User talk:Robert Ullmann/2006

Welcome
Hello, and welcome to Wiktionary. Thank you for your contributions. I hope you like the place and decide to stay. Here are a few good links for newcomers:
 * Wiktionary Tutorial
 * How to edit a page
 * How to start a page
 * Our format guidelines
 * Criteria for inclusion
 * Wiktionary Sandbox (a safe place for testing syntax)
 * What Wiktionary is not
 * FAQ

I hope you enjoy editing here and being a Wiktionarian! By the way, you can sign your name on Talk (discussion) and vote pages using four tildes, like this: &amp;#126;&amp;#126;&amp;#126;&amp;#126;, which automatically produces your name and the current date. If you have any questions, see the help pages, add a question to the beer parlour or ask me on my Talk page. Again, welcome! --Connel MacKenzie 20:03, 13 July 2006 (UTC)

ARchar
Hi! ARchar is a template that we use to make Arabic script easier to display. Most users that come here do not have proper Unicode Arabic fonts. Their computers usually display small Arabic charcters. The ARchar template specifies to use one of the following fonts: Tahoma, Arial Unicode MS, Code2000, Traditional Arabic. So far, I think that Tahoma does the best job. It's the cleanest. But, if a user does not have Tahoma, then the computer chooses Arial Unicode MS and so on. The font size is also increased to 125% so that the characters can be seen clearly. --Dijan 22:37, 16 July 2006 (UTC)


 * We usually use "<". I think it's more formal and is how most printed dictionaries do it.  BTW, thanks for fixing KiSwahili (I don't know why this is listed as an English word.  I've never heard it this way.)  --Dijan 23:22, 16 July 2006 (UTC)

Template:sw-noun
I asked User talk:Vildricianus about creating this and he said to go ahead and try; think I've got it about where I want for the first pass. What do you think? Robert Ullmann 21:49, 17 July 2006 (UTC)


 * I like that it accepts this form:
 * That will allow us to have the template categorize Swhili nouns into classes based on the prefix change. Since there is no common affix to pluralize Swahili nouns, so it makes sense that you have the argument-free invocation produce the uncountable (usually abstract noun?) output. I'm not sure whether the Swahili templates must support the table-format output that the corresponding English templates support, but I would think that consistent output format between all language entries may be desirable. What do you think? Rod (A. Smith) 22:27, 17 July 2006 (UTC)
 * That will allow us to have the template categorize Swhili nouns into classes based on the prefix change. Since there is no common affix to pluralize Swahili nouns, so it makes sense that you have the argument-free invocation produce the uncountable (usually abstract noun?) output. I'm not sure whether the Swahili templates must support the table-format output that the corresponding English templates support, but I would think that consistent output format between all language entries may be desirable. What do you think? Rod (A. Smith) 22:27, 17 July 2006 (UTC)


 * Nouns like mbuzi aren't uncountable; they take count (or measure) words when meaningful: "mbuzi" goat; "mbuzi tatu" three goats. This is common in Bantu words that are not in noun classes, such as M-WA. And there are some irregulars, but not many. I took out the table format to make it easier; I was learning the template syntax (this is the first template I've edited on a wiki). Robert Ullmann 22:41, 17 July 2006 (UTC)


 * Your first template? You sure chose an ambitious place to start! You'll probably find plenty of interesting quirks to stay entertained as you go. :-)


 * Anyway, thanks for the explanation about the countable, non-pluralizable nouns, and it makes complete sense to start without the table format. I like the template name and parameter model, which are probably the most important aspects of template design (so invoking pages can be written just once). Everything else about your template looks good, too, so I'd say to add the table output and announe the template on WT:GP!
 * (Note, there will probably be a push soon to move to the new 3-letter ISO language codes, but even so, template redirects from the 2-letter versions will keep existing invocations working.) Rod (A. Smith) 23:49, 17 July 2006 (UTC)

FYI, I have created an extension that allows template editors to call the extension with a language (e.g. using either "sw" or whatever its 3-letter code is), a part of speech (e.g. "noun"), an inflection class, and a desired inflection (e.g. "pl" for plural) in the template code. That extension will then look up a rule (in the form of a regular expression located in the MediaWiki namespace) for forming the desired inflection.

