As an anonymous user, you can only add new data. If you would like to also modify existing data, please create an account and indicate your languages on your user page.

Meta:International Beer Parlour

From OmegaWiki
Jump to: navigation, search


Nuvola apps chat.png
Start a new discussion
Discussion pages
Strategic discussions
International Beer Parlour

- archives


List of Beer Parlours

International Linguists Beer Parlour

- Inflexions


Questions about words
Articles needing attention
Insect room (report bugs)

- archives


Functionality wanted

- archives


WMF[edit]

See Meta:strategy#Joining or not the Wikimedia foundation. What is the status?  Klaas|Z4␟V:  16:53, 2 August 2014 (CEST)

No news from the foundation. And I don't have enough time to do anything myself. --Kip (talk) 18:47, 17 August 2014 (CEST)
The policy of Omegawiki has been to wait for an adoption by Wikimedia foundation up to the beginning of the 22th century. If nothing has been decided by then by the foundation, then we will think of founding our own organisation and of our funding. Last Monday, we celebrated the 8th anniversary of this "wait and see nothing" policy. --Fiable.biz (talk) 04:41, 17 September 2014 (CEST)
I do not believe that hanging on for the WMF is going to come to any useful result for the funding of OmegaWiki. There are enough people who believe in the possibility and usefulness of the plan to to integrate some or all the information we are collecting here into Wikidata in the foreseeable future. --Purodha Blissenbach (talk) 08:45, 21 September 2014 (CEST)
WikiData is a WMF-project so if this going to happen the policy of 'wait in vain' comes to a happy end. The challenge is IMNSHO to 'beat' the Wiktionary projects or better cooperate with them. At least we support almost three times as many languages.  Klaas|Z4␟V:  09:10, 9 November 2014 (CET)

no english definition for bear cub[edit]

http://www.omegawiki.org/Expression:bear_cub --Bauer (talk) 21:21, 1 November 2014 (CET)

Now yes; kudos to Robykiwi thanks for the notice  Klaas|Z4␟V:  08:50, 3 November 2014 (CET)
Done Done --Purodha Blissenbach (talk) 23:28, 9 January 2016 (CET)

Questionable pages[edit]

After cleaning some 200 SPAM pages from OmegaWiki, there are few pages left which I cannot understand well enough so as to decide about them:

Please delete the SPAM if you can identify it, or let us know here, when a page has been identified as valid. Thank you. --Purodha Blissenbach (talk) 10:50, 12 November 2014 (CET)

Deletion done though I don't understand the first, it contained a suspicious linkSpam to a site registered in Italy - perhaps we should prevent updating outside name-spaces like expression:, meta: etc. or restrict to admins, bureaucrats. Deleting or updating userPages I don't see as my task being not a bureaucrat (yet?) Kind regards,  Klaas|Z4␟V:  11:31, 15 November 2014 (CET)
Done Done --Purodha Blissenbach (talk) 23:28, 9 January 2016 (CET)

Darai Language - cleanup needed.[edit]

There are few pages related to the Darai Language that need content, better linking, or other cleanup, such as being moved to the Darai Language portal. I have coellected them in the temporary Category:Darai Language so as to ease finding them. Since I am unabe to read the script, and I do not know Darai sufficiently, I cannot do it, but possibly be of help. Thank you. --Purodha Blissenbach (talk) 11:10, 12 November 2014 (CET)

Spam and account creation[edit]

We currently have (had) a lot of spam. Most of them come from online servers that are probably infested with bots that can create accounts, or people who are paid for this.

I have tried a few solutions, but the spammers are able to solve recaptcha (actually, it is the one with which I had the most spammers), and mathematical equations, or other simple anti-spam tests. The usernames are somehow similar, but there is a large variety of different IPs behind. Account creation is currently limited to 1 per IP per 24 hours. At the same time I don't want to have a too complicated test that will bother the normal users.

I am now trying the extension ConfirmAccounts [1]. With this, a user who wants an account will have to fill in a short form (you can try it, log out and click create account). It is then up to bureaucrats to approve the account creation. Normally, I should receive an e-mail notification when an account request is made. At the same time I can give the wikidata-omega user rights. I would like to try this for a few time. If it does not work (too much burden for new users?), then I'll revert to the previous state. Up to now, no request has be received in the last 24 hours (except my own test).

And... I have deleted accounts with no edits, to have a more readable recentchanges page. --Kip (talk) 19:11, 25 November 2014 (CET)

The latter may be not a good idea™ since read-only-users with stored preferences are affected, too. --Purodha Blissenbach (talk) 11:14, 28 November 2014 (CET)
I thought about that, but there was no perfect solution. I could only configure that people with edit rights keep their accounts. Tant pis. --Kip (talk) 18:09, 29 November 2014 (CET)
subotimal, but imho fairly good. --Purodha Blissenbach (talk) 15:18, 30 November 2014 (CET)
Done Done --Purodha Blissenbach (talk) 23:28, 9 January 2016 (CET)

Safety pin[edit]

Can someone create a page for "safety pin" with an English definition. Thanks --Bauer (talk) 23:52, 4 January 2015 (CET)

Of course: you can. --Fiable.biz (talk) 03:13, 5 January 2015 (CET)
I will make a lot of mistakes, can't write perfect English.--Bauer (talk) 05:31, 5 January 2015 (CET)
It seems your English is good. You can also use Wiktionary, Wikipedia, Citizendium or any free source: there may be definitions there. But, if you prefer, you can create the expression "safety pin" with a definition in another language you know well, then you can give translations, and translations of the definitions. It's good if you can provide a definition in an international language (in your case, in Russian). Then, you go on the Discussion page of your new DefinedMeaning, and you add a tag like {{Translation needed|rus|English}}, meaning a translation of the definition from Russian to English is needed. The 1st argument "rus" is the ISO 639 code of the language you've written a definition in, while the second argument, "English", is the language into which to translate. Then you click on "Save". And one day, someone will, hopefully, translate your definition. --Fiable.biz (talk) 12:59, 5 January 2015 (CET)
Done - Expression:safety pin :) --Kip (talk) 19:06, 5 January 2015 (CET)
Thanks --Bauer (talk) 02:51, 6 January 2015 (CET)
Done Done --Purodha Blissenbach (talk) 23:28, 9 January 2016 (CET)

