Thanks for bringing this topic up. The use of split memos or split CD parts, split place Detail parts has largely been from the evolution of TMG. It gave some flexibility while not changing the database schema. It became very constraining if the user wanted to output in 2 or 3 languages as each extra language meant that the 9 slots needed to be assigned specifically to particular language. In HRE the concept is to have "named" memos so the split TMG memo would be stored separately as "M1" , "M2" etc The user can also provide memos with user-defined meaningful names. Secondly, as a minus sign "-" can be the first character of a legitimate string value (as in LatLong or a time difference), the exclusion marker in HRE will be another character, probably tilda "~", but otherwise act like the TMG "-". RobinL On 04-May-17 07:50 AM, Michael J Hannah wrote: > John Cardinal asked: >> You quoted the word "generally" twice in statements >> that included Second Site... >> or does the "generally" apply only to TMG? > > That was only intended to apply to TMG. I have always admired how > consistently Second Site (and the TMG Utility) process such things. > Thanks yet again for these programs. > >> Second Site splits the text value... >> After splitting the text into parts, it removes leading >> and trailing spaces.... >> After parsing the text value as described above, >> but before using a part value, Second Site compares the >> leading characters to "-" and "--"... > > In a follow-up message John observed: >> I realized ... that the post applied specifically to the CD value. >> I looked at that code, ... the exclusion processing is different. >> The single- and double-exclusion is applied to the CD as a whole, >> and not to the parts of the CD... I suspect that code was written >> to mimic how TMG behaved at the time it was written. > > Oh dear. This whole combination of split parts and exclusion markers > is entirely too complex. <sigh> I haven't recently done much testing > of this, but I think? that TMG also has this difference between > excluded split memos and excluded split CDs, but reversed? I guess it > is not too surprising that the final version of TMG still has bugs in > this area. > > This just reinforces my strongly held belief that the number of users > who use split memos of any kind are limited, those who use split CDs > are even more rare, and those who use exclusion markers in a later > split part in either split memos or split CDs vanishingly few <grin>. > > Seems to me I have more testing to do for this section of my book. > I guess? I should say thank you? <sigh> > > Michael > The TMG archive is found here: > http://archiver.rootsweb.ancestry.com/th/index/TMG/ > Instructions on how to subscribe to TMG: > http://lists.rootsweb.ancestry.com/index/other/Software/TMG.html > ------------------------------- > To unsubscribe from the list, please send an email to > TMG-request@rootsweb.com with the word 'unsubscribe' without the > quotes in the subject and the body of the message -- Robin Lamacraft, Adelaide, Australia

robin lamacraft wrote, in part: > The storage and the visible to the user do not need to be the same thing. Robin, If your plan is for users to key "~" (or some other character) as the first character in the text field to indicate the value should be excluded, that's not a good design, even if the "~" is not stored in the text field in the database. You seemed to imply that the alternative--which you didn't describe in any detail--would result in UI clutter. I think that's a false dichotomy: there are other solutions that do not introduce clutter. Meanwhile, the "~" introduces clutter and other problems of its own. - The user must know (or find in help) how to hide the value because keying a "~" as the first character is not obvious from the UI. - The user must know (or find in help) how to specify a "~" as the first data character in the value when she does not want the "~" to trigger its "hide this value" meaning. TMG is rife with little rules like this and they contribute to its "difficult to learn" reputation. It's still my favorite genealogy program, but that doesn't mean it's perfect, and the "-" and "--" excluded data triggers are two of the imperfections. I'm sad to hear that HRE will feature similar behavior. John

I followed Lee's suggestion for the filters: Tag type... Abrev equals Census And Any Witness Surname equals [?] END I got an error message: function argument value, type, or count is invalid >From there I get other error messages until I have to close TMG with Task Manager. Looks like I will go with the Focus group and print a journal report with only the census tags. Nancy

I thought I would report failures in loading a project and in restoring a backup in case the information is useful to those who might understand it. I was able to restore from the previous backup, so all is well. The project failure: Project did not load automatically. When trying to load the project manually, the error was “… does not appear to be a valid PJC file”. The backup restore failure: Error was “DynaZip UnZip error: Zip file structure invalid (Can’t understand ZIP file, structure problem).” Using Windows 10 through VMWare on iMac. This is the first time I have been unable to restore from a backup. As this situation is unusual (at least to me), I thought I should report it. Chris Sackett

