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:Functionality wanted/Archive/2009

From OmegaWiki
Jump to: navigation, search

Parts of speech indications[edit]

Not very urgent, but crucial to the good functioning of DefinedMeanings, some indication of which part of speech is described is necessary. Probably a matter more complicated than expected, I guess, because not all languages can be treated as if they all make the same clear distinctions. — Vildricianus 18:53, 21 July 2006 (CEST)

I think what this request aims at is the ability to add additional grammar information, such as word type and gender. I consider this a basic feature that no dictionary can go without. It's most basic benefit would be to help clean up a lot of unclear definitions. For example, it would help to split the DMs in the English entry of Expression:alarm between the verb to alarm and the noun alarm. While these are identical expressions in English, they are not in most (non-isolating) other languages. Confer German Expression:Alarm, Expression:alarmieren (verbalization) and Expression:Alarmierung (substantivation of verbalization).

The complicated thing about Grammar information is, that it needs to be implemented for each language separately. I would suggest to implement English grammar only first, test it, and when it runs properly add grammar of other languages one by one. --Mkill 02:14, 5 August 2006 (CEST)

Once the techical means exist to define grammatical properties and relations, there is no need to wait for English. All languages grammars can be entered simultanously, they are independent of each other. Persons doing the work are (mainly) in disjunct groups, too. -- Purodha Blissenbach 01:36, 6 August 2006 (CEST)
I'd think simply adding some letters before the definition is a good way to indicate word type and gender. Adding it is simple and it is easy to understand. Or maybe the definition and word type should not be mixed.--Troskyldigheten 11:26, 12 January 2009 (EST)

implemented. archived by --Kipcool 11:48, 5 May 2009 (EDT)

DefinedMeaning[edit]

I consider the use of the term DefinedMeaning itself to be a bug. Why?

  • It's confusing and opaque, raising the bar to entry. See: User:spir/Notion
  • It's written in annoying, pretentious and faux-trendy CamelCase™.
  • It looks even worse in other languages. (GedefinieerdeBetekenis anyone?)
  • Clearer and simpler terms are available.
  • The term does nothing to remedy semantic drift, but instead obscures the problem. [Added at 23:53, 28 April 2009 (EDT)]

In the OmegaWiki blog, GerardM recently stated: "[…] concepts are added and we call them DefinedMeanings […]"
So if they are concepts, why not simply call them concepts? It's a perfectly international and self-evident word.
McDutchie 20:15, 27 April 2009 (EDT)

Hoi, there are problems with the notion of a concept. When you talk about the concept horse, in what language is that language ? The notion of what a concept is often culturally determined so concepts do not translate but that is exactly what we do. In order to prevent this drift we have to anchor them in one language. Because in the end it is the definition in that language we can change in order to refine the definition. When you change the definition because it fits better in one language / culture chaos ensues and there is no final arbiter. Thanks, GerardM 16:21, 28 April 2009 (EDT)
That concept definitions in OmegaWiki should be anchored in one language in order to minimize semantic drift is obvious. But that doesn't make DefinedMeanings not concepts. In fact, given their cultural specificity, I would say that this anchoring is precisely what makes them equivalent to concepts.
You seem to imply that the mere use of the term DefinedMeaning somehow helps prevent semantic drift. But a horse by any other name is still a horse, even if you call it EquineEntity. Simply calling a concept DefinedMeaning does nothing to prevent semantic drift (as indeed I am noticing all the time in existing entries); it only obscures the problem. (Added as fifth bullet point.) – McDutchie 23:53, 28 April 2009 (EDT)
A Haflinger is not a Frisian steed.. both are horses. However you are welcome to your opinion. Thanks, GerardM 02:14, 29 April 2009 (EDT)

Those are three distinct concepts, two of which happen to be subsets of the third.

Sorry, with all due respect for the massive amount of work you've done, you are still not making logical sense to me. And since you have been funded by tax euros, I feel I have a right to annoy you further. There must be reasons why this site has less than ten active contributors after years of existence.

I just found the perfect example of what I mean in DefinedMeaning:air (353835). The given definition of "air" is:

An expression or appearance indicating a certain state of mind.

There were six Dutch expressions in this entry, apparently added by native speakers of Dutch. All six were completely wrong because, while clearly related to the definition, they each express a different concept from the one expressed by the definition:

  • humeur (humor): is an internal state or frame of mind itself, not the expression or appearance thereof.
  • stemming (frame of mind): idem.
  • gemoedsgesteldheid (mood): idem.
  • geestesgesteldheid (mood, state of mind): idem.
  • sfeer (atmosphere): is the general emotional atmosphere of a situation, not the expression or appearance of a person's state of mind.
  • verwaande houding (conceited attitude): expresses one specific subset of the concept, not the conept itself which could be simply "houding" (attitude).

So I removed all of these and replaced them with Dutch words capable of expressing the same concept as expressed by the definition: air (air), voorkomen (appearance), houding (attitude), uiterlijk (appearance), manieren (manners).

The point of this is that I was only able to do this by thinking of a DefinedMeaning as a concept. If I had judged it by whether the expressions correspond in some sense to the definition, as the term DefinedMeaning suggests, and as the original contributors had apparently done, then these incorrect entries would have remained. – McDutchie 14:02, 2 May 2009 (EDT)

Whether it is called a DefinedMeaning or a Concept, I don't care. For me it is just a name because we had to give it a name. It can be changed easily in the software, so I can put whatever is the preference of the majority (for now, we have no majority at all). But this is not a bug, this is a name. --Kipcool 15:25, 2 December 2009 (UTC)