Phil   Coordinators were advised there was a problem with the report dated 23 March and it was quickly taken down.   Details of this new reporting system was only communicated to coordinator in the first instance so that hopefully and bugs could be reported back to improve the criteria used to identify possible errant files.   Unfortunately a number of transcribers come up with their own novel ideas on inserting #COMMENT or THEORY lines, if they did it correctly then there would have been no need for the report to be produced   A lot of effort is being taken by my colleague who looks after this side of FreeBMD to identify possible errors in files. Whilst the report of 23 March included files which weren't proper to be include on the flip side they included many files which were erroneous and would have been hidden away except for the fact the report system picked them up.   Emailing  QA-Coordinator is the correct approach where correctly transcribed files are erroneously shown in the report.   Perhaps you can advise exactly how you would 100% correctly identify all erroneous files regarding the use of #COMMENT or #THEORY lines but at the same time exclude all non erroneous files?   Allan Raymond    From: philip clarke <pr.clarke@talktalk.net> >To: 'Allan Raymond' <allan_raymond@btinternet.com>; freebmd-admins@rootsweb.com >Sent: Saturday, 24 March 2012, 9:30 >Subject: RE: Suspect File List > >Allen, > >I had five files listed as suspect on the first list. I checked them and >agreed that I'd made errors, corrected then and resubmitted them prior to >Handoff. They will be updated in time, so I have no problems with them >reappearing in the second list. They will (should) disappear after the >database has been updated. > >However, in the second list, the one dated 23/3/12, I had about 315 files of >mine listed, which is about 75% of what I've submitted. So I started to >check them newest first. I don't accept that the first four were invalid, >but I could be wrong, so I emailed QA-cood as stated in the list, stating >that I regarded them as valid. > >On that limited sample of four files, it looked like valid statements such >as #COMMENT handwritten amendment of crossed out entry, etc, were being >tagged as erroneous. Possibly a #COMMENT(3) handwritten addition of surname. > > >If I do have another 311 files with erroneous use of #COMMENT in them I will >gladly correct them, but why should I have to manual check that number of >files just to validate the FreeBMD software. > >The facility to check files for invalid use of #COMMENT lines is incredibly >useful/valuable to the project. I fully accept its value. However, it aught >to be reasonably reliable/accurate. > >Philip > >  > >-----Original Message----- >From: Allan Raymond [mailto:allan_raymond@btinternet.com] >Sent: 23 March 2012 15:37 >To: freebmd-admins@rootsweb.com >Subject: Re: Suspect File List > >Phil >  >To some extent this report is being tested before making it live. >  >Details of the new report was sent to Coordinators only on 16 March with >various advice given on the purpose of the report and what to do if files >such as yours were shown in error. I'm not aware the report was >generally made known to our transcribers, how did you get to know about the >report? >  >We have identified some of our transcribers are not following >the instructions on how to deal with alias or alternative names and this >report was produced for that reason. Whilst the report include some files >which are compliant with our instructions and hence are shown in the report >in error, these are far outweighed by the number of files which are >correctly included because they are errant. >  >The size of the report being so large and causing problems has been >mentioned by some of coordinators (myself included) seems to be due to the >type of browser being used. This should be a short term problem once the >number of files in the report start reducing due to them being corrected or >excluded if they were shown in the report in error. >  >Allan Raymond     >  >  >  > >From: Phil <phil.osbourn@ntlworld.com> >>To: freebmd-admins@rootsweb.com >>Sent: Friday, 23 March 2012, 9:03 >>Subject: Suspect File List >> >> >>  I    have  just  checked  the  comments-suspect  file  list  at >>  [1]http://www.freebmd.org.uk/SuspectFilesComments.html and find that it >>  erroneously includes all my files with correct alias comments; i.e. it >has >>  picked  out all my "#COMMENT(2)" lines and included the message "Alias >>  comment on only one entry at line". >>  Not only that but the page is so huge that it took ages to load and >crashed >>  my browser on the first load attempt. Aren't changes tested before making >>  them live? >>  Phil >> >> >> > > > > >

