Hi Richard, Dave has already answered the objection with regards to the periods and I'll let that one continue on that thread for the moment ... I'll address the other points within the email.. [email protected] wrote: > > Hello, > Okay, I'll add my thoughts (again). > No, I don't believe the period should be there, except, possibly, for > initials. > It slows down transcription and isn't correct anyway. (Do you think your > ancestors-- those who could write-- signed their name with a period after the > first name?) > That said, the entire point (no pun intended) would be mooted if David, or > one of the programmers, would add two or three lines of code to normalize > forenames as is already done with first names and district names. The > periods would be stripped out during the match logic. The simple point is that we have limited resource otherwise we would look at normalising firstnames. It isnot possible for you to say whether it is one line or 40 lines of code to achieve this wihtout eeing our code and knowing the implications of it. We are aware of the requirement and no doubt will achieve it at some point in the future. We have some things which are currently higher priority taking our time up and we do need to stick to a priority list. The district name normalisation takes a lot of work to keep up to date, although it is now reasonably complete... > Another point-- there's that word again-- on this matter is one I have also > raised in the past. For those syndicates which are double-entering names (as > I believe they SHOULD be doing, but that's another discussion), Whether a syndicate chooses to double enter within their own syndicate is their choice but it will NOT be considered as a double keying from the point of view of FreeBMD. > if one > transcriber enters a forename with a period and another transcribes without, > that counts as a 'no match.' > David, respectfully, I disagree to some extent with your statement that we > shouldn't have syndicates making their own rules. It is far more important > that, WITHIN THE SYNDICATE, everyone is following the same rules. No, it really isn't. There are two ways to run a project like this. One is to allow the syndicates to double enter and guarantee the quality that route, the other is to centrally double enter and match. Because of the variation in source and quality of source we took the decision at the start to double key and match centrally. In that scenario it is very important that the rules that affect matching are central and not distributed amongst Syndicates otherwise the matcing software will continually throw up matches between the two sets of double keyed data. I do understand that IF we double keyed witin syndicates then your point would be true but we are 18 million records into the project and are not going to change now :) > Now, on another matter. Some weeks ago I sent a note asking that entries for > London C and London C. be treated as equivalent. I never received a direct > answer, but judging from the fact that they are still treated as different as > of the current update I gather the answer was 'NO!" No, the answer would be a yes .. but we have to get around todealing with it. It requires a change to a file in our CVS repository and then that code moved live. It will then be available at the rebuild following that. As I said earlier, the problem is one of resource. With holidays and Dave's car crash and the like this has been restricted recently. > This leads to two further questions > 1) Why aren't they treated as aliases of one another? Because we need to make them that way and we haven't added it as one yet... > 2) What is the proper forum for making such a request? I have several more. The best forum is currently to send it to [email protected] however, and Dave will correct me if I am wrong because I haven't looked at the code for a while, when we do a rebuild, all districts that are not in the current matchinglist are pulled out and identified. It is a task we do to then identify the relevant real district for normalisation. This means that we will catch new districts as mis-spellings appear. I know this email sounds like I am rebuffing you on each point, that is not the intention ... your feedback is important but you do also need to take into account that we have limited time in sorting things out :) And. of course... if you have Perl and Unix skills .. feel free to join the programming team ! :)) Cheers Graham > Anyone who wants to see a few should call up marriages, first quarter, 1867, > surname JONES. You'll see several instances where the district names are > obviously the same but are not treated as such. > > Regards, > Rick Elliott > > ============================== > Ancestry.com Genealogical Databases > http://www.ancestry.com/rd/rwlist2.asp > Search over 2500 databases with one easy query!

In message <[email protected]>, Graham Hart <[email protected]> writes >Hi Richard, > > >Dave has already answered the objection with regards to the periods and >I'll let that one continue on that thread for the moment ... > >I'll address the other points within the email.. > >[email protected] wrote: [snip] >> 2) What is the proper forum for making such a request? I have several more. > >The best forum is currently to send it to [email protected] >however, and Dave will correct me if I am wrong because I haven't looked >at the code for a while, when we do a rebuild, all districts that are >not in the current matchinglist are pulled out and identified. It is a >task we do to then identify the relevant real district for >normalisation. This means that we will catch new districts as >mis-spellings appear. Can you expand on this or is it explained anywhere on the site? I read that as meaning that all variations are treated as one - which seems like common sense to me. >I know this email sounds like I am rebuffing you on each point, that is >not the intention ... your feedback is important but you do also need to >take into account that we have limited time in sorting things out :) > >And. of course... if you have Perl and Unix skills .. feel free to join >the programming team ! :)) If I had, I would )-: -- Philip Powell Looking north across the Derwent Valley and Northumberland to The Cheviot