One DefinedMeaning per connotation, please[edit]

Please create a new defined meanings (DM) for each connotation. For instance, there are many synonyms of "to die", but "to die" is the neutral word, while "to perish", "to pass", "to expire" etc. carry connotations, which deserve their own definition, and have equivalent in many languages. Presently, there are 23 English synonyms in DefinedMeaning:die_(6245), and, on the total, already 256 expressions for this DM, so it is quite difficult to find the equivalent in any language. It would be much more convenient to have, for example, a different DM for "to perish" (English), "périr" (French), "perecer" (Castillan), "мөхөх" (Mongolian) etc.. I suggest to add a usage "neutral" to explicitly state that an expression carries no special connotation, and is the default word. In "Semantic annotations" → "Relations", I would add "loose synonym", and add a small piece of code to create automatically such relations by transitivity. For instance, if "to perish" is synonym of "to die" which is synonym of "to snuff it", then the system will create the synonymity relation between "to perish" and "to snuff it". The automatically created relation should be tagged "unchecked". Maybe, if the class of synonyms includes more than 10 words in any language, a human check could be required, to avoid merging unrelated classes by mistake.

Amendments to the Policy to create defined meanings (DMs)[edit]

We need a more precise policy to create defined meanings (DMs). A few examples:

  • should "king" and "reigning queen" be one or two DMs?
  • In Malagasy, the participle "vonòina" ("killed") is not derived from the active verb 'mamòno" ("to kill") and is listed as a different word by the standard reference dictionary of Abinal and Malzac, though the 2 words or forms derive from a common root. Do we need 1 or 2 DMs for this?
  • In English, there are nearly no declensions, so "whose" is considered a different word from "who". In Latin, the equivalent words are just the genitive and nominative of one word. Do we need 1 or 2 DMs in such a case?

Such problems and other ones are going to be set in the "Policy to create defined meanings" to be published officially. It has been proposed for nearly 2 years, but few people voted. All the proposed modifications without substantial opposition will be considered as accepted. You still have a few days to vote. Remember that OmegaWiki is not only for Indo-European languages, please think then vote to the different proposals there: Policy to create separate definedMeanings or not. It is a good idea to read the discussions of that page before voting. OK: it's a bit long, but it's important for a dictionary, isn't it? --Fiable.biz (talk) 03:33, 31 January 2015 (CET)

For me, it is often very hard to deal with this kind of questions as long as some other questions have not been thoroughly discussed and dealt with. Not all are related, but here are my open questions:
  • How to deal with spelling variants, how to properly annotate them, how can we have expressions share their properties beyond spelling without having us to re-enter them for each spelling variant?
  • How to deal with grammatical forms? When English who and whose have relations is genitive of or is nominative of respectively, we need to be able to enter them so as to have useful entries for either lexeme. But currently, we cannot.
  • How to deal with lexemes and expressions the semantics of which are pretty much empty? The Toki Pona expression li is purely grammatical. It is used as a separator inside Toki Pona sentences, but does not convey a meaning on its own. It is not translatable. We can explain it, but there is no Defined Meaning proper since it is lacking a meaning.
  • Definitions are at tims of extremely different precisions, elaboratedness, complexity, or detail, and may or may not belong to specific expert domains. I believe that there is no choice but to have them all, depending on expressions, but it is hard to put them under a simple general rule. A possible solution could be to have language independent domain markers which would have to include a general common speech type of domain, and language dependent register of speech markers, maybe.
--Purodha Blissenbach (talk) 13:14, 31 January 2015 (CET)
  • Spelling variants: for the moment, I suggest we give them up or add them as expressions, but I agree there should be a special treatment. I don't think it should influence the proposed policy: spelling variants are clearly not DMs.
  • Grammatical forms: see International_Linguists_Beer_Parlour/Inflexions. We agree we should do something (And this is, for me, the worse limitation of OmegaWiki.), but so far nothing has been done. This is linked with the lack of will to create a legal entity to govern Omegawiki, get money, hire a programmer etc.. In fact, we will need a rule, but the proposed policy doesn't deal with this question.
  • I don't believe words lack meaning. For instance, in Mongolian, there is a word "нь" whose one meaning is just: "Here stops the subject.", used when there could be a confusion. One could say it carries no meaning by itself, and is just a grammatical marker. But the definition could just be "(Here stops the grammatical subject.)". In Malagasy, there is a word "ve ?" whose definition could be: "(This is the end of a question.)". In English, there is a word "a", not translated in many languages, whose definition could be "(What follows is singular and doesn't refer to any previously mentioned being.)". This word has no plural in most languages, but it has one in French: "des", whose definition could be: "(What follows is plural and doesn't refer to previously mentioned beings.)". Etc.. Note that I include parenthesis in my definitions.
  • There are very precise notions, like "number π", and very loose ones, like "art". So it's normal that some definitions are precise, and some imprecise. Moreover, there are many not so good definitions in Omegawiki, and I think we should work on this. Cannot the "Class membership" field plays the role of your "language independent domain markers"? For instance, there is "science", in it. However, I don't think we need special rules according to domains. But you can propose special rules on the page Policy to create separate definedMeanings or not.
So I do think you should vote on the proposed rules, though they don't pretend to solve all problems. --Fiable.biz (talk) 15:00, 31 January 2015 (CET)
I did already voice my opinion where I could on that page. There are some cases which I am indecisive about. Other cases can imho only be discussed and decided upon once we have clearer ideas about treating derivations, inflections, and so on. --Purodha Blissenbach (talk) 19:28, 1 February 2015 (CET)