This is a general plea to all our transcribers regarding the Suspect File list at: http://www.freebmd.org.uk/SuspectFilesComments.html   Please do not start checking your files against the list or even run the report until advised to the contrary by your Coordinator. The file is quite large and it would be more appropriate for Coordinators to carry out random check of files within their Syndicate to see if the report is significantly reporting more erroneous files than non erroneous files in the report.   Personally I have just checked 7 files at random from different transcribers and from different syndicates and the report correctly identified the files were erroneous.   If transcribers wish to ignore this advice then any abortive work they spend in checking their files shouldn't be blamed on the report.   Time precludes me at the moment in checking more random files but I will continue this exercise later today and supply some of the problems the report is flagging up by giving some examples..   Allan Raymond FreeBMD Coordinator of Syndicates   From: Allan Raymond <allan_raymond@btinternet.com> >To: philip clarke <pr.clarke@talktalk.net>; "freebmd-admins@rootsweb.com" <freebmd-admins@rootsweb.com> >Sent: Saturday, 24 March 2012, 23:01 >Subject: Re: Suspect File List > >Phil >  >Coordinators were advised there was a problem with the report dated 23 March and it was quickly taken down. >  >Details of this new reporting system was only communicated to coordinator in the first instance so that hopefully and bugs could be reported back to improve the criteria used to identify possible errant files. >  >Unfortunately a number of transcribers come up with their own novel ideas on inserting #COMMENT or THEORY lines, if they did it correctly then there would have been no need for the report to be produced >  >A lot of effort is being taken by my colleague who looks after this side of FreeBMD to identify possible errors in files. Whilst the report of 23 March included files which weren't proper to be include on the flip side they included many files which were erroneous and would have been hidden away except for the fact the report system picked them up. >  >Emailing  QA-Coordinator is the correct approach where correctly transcribed files are erroneously shown in the report. >  >Perhaps you can advise exactly how you would 100% correctly identify all erroneous files regarding the use of #COMMENT or #THEORY lines but at the same time exclude all non erroneous files? >  >Allan Raymond  >  > >From: philip clarke <pr.clarke@talktalk.net> >>To: 'Allan Raymond' <allan_raymond@btinternet.com>; freebmd-admins@rootsweb.com >>Sent: Saturday, 24 March 2012, 9:30 >>Subject: RE: Suspect File List >> >>Allen, >> >>I had five files listed as suspect on the first list. I checked them and >>agreed that I'd made errors, corrected then and resubmitted them prior to >>Handoff. They will be updated in time, so I have no problems with them >>reappearing in the second list. They will (should) disappear after the >>database has been updated. >> >>However, in the second list, the one dated 23/3/12, I had about 315 files of >>mine listed, which is about 75% of what I've submitted. So I started to >>check them newest first. I don't accept that the first four were invalid, >>but I could be wrong, so I emailed QA-cood as stated in the list, stating >>that I regarded them as valid. >> >>On that limited sample of four files, it looked like valid statements such >>as #COMMENT handwritten amendment of crossed out entry, etc, were being >>tagged as erroneous. Possibly a #COMMENT(3) handwritten addition of surname. >> >> >>If I do have another 311 files with erroneous use of #COMMENT in them I will >>gladly correct them, but why should I have to manual check that number of >>files just to validate the FreeBMD software. >> >>The facility to check files for invalid use of #COMMENT lines is incredibly >>useful/valuable to the project. I fully accept its value. However, it aught >>to be reasonably reliable/accurate. >> >>Philip >>

Allan, Thank you for your detailed reply. In reverse order: I'm sorry, I don't know how to code that search, but what I can do is check parts of an output list that someone else produces. The latest list only has five suspect files against my name, I've previously corrected them, and they will likely appear on the system when the current maintenance cycle is complete. So there is nothing for me to do at present. If there had been a reasonable number of suspect files on the new list I would have happily rechecked them and responded according (see below). However, I would also like to respond to your comments about transcribers' involvement, I don't regard taking any personal action on files listed as "suspect" as time wasting. If the files had been "wrong" I would have corrected them; if the reason for flagging them as suspect was "unclear" I would have asked for clarification, and if they had been "wrong flagged", that might have highlight a QA of suspect files coding problem. All three actions would have resulted in