Robin, I agree with your decision to use separate, named "parts" for fields that can have multiple values. My GedSite program follows that model when it loads a GEDCOM file. For example, many GEDCOM tags can have text on the tag itself as well as text on a subordinate NOTE tag. GedSite has Memo[Text] and Memo[Note] variables for that purpose. There are other Memo[<part-name>] variations for certain GEDCOM structures. For excluded fields, I would not use a prefix character to record that attribute of a field. The database should have a separate field (separate from the text) for the excluded/not excluded property. Overall, I'd minimize what gets stored in text fields. It seems likely that you'd want to support some formatting codes, and you may also want to support hidden/private/sensitive text. That's probably the most I'd add as metadata that is stored in the text itself. TMG had many fields where there was structure inside the database field value. I suspect that was partly due to the state of database technology in the early 1990s. Using current technologies, putting a lot of structure inside a data field is not a good idea. John

If I understand correctly, mention was made that there aren't very many split CD users. I for one use them a lot in my citation entries, and without their seamless conversion, my data would be an impossible mess. I don't happen to use more than one language, but I'm nevertheless glad to know that HRE will have provision for split CDs. -----Original Message----- From: TMG [mailto:tmg-bounces+joanklince=earthlink.net@rootsweb.com] On Behalf Of robin lamacraft Sent: Wednesday, May 03, 2017 7:58 PM To: tmg@rootsweb.com Subject: Re: [TMG] No text for conditional [CD] Thanks for bringing this topic up. The use of split memos or split CD parts, split place Detail parts has largely been from the evolution of TMG. It gave some flexibility while not changing the database schema. It became very constraining if the user wanted to output in 2 or 3 languages as each extra language meant that the 9 slots needed to be assigned specifically to particular language. In HRE the concept is to have "named" memos so the split TMG memo would be stored separately as "M1" , "M2" etc The user can also provide memos with user-defined meaningful names. Secondly, as a minus sign "-" can be the first character of a legitimate string value (as in LatLong or a time difference), the exclusion marker in HRE will be another character, probably tilda "~", but otherwise act like the TMG "-". RobinL On 04-May-17 07:50 AM, Michael J Hannah wrote: > John Cardinal asked: >> You quoted the word "generally" twice in statements that included >> Second Site... >> or does the "generally" apply only to TMG? > > That was only intended to apply to TMG. I have always admired how > consistently Second Site (and the TMG Utility) process such things. > Thanks yet again for these programs. > >> Second Site splits the text value... >> After splitting the text into parts, it removes leading and trailing >> spaces.... >> After parsing the text value as described above, but before using a >> part value, Second Site compares the leading characters to "-" and >> "--"... > > In a follow-up message John observed: >> I realized ... that the post applied specifically to the CD value. >> I looked at that code, ... the exclusion processing is different. >> The single- and double-exclusion is applied to the CD as a whole, and >> not to the parts of the CD... I suspect that code was written to >> mimic how TMG behaved at the time it was written. > > Oh dear. This whole combination of split parts and exclusion markers > is entirely too complex. <sigh> I haven't recently done much testing > of this, but I think? that TMG also has this difference between > excluded split memos and excluded split CDs, but reversed? I guess it > is not too surprising that the final version of TMG still has bugs in > this area. > > This just reinforces my strongly held belief that the number of users > who use split memos of any kind are limited, those who use split CDs > are even more rare, and those who use exclusion markers in a later > split part in either split memos or split CDs vanishingly few <grin>. > > Seems to me I have more testing to do for this section of my book. > I guess? I should say thank you? <sigh> > > Michael > The TMG archive is found here: > http://archiver.rootsweb.ancestry.com/th/index/TMG/ > Instructions on how to subscribe to TMG: > http://lists.rootsweb.ancestry.com/index/other/Software/TMG.html > ------------------------------- > To unsubscribe from the list, please send an email to > TMG-request@rootsweb.com with the word 'unsubscribe' without the > quotes in the subject and the body of the message -- Robin Lamacraft, Adelaide, Australia The TMG archive is found here: http://archiver.rootsweb.ancestry.com/th/index/TMG/ Instructions on how to subscribe to TMG: http://lists.rootsweb.ancestry.com/index/other/Software/TMG.html ------------------------------- To unsubscribe from the list, please send an email to TMG-request@rootsweb.com with the word 'unsubscribe' without the quotes in the subject and the body of the message