I agree with Purodha. All cases differ. King and queen are not equal nor an emperor and empress etc. Dutch example: secretaris and 'secretaresse both translate to secretary, but are different professions in most contexts. In Italian a female minister (politics) is sometimes called ministro, others use in this case ministra. My €0.02  Klaas|Z4␟V:  15:28, 16 February 2015 (CET)

Languages WMF supports and we have or don't have?[edit]

Example: ace. Is there a list where we can find a complete overview of missing languages here? Of course there are a lot more we support and WMF not (yet?) One or more of us going to Wikimania in Mexico to talk about this?  Klaas|Z4␟V:  11:19, 17 February 2015 (CET)

  • the up to date list of all languages that we have is at Editable languages (automatically generated).
  • there is also a list of all languages that exist on the planet: Help:Language/Checklist for editable languages (some missing probably), with indications of which ones we have, or have not, and why. This is manually maintained, and therefore not up to date.
  • and we have "ace". The language is called Expression:Aceh :) --Kip (talk) 19:33, 18 February 2015 (CET)

Thank you. I added the word in Aceh for goat, because today the Chinese year of the goat started :-)  Klaas|Z4␟V:  15:46, 19 February 2015 (CET)

There are indeed languages missing from Help:Language/Checklist for editable languages, at least several dozens that I know of. The list is likely based on SIL which lists an estimated 10% of the current world languages, imho. --Purodha Blissenbach (talk) 23:32, 2 March 2015 (CET)
yes it was generated automatically from SIL. And the idea is to add missing languages manually when necessary. The SIL list is only there as a starting point :) --Kip (talk) 18:59, 4 March 2015 (CET)
can we automatically update the list, when SIL alters its data? --Purodha Blissenbach (talk) 12:07, 5 March 2015 (CET)

Wikimania 2015[edit]

Perhaps we can participate @ Babel workshop if one or two of us are present in Mexico?  Klaas|Z4␟V:  11:02, 14 March 2015 (CET)

Most likely, I shall not be at the conference. --Purodha Blissenbach (talk) 14:35, 19 March 2015 (CET)
Done Done --Purodha Blissenbach (talk) 23:28, 9 January 2016 (CET)

Problems[edit]

See Meta:Insect_room#Cannot_edit_page for a recent problem, and Meta:Functionality_wanted#Palatinean_language, Meta:Functionality_wanted#Allow_the_Central_American_Spanish_language, Meta:Functionality_wanted#Additional_language_wanted, for a set of likely easy but unsolved requests. Also cannot use Rhinelandic to add translations to, and various other languages that I asked to have added, but I cannot find my requests any more. As per my memory, they where on Meta:Functionality_wanted. --Purodha Blissenbach (talk) 22:16, 1 August 2015 (CEST)

Not sure what to give to Rhinelandic as a ISO ID. proposals are "rheinisch" or "deu-rhine". I didn't find it in linguisticlist. --Kip (talk) 20:47, 17 December 2015 (CET)
Moved to International_Beer_Parlour/Archive20150131#Languages_wanted_for_editing. --Purodha Blissenbach (talk) 21:07, 17 December 2015 (CET)
So, done OK here. --Purodha Blissenbach (talk) 21:07, 17 December 2015 (CET)

Empty pages[edit]

Example: just a title: looks Arab (or Farsi?) to me. Since I don't speak nor read a syllable of those kinds of tongues, I don't know what to do with those.  Klaas `Z4␟` V:  16:31, 23 August 2015 (CEST)

Various bugs?[edit]

You might have noticed several bugs recently? I just discovered accidentally that the disk was full on the server (because of one big log file). Should be better now. --Kip (talk) 21:28, 11 September 2015 (CEST)

Done Done --Purodha Blissenbach (talk) 23:28, 9 January 2016 (CET)

International TranslationDay[edit]

I just read that today 30 September is TranslationDay and celebrated in the entire world. Congrats y'all!  Klaas `Z4␟` V:  14:59, 30 September 2015 (CEST)

Word class in Russia Buryat and in Russian[edit]

From Irkutsk. Could you please add the field "world class" to the language bxr, with nouns, adjectives, pronouns, numerals and verbs? The list is probably not exhaustive. A Buryat just said me their grammar follows Russian grammar, and I believe it, because all educated Russian Buryats speak Russian, there is no school in Buryat, so their grammar classes are from Russian. In the Russian language, the "word class" field includes only adjective, noun and verb. Please complete with numeral and pronoun, according to Wikipedia (I don't know Russian, by the way: I just notice that OmegaWiki's list is even shorter than the one in Wikipedia: https://en.wikipedia.org/wiki/Russian_grammar ). Thank you in advance.--Fiable.biz (talk) 16:32, 9 October 2015 (CEST)

Just added pronoun and numerals to Russian, and the five word classes to Buryat. Also genders. --Kip (talk) 17:52, 13 October 2015 (CEST)
Thank you! --Fiable.biz (talk) 05:45, 14 October 2015 (CEST)

Please take time to fill the field "Part of speech"[edit]

To everybody: please take the few seconds needed to fill the "part of speech" (or "word class") field: it is used to sort the results of any query, and is also useful for automatic translation. I'm not sure all of us realise that some use our work to make opensource machine translations, specially Apertium. See http://wiki.apertium.org/wiki/Getting_bilingual_dictionaries_from_OmegaWiki --Fiable.biz (talk) 16:32, 9 October 2015 (CEST)