a small improvement in quality. I'm fully supportive of your colleague (Barry) making valiant efforts to improve quality and it was not my intention to criticise the process of improving quality of data, but I can't understand why you wish to exclude transcribers from improving quality. The message seems to be only post holders and coordinators have a role to play. Perhaps that is part of the problem and not part of the solution. I must admit, I did not fully understand the instructions on what to do with a "Smith or Jones" name for several months when I started transcribing 18 months ago. To clarify it was obvious what to do with a "Brown or Smith or Jones" name but not a "Smith or Jones" name. I did what I though the instructions required me to do: not "novel ideas" on the use of #COMMENT, as was suggested. Having later found my errors I started correcting the files three or four per day on top of what I normally transcribe. I supplied the extra resource, so my coordinator was not suffering from any misallocation of resource. I was not asked to do this by the QA coordinator and/or the coordinator, and if I'd waited to be told to make the changes, those errors would still be on the system - since those files are not listed as "suspect" I assume that that they do not have novel ideas on inserting #COMMENT or #THEORY lines. Philip -----Original Message----- From: Allan Raymond [mailto:allan_raymond@btinternet.com] Sent: 24 March 2012 23:02 To: philip clarke; freebmd-admins@rootsweb.com Subject: Re: Suspect File List Phil   Coordinators were advised there was a problem with the report dated 23 March and it was quickly taken down.   Details of this new reporting system was only communicated to coordinator in the first instance so that hopefully and bugs could be reported back to improve the criteria used to identify possible errant files.   Unfortunately a number of transcribers come up with their own novel ideas on inserting #COMMENT or THEORY lines, if they did it correctly then there would have been no need for the report to be produced   A lot of effort is being taken by my colleague who looks after this side of FreeBMD to identify possible errors in files. Whilst the report of 23 March included files which weren't proper to be include on the flip side they included many files which were erroneous and would have been hidden away except for the fact the report system picked them up.   Emailing  QA-Coordinator is the correct approach where correctly transcribed files are erroneously shown in the report.   Perhaps you can advise exactly how you would 100% correctly identify all erroneous files regarding the use of #COMMENT or #THEORY lines but at the same time exclude all non erroneous files?   Allan Raymond    From: philip clarke <pr.clarke@talktalk.net> >To: 'Allan Raymond' <allan_raymond@btinternet.com>; freebmd-admins@rootsweb.com >Sent: Saturday, 24 March 2012, 9:30 >Subject: RE: Suspect File List > >Allen, > >I had five files listed as suspect on the first list. I checked them and >agreed that I'd made errors, corrected then and resubmitted them prior to >Handoff. They will be updated in time, so I have no problems with them >reappearing in the second list. They will (should) disappear after the >database has been updated. > >However, in the second list, the one dated 23/3/12, I had about 315 files of >mine listed, which is about 75% of what I've submitted. So I started to >check them newest first. I don't accept that the first four were invalid, >but I could be wrong, so I emailed QA-cood as stated in the list, stating >that I regarded them as valid. > >On that limited sample of four files, it looked like valid statements such >as #COMMENT handwritten amendment of crossed out entry, etc, were being >tagged as erroneous. Possibly a #COMMENT(3) handwritten addition of surname. > > >If I do have another 311 files with erroneous use of #COMMENT in them I will >gladly correct them, but why should I have to manual check that number of >files just to validate the FreeBMD software. > >The facility to check files for invalid use of #COMMENT lines is incredibly >useful/valuable to the project. I fully accept its value. However, it aught >to be reasonably reliable/accurate. > >Philip > >  > >-----Original Message----- >From: Allan Raymond [mailto:allan_raymond@btinternet.com] >Sent: 23 March 2012 15:37 >To: freebmd-admins@rootsweb.com >Subject: Re: Suspect File List > >Phil >  >To some extent this report is being tested before making it live. >  >Details of the new report was sent to Coordinators only on 16 March with >various advice given on the purpose of the report and what to do if files >such as yours were shown in error. I'm not aware the report was >generally made known to our transcribers, how did you get to know about the >report? >  >We have identified some of our transcribers are not following >the instructions on how to deal with