Heiko Evermann wrote:
Hi Gerard,
*Without explicit repetitions, how would you indicate if something is correct in a certain orthography ?
What I would like to have is something like this: Article: Oort {{-nds}} {{-noun-}} # Oort: description in LS of "a place" {{-nds-sass}} {{-nds-harte}} {{-nds-lindow-}} {{-nds-neuber}}
The way it will be available in UW will be more like:
{{-nds}} {{-nds-xxxx-}} {{-noun-}} # Oort: description in the particular (xxxx) orthography for "Oort"
This is a requirement for the UW. You can decide not to give a definition but a definition when given should be in the specific orthography.
# Oort: description in LS of "a way" {{-nds-sass}}
where -nds-harte- has "Aart" instead of "Oort". And there needs to be some room to indicate that some orthoraphies have "Aart" for "Oort" in case of meaning 2.
Every word is known by its orthography/dialect/language so in UW every word will have its own entry for every language. Through the table SynTrans they can be linked to the same Meaning
I would not like to see main entries -nds-sass- etc, because they all translate to the same thing. It would not make sense to multiply the translation tables.
It does make sense. It is essential that we define things properly. The only thing you need to worry about is that what you do is correct. So if you only care about Sass, concentrate on Sass and leave the rest to other volunteers. You do not need to worry about the number of records that will end up in the table SynTrans. They are not multiplied. A word with a single meaning will only have one record in SynTrans.
Besides: in a very high amount of cases these different versions are equal, and it does not make sense to multiply the content. We are writing a dictionary for Low Saxon and not one for Sass-Hamburg, Sass-Oostfreesland, Harte-Oostfreesland etc. And differences are only important where they exist and only in those cases need they be listed and in those cases they need comments as to why the differences exist.
They are not equal because they are about different orthographies and we need to know them as such. Many details may be different depending on what orthography is used, things like the earliest recording of this word in this orthography.
If the UW cannot handle that, it would be nice to adapt it. I think that this is important. It is important to reuse data. If all data that someone has entered for -nds-sass- has to be reentered for -nds-harte- in those cases where both are equal (95% of the entries) then this is just an invitation for trouble, inconsistencies etc. Such redundancies of data must be caught by the design and not by the data.
The essence of a proper relational database is that you store relevant data only once. So there is no need for reusing data because you only have it once for the use that you want out of it. Ultimate Wiktionary has not been designed specifically for nds; it is designed to host all languages, dialects and orthographies. When you can produce a list with the words that differ between Sass and Harte, you can equally provide a list with the words they have in common. Technically it will be possible to create such lists from the UW content when these words have been properly inserted in the database. I really agree with you that redundancies must be caught by the design, but you have to apreciate what the design is ment to do.
Thanks, GerardM