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.

International Beer Parlour/Archive8

From OmegaWiki
Jump to: navigation, search

Two redirects[edit]

Hi, there are two redirects from the logo/site-root and sidebar, that are unnecessary. Could an admin change MediaWiki:Mainpage to read "Meta:Main Page" and MediaWiki:Portal-url to read "Meta:International Beer Parlour". Thanks :) Quiddity 15:10, 30 June 2007 (EDT)

I've done MediaWiki:Portal-url, but not MediaWiki:Mainpage, because then not only the link, but also the text in the sidebar changes to "Meta:Main Page". SPQRobin 19:01, 3 July 2007 (EDT)
Hmm. The root url http://www.omegawiki.org leading to a redirect is the main problem. I don't know enough about mediawiki to know what else to suggest though. Anyone? Quiddity 17:16, 5 July 2007 (EDT)
I know this problem already. There are two possibilities:
  1. Add to MediaWiki:Monobook.css this: BODY.page-Meta_Main_Page #contentSub { display: none; }
    Advantage: The translations are preserved. Disadvantage: The redirect still exists, in spite of that you don't see (Redirected from Main Page) anymore.
  2. Change in MediaWiki:Sidebar "mainpage-url|mainpage" to "mainpage|mainpage-text".
    Advantage: The redirect is fixed. Disadvantage: There are no translations anymore. You have to translate it manually.
SPQRobin 19:18, 5 July 2007 (EDT)

Democratic Republic of the Congo - has it disappeared?[edit]

I was unable to add the Polish translation of "Democratic Republic of the Congo" (eg, Congo-Kinshasa) - when I clicked on the link to the page via the ISO 3166-1 collection and its "missing defined meaning," list I got the message indicating no such page exists.

Same thing when I clicked through using other language collections.

When I clicked on Congo-Kinshasa, I got the same message. When I entered Congo-Kinshasa in the search box, I got to the Expression page for that term, but the links there to Democratic Republic of the Congo all failed.

The Polish translation of this country's name is "Demokratyczna Republika Konga" btw. If someone can identify what's going on with this page, could you also enter this translation? Thanks.

Rsperberg 12:42, 9 July 2007 (EDT)

The original DM DefinedMeaning:République Démocratique du Congo (375093) has been renamed in the past to DefinedMeaning:République démocratique du Congo (375093). without capital D. This now gives an error (This page seems to point to a non-existent DefinedMeaning (concept). Please check the web address). See also User_talk:GerardM#Mixed_definitions. HenkvD 14:15, 9 July 2007 (EDT)
Well, I was able to get to the page to add the Polish name. So my immediate problem is resolved. But the overall problem . . . well, good luck! Rsperberg 15:43, 9 July 2007 (EDT)
The bug seems to be fixed as well. HenkvD 06:13, 21 July 2007 (EDT)
The fix proved problematic, and was reverted. A new fix that will have only one expression associated with a DM will be the eventual result. GerardM 06:42, 21 July 2007 (EDT)

Collection? Or semantic domain?[edit]

I have worked with computers since 1966 and in publishing since 1974, and I am currently studying Khmer. So my primary focus in contributing to OmegaWiki is to provide a strong Khmer resource for basic vocabulary and for computing and geographical terms that are not easily located in the primary Khmer online dictionary (at SEAlang.net/khmer/).

Few computer terms exist already in OmegaWiki, and I have begun to add definitions in English so that I can also attach the appropriate Khmer translation.

In a tool like Fieldworks, I would identify these terms as belonging to a semantic domain, computing. In OmegaWiki, I'm wondering if each defined meaning (see, for example, taxonomy) should be put in a new collection of computing terms, or if there is some other more appropriate mechanism for indicating that these DM's are all used in the same broad semantic area.

If there is a better way to indicate this, can someone point it out to me? If not, can a "Computing" collection be created?

Thanks, Rsperberg 10:19, 11 July 2007 (EDT)

I would say: use Relations like is part of theme of Computing (IT) or more specific terms like Website. HenkvD 14:24, 11 July 2007 (EDT)


Doubt on how to edit: "New exact meaning" and "New definition"[edit]

I suggest these two items to be mentioned in each entry (DefinedMeaning) whenever a entry is opened. Thus far I have resorted to a circumlocution, that is, I add a synonym in the heading "Synonyms and translations", which results in automatically adding a new already existing definition in a given entry. If there is a way of solving this problem, please let me know. Waltter Manoel da Silva wtz 09:43, 12 July 2007 (EDT)

General problem with DefinedMeanings[edit]

Having translated a number of DefinedMeanings into Esperanto, I have come across a general problem, which I don't know how to solve. The problem appears in a number of different forms, but I feel that it's basically the same problem. The problem originates in the fact that different languages cannot be mapped to each other in a bijective way. Here are some of its appearences:

  • The definitions in DefinedMeanings define an expression without using that expression (if we were allowed to use the expression defined in the definition, the definition could just be identical to the expression (e.g. define the expression "horse" by using the definition "horse"), but such definitions would be quite useless). We are supposed to literally translate the definitions into other languages. But when you literally translate a definition into another language, it might happen that you use the expression that that definition is supposed to define, so you violate the principle of not using an expression in its own definition. As an example, look at DefinedMeaning:not (5496). If I translate the definition literally into Esperanto, I get "Ĉiu individuo de certa klaso, sen esceptoj". This defines the Esperanto expression "ĉiuj". But "ĉiuj" is just the plural of "ĉiu", so I use the expression defined in the definition.
  • Literal translations aren't always possible. Look at DefinedMeaning:receive (8487). Both the Esperanto translation (created by me) and the Spanish translation aren't literal translations. This is because in these languages, you cannot form a passive which makes the recipient ("dative object") the subject. A passive in these languages only makes a direct object the subject. So the definition is translated with completely different constructions, involving completely different words. But then it is difficult to say that the translated definitions actually have the same meaning as the originial definition.
  • Words acquire certain fixed "metaphorical" meanings in certain languages, and words are then used with such metaphorical meanings in definitions (e.g. the usage of "framework" in DefinedMeaning:ambito (431196)). When you translate this, you can't translate it literally, because the literal translations of "framework" in other languages are not used in that metaphorical way. But if you use a different metaphor in the target language, you don't really preserve the meaning.

What do others suggest doing in such cases? Should we just ignore the problem, and allow translated definitions with deviating meaning as well as translated definitinos containing the expression defined? But then OmegaWiki could never become as precise as other dictionaries, and would have a major disadvantage compared to Wiktionary. Marcos 13:08, 12 July 2007 (EDT)

The rule of thumb I use to judge a DM translation from language x to language y, is that if I were to translate y back to language x, then the meaning of the original translation would be the same as my translation back from y to x. Of course, this should also be the case for x -> y -> z -> x and x -> y -> z -> y -> x. This is possible if you assume that you can express any thought unambiguously in any language if you use enough words.
When we have the possibility to indicate the source language of a DM, then it will help prevent the creation of long chains of translation such as p -> q -> r -> ... -> z, because we'll know that the source language is, say, p, and we get translations from p -> q, p -> r, p -> s, etc.
The same goes for expressions that are (nearly) atomic and primitive to human language (if such a thing exists), like "not": if x -> y -> x' => x = x', we're good.
Hope all this made any sense! László 07:42, 21 July 2007 (EDT)
Well, your answer was mainly about semantic shift, which wasn't what I was talking about.
At any rate, you make the assumption that one can express any thought unambiguously in any language if you use enough words. But I just think that this assumption is straight-forwardly false. The precise thought expressed by "To be handed something" cannot be expressed in Spanish or Esperanto, no matter how many words you use. Actually, the closest you can get to it is brobably "recibir" in Spanish and "ricevi" in Esperanto. But the problem is that the words we want to define are "to receive", "recibir" and "ricevi", so to define them by "recibir" and "ricevi" would be missing the point of a dictionary. Marcos 14:32, 22 July 2007 (EDT)
When you create a DM for "receive", you also create a Definition. It is this definition that is translated. When there is no word that can be used in translation, you can still translate the definition. This makes it plain what the words that you find in translation mean. Defining a word by an other word is plain wrong in any language; what you do is provide synonyms. GerardM 15:26, 22 July 2007 (EDT)
We are supposed to literally translate the definitions into other languages. I think there is a misunderstanding here. The DefinedMeaning page does not tell us to translate literally, but to create a translation that "adheres as closely as possible to the concept as described by the First Definition". It is impossible to translate definitions word by word (i.e. literally) because naturally languages have different sentence structures and use different expressions, metaphors and so on. The translation is supposed to capture the meaning of the definition. Basically, you describe a concept in another language. --Tosca 17:36, 22 July 2007 (EDT)
Originally it did say that the definition should be translated as literally as possible. I was quite alarmed when I read what you wrote. I have changed the text to make it less ambiguous. Essentially, for the Defining Definition you should NEVER describe a concept in another language, you should translate the original definition. GerardM 03:54, 23 July 2007 (EDT)
Gerard, by saying "describe a concept in another language" I did not mean that you create a completely new definition. It is supposed to be a translation of the original defintion. However, a literal translation is not always the best way to preserve the meaning of the original definition. Translating literally may in fact introduce semantic drift and make the translation sound like a translation, which is bad. So the translator has to make sure that the translation adequately conveys the concept or meaning of the DM and that it is comprehensible for a reader who cannot read the source text. --Tosca 11:29, 23 July 2007 (EDT)
I agree with Tosca that a literal translation is not always the best way to preserve the meaning of the original definition, and that hence the earlier version of DefinedMeaning was better than the most recent version by GerardM.
But even with the earlier version of DefinedMeaning, the problem I am talking about (at least in the first of the three instances mentioned in my first post) is not really solved. I'm not really sure whether anyone here has actually understood the problem, even though I don't think it's difficult to understand (maybe I'm just bad at explaining it).
Maybe you'll understand the problem if I give an abstract example: Suppose there is an expression "Q" in some language, and we create a DefinedMeaning for it, using the definition "YZ". Now suppose that the best translation for "Q" into some other language is "x", and the best translation of the definition "YZ" is also "x" (this can very well happen). Then the current procedure would lead us to have the expression "x" of the second language being defined by "x". This wouldn't be very useful for a user who just speaks the second language, and wants to look up a definition of "x".
I hope you now understand the problem. A possible way of solving it would be to create another DefinedMeaning, with expression "x" and some more useful definition of it, say "wzy". But if we used this solution too much (which would probably happen once we have active contributers in many languages), we would create a multitude of similar DefinedMeanings, which all define the same concept, just using slightly different words in some languages (and having a different First-Definition language).
Another problem with this solution is that a user of the second language looking up "x" would find two definitions: "x" and "wzy". The first would be completely useless to him. A way of solving this problem would be to mark the first DefinedMeaning (which defines "Q" and "x" by "YZ" and "x") as a DefinedMeaning that should be ignored when OmegaWiki is used as a database for a monolingual dictionary of the second language. Marcos 12:58, 25 July 2007 (EDT)
Here's my opinion about that: If x is the best (or only possible) translation of YZ, then YZ is probably not a good definition because it explains using synonyms. Therefore, YZ should be changed. Unfortunately we have a lot of DMs that use synonyms because traditional dictionaries often define like this and people are used to it.
However, there's a problem with changing DMs when they already have lots of translations since most people can't check and rewrite them all. One way around this is to create an alternative definition, but even then you probably can't verify if all given SynTrans still match that alternative definition. That often leaves creating a new DM as the only solution...
In conclusion, it is really important that the first definition is suitable and well thought out. --Tosca 15:36, 25 July 2007 (EDT)
Yes, that is certainly important. But I have come across quite a few DM's with badly chosen definitions :-(
I guess what I should do in future when I find such DM's is to propose an alternative DM (with an alternative definition) to replace the current DM (once it contains more translations), instead of trying to find an ad hoc solution for the language I am translating into. Does that sound right? Marcos 19:06, 2 August 2007 (EDT)
I agree with Tosca's opinion (change the definition) and conclusion (it should be well thought out from the start). Also warn the definition makers inside the template) not' to use synonyms.
Then, for a possible solution, there may be a system of DM definition status. Such as :
  • new definition : please give your opinion about it (and possibly : don't translate yet)
  • approved definition (proved to be good & stable)
  • weak definition : please help by changing/improving it
The definition's status would be computed from users' feedback, number of translations and time. Users could bring opinion from an input template, for instance :
++ (very good) + (rather good) - (rather bad) -- (very bad) [no medium value]
and a checkbox to say "this definition is problematic to translate into my target language".
The question is then to choose the relevant factors of a defintion's status, to ponderate them (relative importance), and to agree upon criteria to change the status : for instance how much good feedback / how many definitions / how long before a definition is said approved ?
There can a set of flags or banners to show visually the definition status.
--Spir 13:17, 21 September 2008 (EDT)
That are good ideas. It's a pity if weak definitions are being translated before they are corrected. Sometimes that happens even if there is an "Attention template" on the talk page.
The question if the definition is problematic to translate is also a very good point.
In addition it would be good to have flags like new/approved for the expressions. Some users add translations from dictionaries without respect for the definition. --Ortografix 17:27, 23 September 2008 (EDT)

