POSEN-L Archives

Archiver > POSEN > 2005-10 > 1129817070


From: Elyssa Kowalinski <>
Subject: Re: [POSEN] Lukasz's Marriage Indexing project
Date: Fri, 21 Oct 2005 00:06:25 +1000
References: <200509152000.j8FK0Fda018975@lists8.rootsweb.com>


Hi folks,

Lukasz is doing the search first-name only for now because he's concerned with the amount of spelling mistakes (from misreadings by us volunteers AND mispellings of surnames by the original recorder) in the files he's being sent. I think he decided that the quickest and easiest way to begin putting the results of the project online would be to run the search by first names, which were far less likely to have been mispelled.

Speaking at least for myself, we're doing the best we can to not make mistakes. If I have trouble reading something you'll note that in the index it's followed by a questionmark. The surname mispellings in the records themselves are a frustrating problem - different priests are spelling the same surname different ways. For example, in a couple of decades of the catholic Chodziez records there's a priest who omits the 'w' in all of the 'traditional Polish' name endings: Karowski becomes Karoski, Wisniewski becomes Wisnieski. In the lutheran Koronowo records I'm doing now
Haas becomes Hass, Hasse and Haass. But to 'correct' them ourselves means that the database becomes innacurate - what if the surname was actually meant to be Hasse?

A wildcard search will pick up on these differences, and that's probably what will be used eventually. But it's up to Lukasz, who's also got things like Real Life to deal with aside from this project. (Note: currently lacking much of a Real Life is why I'm able to do so much indexing).

I've found that using Google search with the words (quotations included) "Poznan Project results" plus the surname you're looking for turns up good results (for a good example, here's one of mine: (http://www.google.com/search?q=%22Poznan+Project+results%22+Karowski&hl=en&hs=vfF&lr=&client=opera&rls=en&filter=0 ) then you can try any new found names that way again or try via the first names index. It would also be handy to try any different spellings of a surname that you have, especially if you get a hit on the original. And finally, please remember that it's only a
starting point - if you find a matching couple then your best bet is to make note of the place the record came from and go and order the film. I can promise you there's a lot more in the original record than just what is in the online index.

Elyssa

wrote:

> Subject: Re: [POSEN] Lukasz's Marriage Indexing project
> Date: Thu, 15 Sep 2005 08:27:20 -0500
> From: "Doug Plowman" <>
> To:
>
> Hello:
> I went to the website and accessed the database. It appears that one can
> only search the database by using the first name and not the surname. Is
> this correct. If this is the case, the database needs a search engine,
> preferably for surname with a wild card search. Just using the first name is
> a difficult and time consuming search.
>
> Doug
> ______________________________
>
> I'm not sure of the details, but Lukasz has some reasons for doing it this way as a temporary measure. I've looked at the extraction template the volunteers use, and it is problematic, in that it uses more than one row for each entry, making conversion to a searchable database difficult. But the same problem would be present for creating the existing Given Name Browsable list, so I presume that a lot of hand work is needed to take the submitted extractions and get them ready for use.
>
> I think (but don't know) that Lukasz does intend to make it searchable in the future. If Google indexes the pages in the future, you'll be able to use the "site:http://www.man.poznan.pl/~bielecki/proj/"; search term, with the surname, to search the existing lists. But that hasn't happened yet.
>
> James

This thread: