria no longer working as well as it did back then. I think these days, any single word search that gets more than 1,000 books.google.com hits is probably OK. But six months from now, that limit will seem silly. Likewise, for questionable terms, (i.e. terms that don't get 1,000 hits right off the bat) we should be requiring at least 15 citations. Some exceptions for narrow jargon uses could be made, (e.g. medical, computing or legal) if supported by other secondary sources, perhaps.

I think our handling of nonces/nonsense (i.e. funner/funnest) needs radical revision, that CFI does not even begin to address right now. We shouldn't have ain't, for example.

The 'three drinks minimum' criteria came about in an effort to protect certain pet terms of contributors here, at the time. It would be interesting to compare the number of results those same terms get now, nine months later. From there, perhaps we could get a comparative measure for tules of thumb. --Connel MacKenzie T C 07:53, 11 June 2006 (UTC)

I am trying to change the color of a link, specifically this one:

WikiSaurus:
		]
		For information on WikiSaurus formatting and how to create a new WikiSaurus entry please see Creating a WikiSaurus entry.

I would like the "WikiSaurus:" in the upper left to remain white while still linking to the category, but my various efforts at using CSS classes to do this have failed. Can anyone give me a leg up? (also, if anyone wants to clean up the fact that this has screwed positioning and allows text to the side of it that would be cool too) - TheDaveRoss 05:15, 5 June 2006 (UTC)

Fixed in Template:wse-header. Don't use classes for this, it's easier to do it on the spot with a <font> tag. It's debilitating, though, that you have to be careful where to put them (i.e. the colon has to be outside of them). Hovering still gives a blue line, but that's not a problem I guess. — Vildricianus 11:53, 5 June 2006 (UTC)

Thanks, I still get the full blue/purple underline but that is just HTML being broken. - TheDaveRoss 15:18, 5 June 2006 (UTC)

I have my preferences set to "Show preview on first edit" and tonight I am getting the red message Sorry! We could not process your edit due to a loss of session data. Please try again. If it still doesn't work, try logging out and logging back in. on any page I try to edit. (And yes, I have tried logging in again, and even rebooting again.) But if you can see this, it's untrue! Has there been a change to the software tonight? --Enginear 22:41, 6 June 2006 (UTC)

But I don't get the message after preview or save, only the first time the edit screen is displayed. --Enginear 22:44, 6 June 2006 (UTC)

Bug. See and followups. — Vildricianus 09:07, 7 June 2006 (UTC)

It now seems to be cured. --Enginear 11:03, 7 June 2006 (UTC)

Hello, in my case the BUG still exist and the followup link you posted does not work :( I am very interested in how to solve this problem when I use my own MediaWiki becouse it occurs all the time.

Can anyone please give further information and maybe provide a workaround? Thanks alot in advance! (83.188.195.59 07:52, 9 March 2008 (UTC))

Moved from Template talk:wotd.

Should a link to {{audio|{{{5|en-us-{{{1}}}}}}.ogg}} be added? Could having an audio file be made a criteria for WOTDs? --Connel MacKenzie T C 17:49, 27 March 2006 (UTC)

I think Dvortygirl has been very good about adding pronunciations for each of these. Is there any opinion, one way or another, regarding audio on Main Page/here? --Connel MacKenzie T C 16:57, 27 May 2006 (UTC)

AOTD? Included somewhere in this template? Yes please! I think there is enough audio to cover a couple of years, so this shouldn't be a problem. —Vildricianus 19:31, 27 May 2006 (UTC)

Done. --Connel MacKenzie T C 05:18, 30 May 2006 (UTC)

I'm afraid we'll have to make it more complicated and not show an audio link when there's no such file. June 7 has a redlink. — Vildricianus 09:45, 7 June 2006 (UTC)

I've added #IFEXIST to {{wotd}} but we'll see in a few hours if it works correctly, across to commons files. --Connel MacKenzie T C 13:54, 7 June 2006 (UTC)

We need two tricks for audio.

