David Gray wrote: > > Dave, > I have seen a similar thread to this one in the archives from last year and > nothing has been done to make the search work since then. Indeed. The search *will* be enhanced. It is just that (at present) enhancements to the search facilities are lower priority than enhancements that aid us getting more records on-line. There are major work items that involve changes that users will never see. We cannot put these items off, because without them FreeBMD simply will not scale beyond 20 Million records. As resource becomes available, the search enhancements *will* happen. > The data did not originate with a period and so is being corrupted by > including the period, not preserved. That rather depends upon what you consider to be "the data". If you consider the original certificates to be the data then you have a point, but from a FreeBMD point of view, the indexes are "the data".... > You said the other day that you had no profound objection to the change > "*IF* we are sure that we can make the instruction > unambiguous". > > What could be more unambiguous than "Do not transcribe the period after the > forename"? People are more likely to follow this instruction than the > alternative. 1) We have to deal with people who have been transcribing it up to now, who will e-mail us bemoaning this change and wanting to know if they have to go through 20,000 entries removing the stray punctuation. 2) It extends the list of exceptions to the "type what you see" rule 3) We can only say this if we are *convinced* that there will NEVER be a case where there is some significance to the period 4) We need to decide whether the period should be retained after an initial. > You also said: > > 1) Discussion on changing the way things are done should take place on > the DISCUSS list rather than here. > 2) If what we lay down does need changing, it is very important that > it is discussed so that we can establish a common policy for all > syndicates. It is extremely damaging if one syndicate decides that > they will establish their own rules separate from the rest. > > >From giving the impression that you would consider the change if people > wanted it, your most recent response gives the impression that you are not > prepared to consider a change, or perhaps I have misread it. If you are not > prepared to consider a change have we wasted our time in discussing it? You misread it! I (and the final decision is not just mine!) am quite happy to consider a change, but the criteria is not that "people want it". The criteria is (and MUST be) that the change makes transcribing easier without any potential harm to the integrity of the data. There are 3 Questions that must be considered; 1) Is this easier for transcribers 2) Are we certain that the period is never significant 3) Are we certain that transcribers will not misinterpret this rule in any way If the answer to all 3 is "yes" then we can relax the rule (note that we don't reverse the rule, because we would then move to a policy that the period is optional and deal with it in code. I suggest that the answer to question 1 is "yes", and that we need to take a very good look at a wide range of scans to reassure ourselves that we can answer yes to 2 Likewise, we need to ensure that we frame any advice to transcribers so that we are happy that the answer to 3 is "yes" I apologize if my apparent scepticism comes over as an unwillingness to consider change. It is nothing of the sort. It is by means of a healthy scepticism that we rigorously test change proposals for suitability. -- Dave Mayall

Dave Mayall wrote: > >You misread it! Thanks for that Dave. >> The search *will* be enhanced. It is just that (at present) enhancements >> to the search facilities are lower priority than enhancements that aid us >> getting more records on-line. Are you suggesting that the search facilities will allow for people to search for a name with or without a period after it? That would be extremely messy. Alternatively, will the search facilities not require a period to be input? In this case why transcribe the period in the first place? >>That rather depends upon what you consider to be "the data". If you consider >> the original certificates to be the data then you have a point, but from a FreeBMD >> point of view, the indexes are "the data".... If the indexes are the data, why do we not transcribe the period after the age in the death indexes or after the volume number? Or for that matter after the forename in the births? It is quite obvious that the first period in the line of separator dots is a full stop. Of course, we can't be entirely certain why the indexes were written or typed with a period after the forenames, but we can make an educated guess that the reason was because it was intended to be seen on the printed page, and not searched for using computers. The period is no more than an indicator that the words, or sentences or 'data' has finished, and is not part of the data itself. >>We have to deal with people who have been transcribing it up to now, who will > >e-mail us bemoaning this change and wanting to know if they have to go through 20,000 >>entries removing the stray punctuation. If I was a betting man, I would put money on the chance that there would be far more records transcribed without the period than with. :-) I have transcribed 19000 records without the period (without