User talk:Sj

Leave notes for me here, or on my page on enWP.

Hi there. Have you seen the article in the Beer parlour? I did it back in January and it showed only about 30% of articles were English entries, properly formatted. I think it is getting better though. What is this "big article" you mention? SemperBlotto 11:36, 15 Mar 2005 (UTC)

I think it's closer to 50% now... even though both english and non-english are growing.

Thanks for your invitation to comment on Unified Wiktionary. You certainly have a lot of enthusiasm for the project!

I have to decline, however. I'm better at organizing and filing. The clear thought required for that careful analysis, though possible for me to do, is not easy for me. And I have to put limits on my Wiktionary time.

Best wishes, --Stranger 23:52, 7 September 2005 (UTC)

No problem, and thanks for the fast response. I think we'll find a clever way to merge databases, in the end; and will be interested to hear your filer's thoughts on whatever schemes are dreamed up. +sj +

Hi Sj,

I already commented on the page for the Unified Wiktionary. I'm a strong proponent of the idea. The bot I'm trying to create though is not focused on exporting to UW. That's just one of the many useful things it's supposed to do. For the moment it's vaporware, alpha level code that I won't let loose on actual wiktionary pages. You can have a look at User_talk:PolyBot and you will find a link to the sourceforge page to have a look at the code. I'll be very interested in commnents and suggestions.

Right now I'm concentrating on a parser. I have a rudimentary export functional for en.wiktionary and nl.wiktionary. There is still a lot of work to many of its parts. What I'm working on right now is the parser for the translations section. I already have the parser working for definitions (what's under the POS header) and synonyms.

Cheers, Polyglot 21:30, 9 September 2005 (UTC)

As for your question whether you can test it on a test wiktionary. It's not very hard to download an SQL dump, set up MySQL and import the dump on your own PC. On Linux it's almost trivial if you're not afraid of get your hands dirty, err, I mean to read some docs on the internet.

The code I'm writing is not connected to the rest of pyWikipediabot for the moment though. The best way to test it, is to look at the files ending in ...test.py. Those are unit tests. Each one testing a specific part of functionality and reporting success or failure.

Connecting it to make use of what pyWikipediabot can do, won't be very hard to accomplish, but I wanted to concentrate on writing functionality that interests me and I don't want to risk the bot going on line before it's ready for prime time, which may still be off for a year or two, who knows? It's slow going all by myself, with a limited amount of time to spend on it. Polyglot 22:08, 9 September 2005 (UTC)

There's a discussion on the BP about renaming the BP to something more "sober". Does UW have a BP (Beer Parlour)? What is it called? Can you reply on the BP? Cheers, --Stranger 19:35, 10 September 2005 (UTC)

One is being developed now... +sj + 20:04, 3 May 2006 (UTC)

This isn't one of OmegaWiki's great strengths, however.

I wonder if there is a list of scripts used to synchronize the data between the projects. Scripted synch is on my mind these days. +sj +

We need a list for words such as:

politics informatics physics mathematics

shears scissors glasses shorts pants

mumps measles rabies

alms thanks kudos

accomodations congratulations news suds

data