1. In general, the loudspeaker image should be clickable, bringing the clicker to the audio file or file description page, instead of to the image. Can be fixed using either CSS or {{click}}. See w:Template:Audio.

   In the interest of consistency, e.g. {{wiktionarysister}}, I think {{click}} is the better of those two options. --Connel MacKenzie T C 19:21, 7 June 2006 (UTC)

   Perhaps. It uses bad tricks, though, being not very compatible. Perhaps CSS hacks will be useful here after all. — Vildricianus 19:34, 7 June 2006 (UTC)

2. In the WOTD, I suggest there be only the loudspeaker icon, clickable then. It's best to get rid of the "Listen".

— Vildricianus 11:40, 7 June 2006 (UTC)

I've conditionally removed "file" in {{audio}}, stopping short of linking the icon as you say. Perhaps a separate {{wotdAudio}} or something? --Connel MacKenzie T C 13:54, 7 June 2006 (UTC)

I'm removing it completely from the template for now. Perhaps back later today. — Vildricianus 19:34, 7 June 2006 (UTC)

#ifexist: can't check Commons files. We may need to force a fourth parameter for the audio. — Vildricianus 20:49, 7 June 2006 (UTC)

Or procedurally ensure that only English words are submitted for WOTD for the next couple months, until it does? Has anyone searched buzilla for the relevant request yet? It's probably on my "votes" list, but obviously could use more votes. --Connel MacKenzie T C 04:40, 9 June 2006 (UTC)

I haven't even logged on to bugzilla. The one option here is that both the WOTD volunteer(s) and the audio volunteer(s) have to maximize communication and make sure not a single day is missed. — Vildricianus 23:22, 9 June 2006 (UTC)

Up. — Vildricianus 21:06, 13 June 2006 (UTC)

Nothing urgent for us, just a notice: . Brion is going to fix up some essential code in the near future, which causes templates that start a table but not finish it to be broken. This is only for HTML tags (so not our {{top}}, {{mid}}, {{bottom}} system), but the wikitable syntax {| , |} will be revised sooner or later as well. — Vildricianus 13:44, 7 June 2006 (UTC)

If I recall correctly, the tutorial points the students to Wiktionary:Sandbox. Wouldn't it be better to point to Special:Mypage/sandbox (erm, or should that be Special:Mypage/Sandbox?) This would prevent us from having to run a Sandbot that clears the sandbox, allow users to keep their very first edits for reference, reduce the amount that new users see spam/vandalism during the tutorial, simplify assumptions about redlinks during the tutorial, etc.

For that matter, the Main Page link to Sandbox likewise could be changed. Perhaps, simply all links to Special:Whatlinkshere/Wiktionary:Sandbox?

E.g. # python replace.py -ref:Wiktionary:Sandbox "[[Wiktionary:Sandbox" "[[Special:Mypage/Sandbox"

--Connel MacKenzie T C 15:33, 7 June 2006 (UTC)

Yes? No? Maybe?? --Connel MacKenzie T C 04:38, 9 June 2006 (UTC)

Absolutely not. The sandbox is for the very beginners. Making it into a complicated special page + subpage which turns out to be on their own page is disastrous for newbies. If Uncle G doesn't return we'll need someone else to write a script. Apart from that, the Tutorial still needs more cleanup I guess. — Vildricianus 23:18, 9 June 2006 (UTC)

The other plus side of having it in one location is that it doesn't take up excess space. If someone shows up and tests something, then leaves never to returns, there is no need to generate their sandbox. - TheDaveRoss 23:21, 9 June 2006 (UTC)

OK then, thanks for the feedback. I'll try and hunt down the Sandbot code, then see if I can run it every hour from toolserver. (Unless someone beats me to it.) Actually, I'll probably experiment on my own box first, then move it to toolserver. Then I'll probably have to set up a request page for other wikis that want it run... {sigh}. --Connel MacKenzie T C 18:42, 10 June 2006 (UTC)

Why, when I edit my usertalkpage (with only the two most recent threads not archived) do I see these after the edit box? I don't use these friggin' sub-sub-sub-sub templates!