I do, too. But
  • given the speed at which I can type and add new words, with the current excessively time consuming interface, I cannot expect any useful machine translation within my expected life time. Adding a word class to an expression takes over a minute per expression.
  • We cannot add missing word classes. Else please point to a description how it can be done.
  • With the lack of a concept to include derived forms (Conjugation, Declination, etc.), we shall likely not be able to supply valid machine translation data for all the languages having such forms.
--Purodha Blissenbach (talk) 11:56, 10 October 2015 (CEST)
Btw., the apertium wiki page you mentioned has typos, and all dictionary file links go 404. Bout one cannot correct them, write to the talk page, log in, create an account, address an administrator, or whatever because all is forbidden. A castle perfectly shielded against communication. --Purodha Blissenbach (talk) 12:15, 10 October 2015 (CEST)
I must admit you are right concerning both the server slowness and the project slowness. To get a better server, we would need money. To get money, we would need an organization. I even proposed statutes for such an association, but the answer of the majority of those who answered was, in 2013: we don't want this, we have been waiting for the WikiMedia foundation to swallow OmegaWiki since 2006 and we will go on waiting, though the WikiMedia foundation never accepted this idea. If enough people have now another idea, we can found an association... Apertium already proposes useful machine translations. Some of them used Omegawiki, but of course inflexions are needed. We could also solve this problem if we had money to hire a programmer...
Regarding the missing word classes: on any expression page or definedMeaning page, click on the link "show" at the right of the expression you want to give the word class. Then click on "Edit", then on "Option values"→ Property→"Part of Speech" and then choose the part of speech in the rolling list, and don't forget to click on "Save" at the bottom of the page. You don't even need to log in for this. --Fiable.biz (talk) 13:18, 12 October 2015 (CEST)

 

@Fiable.biz: you misunderstood my question.
How do we add to the rolling list?
How do we sort the rolling list, so it becomes a bit more useful?
How do we get the many missing languages for which we could add translations?
I am a programmer, and I have given up years ago to work on Omegawikis software after having spent my holidays on futile attempts to even install an instance for testing. I would work (slowly) improving Omegawiki, if I had a chance. This is not a questiion of money, it is a question of noone interested, willing, capable, or knowledgeable enough to give little hints, or do (presumably) easy things.
I am really pissed off. Sorry. --Purodha Blissenbach (talk) 14:46, 12 October 2015 (CEST)
(1 reply and further discussion moved below) --Purodha Blissenbach (talk) 08:57, 14 October 2015 (CEST)
If someone had time to program a small bot, it should be possible, with the omegawiki API, to create a bot that finds part of speeches that are already available in OW and give them automatically to the synonyms and translations of that word. This should already reduce the task that has to be done manually. For operating a bot you don't need to install an instance of omegawiki, but you need some knowledge of perl or python, and a bit of sql too. --Kip (talk) 17:57, 13 October 2015 (CEST)
You cannot do that automatically, you must manually approve or decline each instance. --Purodha Blissenbach (talk) 18:11, 13 October 2015 (CEST)
That means, your bot must:
  1. ask for a language,
  2. find all available part of speech values for that language,
  3. find expressions of the language that do no have "part of speech" specified,
  4. for each expression,
  5. find the definitions of the expression in the language,
  6. determine by some language dependent heuristics, what part of speech this word has,
    • maybe find translations of that word having a part of speech set, and use the most often set value as a hint,
  7. display the expression, its definitions, the list of all part of speech values for the language along with radio buttons with the heuristically determined one checked as default,
  8. get input from the user (including skip and stop)
  9. alter the part of speech setting,
  10. go to to next expression, if not stopped.
For several steps, I do not know how a bot could accomplish that. --Purodha Blissenbach (talk) 18:30, 13 October 2015 (CEST)
I think for other things too it would be a good idea to introduce the notion of proposal, to be checked. These proposals could be from a bot, but also for a person working with a language which is not his mother tongue. For instance, I would prefer that the definition I write in Mongolian be proposals and only those I could have checked by a Mongolian be considered as trustable. --Fiable.biz (talk) 05:31, 14 October 2015 (CEST)
  • Agreed. I would like it often, too, but we cannot make it happen atm. We can, however, make a wishlist, or file phabricator tasks, saving our ideas for future treatment. --Purodha Blissenbach (talk) 08:57, 14 October 2015 (CEST)

Hacking OmegaWiki[edit]

I tried once more to get a testbed installation of OmegaWiki. If you like to try it:
https://tools.wmflabs.org/owintes/ow3/index.php?title=Extra%3AAll+Sigge&from=&to=&namespace=16

I took a fresh MediaWiki from git, plus composer, plus the stuff that composer automatically installs with MediaWiki, plus a few skins, plus the extensions Polyglot and Wiki Lexical data. There were several obstacles that should need analysis and treatment. My fixes were only meant to get it up and running fast, they are not of general use, I think. By the way, although I did check everything out from the gerrit repository with the idea to be able to upload patches, gerrit does not allow to upload patches, and installing "git review" in the repositories did not work either (it stalls and echos any input once I typed the key passphrase.

Here are the highlights:

  • MediaWikis install script did most of its job but did not give the initial user any rights, nor did it create a Main Page. Instead it made the web server emit "Invalid request".
    • Remedy: I ignored that.
      • I manually added the user to all four existing groups (as per Special:ListGroupRights) via a database command.
  • WikiLexicalDatas install script choked in ./OmegaWiki/WikiDataAPI.php barking at a missing class.
    • Remedy: I found the file where the class is defined and added require_once-d it in ./OmegaWiki/WikiDataAPI.php and rerun the install script successfully.
      • Now the Website did not run any more complaining that it could not find the file just require_once-d.
        • I commented the require_once out.
  • Trying to access one of the Expression pages that were created by the install script, or trying to open a non-existing expression page, I get a lot of error messages and traces piled upon one another that are thus unreadable in my browser. The topmost error complains about a missing index "uw" somewhere. "uw" is nowhere in the configuration.
    • Remedy: after finding out that there was only one entry in the array in question and its key was "ow3" for whatever reason, there was a section of globals defined in App.php not documented in the page https://www.mediawiki.org/wiki/Extension:WikiLexicalData but setting various values to "uw", I copied the section to Localsettings.php, altered all occurrences "uw" to "ow3".
  • Not permitted to edit an expression, though user is in all groups shown on Special:ListGroupRights. Error message says: You are not permitted to edit pages in the dataset "". Please see our editing policy.
    • Remedy: Alter name of right 'editwikidata-uw' to 'editwikidata-ow3' everywhere (but App.php where defaults are set)
  • the [save] button does nothing for new/edited expressions/spellings/meanings, etc..
  • Importing the database dump of OmegaWiki takes long time but has no effect on the actual database.