realising what the official line was, I hasten to add). >> There are 3 Questions that must be considered; >> 1) Is this easier for transcribers Absolutely!!! It has already been pointed out that those using SpeedBMD or WinBMD would find it extremely awkward to add the period and it would slow down the transcription rate. I can't speak for those who are not using these utilities, but I would have thought that it would be easier not to include the period as against including it. After all, on the average page, it would mean aprrox. an extra 375 keystrokes. >> 2) Are we certain that the period is never significant We can never be absolutely certain of anything. As we will all have seen when transcribing, there are some very strange names out there. :-) There may well be some parents who have been daft enough to insist that a full-stop be part of their child's name.:-). What is certain however is that the number would be infinitesimal compared to the vast number without. By considering the the period significant because a small number (perhaps none) has it as part of the name ignores the many millions that do not. >> 3) Are we certain that transcribers will not misinterpret this rule in any way It is difficult to see how it can be misinterpreted, but you can never be certain. >> I apologize if my apparent scepticism comes over as an unwillingness to consider change. >> It is nothing of the sort. It is by means of a healthy scepticism that we rigorously test >>change proposals for suitability. No appologies needed Dave. I would like you to know that I am one hundred percent committed to this project and I intend to see it through to the end (of myself or the project whichever comes first) :-) whatever decision you finally come to. I have a great respect for you and your colleagues for the tremendous work you put in and I am proud to be part of it. David Gray Heysham, Lancashire Norton AntiVirus protected

Quoting David Gray <[email protected]>: > Are you suggesting that the search facilities will allow for people to > search for a name with or without a period after it? That would be > extremely > messy. Alternatively, will the search facilities not require a period to > be > input? In this case why transcribe the period in the first place? Hypothetically, I am suggesting that the period will be excluded from the data for search purposes (but may well be retained for display purposes). One possible reason is that the period *may* have some significance that we don't know about. Of course, we can't be > entirely certain why the indexes were written or typed with a period > after > the forenames, but we can make an educated guess that the reason was > because > it was intended to be seen on the printed page, and not searched for > using > computers. The period is no more than an indicator that the words, or > sentences or 'data' has finished, and is not part of the data itself. It is the words "we can make an educated guess" that set the alarm bells ringing. When deciding what goes into the RAW data, there is no room for educated guesses. There is room for making educated guesses in building the database from that raw data. The reasoning is that if we say "don't transcribe that" and it later turns out that it was important, we can't do a thing to get it back. If we say transcribe it, then tell the program to ignore it, then should we ever find out that it is important we can just tell the program to stop ignoring it. > >>We have to deal with people who have been transcribing it up to > now, > who will > > >e-mail us bemoaning this change and wanting to know if they > have to > go through 20,000 > >>entries removing the stray punctuation. > > If I was a betting man, I would put money on the chance that there would > be > far more records transcribed without the period than with. :-) There will still be a lot of transcribers who have done it one way and will be up in arms because they think they have done it wrong and wasted many hours in the process. It doesn't matter how you try to explain it, it would mean a lot of effort soothing troubled brows. >I have > transcribed 19000 records without the period (without realising what > the > official line was, I hasten to add). > > >> There are 3 Questions that must be considered; > >> 2) Are we certain that the period is never significant > > We can never be absolutely certain of anything. As we will all have > seen > when transcribing, there are some very strange names out there. :-) > There > may well be some parents who have been daft enough to insist that a > full-stop be part of their child's name.:-). I was thinking of the possibility that "." may indicate that there are other forenames that have been truncated from the index (as an example) > >> 3) Are we certain that transcribers will not misinterpret this > rule > in any way > > It is difficult to see how it can be misinterpreted, but you can never > be > certain. Quite so. Some people can misinterpret anything. FWIW, I am at present somewhat persuaded of the merits of this argument, and I'm going to spend a couple of days looking at some images. This is despite the fact that life would be easier for me if I stuck to the present line, and told those who were advised to omit the periods to complain to those who gave them the advice :-) -- Dave Mayall ------------------------------------------------- This mail sent through UK Online webmail