   * Template:foreach
   * Template:foreach/pass1
   * Template:lookfrom
   * Template:temp
   * Template:x0
   * Template:x1

Someone please get them off my user talk page! --Connel MacKenzie T C 18:39, 10 June 2006 (UTC)

You do use lookfrom and temp. The others I've removed from {{temp}}. Davilla 19:15, 10 June 2006 (UTC)

Bless you. Thanks! --Connel MacKenzie T C 19:20, 10 June 2006 (UTC)

No, really, the one issue is Template:Recent changes. Please, figure out what it is and remove it! :-) — Vildricianus 19:36, 10 June 2006 (UTC)

I found the program ^wiki7 that I threw together a while back, and updated the BP "all headings" list. Improvements/suggestions welcome. Someone please rewrite it in python, to pull the pages dynamically, to get the level two headings. --Connel MacKenzie T C 19:18, 10 June 2006 (UTC)

I was thinking that a tool which would help in the purging of "wanted" lists would be great, something which could check whether a link was blue or red and remove the containing line if it were blue (or red I guess) would make things much easier. That "All Wiki Browser" might be suited for the task, but one of Connel's type tools could also be very cool. - TheDaveRoss 16:14, 12 June 2006 (UTC)

Which "wanted" lists do you mean? The two atop Special:Recentchanges? That sounds like a good exercise for someone learning Python, who has "spare time." --Connel MacKenzie T C 18:15, 13 June 2006 (UTC)

w:Template:Opentasks is being maintained by a bot, but I don't know whether that's with a time schedule (probably) or with something checking links or the like. — Vildricianus 18:47, 13 June 2006 (UTC)

I use Opera 8.5 in Windows 98, and have my cookie handling set as:

• 3rd Party -- "Accept only cookies set to the server itself"

• Accept cookies with incorrect paths
  • Warn me about incorrect cookie paths

Whenever I want to edit in Wikipedia I get a box warning me that "A path in a cookie does not match the page address" and "You might want to ask the site's Webmaster to set legal cookies". This happens on every refresh, ie: preview.

I must say now that this cookie problem does not present in Wikipedia or Commons -- it seems to be specific to Wiktionary, although it may happen on other Wikis which I haven't looked at yet.

So, since incorrect paths in cookies indicate potential security problems, please can cookie handling be regularised to pacify standards-compliant browsers? Please?  Gordon | Talk, 05:54, 2 January 2007 (UTC)

This is probably a better place to ask, than bug reports. Can you tell me which cookie it was in particular? Looking at MediaWiki:Monobook.js#Cookies I don't immediately see where cross-site cookies are being introduced. --Connel MacKenzie 06:58, 2 January 2007 (UTC)

The one I just generated contains the text "edittoolscharsubset=0;path=/wiki;expires=Thu, 01 Feb 2007 09:47:07 GMT". And this one from the preview says "edittoolscharsubset=0;path=/wiki;expires=Thu, 01 Feb 2007 09:55:58 GMT".  Gordon | Talk, 09:57, 2 January 2007 (UTC)

As mentioned below in "Strange tooltip", I don't have MediaWiki:Monobook.js in my cache. I do have "User:GordonE/monobook.js" in the cache, but that is only 26 bytes and isn't even js: all I see is little boxes in NotePad. (I've started refusing these cookies, and it doesn't seem to make any difference to my editing. Are these cookies really truly necessary?)  Gordon | Talk, 11:29, 5 January 2007 (UTC)

Has anyone filed a bugzilla: report for the overagressive JS caching issue yet? --Connel MacKenzie 16:46, 11 January 2007 (UTC)

Often, when hovering over headings in articles (but not talk: or Wiktionary: pages), the message '"" is not a standard Wiktionary heading' appears. For example, this message appears when hovering over the word etymology in the article waffle. Is this a problem with my computer or the Wiktionary articles? Smurrayinchester 14:24, 4 January 2007 (UTC)

It is a problem with WM software that is forcing the older, cached version of MediaWiki:Monobook.js, I believe. --Connel MacKenzie 17:27, 4 January 2007 (UTC)