Thanks for all of the suggestions. I tried to set up a filter like Lee said, but I had Prin1 first and census as the secondary filter. I must have done something wrong because TMG did not like it at all. I got an error message then I couldn't close it and couldn't get back to TMG, had to close it with Task Manager - so I asked for help. No Donald, I do not have my census tags set up like you anymore. I did at one time but wasn't happy with it. Sally thanks for the suggestion of using a focus group. I haven't used focus groups much and I forget about them. I can see the practical use of your suggestion. Thanks for all the help, Nancy On Behalf Of Donald Range A couple of added comments on Lee's suggestion: If you have set up your census tags as I have, with the head of household as Principal 1 and other household members as witnesses, the "Any Witness" filter will cause everyone in any household in a census tag with at least one person having the specified surname to be included in the report. The filter will include all name variations, including any Name-marr tags. So for example if you specify the Black surname a woman Mary Smith who marries Joe Black then later marries Sam Spade and has son John Spade and appears with her Spade family in a census tag will cause Mary, Sam, and John Spade to all be included in your Black report. If that's too broad for your purpose you could use "Principal 1" instead of "Any Witness" in the first term of the filter. Note that filtering for "Principal 1" will still cause households with a person who has a name variation such as Name-marr that matches your specified name to be included. Typically this would occur when a divorced or widowed woman with matching maiden name is heading a household and therefore listed as Principal 1. In the report options include the Date and Prin1 Last, Given fields as output columns. You can specify the sort order with Date first to have your report completely date ordered, or specify Prin1 Last, Given to be the first sort in order to have it grouped by person's name. Lee Hoffman/KY wrote: > > There are a couple of ways to get this -- 1. Run a List of Events > reports using the filter: > Tag Type ... Label =Equals CENSUS AND > Any Witness... Surname =Equals [?] END > Each time you run this report/filter the [?] will cause TMG to ask for > the search surname. > > > 2. Run a List of Witnesses report using the filter like: > > A> > Witness Is Not a > Principal AND > Tag Type ... Label =Equals CENSUS AND > Any Witness... Surname =Equals [?] END > Each time you run this report/filter the [?] will cause TMG to ask for > the search surname. > > -OR- > > B> > Witness Is a > Principal AND > Tag Type ... Label =Equals CENSUS AND > Any Witness... Surname =Equals [?] END > Each time you run this report/filter the [?] will cause TMG to ask for > the search surname. > > Note that the List of Witnesses report will either give you just > persons that are Principals or that are just Witnesses. The List of > Events report will give both Witnesses and Principals (that is all > witnesses). > > In all cases, set the sort order in the [Options...]=>Output Columns tab. > > Sally wrote > > Couldn't you just use a focus group of people surnamed Black? Then create a journal report including only census records. I use census records as part of my customized journal reports, > and they are in order by sort date. Exactly what I want!

Michael, Sorry to add to the software research pile! If you discover any differences between SS and TMG, please let me know. I am probably going to change SS to allow exclusion markers in the CD parts because I think TMG allows them. If there are other differences, I'll consider making additional changes. I will probably avoid any changes that would match unexpected and/or inconsistent split-field behavior in TMG, but I think there are less of those in TMG v9.05 than there were in prior versions. John

Michael, The process I described in my last reply applied to memo fields in general. I realized after writing it that the post applied specifically to the CD value. I looked at that code, and while the treatment is very similar--it uses the same functions to process the data--the exclusion processing is different. The single- and double-exclusion is applied to the CD as a whole, and not to the parts of the CD. I think I mentioned that in an earlier reply in this thread. I suspect that code was written to mimic how TMG behaved at the time it was written. John

Michael, You quoted the word "generally" twice in statements that included Second Site, for example, "TMG (and Second Site) will "generally" output correctly". Did you mean to imply that Second Site's treatment of split fields is not completely predictable, or does the "generally" apply only to TMG? If you think that Second Site is not predictable, please explain further, or point me to something on your site, because as far as I know, it is completely predictable. Second Site splits the text value (memo, or whatever) into parts based on the position of the "||" characters. It does not get confused by completely empty parts, so "a||||c" has "a" in part 1 and "c" in part 3. After splitting the text into parts, it removes leading and trailing spaces. So, these are equivalent: "a||||c" "a|| ||c" " a || || c " Ignore the quotes in the examples; those are there to make the start and end of the text field visible. In all the cases above, part 1 is "a", part 2 is empty, and part 3 is "c". After parsing the text value as described above, but before using a part value, Second Site compares the leading characters to "-" and "--", the single- and double-exclusion markers, respectively. Second Site ignores the part if it starts with "--". If it starts with "-", it excludes the part unless "Show Excluded Data" is checked. John