There are two options for how we could incorporate the extension in to. One is to list "M-WA" as an inflection class for Swahili nouns with the plural-forming rule. Editors of the lemmata (singular Swahili nouns in this case) can then allow the extension to generate the plural by specifying ".

Another option is available if most Swahili nouns that begin with "m" form their plural by replacing "m" with "wa". If that is the case, we could just create a regular rule (i.e. without a named inflection class), which editors of Swahili noun entries could invoke as " " (not there is no inflection class).

Anyway, I'm pretty excited about the extension because if people like it and we install it here, it may simplify entries for lemmata of many languages. Please offer any feedback you might have. Rod (A. Smith) 01:02, 6 August 2006 (UTC)


 * Very interesting; when I started looking at the template I was wondering if there were string functions so I could do just about exactly that (I was thinking of ).


 * One can't assume m->m-wa, there is also an m-mi class, (mchezo sport, michezo sports) and a lot of the nouns without plural forms start with m. So we'll always need the class, but that is no problem.


 * In Swahili (as in other Bantu languages) verbs and adjectives take noun-class concordance prefixes. So the same classes can be used with your extension in sw-adj and sw-verb. There are also tense markers, so I am not diving into that just yet; I'm noodling on just what it should look like. Thanks! (do you watch talk pages you've added to, or should I be replying on yours? I like keeping it together.) Robert Ullmann 10:28, 8 August 2006 (UTC)

Keeper
Of course it does raise the question if Keeper fulfills the CFI, but it also separates out that question from the mess that was at keeper imo. And I truly do not think that it belongs in keeper in a case sensitive wiktionary. Frankly, I did not have the energy to take on the Harry Potter fans today... *smile*--sanna 09:29, 19 July 2006 (UTC)


 * I have noted the use of and  instead of cattag, and will use those in the future. I have read the RfC, have I violated the discussion there in any way?--sanna 09:52, 19 July 2006 (UTC)


 * Ah, I misunderstood you. Thank you for the note at RfC *smile* --sanna 09:56, 19 July 2006 (UTC)

Sami languages
There are actually several Sami languages: Northern Sami (by far the largest one, a few ten thousand speakers and the language people use when they speak about the Sami language), and smaller ones like Inari, Skolt, Lule and Kildin with hundreds or less speakers. They are more or less mutually unintelligible (in addition, Kildin &amp; Akkala use cyrillic alphabet). All of them have own ISO codes, so they qualify as separate languages in Wiktionary (see List of ISO 639-3 codes). Kildin Sami's ISO code is sjd. I don't have any dictionaries for those smaller languages, so they'll probably remain curiosities, but it would be nice to see a decent list of for example Skolt words...--Jyril 22:24, 23 July 2006 (UTC)

Re: Egyptian languages
Thank you so much! I should have guessed that Wiktionary would support all manner of language functions. I will definately go back through and add the hieroglyphic text. It is indeed much better to see the word than just describe it. CAmbrose 15:30, 24 July 2006 (UTC)

It's a shame that the template didn't work, it would be awesome to cycle through the Egyptian words like going through a dictionary. Even if there was no bug though, it still wouldn't quite work. For example pr-aA has two different groups of glyphs despite the words being transliterated and said the same. I do like your formatting much better though. Thank you for all your help! CAmbrose 00:22, 25 July 2006 (UTC)

Hayeren Template
On most of the sister sites of Wikipedia, Armenian characters were not being displayed properly. I didn't try it out here because I figured it is the same, but it appears to work. I'll revert my uses of it. Feel free to delete it. - Rappo 22:39, 15 August 2006 (UTC)
 * The problem is evident now. For example, this letter: շ should actually appear as  and also ք ->  (since the template doesn't exist anymore, just put your cursor over the link and the correct form of the letter will appear).  Any ideas?
 * Nevermind, it seems to depend on the computer Rappo 22:22, 16 August 2006 (UTC)

Re:あくぎ et al
Japdef could've come from any of the many Japanese words I've seen it used on here. I just use it because its fancy, sorta. 悪戯 can be pronounced akugi or itazura, so I figured Furigana would be better? I've been using the janoun template for the kanji entries, then furigana for hiragana entries. Is that bad? Ric | opiaterein 15:23, 19 August 2006 (UTC)
 * So for things like kaku that have...a lot of meanings, should all the meanings be organized by part of speech? Ric | opiaterein 15:42, 19 August 2006 (UTC)