This seems to be the same problem reported above in "Tooltips for standard headings". I don't know about User:EncycloPetey, but I don't have MediaWiki:Monobook.js in my cache -- the nearest I have is "User:GordonE/monobook.js" which is only 26 bytes and isn't even JavaScript (it comes up as little boxes in NotePad). So how can we get rid of the problem?  Gordon | Talk, 11:18, 5 January 2007 (UTC)

MediaWiki:Monobook.js is the en.wiktionary.org site-wide default Javascript file. I'll disable this entire section for now, as Hippietrail seems unavailable to fix it. If the media wiki software continues to over-aggressively cache the file though, it may still be a problem. --Connel MacKenzie 22:12, 5 January 2007 (UTC)

Toggling the Special:Preferences tab's "Disable page caching" checkbox allowed me to see the newer version (with it completely disabled) but as soon as I log out, it returns to the faulty version. Someone mentioned having some success by manually clearing the browser's Javascript cache, but something is really amiss. Tools, Options, Privacy, Cache, eliminates the local cached copies, but I get the same old version on the next visit? Likewise, hunting down the .js files in the filesystem and eliminating them has the same magical, resurrection properties on the next page view here, for me. If anyone knows more about the JS caching changes of the last month or two, I'd appreciate the information. --Connel MacKenzie 22:33, 5 January 2007 (UTC)

Has anyone filed a bugzilla:new bug report on this yet? --Connel MacKenzie 16:43, 11 January 2007 (UTC)

Many visiting Wikipedians (Wiktionary newbies) make the mistake of thinking that a redirect to Wikipedia either 1) works or 2) is useful.

In fact, putting a redirect on a 'User talk:' page merely prevents Wiktionary-relevant information (such as {{pediawelcome}}) from being given to the newbies.

More problematic, is that anything after the redirect (such as a section added using the button) is not rendered.

Anyone have ideas on how to address this problem (technically, or procedurally, or via policy)? --Connel MacKenzie 18:02, 10 January 2007 (UTC)

That is, I suspect any technical solution I devise could be evaded, so a policy approach might be better. --Connel MacKenzie 16:41, 11 January 2007 (UTC)

Is this Import log a new feature? Also, we seem to have another log: Bot log. What does the Import log do? Something to do with this page? A feature for transwiki? --Dangherous 14:39, 29 June 2006 (UTC)

Read Beer parlour. As for the bots, yes, bureaucrats can now set bot flags. I don't think we'll ever use it :-). — Vildricianus 15:10, 29 June 2006 (UTC)

No, not beer parlour. The Importpages stuff is described a little on meta:, with Wikimedia Foundation's decision to shut it off here. ("Here" at that time, meant Wikipedia, apparently. Since then, various sister projects have been split off, as well as other language wikis.) It seems to be a part of WikiHistory or perhaps WikiFolkLore, now, to think that there ever was an easy way to upload pages in bulk. --Connel MacKenzie T C 00:16, 30 June 2006 (UTC)

Yes, beer parlour, where there is a link to the most recent mail about it. Your description is horribly outdated. :-) — Vildricianus 15:49, 30 June 2006 (UTC)

Wow. I'd like to know a little more about how it gloms the user information over. They have "Wiktionary" contributions, but they aren't Wiktionary users? I guess that makes it, that much more tricky, to determine if a userpage is for a user, or not.

Will the next version of Import: also bring those users' passwords and preferences, a la single-sign-on? --Connel MacKenzie T C 09:42, 1 July 2006 (UTC)

Worth one's salt. They don't become users. Links go to their contribs pages, which do have the contribs they made to the article on Wikipedia. — Vildricianus 09:48, 1 July 2006 (UTC)

• Fixed section title to facilitate archiving. --Connel MacKenzie

If you see Special:Rhymes, there are loads of "broken pages". Could we get a bot to just delete all the pages with "broken" in their names? This seems aas though it were a hangover from the new namespaces. --Dangherous 15:33, 29 June 2006 (UTC)

Not a bot no. Perhaps I'll ask brion to run a batch delete script. Right now they do no harm. — Vildricianus 16:00, 29 June 2006 (UTC)

Done. 3477 pages zapped in a blink. — Vildricianus 17:10, 30 June 2006 (UTC)

Fixed title to facilitate archiving. --Connel MacKenzie

I think it would be cool for all the templates that use span id=transitivity to hide themselves, to also have span id=abbr transitivity so that "transitive" becomes "v.tr", "intransitive" becomes "v.it", "countable" becomes "C", "uncountable" becomes "U" etc.