Harry Potter[edit]

I have finished Harry Potter, things can go back to normal for me ... GerardM 02:33, 22 July 2007 (EDT)

Presentations Wikimania 2007[edit]

I have given two presentations at Wikimania .. enjoy :) GerardM 20:03, 3 August 2007 (EDT)

(The above links are to wiki HTML pages, not to PDF content. This confused my browser, which tried to display the HTML with a PDF reader. The trick is to follow these links as HTML, even though they end in ".pdf". The resulting HTML pages have an HTML link to the PDF files themselves. Hope this helps anyone else with a confused browser. Those of you with a smarter browser, sorry for the distraction. JimDeLaHunt 17:16, 4 August 2007 (EDT) )

partnership, information accessibility and Licensing[edit]

I see today that we can switch selection between three Data-Set and I wonder what is the exact licensing of the non-Community ones. From some GerardM wikimania talk material I read :

« 

  • Databases provided by external partners
    • each AD has its own set of tables
    • copyright remains with the external partner
  • Data is editable in the Community Database
    • first copied by AD user, then edited by CD user
    • license of data is CC-by / GFDL
  • Concepts can be linked across databases
    • this connects concepts across domains

»

but do not really understand exactly what it does mean.

I think OmegaWiki must focus on the community data-set contributions and give hint about the exact policy on those different contents.

http://www.omegawiki.org/index.php?title=Expression:abeja&dataset=umls does indicate GFDL/CC-BY at the bottom. But http://www.omegawiki.org/index.php?title=Expression:abeja&action=edit&dataset=umls refers to an empty Meta:Permission_policy

I think too that a more prominent link to Licensing policy should be made near the save submission button of data pages (likewise the text on normal wiki pages).

luna 11:52, 8 August 2007 (EDT)

The content that has been made available in the UMLS and Swiss-Prot datasets is of significant value. Their value is particularly apparent in a scientific setting. Within an Open Access setting it is significant that the Wiki for Professionals provides its services free of charge. Given that there is a lot of annotation to do on the content, it is likely that much of the content will become available under a CC-by and/or GFDL license.
When it comes to copyright, each contributor has the copyright to the data they produce in OmegaWiki for as far as copyrights apply. It is a requirement to allow for the licensing terms that apply to the data within. As Authoritative Databases contain completely separate files, it is legal and possible to make the data available with both separate copyright and licensing. GerardM 14:25, 8 August 2007 (EDT)
I think this content is indeed useful, my point is more about a proper labeling of which data is under which licensing terms. The user must know which are the possibility for using the data. The contributor should be incited to be coherent with the project licensing policy (GFDL/CC-BY). luna 06:17, 9 August 2007 (EDT)
I am really happy to have this data available at this time. As the relation between these contributing organisations matures, this issue will be addressed. At this moment they do not license the data with the licenses we take for granted. This is partly due to the maturity of these databases; they are older then any of the licenses we prefer. GerardM 05:24, 11 August 2007 (EDT)

Linking the different databases[edit]

I have been toying around with the functionality that allows us to connect content in multiple databases. I have been working on some chemical elements and some animals. You connect things by connecting on the DefinedMeaning.. The results can be found in this collection. GerardM 05:21, 11 August 2007 (EDT)

Language selection in the sidebar[edit]

The one thing that distinguishes OmegaWiki from other wikis is that the language of the labels used in the content is configurable for the supported languages in OmegaWiki. This feature is not available to anonymous users. It is for this reason that we want to add a language select in the sidebar. This will set a cookie and thereby make it the default for all users.

One of the things that is currently not localised are many of the labels in the screens. These labels will become messages in a next release of the software. This means that the user interface can be completely shown in the selected language. GerardM 05:05, 24 August 2007 (EDT)

What about automatic preset of this cookie with HTTP content-negociation information ? luna 05:57, 24 August 2007 (EDT)

Thank Christ![edit]

I finally realized that I could get to this page directly through firefox and not via that cocking awful 'desktop' fiasco. OK, so there are some nice concepts, and those concepts are implemented through the 'desktop' - but its functionality sucks! Anyway... I am currently confused... what is the relationship between this wiki and that ... thing?

I tried searching 'knowlet space' for the word 'protein' but there were zero hits. What gives?