 * Yes, absolutely. And the pos headers get repeated if there are (for example) two different nouns (as opposed to one noun with two different senses). (can we keep the conversation thread on one page? ;-) Robert Ullmann 15:47, 19 August 2006 (UTC)


 * I've always thought keeping it on one page was a little difficult, because one of us isn't going to be aware of new messages. Anyway, I'll try to keep this in mind. :) Ric | opiaterein 15:51, 19 August 2006 (UTC)

Thanks
Thanks so much with your help, and to other Wikitionary users. I am glad you are following and correcting what I have contributed today. I have always been curious and interested in how to a dictionary is put together. I imagine lots of people working together, with different fields of expertise and different skills. I hope to add more to this project soon.
 * Another thing: is it OK to write a blog on my page? I doubt the site is for blogs, so will it get deleted here? I like the idea of not having to sign in to write a blog. --Feet first 22:47, 24 August 2006 (UTC)

re:kansen as an example
What I couldn't quite put into words before was that people who don't know the difference between kanji, hiragana, katakana and romaji should learn something else about Japanese before trying to learn words and put them together in sentences that are incorrect. Other wise we might have people saying "watashi desu mike," not realizing that ha is the article that marks the subject, with the verb coming at the end of the sentence. So having a dictionary format that makes it 'easier' for people who know less about Japanese...hmmmm....see what I mean? It's not helping them.

So that aside, you do realize that the format you've used is the same that is used by that "japdef" template? :) Ric | opiaterein 12:25, 26 August 2006 (UTC)


 * Yes, I'm paying attention. Everything that people have done is an expression of what they think should be done, modulo what they think is possible or allowed. Robert Ullmann 12:33, 26 August 2006 (UTC)


 * The only thing I don't like is that it's in a numbered list... the words are all different, even though they're pronounced the same. I liked it the way I had it originally. lol Ric | opiaterein 18:17, 26 August 2006 (UTC)

English meanings in your recent AWB changes
The English meanings don't display in the page after your changes..

ex: 依

They are in the text in edit mode, they just don't show up in the display. (unsigned comment by Versageek)
 * See WT:BP discussion, we are debating whether the common meanings are useful, and whether they are a copyvio. At this exact moment, the template isn't displaying them. Robert Ullmann 15:19, 20 October 2006 (UTC)


 * ok, as long as it's intentional :) --Versageek 15:29, 20 October 2006 (UTC)

Yep
Yes, but a CheckUser would have to verify that. There are occasionally people who join our project from Wikipedia or other language Wiktionaries -- or who just took their time as an anon before being coaxing into creating an account (like our top Slovene and Portuguese contributors). --EncycloPetey 23:43, 24 October 2006 (UTC)

Danton

 * What do you mean you like "Ruedanton"? You mean as a nickname? Why do you like that name so much? That was just the name of the street my local church was in. Beautiful church too. --Pickled onion 23:53, 24 October 2006 (UTC)
 * p.s. I'll get bored of WT soon. My new job starts next Monday. Until then I'll help out the WT project. The words aren't gonna descrive themselves, are they. ;)--Pickled onion 23:53, 24 October 2006 (UTC)

Thanks for fixing my screwup on top2 template, and sorry! :-(
Hello,

I see that I mucked up the template with some extra carriage returns. :-( Thanks for fixing it, and sorry about that; I only just found Wiktionary:Index_to_templates, and didn't realize until then that there was a column difference between top2, top3, top4, etc. (as there was no indication on their pages or their talk pages.) I wasn't intending to make a change, but rather to document what seemed to be an undocumented tag that was nonetheless in frequent use.   I'm guessing the distinction is to make the CSS more manageable or something...is there any other difference between them that I should be aware of?  --Jeffqyzt 01:40, 25 October 2006 (UTC)


 * I'm not at all sure you introduced the extra return (there was really only one). But it is gone. top2 sets up the first column at a fixed width of ~48%, top3 at ~30%, top4 at ~22% or so. Then the mids do the same; that's why mixing them is messy. Robert Ullmann 01:49, 25 October 2006 (UTC)