Hippietrail probably has the full list somewhere, right? Should we start categorizing templates by the spans they use, using noinclude/category... ? --Connel MacKenzie 06:10, 22 July 2006 (UTC)

Well I don't know that there is a full list as such but we can certainly start making one. The dictionaries I've been using lately use vt, vi, and vti. All of the dictionaries I've seen which support countability use C and U in a square box or amd . As for naming the span IDs we need to adopt software engineering practices and have extensible naming schemes. We should nut out what exactly would be best but something like attr_xxx for word attributes or abbr_xxx for abbreviations is the lines along which to start thinking. — Hippietrail 06:38, 22 July 2006 (UTC)

You've mentioned that a couple times. Please expound on the concept with more examples. I'll go with the flow, if you let me know what makes sense to you. BTW, should the gender templates fall under the same rules as the transitive/instransitive/countable/uncountable for abbreviation masking? They are the exception to the rule, currently. --Connel MacKenzie 05:40, 23 July 2006 (UTC)

That's the sort of thing I'll turn into spaghetti if I try to make a document but not in a live conversation. I'm on IRC right now hoping to cover it with you. — Hippietrail 06:01, 23 July 2006 (UTC)

Erm, no, that was why I was suggesting organizing/grouping them with a category. --Connel MacKenzie 18:48, 26 July 2006 (UTC)

Adjusted section title to facilitate archiving. --Connel MacKenzie

This facility is difficult to browse. Could it have a "Display Users starting at" box like Special => All pages? SemperBlotto 08:41, 10 August 2006 (UTC)

You know, that's a good idea. Beobach972 19:17, 10 August 2006 (UTC)

Well then, please create yourselves bugzilla accounts and VOTE for bugzilla:1866. It has been around for about 7,000 wiki-years. With single sign-on coming soon, this will quickly become a big problem for us. --Connel MacKenzie 00:48, 11 August 2006 (UTC)

Resolved. I've asked Brion, took a couple of minutes. — Vildricianus 19:52, 15 August 2006 (UTC)

Adjusted section title to facilitate archiving. --Connel MacKenzie

See Wiktionary:Beer parlour archive/August 06#Redirects

Signing to facilitate archiving. --Connel MacKenzie 07:36, 16 August 2006 (UTC)

Wikipedia seems to be first character case sensitive now (See e.g. w:iPhone). So, could the {{wikipedia}} template be simplified to not convert the first character to uppercase please? SemperBlotto 14:48, 12 January 2007 (UTC)

Ack! Will we have to go through all the pages that call the {{wikipedia}} template and check case sensitivity? --EncycloPetey 15:59, 12 January 2007 (UTC)

Looks like it's still smashing case: your link goes to http://en.wikipedia.orghttps://dictious.com/en/IPhone but some JS magic changes the page title and article header. See w:Template:lowercase. Cynewulf 16:19, 12 January 2007 (UTC)

Would it be possible for edits to User and User Talk pages by the user concerned to be marked as patrolled automatically? One less chore for sysops. SemperBlotto 11:00, 27 October 2006 (UTC)

Done. --Connel MacKenzie 20:24, 14 January 2007 (UTC)

Based on recent patrolling experiences, I have created a template {{notwikipedia}}. The template displays as follows:

Please note that this is not Wikipedia. You are currently on Wiktionary, a dictionary web site. If you would like to add a word and its definition, then you are welcome to do so. Encyclopedic articles, such as the one you entered belong on an encyclopedia web site, such as Wikipedia. Thanks.