--Dan 06:48, 15 September 2007 (EDT)

Hello. By starting your question by saying everything sucks, you'll most probably not receive any answer. PS: there is no user known as Christ. Kipcool 07:27, 22 September 2007 (EDT)
To my mind, a real Wiki is much too loose for a good collaborative dictionary. This project, though using Wiki, is quite different, and keeping "wiki" in the name is questionable. It's here for historical reasons and they can't change their name every year. --Fiable.biz 23:54, 5 September 2009 (UTC)

Latin language & taxonomic names[edit]

I have been asked REALLY REALLY urgently and nicely to allow for the editing of content in the Latin language. I did not want to because the Latin used in taxonomic names does not quality as such and I do not want "Homo sapiens sapiens" and "Equus caballus" as a Latin phrase. They have nothing to do with how people speak Latin.

So we can add Latin, but until we have a solution for the taxonomic names they are not to be included as being Latin. GerardM 05:55, 22 September 2007 (EDT)

Thank you very so much :-) Kipcool 07:22, 22 September 2007 (EDT)
Vielen Dank. :-) --Tosca 12:01, 22 September 2007 (EDT)

If we had a solution for taxonomic names it would be easier to avoid creating duplicate DMs for biological species as DefinedMeaning:Buchfink (543457) and DefinedMeaning:vink_(653000). --Ortografix 15:49, 4 March 2008 (EST)

Can't scientific standard terminology be it's own separate language? This would work not just for Equus Caballus, but also for H20 and probably some math stuff too. Of course, biological taxonomy names are the biggest win, because it is probably much easier to accidentally duplicate the DM for pig than for caffeine. Homunq 10:38, 18 February 2009 (EST)
Probably we have to wait a few decennia for a solution. --Ortografix 08:20, 15 March 2009 (EDT)

External collaboration[edit]

Hi, I just signed up here at OmegaWiki, and hope this project will flourish and thrive as well as it should! Just out of curiosity, are there any ideas how we can collaborate with The Rosetta Project? This venture came into existence due to various inefficiencies and redundantly repetitive efforts related to cross-language resource utilization, and the other pertains to "a global collaboration of language specialists and native speakers working to build a publicly accessible digital library of human languages", as stated on their About Us page. Seems to me the two should be a perfect match, and much can be gained from working together!

What is a vegetable?[edit]

Now we have the ability to add DM's to (as of yet non-user-editable?) classes, and one of them is "vegetable". You'd say it's a pretty straightforward class, but on closer inspection, DefinedMeaning:vegetable (3953) is defined as the plant bearing a certain kind of fruit. However, I'd say the fruit itself can be called "vegetable" at least as well as the plant that it came from (if not better). Now we have the odd situation that DefinedMeaning:aubergine (420897) (the fruit) is not in the "vegetable" class, but DefinedMeaning:eggplant (507273) (the plant) is. As I see it, there are two ways to proceed:

  • add the fruit DM to the vegetable class too, and change the description of DefinedMeaning:vegetable (3953) to include the fruit by itself, or alternatively:
  • add another class called "fruits of vegetables", and add all the fruits to that seperately

I understand this is just a proof of principle, but it is only by understanding how to cope with trivial examples like this, that we can understand how to do it "for real". The cause of the problem here is, that the "vegetables" class is not what I expected it to be.

Curious about what you are thinking! László 09:42, 1 October 2007 (EDT)

The words "fruit" and "vegetable" have both culinary and botanical meanings. In botany, a fruit is the seed-bearing part of the plant, and the vegetable kingdom includes all plants. In culinary usage, a fruit is sweet, and is cooked like other fruits, so rhubarb stalks made into pie are a fruit, while tomatoes and eggplant are vegetables. In culinary usage, a strawberry is a fruit, while in botany only the little yellow seed pods are fruits, and the sweet red part is an enlarged stem. In law, the US Supreme Court famously sided with the cooks against the scientists, determining that tomatoes were to be taxed as vegetables.--Mokurai 01:55, 30 November 2007 (EST)

Searching for Uppercase[edit]

Hi,
when I'm searching for upper-case words--as they are common in the german language--I just find the small written ones. I think that is a bug that needs some fixing. MovGP0 18:27, 6 October 2007 (EDT)

We are working on revamping the user interface. This is one of the things that I expect to see improved. GerardM 18:40, 6 October 2007 (EDT)
Until it is fixed you can add 'Expression:' in the search box to find the upper-case-word, so use 'Expression:Weiß' instead of just 'Weiß' HenkvD 18:23, 7 October 2007 (EDT)

APIs[edit]

Hi. What's the recommended way to use omegawiki programatically to retrieve translations for a word? Ojw 13:19, 19 October 2007 (EDT)

At the moment we do not provide APIs for that. We are at the moment asking for what people like in such an API. I very much want support for standards like TBX, LMF, OWL to be implemented. GerardM 13:29, 19 October 2007 (EDT)
Personally, the simpler the better. e.g. being able to download http://www.omegawiki.org/API/DefinedMeaning:river_(3253) and get a list of lines with "en:A stream of water, fr:Courant d'eau..." would be perfect. Ojw 14:16, 19 October 2007 (EDT)
If you just want the raw data, you can download the daily database dump as described in Development. Of course, if you instantly need up-to-date data on-demand, an API is very necessary. A little side project that I have would eventually depend on an export API, so I am following this topic with interest. OWL sounds like a very suitable format that is simple enough to easily parse but feature-rich enough to include all pertaining data. It shouldn't be too hard to rip the existing OWL export functionality from Semantic MediaWiki (I say this without even looking at its source, so YMMV). László 14:40, 19 October 2007 (EDT)

Hug/embrace[edit]

I'm fairly new with OmegaWiki, and stumbled on a problem and I would like to take this oppurtinity to learn.

I saw that Expression:embrace and Expression:hug share the same meaning, and I disagree with that. First of all, there's also the word embrace in the other context: "Country X embraces Country Y as a new partner", or "I embrace you as my new daughter", and second of all, it's just an other word, or am I wrong?