Unblock notice
I am obligated to inform you that I have unblocked User talk:144.124.16.33. I hope you can agree with the change. No action is necessary on your part. DAVilla 11:28, 25 October 2006 (UTC)


 * Two edits to these words in an hour? Okay, not proof, but it sure fits the pattern of a vandal who comes back to check his work and tries something else when it has been reverted and blocked. We'll see what happens ;-) Robert Ullmann 11:42, 25 October 2006 (UTC)


 * I should explain that I have this word on my watchlist (on account of the quotations), and that the definitions given regularly migrate their way from Urban fictionary. Also the writing style of the two contributions was very different, or otherwise I wouldn't have been quite as confident. Of couse certainty is never achievable. DAVilla 19:24, 26 October 2006 (UTC)

AWB changes - problem
I've taken a closer look at some of the changes you're making with AWB. Sadly, I'm finding more major problems.

You cannot include both the inflection and definition in the same sub-template. It seems these templates have undergone some significant reworking without adequate review. Definition lines must start with "#". Even when referring to some other definition. Since CKJV entries normally have mere translations entered for their definition lines, it seems someone has taken the unacceptable shortcut of combining the inflection information with the definition information. This doesn't just parse incorrectly for my tools; it also means that 'dict' rfv2229 information cannot be extracted.

Please, let's get this major problem straightened out, before moving on with your NanshuBot conversion(s). --Connel MacKenzie 15:38, 25 October 2006 (UTC)


 * What "sub-template" are you referring to? Robert Ullmann 15:41, 25 October 2006 (UTC)


 * What "sub-template" are you referring to? Robert Ullmann 15:41, 25 October 2006 (UTC) (and lets keep this conversation together on my talk page, I trust you are watching it! ;-)
 * OK, please update the timestamp of the link on my talk page if I miss a reply. --Connel MacKenzie 15:49, 25 October 2006 (UTC)


 * Am I reading this wrong? You use in the first language section,  and  both which look like they are combining definitions into the inflection line information, which need to be separate templates (two for Han char, two for Han ref,) to be forwarded as dict information.  --Connel MacKenzie 15:49, 25 October 2006 (UTC)


 * Han char has the "common meaning" which is not a definition of the character (it might be more like etymology for some of the languages). We really don't want it showing up as a definition. We may not want it showing up at all. (that is TBD). The individual language sections will (and a number of them do) have definitions in standard format.


 * Han ref has no definition information at all ... it is dictionary references. So I don't know what you mean?


 * Oh, can you give me a ref to what you mean by "forwarded as dict information"? Robert Ullmann 16:01, 25 October 2006 (UTC)


 * If a symbol can't be described in any way at all, then it shouldn't have a Wiktionary entry. The occluded "common meaning" information clearly is the closest thing to a "definition" these entries have - if that duplicates etymological information, that is fine, but the place for the definition information is on the definition line.  If a separate template there (?) then "hides" that information, that is fine...but it needs to be on a definition line.  Preferrably, with  taking its place.
 * Yes, we keep it for some because we have no other definition info. But it is not a defintion of the meaning of the character in a language which is the only thing that makes any sense. Some of them have proper language definitions with POS headers and inflection lines, of course we could use more.


 * Are you asking what rfc2229 'is?' The interface is not working today, but that has been a long term goal for quite some time now.


 * Pardon, you said "rfv2229", which made no sense to me? (And speaking as the author of numerous RFCs, 2229 is beyond obsolete, it even lacks a language selector. And a ridiculous number of other things. Who could possibly care? If it even referred to an XML format for a standard dictionary entry it might be worth looking at.)


 * Note that that isn't the only thing that is affected by improper formats. Almost all of my 'cleanup' todo lists will now list your entries, as they are no longer identifiable as "garbage NanshuBot entries."  So you are about to flood my lists.


 * Because they aren't "garbage NanshuBot entries" any more? We can have proper WT:ELE sections with languages and all? There isn't any junk in front of the first L2 header? And they now need language fixups as much as any other entries in the wikt? (And what would your lists be flooded with? Are there other things to fix?)