alias or alternative names and this >report was produced for that reason. Whilst the report include some files >which are compliant with our instructions and hence are shown in the report >in error, these are far outweighed by the number of files which are >correctly included because they are errant. >  >The size of the report being so large and causing problems has been >mentioned by some of coordinators (myself included) seems to be due to the >type of browser being used. This should be a short term problem once the >number of files in the report start reducing due to them being corrected or >excluded if they were shown in the report in error. >  >Allan Raymond     >  >  >  > >From: Phil <phil.osbourn@ntlworld.com> >>To: freebmd-admins@rootsweb.com >>Sent: Friday, 23 March 2012, 9:03 >>Subject: Suspect File List >> >> >>  I    have  just  checked  the  comments-suspect  file  list  at >>  [1]http://www.freebmd.org.uk/SuspectFilesComments.html and find that it >>  erroneously includes all my files with correct alias comments; i.e. it >has >>  picked  out all my "#COMMENT(2)" lines and included the message "Alias >>  comment on only one entry at line". >>  Not only that but the page is so huge that it took ages to load and >crashed >>  my browser on the first load attempt. Aren't changes tested before making >>  them live? >>  Phil >> >> >> > > > > >

Phil I also wouldn't know how to code the search criteria to ensure we only capture only erroneous files but exclude those which are correct. As you mentioned they aught to be reasonably reliable/accurate I thought you had an inkling of how to disregard correct entries from appearing in the report? I think what I had said was that the report was only conveyed to coordinators so that they feed back any bugs in the reporting system. It was you who mentioned you didn't wish to have to manually check through a large number of files just to validate the FreeBMD software. I was supporting you in this by saying it was why it was first sent to Coordinators so they could carry out a random check of files to see how valid the reporting system was in picking up only erroneous files. Once any bugs had been ironed out the report could be freely available to transcribers so they wouldn't have to check through large numbers of possibly correct files. I would hope later today to provide details of some files which contain novel ideas on insertion of the #COMMENT or #THEORY, I have no reason to suppose any of these files will be yours since I will just look at a random selection of files. Allan Raymond ----- Original Message ----- From: philip clarke To: 'Allan Raymond' ; freebmd-admins@rootsweb.com Sent: Sunday, March 25, 2012 12:41 PM Subject: RE: Suspect File List Allan, Thank you for your detailed reply. In reverse order: I'm sorry, I don't know how to code that search, but what I can do is check parts of an output list that someone else produces. The latest list only has five suspect files against my name, I've previously corrected them, and they will likely appear on the system when the current maintenance cycle is complete. So there is nothing for me to do at present. If there had been a reasonable number of suspect files on the new list I would have happily rechecked them and responded according (see below). However, I would also like to respond to your comments about transcribers' involvement, I don't regard taking any personal action on files listed as "suspect" as time wasting. If the files had been "wrong" I would have corrected them; if the reason for flagging them as suspect was "unclear" I would have asked for clarification, and if they had been "wrong flagged", that might have highlight a QA of suspect files coding problem. All three actions would have resulted in a small improvement in quality. I'm fully supportive of your colleague (Barry) making valiant efforts to improve quality and it was not my intention to criticise the process of improving quality of data, but I can't understand why you wish to exclude transcribers from improving quality. The message seems to be only post holders and coordinators have a role to play. Perhaps that is part of the problem and not part of the solution. I must admit, I did not fully understand the instructions on what to do with a "Smith or Jones" name for several months when I started transcribing 18 months ago. To clarify it was obvious what to do with a "Brown or Smith or Jones" name but not a "Smith or Jones" name. I did what I though the instructions required me to do: not "novel ideas" on the use of #COMMENT, as was suggested. Having later found my errors I started correcting the files three or four per day on top of what I normally transcribe. I supplied the extra resource, so my coordinator was not suffering from any misallocation of resource. I was not asked to do this by the QA coordinator and/or the coordinator, and if I'd waited to be told to