Okay, what I first did was to remove the original dutch equivalent of embrace (omhelzen) and to add a new word: knuffelen (to hug). Is there an easier method to do this?

Now I'm kinda stuck. First of all, it's easier that we assume I'm right, just for learning purposes ;) So if I'm right embrace should have another definition and doesn't belong in that DefinedMeaning? So should I delete embrace and create a new expression? And what happens if I uncheck identical definition?

Thanks in advance :)

Yorian 09:04, 28 October 2007 (EDT)

Hoi, a word can have multiple meanings. And as a consequence we need multiple DefinedMeanings in OmegaWiki. So when a word has a different meaning in a different context, it does not mean that the existing DM has to be inclusive of that meaning. GerardM 09:28, 28 October 2007 (EDT)
Hm, when I try to set a gender, my only option is undefined? Yorian 09:50, 29 October 2007 (EDT)
A gender is language specific and must like the part of speech be defined. GerardM 11:32, 29 October 2007 (EDT)
Okay, so for each language this has to be defined, and this has yet to be done with bokmål? How can I add this or where can I ask to add this? Yorian 12:57, 29 October 2007 (EDT)
Hoi, you do not define per language. You define a sense for a word. As long as there is sufficient a match you do not need to define for a language. GerardM 16:32, 29 October 2007 (EDT)
Gerard, he is talking about adding the gender annotation to bokmål expressions. When you try to do so, the dropdown only contains an "undefined" option, not the "male" and "female" options. The DM's of "feminin" and "masculin" already have the bokmål translations in place ("hunkjønn" and "hankjønn"). Either this is a bug or you have to do something to enable these as annotation options. László 16:46, 29 October 2007 (EDT)
That's exactly what I mean :) Yorian 09:06, 30 October 2007 (EDT)
I have done some sleuthing and I found out that the PHP file directly responsible for the "undefined" line is SpecialSelect.php (source code), which is AJAXly invoked like this after you select "gender" for annotating a Bokmal word (note the lack of output, even though a list of names is expected). The comments in the PHP code would suggest that this may indeed be the cause of the problem ;) László 10:57, 30 October 2007 (EDT)
As I said earlier, like part of speech it needs to be configured per language .. This is done on the lexical item. Thanks, GerardM 11:49, 30 October 2007 (EDT)
Maybe I'm dumb, but I still don't have any idea where I can change it? Can I change it? Yorian 09:04, 31 October 2007 (EDT)
You can change it: edit lexical item, open the section "Class attributes" and click on "Options »" in the line "... grammatical gender ...". Then can add options (one by one) at the end of the drop down list. --Ortografix 17:52, 31 October 2007 (EDT)
Ah, this worked, I've added the genders for Bokmål :) I wonder though, in bokmål I can choose between gender and gender (kjønn and kjønn) when I edit the annotation of a word, and when I choose a gender of an English word I can even choose between gender, gender, grammatical gender and grammatical gender. Is this a bug? Which should I choose? Yorian 22:26, 31 October 2007 (EDT)
Any will yield the same result. But at this point this duplication shouldn't be happening anymore... :) Malafaya 03:52, 15 February 2009 (CET)

Activity[edit]

I wonder how active this project is. On the editorial side it's really active, there are a lot of recent changes, especially since people have to be approved in order to edit, and it's kinda difficult.

But on the other sides? I can see GerardM is really busy, but other founders of this project? And on development side? I'm just curious, I think this is a fantastic project and more people should support it :) Yorian 22:58, 31 October 2007 (EDT)

Database[edit]

I try to download this file, but I only get a 365 byte file? Yorian 10:46, 1 November 2007 (EDT)

I will ask for someone to have a look at it.. GerardM 11:26, 1 November 2007 (EDT)

collections and class attributes[edit]

At this moment adding collections and class attributes is painfully slow. The developers have been notified and are looking into it.... a not so happy GerardM 19:53, 1 November 2007 (EDT)

About specific languages[edit]

Bel in the main i18n template[edit]

Please add the bel portal to the (protected) main i18n template. Thanks! Yury Tarasievich 15:11, 2 December 2007 (EST)

To be more precise, please add the bel language entry (па-беларуску) to the Template:Mainpage/Translations. Yury Tarasievich 15:14, 2 December 2007 (EST)

Hoi, you can make the change, you have the power :) GerardM 16:18, 2 December 2007 (EST)
Joy. I didn't even realize I'm in the "power group", and there seems to be nothing on me in the logs. Must be divine intervention. Thanks all the same. :))) Yury Tarasievich 16:58, 2 December 2007 (EST)

31 Mayan languages[edit]

I have just obtained a CC-BY-SA license for the material in "The Maya Site"... this is >40,000 entries in >30 Mayan languages from >70 source dictionaries spanning >400 years. I want to upload this material to OmegaWiki, and then I want it to be usable (editable) by Mayan 5th graders using XO laptops or other computers, in schools with relatively poor internet connections. (I myself am a programmer and a teacher, but only a very very amateur Mayanist.) I'd love to be able to chat in realtime with some OmegaWiki programmers / users about the many issues this raises. (This note is only a very preliminary announcement, don't worry, I will be expanding it). You can contact me via my user page or the "email this user" function. Homunq 10:25, 18 February 2009 (EST)

Issues I've identified so far:
  • Metadata (source - both for acknowledgement and for dialectical variations, which will probably be too varied and unexpected to typify in advance) on words, not on definedmeanings. From looking at the ERD, it seems that this would go in the uw_syntrans_??? table, but how is that table used?
  • I need to add the Mayan language codes. But ethnologue has 69, whereas the data (and the traditional names) only recognize around 30.
It turns out that thanks to Nora England the ethnologue mess has been cleaned up, here is the list of language names and codes. Note that these are the generally agreed on names, to be used as the English names (although some English sources drop the final o's as being just a Spanish gendered ending); in several cases, they are not the languages' autonyms, for instance both Tzotzil and Tzeltal speakers refer to their own language as "Batsi(l) k'op".
The list below is all the language codes that still exist, and it is the exact list of languages according to the dictionary data as well. A few of these languages are actually extinct or close to it, but AFAIK, the only one officially so is Chicomuceltec.