Future of OmegaWiki[edit]

(continued from above)

Ah! For modifying the list, OmegaWiki has a semi-automatic auxiliary server called "Kip": you write your query (not even in SQL format: human language is accepted) on the international beer parlour and wait a few days: that's done. :-) It is of course not enthusiastic to work for a project waiting to disappear. It would be much more interesting to go ahead and define other goals than waiting the Wikimedia foundation pillage Omegawiki. Apertium has sponsors: they didn't wait for authorization from the WikiMedia foundation for this. --Fiable.biz (talk) 05:31, 14 October 2015 (CEST)
  • I could do so myself years ago. That's why Colognian has parts of speech. The functionality was removed. If I cannot do it myself, it is not going to happen :-( see dozens of request to enable languages waiting for years now. --Purodha Blissenbach (talk) 08:57, 14 October 2015 (CEST)
Did I understand Fiable well that it's better to join a company than a non-profit organization with thousands of hard working editors who actually created us? What would be nicer than going back to our roots before the end of our existence? I don't believe it will happen. We get more followers on our @OmegaWiki twitter-account. Even favorited, listed and retweeted.  Klaas `Z4␟` V:  09:11, 18 October 2015 (CEST)


Given the fact that OmegaWiki currently runs on MediaWiki Version 1.22, which is close to becoming outdated, and having had a look into the code base of the extension WikiLexicalData which adds most of the functonality of OmegaWiki to MediaWiki, I would like to refactor the codes base. Goals:

  • Better separate the more general structured database part (currently named LexicalData) from the applicatioin part (currently name OW or Omegawiki*) with the perspective to develop then into distinct MediaWiki Extension.
  • Make the directory structure clearer, easier to understand, and conformant with new capabilites of MediaWiki.
  • Unify some things which are currently unnecessarily different for legacy reasons.
  • Get rid of some deficiencies and drawbacks, such as language manager activities not listed in the recent changes, etc.
  • Likely more, popping up along the path.

Comments? --Purodha Blissenbach (talk) 17:31, 18 October 2015 (CEST)

In case you are interested, depending on when it is updated, we have a Doxygen generated documentation at http://www.omegawiki.org/downloads/manual/index.html -- 向榮 /Hiong3-eng5/ (talk) 04:16, 27 October 2015 (CET)
Actually, from my pov, if we seperate WLD with OW, WLD would be the base programs plus language class. Expressions and DM are obviously OW. Transactions I don't know. But currently, features are needed most. --向榮 /Hiong3-eng5/ (talk) 05:04, 6 November 2015 (CET)
http://www.omegawiki.org/downloads/manual/md__omega_wiki.html and http://www.omegawiki.org/downloads/manual/md__wiki_lexical_data.html are my layout of the current state of our software, that is, as of last year. I don't understand it all yet, but it may help you see some structure. Hope this helps. --向榮 /Hiong3-eng5/ (talk) 05:12, 6 November 2015 (CET)

trazioni alla sbarra[edit]

what language is trazioni alla sbarra?

Under DM:pull_up_(1566898) it is English. - 向榮 /Hiong3-eng5/ (talk) 04:02, 22 October 2015 (CEST)

Seems to come from a Romance language, but I cannot tell which. --Purodha Blissenbach (talk) 09:09, 22 October 2015 (CEST)
Italian -- 13:57, 30 October 2015 Patio (Talk | Contributions)
Done Done} --Purodha Blissenbach (talk) 11:26, 9 January 2016 (CET)

MediaWiki software update?[edit]

Kip, can I request an upgrade to the latest MediaWiki software. I found the hook for recent changes, two of them actually, one deprecated and another one the latest, if I use the original hook, it will not work later, when the hook is removed. If I use the latest one, it won't run with the current MediaWiki software. --向榮 /Hiong3-eng5/ (talk) 06:42, 6 November 2015 (CET)

Hmm, the hooks I found are not what I thought they were. Never mind. --向榮 /Hiong3-eng5/ (talk) 11:06, 6 November 2015 (CET)

Ok. I can still do a MediaWiki update if you like. What version are you using? --Kip (talk) 19:46, 6 November 2015 (CET)
Actually, I am using an older version right now :) I'll update my version and check the code first if it breaks down or something. Will keep you posted. --向榮 /Hiong3-eng5/ (talk) 01:46, 7 November 2015 (CET)

Anniversary[edit]

I see oldest update of Main Page was 10 years ago next week. Will we celebrate like WMF does with 15? If so when and how?  Klaas `Z4␟` V:  17:25, 20 December 2015 (CET)

I had a simiar idea but no idea how and what to do. --Purodha Blissenbach (talk) 18:10, 20 December 2015 (CET)
A certain GZWDer @ WikiData had put the inception of OmegaWiki at 27 December 2005. 向榮 /Hiong3-eng5/ (talk) 09:46, 5 January 2016 (CET)

How to define language dependent grammatical annotations?[edit]

Assuming, I have defined something like part of speech for one of the newly accessible languages (Yes, currently, it does not work), then I would e.g. like to define for that language, that once a word is set to noun, another selection such as grammatical gender can be choosen for it as well. How do I do that? --Purodha Blissenbach (talk) 19:00, 25 December 2015 (CET)

You can take example on DefinedMeaning:German_(5688)
  • Add to the language the class "language"
  • Then you can add class attributes. For languages, it is always at level Syntrans, and usually a list of options.
But seeing your other bug report, you already did this?
We don't have as of yet a way of telling that "grammatical gender" is available only if part of speech is set to "noun". --Kip (talk) 23:47, 14 January 2016 (CET)

How to create collections?[edit]