make the changes, those errors would still be on the system - since those files are not listed as "suspect" I assume that that they do not have novel ideas on inserting #COMMENT or #THEORY lines. Philip -----Original Message----- From: Allan Raymond [mailto:allan_raymond@btinternet.com] Sent: 24 March 2012 23:02 To: philip clarke; freebmd-admins@rootsweb.com Subject: Re: Suspect File List Phil Coordinators were advised there was a problem with the report dated 23 March and it was quickly taken down. Details of this new reporting system was only communicated to coordinator in the first instance so that hopefully and bugs could be reported back to improve the criteria used to identify possible errant files. Unfortunately a number of transcribers come up with their own novel ideas on inserting #COMMENT or THEORY lines, if they did it correctly then there would have been no need for the report to be produced A lot of effort is being taken by my colleague who looks after this side of FreeBMD to identify possible errors in files. Whilst the report of 23 March included files which weren't proper to be include on the flip side they included many files which were erroneous and would have been hidden away except for the fact the report system picked them up. Emailing QA-Coordinator is the correct approach where correctly transcribed files are erroneously shown in the report. Perhaps you can advise exactly how you would 100% correctly identify all erroneous files regarding the use of #COMMENT or #THEORY lines but at the same time exclude all non erroneous files? Allan Raymond From: philip clarke <pr.clarke@talktalk.net> >To: 'Allan Raymond' <allan_raymond@btinternet.com>; freebmd-admins@rootsweb.com >Sent: Saturday, 24 March 2012, 9:30 >Subject: RE: Suspect File List > >Allen, > >I had five files listed as suspect on the first list. I checked them and >agreed that I'd made errors, corrected then and resubmitted them prior to >Handoff. They will be updated in time, so I have no problems with them >reappearing in the second list. They will (should) disappear after the >database has been updated. > >However, in the second list, the one dated 23/3/12, I had about 315 files of >mine listed, which is about 75% of what I've submitted. So I started to >check them newest first. I don't accept that the first four were invalid, >but I could be wrong, so I emailed QA-cood as stated in the list, stating >that I regarded them as valid. > >On that limited sample of four files, it looked like valid statements such >as #COMMENT handwritten amendment of crossed out entry, etc, were being >tagged as erroneous. Possibly a #COMMENT(3) handwritten addition of surname. > > >If I do have another 311 files with erroneous use of #COMMENT in them I will >gladly correct them, but why should I have to manual check that number of >files just to validate the FreeBMD software. > >The facility to check files for invalid use of #COMMENT lines is incredibly >useful/valuable to the project. I fully accept its value. However, it aught >to be reasonably reliable/accurate. > >Philip > > > >-----Original Message----- >From: Allan Raymond [mailto:allan_raymond@btinternet.com] >Sent: 23 March 2012 15:37 >To: freebmd-admins@rootsweb.com >Subject: Re: Suspect File List > >Phil > >To some extent this report is being tested before making it live. > >Details of the new report was sent to Coordinators only on 16 March with >various advice given on the purpose of the report and what to do if files >such as yours were shown in error. I'm not aware the report was >generally made known to our transcribers, how did you get to know about the >report? > >We have identified some of our transcribers are not following >the instructions on how to deal with alias or alternative names and this >report was produced for that reason. Whilst the report include some files >which are compliant with our instructions and hence are shown in the report >in error, these are far outweighed by the number of files which are >correctly included because they are errant. > >The size of the report being so large and causing problems has been >mentioned by some of coordinators (myself included) seems to be due to the >type of browser being used. This should be a short term problem once the >number of files in the report start reducing due to them being corrected or >excluded if they were shown in the report in error. > >Allan Raymond > > > > >From: Phil <phil.osbourn@ntlworld.com> >>To: freebmd-admins@rootsweb.com >>Sent: Friday, 23 March 2012, 9:03 >>Subject: Suspect File List >> >> >> I have just checked the comments-suspect file list at >> [1]http://www.freebmd.org.uk/SuspectFilesComments.html and find that it >> erroneously includes all my files with correct alias comments; i.e. it >has >> picked out all my "#COMMENT(2)" lines and included the message "Alias >> comment on only one entry at line". >> Not only that but the page is so huge that it took ages to load and >crashed >> my browser on the first load attempt. Aren't changes tested before making >> them live? >> Phil >> >> >> > > > > >