I am not 100% sure of the below, but this is the best I can do. There are 3 versions: "technical autonym" (related languages, such as Tzeltal and Tzotzil or K'iche', Achi', and Kaqchikel, often have identical colloquial autonyms, so native speakers have a more precise word to distinguish them); English; and Spanish; in that order. The autonym is in some cases two words.

yoko ochoco   Chontal chontal  [chf] (Mexico)  
ch'ol Chol chol  [ctu] (Mexico) 
ch'orti' Chorti Chortí  [caa] (Guatemala) 
tseltal k'op   Tzeltal Tzeltal   [tzh] (Mexico) 
tsotsil k'op   Tzotzil Tzotzil   [tzo] (Mexico) 
chikomuseltek Chicomuceltec Chicomuceltec  [cob] (Mexico)  
teenek Wastek huastec  [hus] (Mexico) 
chuj Chuj chuj  [cac] (Guatemala) 
tojolab'al Tojolabal tojolabal  [toj] (Mexico) 
jab' xub'al   Jakaltek jacalteco [jac] (Guatemala) 
q'anjob'al Kanjobal kanjobal [kjb] (Guatemala) 
conob Akatek acateco [knj] (Guatemala) (NOT ISO name; official name is still the outdated “western  kanjobal”)  
mocho' Mocho mocho  [mhc] (Mexico) 
awakateko Awakateko aguacateco  [agu] (Guatemala) 
ixil Ixil ixil [ixl] (Guatemala) 
qyol mam    Mam mam [mam] (Guatemala) 
b'a'aj Tektitek tectiteco [ttc] (Guatemala) 
q'eqchi' Qeqchi kekchí [kek] (Guatemala) 
poqomam Poqomam pocomam [poc] (Guatemala) 
poqomchi' Poqomchi pocomchi [poh] (Guatemala) 
kaqchikel Ch'ab'äl   Kaqchikel Kaqchiquel [cak] (Guatemala) 
achi' Achi achí [acr] (Guatemala) 
k'iche' Kiche quiché [quc] (Guatemala) 
tz'utujiil Tzutujil tzutujil [tzj] (Guatemala) 
sakapultek Sakapultek sacapulteco [quv] (Guatemala) 
sipakapense Sipakapanse sipacapeño [qum] (Guatemala) 
uspantek Uspantek uspanteco [usp] (Guatemala) 
itza' Itza itzá [itz] (Guatemala) 
mopan Mopan mopán [mop] (Belize) (secondary ISO name)
lacandon Lacandon lacondon [lac] (Mexico) 
maya Yucatec yucateco [yua] (Mexico)  (secondary ISO name)  

... more to come, gotta go. Homunq 17:07, 18 February 2009 (EST)

  • I need the local name as written locally, the English name and the code in order to enter them in OmegaWiki. Thanks, GerardM 05:31, 21 February 2009 (EST)

Mongolian[edit]

Why isn't it possible to enter an expression or a translation in Cyrillic Mongolian? Note that Mongolian is one language written everyday with 3 scripts. 2 official scripts: the Mongol-Uigur (vertical) script, official in Inner Mongolia (China) and taught in high school in independent Mongolia, and Cyrillic, official in independent Mongolia, plus unofficially in Latin script in nearly all (very many) SMS and many e-mails. --Fiable.biz 11:00, 6 September 2009 (UTC)

According to Wikipedia and more relevantly to the ISO-639-3 standard, there are three Mongolian languages. These languages may be written in multiple scripts. OmegaWiki allows any language that is recognised by the ISO as a language. When you want to have support for your Mongolian language, identify the one you will be contributing for and we can add it to the list of languages. Thanks, GerardM 17:42, 3 October 2009 (UTC)
After talking with Kipcool I have added Halh Mongolian (khk) as a language to OmegaWiki. Thanks GerardM 17:55, 3 October 2009 (UTC)

Dialects, regionalisms[edit]

  • If I say "he he" instead of "tsia" to say "no", it's highly probable I'm a Madagascan of the Tsimihety tribe, but deciding for that that Tsimihety is a full language different from Malagasy would lead to define at least as many languages as provinces in Madagascar, with only a few hundred words differing, and the other dozens of thousands being common. (regionalism)
  • If I pronounce the French word "jaune" (yellow) with an open "a" instead of a close "o", I'm very probably form Quebec, but not only Quebec language cannot be regarded as a fully independent language (I'm French, I've never been in Quebec in my life, but the words I don't understand when a Quebecker speaks are rare. "jaune" is an example I made repeated 3 times and eventually only understand thanks to its translation into... Mongolian.), but French speakers around the world would say it's the same "word" with 2 pronunciations. (regional pronunciation)

So dialectal words should be allowed in OmegaWiki, but not required for all words, with the possibility to state explicitly that there is no proper word in the dialect. If there is one, it should be said if the word is used competitively with the main stream's word, or if it completely replaces it. If the difference is just in pronunciation, then a specific definition shouldn't be asked for. --Fiable.biz 14:48, 26 August 2009 (UTC)

Regionalism is already implemented. We have French for the language spoken everywhere, and the languages French (France), French (Canada), French (Belgium) and French (Switzerland) for the regionalisms. The same goes for English (Uk or USA) and Portuguese (Portugal or Bresil). --Kipcool 08:58, 2 September 2009 (UTC)

valencià[edit]

Hola Gerard, ik zou je willen vragen of het mogelijk is om het Valenciaans (Valencian) mee in de talenlijst op te nemen. Ik woon hier in het Valenciaanse deel van Spanje en kan toch al redelijk wat woordjes toevoegen. Op voorhand bedankt. MARCEL MARCEL 12:25, 28 October 2007 (EDT)