In the context with the newly available languages (thank you, Kipcool!), I shall upload several word lists. These are from word books and so as to properly credit the word book authors who gave the permission to upload, and so as to make clear where the data came from, I need to create collections for them, and put the words into them with references to the pages or columns where they are to be found. Yet I am clueless as to how I could make something selectable as a collection (or class, for that sake) once there is a DefinedMeaning for it. Btw., there are other kinds of collections needed as well. --Purodha Blissenbach (talk) 19:20, 25 December 2015 (CET)

There is a special page for making collections. May I have access to it, please? --Purodha Blissenbach (talk) 22:09, 27 December 2015 (CET)
Sorry I had not read this question earlier.
Creating a collection is fast (using the special page as you said), but not so easy, because you need to know the database structure and understand what the "collection type" are. I mean it is ok for you probably, but I don't want to make it accessible to any sysop.
So probably the best would be to make you bureaucrat. An active bureaucrat would be nice, and you wouldn't have to wait for me to fulfill your requests (since I am not too much active). If it is ok with the others.
On the other hand, I think anyone can create a class, this is done by putting a DM into the collection called Community Database class --Kip (talk) 23:39, 14 January 2016 (CET)

Language needed: German as used in Germany.[edit]

We need the language German as used in Germany, ISO code de-DE.

There is at least one word, Groschen, which is used in Germany meaning 10-Pfennig-Stück or 10 Pfennig, while in Austria, it means a tenth of an Austrian shilling. Astonishingly, I found Deutsch (Österreich) and Deutsch (Schweiz) but not Deutsch (Deutschland) in the list of available languages. --Purodha Blissenbach (talk) 11:28, 29 December 2015 (CET)

You're probably right, but I think they mean with German what you call German as spoken in Germany. Dutch has the same "problem". It's spoken in Netherlands (12 provinces), some of the provinces of Belgium, a tiny piece of France, Suriname, the Antilles and very old people in Indonesia. Examples of English (England, USA, Canada Australia, India etc.), French (CA, CH etc.) and peobably many other.  Klaas `Z4␟` V:  12:08, 29 December 2015 (CET)
For French, we have:
and for English, we have British and American English plus English. That is how it should be. Have the general language and have (small) deviating subsets for certain areas or other types of variants. That is as it should be for German as well. The Dutch of Belgium is named Flemish as far as I know. --Purodha Blissenbach (talk) 12:51, 29 December 2015 (CET)
Btw., as I understand it, if some expressions is not listed in, say, en-GB, you would look for it in en if you are writing British. Leave aside that the current coverage of en-GB may be incomplete, we have no way to mark that an expression in en is simply not used in en-GB. --Purodha Blissenbach (talk) 12:51, 29 December 2015 (CET)
I fond out that the Dutch of Belgium (nl-BE) and (West) Flemish (vls) are somewhat similar but not the same. --Purodha Blissenbach (talk) 13:35, 5 January 2016 (CET)
I am not too much fond of having these many languages for regional variety, because it is basically the same language. We have already the option to add the region (or country) where a word is used as an annotation (annotation "area" in English).
The advantage I see with an annotation is:
  • if a word is used in e.g. German of Austria and German of Switzerland, just enter the word once and put the two regions in the annotations (instead of creating the two words as synonyms. Could be more than two...)
  • if a word is used only in part of a country (not sure if I can give example in German because there are also local dialects which are really different languages, but in French, some words are used only in Nord-Pas-de-Calais, or near Lyon, or in Bretagne) then the annotation system is more flexible than defining a separate language for each region. For English there are also a few words that are used only in e.g. Scotland (not talking about Scottish language here), or in some parts of Australia, etc. I don't consider that English (Australia) and English (UK) should be treated differently in the database than English (north of Australia) and English (Scotland).
  • and also there is the problem that the omegawiki software does not know the hierarchy of languages and will show German and German (Germany) in two separate tabs, but that issue could be fixed with some coding. --Kip (talk) 19:18, 5 January 2016 (CET)
I basically agree with you. But currently, I think, being consistent is more important than doing something which is only partially working or insufficiently supported. If need be, that can be changed, and I believe it will. Since that has to be done programmatically anyways, it does not really matter how many languages/regions are affected. Thank you for your consideration. --Purodha Blissenbach (talk) 22:09, 5 January 2016 (CET)
So I added "Deutsch (Deutschland)", but the idea was rather to delete the other ones and replace them by an annotation (for consistency) ;) --Kip (talk) 19:06, 8 January 2016 (CET)
Done Done Thank you! --Purodha Blissenbach (talk) 09:40, 9 January 2016 (CET)

Language requests[edit]

I would like to have these languages added:

code name DefinedMeaning
nds-hoetter Hötter Platt (1604293) Done Done
pfl-lowrhl Pälzersch (1604286) Done Done

--Purodha Blissenbach (talk) 23:20, 9 January 2016 (CET)

Done --Kip (talk) 23:25, 14 January 2016 (CET)


code name DefinedMeaning
lim-gemmer Jömelejer Platt (1605090)
deu-kerkra Kirchröadsj Plat (1605114)
lim-krefel Krieewelsch Platt (1604286)

--Purodha Blissenbach (talk) 17:01, 25 January 2016 (CET)

Collection requests[edit]

I would like to have these collections added:

name DefinedMeaning
iana language subtag (1604302) 15px incorrect DM (1604382)
ISO 639-2/B codes (1601607)
ISO 639-2/T codes (1601611)

--Purodha Blissenbach (talk) 00:28, 10 January 2016 (CET)