 * Additionally, the combination of the inflection information and the definition information breaks all of Patrick Stridvall's toolserver tools. Additionally, the combination breaks all JS cleanup tools.  Additionally, the combination makes the entries incomprehensible to English speakers.  Additionally, the combination is in direct conflict with WT:ELE.  Additionally, templates cannot be transcluded in monthly XML dumps.  Additionally, multiple definition lines cannot be entered with your template scheme.  Additionally, multiple definition lines cannot be accurately qualified by italbrac/cattag type definition line qualifiers.


 * Again, there is no inflection or definition line information in either template; those only occur in language sections. Definition lines don't belong under a "character" header outside of a language; it is nonsense. No tools should be extracting any language information from the NanshuBot stuff, as you say yourself, the format (at least) is garbage. It is at least recoverable in some cases.


 * I could go on and on, about the dozens of other things you are inadvertently breaking. --Connel MacKenzie 16:22, 25 October 2006 (UTC)


 * Anything built on NanshuBot entries is GIGO ;-) eh?

What would you have us do? You complain that the NanshuBot stuff is garbage, and then you complain when someone takes out the garbage?! If you want tools built with dependencies on the garbage to keep working, then we have to keep the garbage, right?

Of course anything built on the Nanshu junk has to be fixed, including your stuff. What else could you possibly expect?

Tautology: Any tool dependent on the NanshuBot garbage will need fixing when the NanshuBot garbage is fixed.

Do we fix it or not?

Best regards -- Robert Ullmann 18:16, 25 October 2006 (UTC)

technical
Take a look at 間. I'm picking on this precisely because it is subtle. Note that the definitions (of two different words) in Japanese don't match the so-called "common meaning", even though one is close, but distinctly different. The Chinese languages and Korean lack definitions; they are needed. The "Translingual" section has no information that is useful in 2229 or any tool extracting definitions. It does have some that is partly useful to humans looking at the entry given that none of the Chinese languages have sections in this entry. We have choices: we can present the "common meaning" in a way that is not represented as a definition (either visually, or to a tool); or we can delete it; or we can present it (incorrectly) as a valid definition of the (somehow) abstract character 間. (Oh, and there is other stuff that shows up in "common meanings" that isn't even "meanings" ;-)

Nanshu did this right. He took a field from the Unihan database called "kDefinition", and called it "Common meaning" because definition it is not.

For your tools: IMHO, you should treat L3 headers "Han character", "Kanji", "Hanzi", and "Hanja" as valid (at least for now). Treat "Compounds" as valid at L4 (only under one of those, if you like); it makes much more sense than "Related terms"; often the single character isn't a "term" in the first place; it is specified by the WT:ELE language variations.

Tell me what else is flagged; I am trying to fix anything/everything I can automatically!

I don't know what predicate you are using to identify the Nanshu entries, but if I am breaking it, I can suggest an improvement: use the Han and Han-Ext-B character ranges? (One simple way is to use char >= '一' (4E00) that ID's the Han range pretty well. I'm not doing anything with Ext-B which is lower.)

By all means flag the entries/languages that don't have any POS header or definition lines. (If you want me to introduce some stub template I can, but I can't invent the POS header or inflection line/template.)

What else? Robert Ullmann 19:07, 25 October 2006 (UTC)


 * The main thing I am concerned about is that you say "POS header or definition lines." My point is that both are required, separately.  If the definition line contains only  and the commented-out "Common meanings" that would be fine by me...optimal, really.  But to not have any definition line (or to have that "common meanings" included as part of the inflection template) is not just wrong, but also more difficult to correct, later.


 * I just don't understand why you keep saying "inflection template"! What "inflection template"? (Han char isn't; Han ref certainly isn't...) Yes, language sections need those. "Translingual" isn't a language. (almost by definition I would think ;-) What definition lines? Yes, If you want me to muck into languages, like Mandarin or Korean. But mostly I was leaving those alone.


 * Do you want me to create for you to use on those definition lines?  --Connel MacKenzie 19:16, 25 October 2006 (UTC)


 * Um, for where? It makes no sense inside a language section (which has definitions); what is it going to do in the non-language Translingual section? The AWB thing doesn't do anything inside language sections except to fix the kanji template(s) inside Japanese (add rs parameter, subst so the Readings header isn't inside the template)

''It makes no sense to have definitions in the Translingual section. Han characters do not have definitions outside of a particular language.''