Hoi, Valencian is according to Ethnologue a part of Catalan-Valencian-Balear We already support Catalan. GerardM 11:46, 30 October 2007 (EDT)
Isn't it possible to add "Catalan (Valancian)" and maybe also options for the Balearic dialects? As we also have Austrian and Swiss German, American and British English, Bairisch and Kölsch, it would just be fair to support Valencian and Balearic terms, and this way we avoid the necessity of having identical terms in both Catalan and Valencian. I hope the Valenciàns can accept such a solution. Andreas 14:11, 21 November 2007 (EST)
Somebody could at least say no, because I have some Balearic terms to add :-) Andreas 22:45, 25 November 2007 (EST)
Well, the answer is more like "not now" as we do not have a proper way of annotating this yet. GerardM 02:56, 26 November 2007 (EST)

Eastern Yiddish[edit]

Hoi, on request from Gangleri I have added the Eastern Yiddish language as an editable language. GerardM 06:49, 12 November 2007 (EST)

Molisano[edit]

Hi, I'm new here. I am a native speaker of Molisano, the Italic language ("dialect") spoken in the Molise region of Southern Italy. I see that here on OmegaWiki, as on Wikipedia, my language is grouped with Neapolitan. "Neapolitan" in this sense covers a group of related southern Italian languages. The problem is that my dialect or language, spoken in Molise on Italy's East coast is quite different from Neapolitan, spoken on the West coast. It is, however, similar to the dialects spoken in neighboring Abruzzo and northern Puglia, all along the East coast.

The differences are lexical and pronunciational, to the point where mutual intelligibility is not a given. Molise was the site of migrations (like much of Southern Italy) from Albania (see Arberesh Language) and is also the only region in Italy to have received Slavic migrants (Molise-Croats) who have maintained their language until today. These two major influences have further distinguished the Eastern and Western "Neapolitan" languages. Furthermore, the language of the city of Naples, which was once capital of its own kingdom, has a long written history. The dialects or languages of Molise and Abruzzo are only beginning to be written and standardized. Because the South-Eastern Italian languages have a very different sound from the Neapolitan and South-Western Italian ones, the written form is quite alien.

Projects like this have been created to give languages the chance to flourish rather than be crushed by more influential neighbors. Neapolitan has been able to undergo a renaissance rather than lose out to Italian. However, at the same time I am witness to my own language dying out because people only have the choice to speak Italian or Neapolitan, both of which are euqally foreign, rather than our own language. I wanted to start a wikipedia in Molisan, but since the writing system is not well-established, it was simply too daunting. When I found OmegaWiki, I thought I was in luck since it allowed the development of an online community. Of course that's of no use to me since my language isn't available and I can't really speak or write Neapolitan to the point where I could participate.

What I would like to see would be for the "Neapolitan" language to be divided into Western and Eastern dialects, with the Western staying "Neapolitan" and the Eastern being called "Molisan" (possibly with the three-letter code mol?). I really don't know what it entails to have a three-letter code recognized, but I don't believe that those three letters determine whether a language exists or not. If anyone could give me more information on how to proceed, I'd really appreciate it. Thank you very much for your consideration. --Larenese 06:51, 14 November 2007 (EST)

I am discussig some issues with colleagues on such themes (the same problem arises for a number of less resourced languages since there are no real lines that can be drawn where one language starts and ends) and I am still not really back to work - so sorry if it will take time to answer each point here. In the meantime, please collect names of books, where one can find them (libraries), ISBN codes etc. in Molisan the Molisan Grammar rules (there should be books around), Dictionaries etc. We know about these issues and that is why on nap.wikipedia we allow for all versions of writing, provided that the pages is properly tagged. Probably we will get a solution like nap-aaa, whereby aaa stands for an IANA subtag indicating the dialect. You would need to apply for a proper ISO 639-3 code to get a distinct three letter language code and that would mean that you need to meet certain criteria. If this is your goal, of course we encourage you to go that way as well. Thanks for your patience! --Sabine 09:28, 15 November 2007 (EST)
Hi! the problem you expose is common to most non-normative languages of the planet. There are different solutions:
  1. issueing a new language code within ISO 639-3 (possible but not very quick)
  2. choosing a language subtag.
This second road has the advantage of being much quicker from a bureaucratic point of view. Basically it sets a local variant (some would say a dialect, or a flavour) of an existing language (in this case, of Neapolitan). AFAIK the problem for using this on OW is technical: everything we have is based on ISO 639-3 codes, most notably such is the case with the UI. So we might be reasonably quick in giving you the chance of using a proper label for Molisan content (that would not include Pugliese and even Abruzzese, if I get you right), but you'll still to need to obtain a legal language subtag for it, by specifying some reference text as a vocabulary and a grammar. Be aware than when conflicting versions/scripts etc exist for a single dialect, tomorrow someone may require a subtag that actually refers to the same geographical/linguistic area you are addressing, while using different standards (grammars, vocabulary, etc). I'm afraid that giving you the possibility of setting up a proper Molisan UI will be MUCH longer, though. The software on which OW is based is mediawiki (same as wikipedia), so until THAT doesn't start to allow dialects in the interface (which is something we do not control ourselves) it will be simply impossible to do it here. --Bèrto 'd Sèra 13:30, 15 November 2007 (EST)
Thank you both. Your answers are very helpful. I don't think there exists enough of a literary corpus (if there is any literature at all - i know only of maybe one or two substantial examples) to apply for a code officially. It would probably be better to go with a sub-tag, as you suggested. However, the problem then arises of what is included, that is, how open it is to be. I am uneasy about drawing a line between Molisano and southern Abbruzzese. I have travelled in Abbruzzo and spoken my own dialect and been taken for Abbruzzese.I had no trouble at all understanding them (this was near Pescara). I'm by no means an expert. I know my own dialect reasonably well, and I know that Napuletane is rather different, though, when spoken carefully, it is intelligible. About Pugliese dialects I know very little (I imagine either that they are similar to Molisano or that southern Molisani towns speak some Pugliese dialect). The most distinction I can make out would be between West and East "Neapolitan". For the Western side, I understand that there is a substantial difference between what is spoken in Napoli itself and in Campania outside Naples. However I really know very little about this.
The main thing is that I know my dialect does not sound like Napoletano, and compromises must be made when communicating with Napoletani, whereas I made little or no adjustments to my speech in my limited travels in Abbruzzo. Also, Molisano has no real written form of its own, however it is not well represented by written Neapolitan (which has a long history). These factors lead me to tentatively suggest some distinction between the Neapolitan spoken on the Tirreno and on the Adriatic. The differences may also be due to varying influences. The Western dialects have been influenced more by the closer Spanish, Catalan, Sardinian (?), Sicilian (?), etc... Whereas the Eastern dialects have been exposed to Albanian, Slavic, and also Greek influence.
All this is to say that I very much support and appreciate the efforts to encourage the language(s) of Southern Italy. I am very hesitant about creating further divisions which may weaken the overall vitality of the language or dialect continuum. At the same time, there are inevitable differences between Neapolitan proper and my own dialect. If you notice, these are basically at opposite shores of Italy, and there are the Appenine running down the middle. So, if others can support the factuality of it, it would seem logical to have East and West Neapolitan separate but both under a common Neapolitan banner. I would not, however, wish to impose my dialect of speakers in other areas, such as Abbruzzo or Northern Puglia, if they consider their dialects sufficiently different. Basically I just thought I'd put the issue out there so that perhaps with future expansion it can be discussed further.
I am currently conducting my own research into the dialect of Larino, my town, and am in contact with other interested parties. I think for now it would be possible to continue working under Neapolitan, as long as we can use our own dialect and experiment with our own writing system until it is substantial enough to discuss a separate indentity.
Thank you both very much for taking the time to answer :-) --Larenese 02:53, 23 November 2007 (EST)