Allan, Unfortunately, my confident words came back and bit me. Two of my suspect files contained the lines of the type #COMMEMT(2) entry reads Smith or Jones for Spouse's name, sandwiched between a Smith and a Jones data line, i.e. "Comment" spelt with three Ms and no N. Possibly, these two uses seem to fall into the heading of novel ideas on insertion of the #COMMENT or #THEORY. One has now file has now been corrected, the other will be soon. I also found some false positives, i.e. of the form: --, John, Jones, Birmingham, 9c, 1066 --, John, Jones-Smith, 9c, 1066 NOT: "--, John, Jones or Jones-Smith, Birmingham, 9C, 1066" keyed as seen or keyed twice with no COMMENT(1) or COMMENT(2) line in between. Since the computer can't read the scan, only the transcripts, it picked up a "valid error" - even if it was in these particular cases wrong. Philip -----Original Message----- From: Allan Raymond [mailto:allan_raymond@btinternet.com] Sent: 25 March 2012 16:43 To: freebmd-admins@rootsweb.com Subject: Re: Suspect File List Phil I also wouldn't know how to code the search criteria to ensure we only capture only erroneous files but exclude those which are correct. As you mentioned they aught to be reasonably reliable/accurate I thought you had an inkling of how to disregard correct entries from appearing in the report? I think what I had said was that the report was only conveyed to coordinators so that they feed back any bugs in the reporting system. It was you who mentioned you didn't wish to have to manually check through a large number of files just to validate the FreeBMD software. I was supporting you in this by saying it was why it was first sent to Coordinators so they could carry out a random check of files to see how valid the reporting system was in picking up only erroneous files. Once any bugs had been ironed out the report could be freely available to transcribers so they wouldn't have to check through large numbers of possibly correct files. I would hope later today to provide details of some files which contain novel ideas on insertion of the #COMMENT or #THEORY, I have no reason to suppose any of these files will be yours since I will just look at a random selection of files. Allan Raymond ----- Original Message ----- From: philip clarke To: 'Allan Raymond' ; freebmd-admins@rootsweb.com Sent: Sunday, March 25, 2012 12:41 PM Subject: RE: Suspect File List Allan, Thank you for your detailed reply. In reverse order: I'm sorry, I don't know how to code that search, but what I can do is check parts of an output list that someone else produces. The latest list only has five suspect files against my name, I've previously corrected them, and they will likely appear on the system when the current maintenance cycle is complete. So there is nothing for me to do at present. If there had been a reasonable number of suspect files on the new list I would have happily rechecked them and responded according (see below). However, I would also like to respond to your comments about transcribers' involvement, I don't regard taking any personal action on files listed as "suspect" as time wasting. If the files had been "wrong" I would have corrected them; if the reason for flagging them as suspect was "unclear" I would have asked for clarification, and if they had been "wrong flagged", that might have highlight a QA of suspect files coding problem. All three actions would have resulted in a small improvement in quality. I'm fully supportive of your colleague (Barry) making valiant efforts to improve quality and it was not my intention to criticise the process of improving quality of data, but I can't understand why you wish to exclude transcribers from improving quality. The message seems to be only post holders and coordinators have a role to play. Perhaps that is part of the problem and not part of the solution. I must admit, I did not fully understand the instructions on what to do with a "Smith or Jones" name for several months when I started transcribing 18 months ago. To clarify it was obvious what to do with a "Brown or Smith or Jones" name but not a "Smith or Jones" name. I did what I though the instructions required me to do: not "novel ideas" on the use of #COMMENT, as was suggested. Having later found my errors I started correcting the files three or four per day on top of what I normally transcribe. I supplied the extra resource, so my coordinator was not suffering from any misallocation of resource. I was not asked to do this by the QA coordinator and/or the coordinator, and if I'd waited to be told to make the changes, those errors would still be on the system - since those files are not listed as "suspect" I assume that that they do not have novel ideas on inserting #COMMENT or #THEORY lines. Philip -----Original Message----- From: Allan Raymond [mailto:allan_raymond@btinternet.com] Sent: 24 March 2012 23:02 To: philip clarke; freebmd-admins@rootsweb.com Subject: Re: Suspect File List Phil Coordinators were advised there was a problem with the report dated 23 March and it was quickly taken down. Details of this new reporting system was only communicated to coordinator in the first instance so that hopefully and bugs could be reported back to improve the criteria used to identify possible errant files. Unfortunately a number of transcribers come up with their own novel ideas on inserting #COMMENT or THEORY lines, if they did it correctly then there would have been no need for the report to be produced A lot of effort is being taken by my colleague who looks after this side of FreeBMD to identify possible errors in files. Whilst the report of 23 March included files which weren't proper to be include on the flip side they included many files which were erroneous and would have been hidden away except for the fact the report system picked them up. Emailing QA-Coordinator is the correct approach where correctly transcribed files are erroneously shown in the report. Perhaps you can advise exactly how you would 100% correctly identify all erroneous files regarding the use of #COMMENT or #THEORY lines but at the same time exclude all non erroneous files? Allan Raymond From: philip clarke <pr.clarke@talktalk.net> >To: 'Allan Raymond' <allan_raymond@btinternet.com>; freebmd-admins@rootsweb.com >Sent: Saturday, 24 March 2012, 9:30 >Subject: RE: Suspect File List > >Allen, > >I had five files listed as suspect on the first list. I checked them and >agreed that I'd made errors, corrected then and resubmitted them prior to >Handoff. They will be updated in time, so I have no problems with them >reappearing in the second list. They will (should) disappear after the >database has been updated. > >However, in the second list, the one dated 23/3/12, I had about 315 files of >mine listed, which is about 75% of what I've submitted. So I started to >check them newest first. I don't accept that the first four were invalid, >but I could be wrong, so I emailed QA-cood as stated in the list, stating >that I regarded them as valid. > >On that limited sample of four files, it looked like valid statements such >as #COMMENT handwritten amendment of crossed out entry, etc, were being >tagged as erroneous. Possibly a #COMMENT(3) handwritten addition of surname. > > >If I do have another 311 files with erroneous use of #COMMENT in them I will >gladly correct them, but why should I have to manual check that number of >files just to validate the FreeBMD software. > >The facility to check files for invalid use of #COMMENT lines is incredibly >useful/valuable to the project. I fully accept its value. However, it aught >to be reasonably reliable/accurate. > >Philip > > > >-----Original Message----- >From: Allan Raymond [mailto:allan_raymond@btinternet.com] >Sent: 23 March 2012 15:37 >To: freebmd-admins@rootsweb.com >Subject: Re: Suspect File List > >Phil > >To some extent this report is being tested before making it live. > >Details of the new report was sent to Coordinators only on 16 March with >various advice given on the purpose of the report and what to do if files >such as yours were shown in error. I'm not aware the report was >generally made known to our transcribers, how did you get to know about the >report? > >We have identified some of our transcribers are not following >the instructions on how to deal with alias or alternative names and this >report was produced for that reason. Whilst the report include some files >which are compliant with our instructions and hence are shown in the report >in error, these are far outweighed by the number of files which are >correctly included because they are errant. > >The size of the report being so large and causing problems has been >mentioned by some of coordinators (myself included) seems to be due to the >type of browser being used. This should be a short term problem once the >number of files in the report start reducing due to them being corrected or >excluded if they were shown in the report in error. > >Allan Raymond > > > > >From: Phil <phil.osbourn@ntlworld.com> >>To: freebmd-admins@rootsweb.com >>Sent: Friday, 23 March 2012, 9:03 >>Subject: Suspect File List >> >> >> I have just checked the comments-suspect file list at >> [1]http://www.freebmd.org.uk/SuspectFilesComments.html and find that it >> erroneously includes all my files with correct alias comments; i.e. it >has >> picked out all my "#COMMENT(2)" lines and included the message "Alias >> comment on only one entry at line". >> Not only that but the page is so huge that it took ages to load and >crashed >> my browser on the first load attempt. Aren't changes tested before making >> them live? >> Phil >> >> >> > > > > >