See 字. The pictograph is a roof over a child, (see the Seal script in the entry), meaning to give birth. The ideographic character means to give birth, to marry, to be an educated adult. That was then used to symbolize writing (education) and written characters. (And also read in Japanese as part of a village, perhaps neighborhood.) Giving 字 a "definition" of "letter, character, word" is like saying "run" means "a score in cricket; baseball". Not wrong, not useless ... but, um?) Robert Ullmann 19:53, 25 October 2006 (UTC)

The Translingual section does not have parts-of-speech, inflections, or definitions. It can't. None of them make any sense. The language sections can, and do when they have been added. And the language sections should. And there is nothing conflicting with that. Robert Ullmann 20:07, 25 October 2006 (UTC)


 * I'm missing something big here, and it is disturbing. Why do you have "common meanings" listed in the Translingual section, then?  --Connel MacKenzie 05:50, 26 October 2006 (UTC)


 * Where else are we going to put it? I'm very open to ideas. I had simply made it non-displayed; but at least two people missed it because we don't have definitions for the Chinese languages yet for a lot of common characters; I've used it myself, knowing its limitations. It really ought to go away. It would be better if Nanshu hadn't picked it up, but at that time there were no language POS sections for CJKV at all!


 * I could stash it under Chinese/Hanzi, with the other stuff that needs separation into (at least) Mandarin, Cantonese, and Min Nan...
 * Hmmm, maybe not such a bad idea. It (common meaning) is usually pretty close to the Mandarin definition. Might be a starting point for real POS entries.


 * Do note that my ruleset is very carefully idempotent; if I run it on an already formatted entry it does nothing; if I add a rule and run it again, it does exactly what it would have done if the rule had been there in the first place. (the reverts you may have noted are when the ruleset didn't meet this criterion, and I had to fix something ...;-) What this means is that it is easy to improve things; just run it again through Category:Han characters Robert Ullmann 12:02, 26 October 2006 (UTC)


 * So, by your conclusion, we shouldn't have @ containing a ===Symbol=== section within the Translingual heading? --Connel MacKenzie 06:48, 26 October 2006 (UTC)


 * It's a symbol. You could probably give 一 a translingual definition as a symbol or numeral as well; but that doesn't apply to the 21000+ characters that aren't symbols or numerals. (Compare a, which is given a definition in Translingual only as a symbol. The letter-usage is, of course, in a dozen+ language sections.) Note that a big difference is whether the character is used at all in the language. The majority of un-common characters are/were only used in Mandarin (or what we now call Mandarin). A lot of common characters are used in very distinct subsets of the 16 or so CJKV languages. Robert Ullmann 12:02, 26 October 2006 (UTC)

example
Please look at 上.

If we put the "meanings" there as definitions (where they aren't too bad), and fix the headings a bit ...

Now they will show up to your tools as Chinese language definitions, not some unknown language.

I've looked through your todo lists, and I don't see how this would flood them (presuming Kanji, Hanzi, and Hanja are considered valid L3 psuedo-POS headers). There are other things that will show up, but properly, as they should. (一 has an L2 Etymology section ;-)

This keeps the "definitions" from being directly associated with Japanese and Korean where they are often misleading. (I don't know enough about Vietnamese usage.) Robert Ullmann 13:17, 26 October 2006 (UTC)


 * That is much closer to addressing my formatting complaint. But the Translingual section still has no "#" lines.  (That was my original, perhaps poorly worded complaint.)


 * I think my argument of flooding the /todo lists was a bit of a stretch. The NanshuBot entries will remain excluded until I turn those checks back on.  But I had hoped to do so, as a result of these entries being cleaned up.


 * The main /todo list I'm concerned about is /todo3. It looks for "#" lines within a recognized "POS" section.  From all that you've said in the past few days, it seems to me that I can never consider "Han character" to be a valid POS section.  I've added "Kanji", "Hanzi", and "Hanja" as POS headers (although I remain doubtful that they should be considered such) so that when I do turn "the Nanshu"s back on, it should show only those that are most defective.


 * I'm also still not convinced that "Mandarin" was decided upon, for the Wiktionary name for the Chinese language. The target audience is still the English speaker, not the dialect expert.  But that is quite a different conversation.