Done --Kip (talk) 23:28, 14 January 2016 (CET)
It does not ask for a DM when creating the collection. --Kip (talk) 19:03, 15 January 2016 (CET)
  • Collection name:
  • Language of name:
  • Collection type:
  • Dataset:
Is there a selection list once the name is entered? If so, this is the place to select the correct DM. Sigh. --Purodha Blissenbach (talk) 23:08, 15 January 2016 (CET)
No there is no selection list. I looked in the code, it takes an expression and a language id, and takes the first DM id that corresponds to it.
I will have to change it manually in the database then. Would be nice to replace the current Special page for adding collection with a dm. --Kip (talk) 23:37, 26 January 2016 (CET)
Recorded at the insectroom. I shall look into it as soon as possible. --Purodha Blissenbach (talk) 16:41, 27 January 2016 (CET)
I changed the DM id in the database, should be fine now (please check I don't know of a DM in that collection). --Kip (talk) 18:44, 28 January 2016 (CET)
For the time being, the only solution I have is to create an expression, DM, and collection in one go by entering a name that certainly does not exist yet. Both expression and definition need to be changed, then, thereafter. For the reason, see the insectroom. --Purodha Blissenbach (talk) 21:16, 30 January 2016 (CET)

Presentation of OmegaWiki[edit]

A presentation of Omegawiki is scheduled for monday January 18th, 2016, at the University of Cologne, Germany, in a seminar "Applied Computer Linguistics" of Prof. Dr. Jürgen Rolshoven. If you want to attend it, you can. Come to the gray building called "Philosophikum" vis-a-vis the main university building, room PC72 on the ground floor at 14:00 o'clock sharp, and kindly ask Prof. Rolshoven to allow you in as a "Hospitant". There will be two presentations held in a row in almost two hours. They are usually in German, but I can switch to English on demand. Slides and brief English annotations are available at http://ald15.purodha.net/. --Purodha Blissenbach (talk) 15:13, 13 January 2016 (CET)

How was it? Did you speak about of our (and other linguistic sites') presence in social networks as well?  Klaas `Z4␟` V:  09:33, 20 January 2016 (CET)
No - first, I am not a user of such networks and I know nothing in that field which I could convey, second, time was limited, and I had to concentrate on the more technical and linguistic subject matters. There will likely be a discussion on some aspects of the things I told on February 15th, same place, same time. If I have a chance, I can drop something on it, if I know what to say. --Purodha Blissenbach (talk) 16:25, 25 January 2016 (CET)
Just tell the interested audience there exist blogs and an @OmegaWiki twitter-account. More info: Sabine and/or me. The former is rarely busy with it anymore. Anyone else cares to help BTW?  Klaas `Z4␟` V:  13:06, 9 February 2016 (CET)

Writing an open word translation client?[edit]

I'm a programmer interested in participating in the creation of a translation dictionary client for OmegaWiki. Plug in a "From" language and a "To" language, search for words, get results. Are there ongoing projects like this? Would it be worth it? --Bistenes

In my opinion, yes. I know of ideas in this field, but not of any programming already having begun. I think, this would be a typical application using the API, but off the top of my head cannot tell you precisely, if that is possible already, and exactly how. Another possible approach might be to use the suggest interface which can at least get you lists of something from OmegaWiki, but off the top of my head I do not know where it is documented. Do not hesitate to ask, I can help you with many things, including to add exactly the kind of API function you need. Thank you a lot for your interest! --Purodha Blissenbach (talk) 16:51, 25 January 2016 (CET)
Our data are accessible with DictionaryForMIDs http://dictionarymid.sourceforge.net/ - they are actually not using the api, but downloading monolingual and bilingual dictionary files that we create on demand (any user can create).
I don't think there is anything else, but yes it would be worth it.
So you have the choice between using the api, and using offline data (monolingual and bilingual as simple CSV text files)
An help page about the API is Help:OmegaWiki API. The help that is accessible via http://www.omegawiki.org/api.php is normally up to date, but less user friendly (and other stuff that are general for mediawiki are mixed in there as well, search for "action=ow")
Please contact User:Hiong3.eng5 (preferably by e-mail), he is the one who did most of the API, and who worked with dictionarymid, and who would have maybe started something else in that direction. --Kip (talk) 23:49, 26 January 2016 (CET)
Hi Bistenes, I am Hiong3.eng5, just tell us what you need, and we will try our best to improve the API to suit your needs. 向榮 /Hiong3-eng5/ (talk) 05:40, 29 January 2016 (CET)

Server time[edit]

The server time of OmegaWiki seems to be ten minutes in the future. --Purodha Blissenbach (talk) 13:05, 22 February 2016 (CET)

18:20, 6 March 2016 (CET) test Yep, like the clock in some bars for people who have to catch a bus or train and leave the last moment which is usually too late :-) 18:23, 6 March 2016 (CET)
I know one of those ː-) --Purodha Blissenbach (talk) 20:32, 6 March 2016 (CET)
I just changed some things in the system, so that hopefully the time is now corrected regularly based on some central server. At least the current time is correct now. --Kip (talk) 18:35, 5 July 2016 (CEST)

Links to WMF-projects[edit]

Sometimes we link to Wikipedia, but I see Wikidata as well. Perhaps even other projects of Wikimedia Foundation? Is this a preferred behavior to keep or is it better to choose for only one?  Klaas `Z4␟` V:  08:27, 2 July 2016 (CEST) (p.s. many here know me as 'Patio' my old userName)

At least Wikivoyage comes to mind with toponymes etc., but also Wikiversity and Wikispecies.
I just do not know how to add the appropriate fields. -- Purodha Blissenbach (talk) 14:23, 2 July 2016 (CEST) Purodha.
Regarding "Sometimes we link to Wikipedia, but I see Wikidata as well". Actually, now we only link to Wikidata. The link to Wikidata is automatically transformed into a Wikipedia link if there is one in the user language (so each user will see a different link). Otherwise it stays a Wikidata link.
We also have link to Wikimedia Commons (to link to categories).
We can either link only to Wikipedia, which then gives links to other projects, or we can link to all WMF projects (then we need more time to add them and maintain them). If Wikiversity and Wikispecies and Wikivoyage use Wikidata for their interwiki links, then the maintaining is not so much trouble since they would have a unique ID. I don't know how it is done currently. --Kip (talk) 18:41, 5 July 2016 (CEST)
If I understood you well Kip: when WikiData links to a project, we can use those link(s) easily?  Klaas `Z4␟` V:  07:37, 6 July 2016 (CEST)
yes. It just needs a small adaptation of the code that exists for wikipedia. --Kip (talk) 18:43, 8 July 2016 (CEST)