Ladin (lld)[edit]

Hello. I just wanted to mention that it may not be good to have Ladin (lld) as one single language option. It is spoken in several local variants, some of which are standardised. Additionally there is Standard Ladin, which is a literary language which in future should be used for official issues. We should at least have the five major idioms which have a standard writing and the standard. Some of the Ladin terms in OmegaWiki are in idiom Fascian, others are in Standard. Can we do anything, although they all use the lld code? Otherwise in the end we get all mixed up. Andreas 12:33, 30 November 2007 (EST)

New collection[edit]

Gangleri asked for a collection with the class attributes. I have created the Community class attributes collection. When your languages has its translations, it will ensure that the annotations can be shown in your language. Thanks, GerardM 20:16, 17 November 2007 (EST)

Corrupted talk/user pages and templates[edit]

Apparently most (or all) pages except Expressions and DM that were changed between 2007/Aug/10 and 2007/Aug/14 are corrupted, see Insect room. Users should check their edits in that time. --Ortografix 20:46, 21 November 2007 (EST)

I made both Erik and Berto aware of this issue .. I have not heard back from them. Thanks, GerardM 20:53, 21 November 2007 (EST)

Checked by native speaker[edit]

I propose to introduce a flag "confirmed by native speaker" for definitions and translations/synonyms. The default value should be "not confirmed". What do you think? --Ortografix 20:39, 3 December 2007 (EST)

For many languages there are big issues with orthography. Often there are multiple dialects that are written as they "sound". The result is that a native speaker is not necessarily the source of the "best" way to write. There are currently tools that deal with quality anyway. There are things that are detrimental to the quality that need fixing (imho) desperately but it is not given priority.. things like the ability to choose a different identifying expression when it is clearly wrong. The prevention of renames for DefinedMeanings and Expressions .... the list is sadly long. GerardM 20:57, 3 December 2007 (EST)
At least a native speaker can judge if the expression exists in his language and if it matches the definition. --Ortografix 21:03, 3 December 2007 (EST)

Interface not working properly?[edit]

Where's the proper place to report the problems with the software interface? I see no trac/bugzilla here or anything of the kind?..

I'm using the linux versions of Opera 9.24 (latest stable), and also checked in Firefox 2.0.0.4 and Konqueror 3.5.7, and I can't add the language to the Multiply meanings (in the attempted new Expression). The dropdown entry field appears, I enter my three-letter language code (bel), the dropdown expands to show the very first page of options (Afrikaans to Asturian), then dropdown disappears leaving no effect on the page. In Firefox, sometimes dropdown shows me the "Belarusian" and then disappears, before I have a chance make a move with a mouse. In Konqueror, I'm able to get dropdown with my Belarusian option which stays put until I click it, however, entered language option somehow "blanks itself" even if I scroll page around a bit or try to access the language choice in Definition section.

Opera error console tells:

JavaScript - http://www.omegawiki.org/index.php?action=edit&preload=&editintro=&title=Expression%3A%D1%87%D0%B0%D0%BB%D0%B0%D0%B2%D0%B5%D0%BA&create=Create+expression
Unknown thread
Error:
name: TypeError
message: Statement on line 83: Could not convert undefined or null to object
Backtrace:
  Line 83 of linked script http://www.omegawiki.org/extensions/Wikidata/ OmegaWiki/suggest.js
    table.parentNode.replaceChild(newTable.firstChild, table);
  At unknown location
    [statement source code not available]

While Firefox 2.0.0.4 error console tells:

Warning: Unknown property 'zoom'.  Declaration dropped.
Source File: http://www.omegawiki.org/extensions/Wikidata/OmegaWiki/suggest.css
Line: 28

I don't know where's the error console in Konqueror. —Yury Tarasievich 07:05, 7 December 2007 (EST)

You do not select a code by typing its code. You select it by typing its name. Also depending on the language that you use for your userinterface, the script can be Cyrillic. (This is in the expression or defined meaning page) Thanks, GerardM 07:38, 7 December 2007 (EST)