 * I think the approach you've just suggested will be very helpful. I do not wish to delay you further.  But perhaps in light of all the changes, you should re-summarize them on WT:BP?  --Connel MacKenzie 20:57, 26 October 2006 (UTC)

Conversation tone
It looks like this may take some extended discussion. Can you please look at Grease pit and consider connecting to that IRC-via Java+CGI interface? When we conclude the conversation, we can post it all here - but the WMF servers are quite slow right now, and that style of talk is probably much more productive. More importantly, is likely to be less time-consuming for both of us, so you can get moving again, once the templates are corrected/split. --Connel MacKenzie 15:56, 25 October 2006 (UTC)


 * After connecting to the gateway, use "/join #wiktionary" to join the Wiktionary conversation channel. --Connel MacKenzie 16:00, 25 October 2006 (UTC)


 * You can just tell me what is going on? They aren't inflection OR definition templates. So I need a real description of why either is relevant. And I can't play on IRC; I am doing several things at once, and must be able to timeshare on my terms? It can wait a bit. Robert Ullmann 16:05, 25 October 2006 (UTC)


 * IRC is a more efficient tool - that is the only reason I suggested it. You have three open ended questions (above) already - delving into the topics in a conversational manner reaches mutual understanding much faster.  And this WMF server is slow (and prone to edit conflicts) while conversing here.  As I suggested, a conversation between you and I can be reposted here when complete.


 * I assure you, being flexible in the tools you use is not "playing." Since you are content with stopping for now, I'll reply above to those points.  --Connel MacKenzie 16:11, 25 October 2006 (UTC)


 * As you can see from my initial response above, trying to answer 50 questions (because I don't know where the misunderstanding lies) gives the conversation an unnatural, unintended hostile tone, that in a more direct conversation would simply not be present. --Connel MacKenzie 16:27, 25 October 2006 (UTC)


 * I don't interpret it as hostile at all. We are discussing technical questions and processes, right? Not anything personal of course. What is the problem? (yes, someone who is accustomed to making ad hominem arguments themselves might misinterpret, but who cares! My talk page ;-)) Robert Ullmann 18:22, 25 October 2006 (UTC)

zh:forms
I don't do that, because there's already a link to the other form on the page. I can get there through another link. Why link to it more than once? Also, I like seeing the individual characters without having to go through other pages to get to them. Ric | opiaterein 19:42, 25 October 2006 (UTC)

John Harrison Higns
I recently got this on my talk page, and it apparrently as deleted by you. Any idea what's going on? In particular, has JHH now got satisfaction? Please respond. Andrew massyn 17:17, 27 October 2006 (UTC) Hello Andrew Massyn I would just like to inform you that I recently created a Wiktionary account and I noticed that my user page had already been made by a user under the name of John Harrison Highns and it was filled with untrue information. I have still keept it as evidence on my user page, if you want more information have a look at the user page at User: Konstable --Konstable 09:52, 26 October 2006 (UTC)
 * John Harrison Highns
 * The above guy was not me, but now I am me... Confusing I understand. I didn't even know you could change accounts like that.  Here is some confirmation on en: --Konstable 12:48, 28 October 2006 (UTC)

AWB edits marked minor?
Is it possible for you to set your Special:Preferences to mark all AWB edits as minor, by default? --Connel MacKenzie 15:07, 28 October 2006 (UTC)


 * Done. It would be really nice to run under a bot flag, but when I suggested that, the conversation degenerated into anything but whether it should have a bot flag .... Robert Ullmann 15:12, 28 October 2006 (UTC)


 * I am all too-familiar with that phenomenon. Thank you.  (Sorry about the talk page message.  I'll stop now, for a while.)  --Connel MacKenzie 15:17, 28 October 2006 (UTC)


 * That's okay. AWB is good; you change this page and it stops and asks you to look at the message ;-) Right now, I'm unfolding the common meaning into definition lines again. (It is good to collect them and then unwrap them again; the entries use all manner of #, *#, **, and a few others ...) Robert Ullmann 15:20, 28 October 2006 (UTC)