Putsch[edit]

Is there a native German willing to take a look followed by necessary update(s) to these lemmata? Danke scön bij voorbaat ;)  Klaas `Z4␟` V:  13:17, 18 July 2016 (CEST)

ː Done --Purodha Blissenbach (talk) 19:06, 18 July 2016 (CEST) Dank(e)!Grazie!Thanks!Merci! ;-D  Klaas `Z4␟` V:  11:45, 6 August 2016 (CEST)

General question about Chinese/Cantonese/Mandarin/etc[edit]

When I copy a word from http://zh.wikipedia.org as a nitwit of the script, what language should I choose? Mandarin (simplified) was just a guess in my latest change. Is that (mostly/always?) right?  Klaas `Z4␟` V:  07:45, 8 August 2016 (CEST)

I do not know. Technically, the script should most often or aways be determineble from the input and could then be set automatically. But I doubt that this is done. --Purodha Blissenbach (talk) 09:31, 8 August 2016 (CEST)
By default, the script that is shown should be "Mandarin (simplified)"
The script can be changed with the dropdown menu, the third tab after article / discussion. To be sure you can select the first one from the list, "大陆简体" (zh-cn) which is simplified Chinese.
It used to be that you could select traditional Chinese as well, but now this has been divided into the different dialects/regions that use traditional characters.
热点 is simplified indeed. The traditional is 熱點.
More info there https://en.wikipedia.org/wiki/Chinese_Wikipedia#Origin_of_edits . And http://www.chine-nouvelle.com/outils/convertisseur-simptrad.html for a converter. --Kip (talk) 21:10, 8 August 2016 (CEST)
Ok, I have no further clue as to how to answer your question. --Purodha Blissenbach (talk) 10:38, 9 August 2016 (CEST)

Annotations[edit]

Hi, I do not get "word class" as property under option values. I wanted to add some ... I had a look at my old edits and there it still shows up, even if not in the list. I hope you do understand what I mean :-) Thanks for any hint. --Sabine (talk) 16:00, 16 August 2016 (CEST)

Do you mean Part of Speech which is in Option Values?:
     ▼ Option values
	Property         Option
Add	part of speech   noun 
A result of my search for an answer you see here.
In namespace DefinedMeaning I do not see it either
▼ Semantic annotations
Property                Value
Wikidata ID             q43450
Commons category        Gymnastics
Image from Commons      Gymnastics brokenchopstick.jpg
topic                   sport
Is this intentionally, a bug or what? Making things less transparant (read: more complicated) for "normal" (luckily I'm not ;-}) people?
HTH,  Klaas `Z4␟` V:  10:30, 17 August 2016 (CEST)


"word class" / "part of speech" is defined as a lexical annotation (language dependent) and not a semantic annotation (language independent). These are two categories of annotations that are separate on the pages (lexical is shown at the top, semantic at the bottom)
When you edit an Expression:... page, the lexical annotations are shown on top. You can edit the part of speech there and save.
When you edit a DefinedMeaning:... page, the lexical annotations are not shown (because the DefinedMeaning is language independent). However, in the "synonyms and translations", you can find the word you are looking for and edit its annotations - right-most column. Word class is normally there.
--Kip (talk) 22:31, 18 August 2016 (CEST)
So in nameSpace DefinedMeaning it has to be done for all languages separately because a word can be in one language a noun and in another e.g. a verb? My example above is well chosen. Basically it is a verb, used as a noun factually. German is the proof: same word written with uppercase T. Do you guys really expect from us to do multiple work where it is rarely (perhaps 'never'?) necessary?  Klaas `Z4␟` V:  09:45, 19 August 2016 (CEST)
With international words I do not find the option 'part of speech'. Why is that and can it be repaired or did I miss something?  Klaas `Z4␟` V:  09:42, 22 August 2016 (CEST)
Part of speech is language dependent, at least in OmegaWiki. Example, the English preposition may in fact be a post position in the Chinese language. If you know the international language's 'part of speech', you must tell Kip what expressions to include to the international language's 'part of speech', and he will add it. It would be cool if the WikiLexical software knows how to auto add part of speech whenever a part of speech is the same for all language (aren't nouns always nouns wherever we live?), or for languages that share the same part of speech rules. --向榮 /Hiong3-eng5/ (talk) 08:12, 9 September 2016 (CEST)
huh ... it seems I asked one of "those" questions ... I'll look into this again this evening or tomorrow, when I am at home. Thank you for now!!!--Sabine (talk) 17:18, 23 August 2016 (CEST)

andorran franc[edit]

Please read this voice: DefinedMeaning:franc_andorran_(897090) I talk here because, IMHO, the andorran franc don't exist, even in the past.

-SebastianoPistore (talk) 12:52, 25 August 2016 (CEST)

There were Andorran pesetas and francs  :-P  Klaas `Z4␟` V:  09:46, 29 August 2016 (CEST)
I verified, it's a very complex question: "andorran" franc and "andorran" pesetas were french franch and spanish pesetas used in Andorra: no difference between french franch and spanish pesetas. Because Andorra hasn't no own money. When Spain and France used Euro, Euro go in Andorra ("franch" Euro and "spanish" euro). From 2014 http://euros.ad/historia-i-dissenys Andorra start minting own euro. https://fr.wikipedia.org/wiki/%C3%89conomie_de_l%27Andorre read "La monnaie" https://es.wikipedia.org/wiki/Econom%C3%ADa_de_Andorra read "Moneda"--SebastianoPistore (talk) 12:28, 29 August 2016 (CEST)

Typo in DM[edit]

When there is is a typo in a DefinedMeaning I try to repair it in the URL, but the system comes back with 'create' so I can not see if I succeed. What to do? Example I tried to correct pessmism to pessimism  Klaas `Z4␟` V:  14:37, 25 September 2016 (CEST)