Hoi, To explain to the people on the Wiktionary mailinglist where this comes from, there is a huge debate on the Wikipedia-l mailinglist about having a seperate English and American English wikipedia.
In the plans for Ultimate Wiktionary there are three ways in which words can be destinguished as being of a particular orthography. I will describe these here and hope to use the energy of this discussion for this question that needs a resolution at some stage.
1) English, American English and other orthographies are treated as seperate entities. This means that all words need to exist for each orthography/dialect. On the plus side it means that descriptions like etymology and meaning will be in this one orthography as well. This is also the most easy method to provide information for a spell checker.
2) We treat these variants as belonging to a specific "spelling authority". This means that one word needs to be only once in the database. It means that the meanings and etymologies etc can be in any of the orthographies.. It means that you cannot record the relations between the words of these different orthographies/dialects. When words are properly identified, it means that we can use the information for a spell checker. It does not clearly help you understand what Meanings exist in a particular varietion of English. This is in my opinion the weakest option as it does not allow you to identify which meaning is true for a particular version of English.
3) We can label Meanings as belonging to one of these particular orthographies. When words are properly identified, it means that we can use the information for a spell checker.
In my opinion the number 1 option is technically the best solution. Going for this option is propably less problematic then breaking the en.wikipedia.org into pieces. Going for this option seems like a lot of duplication. It does however provide us with the possibility to be more precise in what makes English different from American, Australian etc.
Please let me know what you think and particularly why.
Thanks, GerardM
Can't we just have a list of checkboxes for each dialect of a given language for each meaning of a word? For words like "to be" or "dog" we could just check boxes for every dialect, while we wouldn't for words like petrol or gasoline...
Pawe³ Dembowski wrote:
Can't we just have a list of checkboxes for each dialect of a given language for each meaning of a word? For words like "to be" or "dog" we could just check boxes for every dialect, while we wouldn't for words like petrol or gasoline...
How many different orthographies / dialects are there for English. I would not dare to presume that a word is truly shared between the less well known versions of English. There are also the "true" dialects like Geordie that have to be considered. Creating check boxes assumes in a way that the editors /know/ these different versions of English well enough. Technically it can be done, but the spelling of the text in a meaning, an etymology needs to be adapted anyway. You have to realise that certain meanings do not travel well. It is therefore not only the orthography but also the Meanings of a word that needs to be considered.
Consequently, a simple check box is problematic in itself.
Thanks, GerardM
wikipedia-l@lists.wikimedia.org