Thanks to Kappa for fixing the typo I made in the original edit. --EncycloPetey 01:54, 11 January 2007 (UTC) (Well, not the original edit; that one I forgot to save before clicking on a link that subsequently caused all my work to evaporate... or sublimate, or whatever it is that such ethereal electronic texts do when they vanish.)

The template assumes that they know the difference between Wikipedia and Wiktionary, which is not necessarily the case. I'd suggest the following wording, which just points to them to Wikipedia and provides links about what Wiktionary hosts:

Hello {{subst:PAGENAME}}. I have removed your recent contributions because they do not fit Wiktionary's mission as a multilingual dictionary. Encyclopedic content is more appropriate on Wikipedia, a wiki encyclopedia. See the criteria for inclusion and the style guide for information on what information Wiktionary hosts. Thanks.

—{admin} Pathoschild 04:07, 11 January 2007 (UTC)

I wouldn't mind seeing a {{#if:{{{1|}}}|] (see also: ].)||}} stuck in there somewhere. --Connel MacKenzie 16:39, 11 January 2007 (UTC)

Maybe a link to ] too. --Connel MacKenzie 16:40, 11 January 2007 (UTC)

How about this (variable text underlined)? The last cell is the template code.

—{admin} Pathoschild 00:50, 12 January 2007 (UTC)

On the rare occasion that I "follow procedure" and issue an actual warning thingy, I consistently do not remember which parameter is what. So, I meant a default link to RFD or the page or both. But what you've put together is really good. So, I dunno. Thanks for doing it, though. --Connel MacKenzie 05:00, 12 January 2007 (UTC)

I use TemplateScript to remember parameters for me. :) —{admin} Pathoschild 07:14, 13 January 2007 (UTC)

It is not clear to me what that does. Is that like javadoc? Or more like the thing that we wanted to add for pseudo-preload JS button insertions, for English words? --Connel MacKenzie 20:59, 14 January 2007 (UTC)

Also, I think it is fair to say that you are probably the only one on en.wiktionary that has heard of that tool before now. And 99.9% of our users refuse to get too advanced with JS tools. --Connel MacKenzie 21:01, 14 January 2007 (UTC)

OK, so the Tea Room is overdue for archiving right? I tried to archive the April discussion, but was unable to add it to the archive because of the spam protection filter. I have never encountered this before, and don't know how to proceed. I could not find the dreaded "edoctusdigital" in the text I was trying to archive. --EncycloPetey 03:32, 14 January 2007 (UTC)

That usually means the prohibited text is already in the archive. I'll look now. --Connel MacKenzie 18:24, 14 January 2007 (UTC)

Removed. Please continue now. --Connel MacKenzie 18:26, 14 January 2007 (UTC)

There's a problem with the ordering of Ancient Greek. When words are put into their various noun/verb/etc. categories (see, for example, Category:Ancient Greek proper nouns) the words with accents and breathing marks get separated from the rest. The solution to this is put a pipe and then put in a version of the word without accents and spiritus (eg. Category:suchandsuch|nonaccented form). So, here's where it gets more complicated. We have a template, Template:grc-noun, which puts in a standardized heading at the part of speech. In addition to displaying this stuff it also automatically adds the Ancient Greek noun category. What I'm attempting to do is add a sixth parameter which would be the non-accented form. I did this (check the most recent edit on Template:grc-noun), and it works. However, there is a slight problem with it. All of the nouns which have the template, but don't have the sixth parameter filled in, are ordered under { (see Category:Ancient Greek nouns). Does anyone have any solution whereby we could include this parameter, and yet keep it optional? Thanks. Cerealkiller13 10:30, 13 January 2007 (UTC)

I've made two attempts, but I think this one is beyond my meager skills. --EncycloPetey 17:58, 13 January 2007 (UTC)

I made one minor (I think it was minor) correction, but statisics show the job queue is stuck again...so until it clears, I don't know if the change had any effect, or not. --Connel MacKenzie 18:23, 14 January 2007 (UTC)