Redirects
Hi, thanks for the note and the welcome. I wouldn't have tried redirecting except that I had run across some (dubious) examples of it such as gugyeol. However, now you've got me interested, and I note that Redirects currently states that: "For languages other than English, especially the more inflected ones, flexibility is retained on this matter, and creating redirects to the basic form of a word is still a better option than keeping a red link." So maybe redirects are OK for a stem form such as 없었, which never appears on its own? I'm not really sure, so I'll follow your advice and avoid creating them in the future. Cheers, -- Visviva 15:27, 29 October 2006 (UTC)

rfd and delete tags.
Thanks for that tip, I did not realise there was a difference, but have found the page listing correct stubs for future use. Thanks again. --Dmol 21:57, 29 October 2006 (UTC)

rs parameter in  ?
I have attached   to 恩 article, but a radical of this kanji seems to be informed uncorrectly. Is it really alright not to inform rs parameter? --Izumi5 14:00, 2 November 2006 (UTC)


 * After my code runs, it should be okay. Check it now Robert Ullmann 14:57, 2 November 2006 (UTC)


 * I got it now, and thanks for posting at my talk page. --Izumi5 14:30, 3 November 2006 (UTC)

Stroke order
Hi, if possible, could you have your AWB distinguish a stroke order image tag in a Han character entry from the text explanation? Either automatically or by hand, marking with a simple, regular comment such as " " will make things much easier for us to find and fix the entry later. Cheers, --Tohru 15:36, 3 November 2006 (UTC)


 * Great. I felt relieved. --Tohru 16:04, 3 November 2006 (UTC)

Usage notes
I see how you did that (on jerk). I wasn't sure if a bullet was the correct method for usage notes which is why I left it as that and changed the order, which wasn't very satisfactory. Moglex 13:28, 4 November 2006 (UTC)

UllmannBot or you?
Weren't you going to use the other account for these edits? --Connel MacKenzie 22:28, 3 November 2006 (UTC)
 * UllmannBot has its bot flag, now. Please use it responsibly and give me a shout if there's anything else. Dvortygirl 19:53, 7 November 2006 (UTC)

Please review my rollbacks
I just rolled these back, as very suspicious looking (and a username containing "suck", therefore blocked.) But my Korean knowledge is non-existant. Still, this looked like really bad edits.

http://en.wiktionary.org/w/index.php?title=%E4%BA%BD&amp;diff=prev&amp;oldid=1750876

http://en.wiktionary.org/w/index.php?title=%E4%B8%B7&amp;diff=1750879&amp;oldid=1750657

Please review and correct as needed. TIA. --Connel MacKenzie 04:57, 6 November 2006 (UTC)


 * these are fine for content, although whether we want Kwukyel as yet another L3 header is another thing ;-) (And the wikipedia reference has to move.) "Dustsucker" is just slang for a vacuum cleaner ... Robert Ullmann 16:13, 6 November 2006 (UTC)

Your posting to my "discussion" page
Hi, I'm not sure I understand your posting to my "discussion" page. I'm simply trying to help refine and make our entries as clear and usable as possible, but you don't seem even to have considered my suggestions. Whatever terminology you'd like to use is fine but the template as it stands now does not provide maximum clarity and usefulness, for the reasons I stated earlier. I wish I could understand the workings of the templates but unfortunately I don't. Badagnani 22:58, 12 November 2006 (UTC)


 * I only found out about your "suggestions" from noticing something on Connel's talk page, where he had reverted your removal of a template. (And I know you know who created the template, and where his talk page is.)


 * I am always open to suggestions, but I will follow the standard style of en.wikt entries: transliterations/romanizations, regardless of language, go on the inflection line. They certainly do not go as bullets or tables in between the inflection line and the definitions, nothing is permitted there. Robert Ullmann 23:23, 12 November 2006 (UTC)

More on Redirects
Thanks for shouting, but what about the lowercase instances where when I put in "gamecube" it doesn't go to "GameCube"?


 * It is the first result from the search box. Anywhere else? (should be linked as gamecube if you want to write it in lc) Robert Ullmann 01:08, 13 November 2006 (UTC)

idle curiosity question
Are you the same Robert Ullmann acknowledged in 2119? (It came up in a discussion on the WikiEN list.) —scs 14:07, 13 November 2006 (UTC)


 * "Brian Ritchie". Heh.  (But are you sure it wasn't Rob Thompson?) —scs 14:24, 13 November 2006 (UTC)


 * I wondered how long it was going to take you to notice that! :-) (Perhaps my "Rob Thompson" jibe was too subtle.) —scs 13:55, 14 November 2006 (UTC)