It doesn't appear to have worked. Category:Ancient Greek nouns still has a bunch of nouns sorted under "{". Thanks both of you for the attempt. Anyone else up to the challenge? Cerealkiller13 19:42, 14 January 2007 (UTC)

Oh, and one little clarification. While I stated earlier that it worked when the sixth paramater was filled in and didn't when it wasn't, it turns out I was incorrect. I filled in the sixth parameter in κόσμος as a test case, and it just moved it to the bottom of the { list, where I couldn't see it at first. Cerealkiller13 19:54, 14 January 2007 (UTC)

All better now? --Connel MacKenzie 20:15, 14 January 2007 (UTC)

Seems to be. I just tested ἄνθος which now alphabetizes with unmarked α, and the α-words without a template show up there as well. However, now that the template does what it's supposed to do, I've realized that all grc words will need the sixth parameter if we want the list in alphabetical order within a particular initial letter list. Notice under α that the list runs ἄνθος, αἰών, αἷμα; this is reverse alphabetical order. It happens because only ἄνθος has a sixth parameter (ανθος), while the others are still being alphabetized according to letters with diacriticals. So much for simple fixes. --EncycloPetey 20:43, 14 January 2007 (UTC)

Well then, "simply" remove the "|" after "{{{6" and the ones that don't have the parameter will get sorted out for you...  :-)   --Connel MacKenzie 20:54, 14 January 2007 (UTC)

It looks beautiful. Thanks Connel. No, it's not a quick fix, however, once αἰών and αἷμα have the proper ordering thing done (which they now do), the ordering works perfectly. I'll go through the nouns and proper nouns as time and sanity allow and get them all ordered properly. Thanks again for the help. By the way, may I ask why the pipe after the {{{6 fixes it? I'm a little naïve when it comes to programming language, and I really need to learn. Anyone who attempts to explain this to me will likely have to use very small words to get the desired result (me understanding). Cerealkiller13 21:11, 14 January 2007 (UTC)

Well yes, but this being Wiktionary, we're not supposed to use small words.  :-) Numbered or named parameters can be "passed" into a template. Within the template, they are wrapped in three "{}" squiggly-bracket pairs. (Note: Special keywords only get two, but variables get three.) Any variable can have a "default" supplied by the template itself...using the "|" pipe character. So, {{{1}}} will appear the same as {{{1|Four score and seven years ago}}} unless the person calling the template forgot the parameter. The {{{1|}}} trick is, simply, to substitute nothing instead of the text {{{1}}} (when the parameter is mistakenly omitted.)

NB: If no parameter is passed, and the template has "{{{1}}}", the parser function goes into "debug" mode and outputs "{{{1}}}" so you can tell something is wrong. --Connel MacKenzie 21:40, 14 January 2007 (UTC)

The flip side of all this, is the nasty category sorting. If no piped sort-key is supplied, or if the piped sort key is blank, the {{FULLPAGENAME}} will be used as the sort key (alas, instead of simply {{PAGENAME}}.) Since the template was always supplying "{{{6}}}" as the sort key, they all were sorting under "{" in the order the entry was last "touched" by the job queue process. (i.e. completely random.)

--Connel MacKenzie 21:40, 14 January 2007 (UTC)

A few quick notes on the template (if anyone cares). 1. Ancient Greek proper nouns have their own template which is identical except that it adds the category Category:Ancient Greek proper nouns instead, and has been updated with the naming convention. 2. I doubt anyone would, but just in case, please don't institute the new ordering scheme into the proper nouns yet, as I've been going through them in the order that they currently are in and cleaning them up, and I'll lose track of which ones are left if someone orders them properly right now. 3. For whatever reason, the template puts a space in, and so the definition must go directly beneath it in the editing window. If you leave an empty line, the entry looks goofy (for example ἄνθος). Cerealkiller13 21:21, 14 January 2007 (UTC)

There is a satanic extraneous linebreak between </includeonly> and <noinclude> causing the goofiness. --Connel MacKenzie 21:25, 14 January 2007 (UTC)

Was that all? Well, the evil has been purged. We can all sleep better at night now, knowing Wiktionary is no longer host to demonic influence. Cerealkiller13 21:32, 14 January 2007 (UTC)

Whew! --Connel MacKenzie 21:48, 14 January 2007 (UTC)

• By the way, the sort-key enborkenedness mentioned above, is why all subcategories should always supply "*" as the subcategory sort key - especially if the parent category has more than 200 items. --Connel MacKenzie 21:48, 14 January 2007 (UTC)

Sorry to spoil the joyous celebrations, but look at ἄγγελος. Why is it doing this? Entering Category:Ancient Greek nouns| into the editing window still works. Dammit! Cerealkiller13 08:12, 15 January 2007 (UTC)

Why does that one have only 5 parameters? Wow, I have not seen that error before. I added the 6th parameter, and it of course worked. Removing it cause the error to repeat though. Very odd. --Connel MacKenzie 09:25, 15 January 2007 (UTC)

Well, I'm going to revert Template:grc-noun back to its original state. If anyone has any ideas on how to get it to work, by all means give it a shot, but I don't want those entries looking ridiculous indefinitely. By the way, is there some page on wiktionary, or perhaps wikipedia that has the rules for all the formatting, or is it simply html or something else? Cerealkiller13 20:01, 15 January 2007 (UTC)

Also, does anyone know of any problems with some of the entries having six paramaters in a five paramter template? They all seem to look alright, but if it's destabilizing the Wiktionary servers or something like that, I'll go through and fix them. Cerealkiller13 20:05, 15 January 2007 (UTC)

Erm, no, I just meant it was an unusual error (no doubt, due to the extended unicode for the accented character that you are trying to mitigate!) I think the 6th parameter is fine - and obviously should be used. --Connel MacKenzie 20:12, 15 January 2007 (UTC)

The template documentation is all on meta:. m:Magic words is one place to start. --Connel MacKenzie 20:13, 15 January 2007 (UTC)

Well, as it turns out, Mr. Ullmann had the fix, and it seems to be working now (for the love of God it better be). Three cheers for him. And thanks for the magic words. Cerealkiller13 20:33, 15 January 2007 (UTC)

Sorry for the delay. The logo contest finalised and the logo is now ready; the image Image:WiktionaryEn.svg is uploaded to Commons, but needs uploading over the default Wiktionary logo by an admin. The favicon chosen, Image:Wiktfavicon en.svg, is also uploaded; the file appears very small, but is scalable for use in Sister Projects etc. Smurrayinchester 22:22, 15 January 2007 (UTC)

Also, logos for the French, Chinese, Japanese, Hebrew, Russian and Korean Wiktionaries are also ready (see Commons:Category:Wiktionary). Could any speakers of the above languages please inform the above Wiktionaries? The logo has already been set up on the Vietnamese Wiktionary. Smurrayinchester 16:15, 16 January 2007 (UTC)

one: we would have to have a vote here, since this logo is seriously disliked by a number of people, who think the current logo is just fine

two: we would need a new logo that is not a blatant violation of the Hasbro/Spears trademarks Robert Ullmann 15:04, 17 January 2007 (UTC)

Not sure when this changed, about when I was in Rwanda last week ... the definition of the history tab was changed on the wikt, maybe here, maybe by meta. Anyway, it breaks the code in pywikibots. It will raise NoPage on any page get call. To fix, edit wikipedia.py, in function getEditPage, change:

RversionTab = re.compile(r'<li id="ca-history"><a href=".*title=.*&action=history">.*</a></li>')

to

RversionTab = re.compile(r'<li id="ca-history"><a href=.*title=.*&action=history.*</a></li>')

I don't know if this is specific to en.wikt, and I'm not reporting this on SourceForge: their bug procedure is beyond ridiculous. Robert Ullmann 16:51, 16 January 2007 (UTC)

Yes, an extra DIV has been added at the WikiMedia software layer, wrapping the "tab" headings for some reason. Do you syncronize with SVN daily? I'll point the #pywikipediabot channel to this conversation, now. --Connel MacKenzie 07:32, 17 January 2007 (UTC)