John Cardinal asked: > You quoted the word "generally" twice in statements > that included Second Site... > or does the "generally" apply only to TMG? That was only intended to apply to TMG. I have always admired how consistently Second Site (and the TMG Utility) process such things. Thanks yet again for these programs. > Second Site splits the text value... > After splitting the text into parts, it removes leading > and trailing spaces.... > After parsing the text value as described above, > but before using a part value, Second Site compares the > leading characters to "-" and "--"... In a follow-up message John observed: > I realized ... that the post applied specifically to the CD value. > I looked at that code, ... the exclusion processing is different. > The single- and double-exclusion is applied to the CD as a whole, > and not to the parts of the CD... I suspect that code was written > to mimic how TMG behaved at the time it was written. Oh dear. This whole combination of split parts and exclusion markers is entirely too complex. <sigh> I haven't recently done much testing of this, but I think? that TMG also has this difference between excluded split memos and excluded split CDs, but reversed? I guess it is not too surprising that the final version of TMG still has bugs in this area. This just reinforces my strongly held belief that the number of users who use split memos of any kind are limited, those who use split CDs are even more rare, and those who use exclusion markers in a later split part in either split memos or split CDs vanishingly few <grin>. Seems to me I have more testing to do for this section of my book. I guess? I should say thank you? <sigh> Michael

At 5/3/2017 13:37, nancy wright wrote >I want a report that shows the census records for each person. I want the >report for all persons surnamed BLACK, but later I would like to get this >report for persons of other surnames. >If possible I would like the census' to be listed in order. I don't know if >that can be done - I just use Census, not (for example) census1930, etc. There are a couple of ways to get this -- 1. Run a List of Events reports using the filter: Tag Type ... Label =Equals CENSUS AND Any Witness... Surname =Equals [?] END Each time you run this report/filter the [?] will cause TMG to ask for the search surname. 2. Run a List of Witnesses report using the filter like: A> Witness Is Not a Principal AND Tag Type ... Label =Equals CENSUS AND Any Witness... Surname =Equals [?] END Each time you run this report/filter the [?] will cause TMG to ask for the search surname. -OR- B> Witness Is a Principal AND Tag Type ... Label =Equals CENSUS AND Any Witness... Surname =Equals [?] END Each time you run this report/filter the [?] will cause TMG to ask for the search surname. Note that the List of Witnesses report will either give you just persons that are Principals or that are just Witnesses. The List of Events report will give both Witnesses and Principals (that is all witnesses). In all cases, set the sort order in the [Options...]=>Output Columns tab. Lee

Oops, I hit "Send" too quickly. I have found in my list of bugs some cases where the existence of only a single exclusion marker in a split memo part will cause problems. While "most" of the time it will work, it would be safer if that character is not used as the placeholder. I would still like to hear from you, Cheryl, to see if your example of the dash being output is different from the cases I have seen. Michael

Cheryl observed the output of an exclusion marker as the only value in a split citation field, and wondered whether an "empty" split field should contain a space character. I did a number of tests concerning this issue as part of describing this feature in my on-line book. See especially the Citation Elements section of the Source Elements chapter: http://www.mjh-nm.net/SRCELEMS.HTML#CitationElements And a separate lengthy discusson of Empty Split Memo Parts in the Tag Sentences chapter: http://www.mjh-nm.net/TAGSENTS.HTML#EmptyMemoVariables Summarizing the points made in those sections: A single space should be included as the place holder for any empty split part other than the first. As noted by others, HELP specifies the need for the space, and there is (at least) one known remaining bug where the absence of the space will produce incorrect output in some TMG reports. While TMG (and Second Site) will "generally" output correctly if there is nothing as a placeholder, I always recommend including the placeholder space to avoid any remaining bug(s). Likewise an empty first split part should also have a placeholder due to some reported bugs, but that part requires a different special placeholder character as described in my book. Again TMG (and Second Site) will "generally" output correctly if there is nothing as a placeholder for the first split part, but I also find it safer to include that special placeholder character. Using the final TMG version 9.05 I could not reproduce your output of the single dash "-" if that character is the first or only character between the vertical bars defining CD2. However, if the dash is not the very first character in CD2 (e.g. is preceded by a space character), then the dash will output. In my testing the conditional markers are observed for split citation detail parts. So effectively a single dash does seem to work as the sole placeholder character, even for the first split part, and I should update my book to reflect that. Cheryl, Before I update my book, could you describe in more detail what report you were running, and its options, which produced the output of the dash you described? If there is some circumstance which will output the dash, I should mention that as a reason not to use it. Michael

A couple of added comments on Lee's suggestion: If you have set up your census tags as I have, with the head of household as Principal 1 and other household members as witnesses, the "Any Witness" filter will cause everyone in any household in a census tag with at least one person having the specified surname to be included in the report. The filter will include all name variations, including any Name-marr tags. So for example if you specify the Black surname a woman Mary Smith who marries Joe Black then later marries Sam Spade and has son John Spade and appears with her Spade family in a census tag will cause Mary, Sam, and John Spade to all be included in your Black report. If that's too broad for your purpose you could use "Principal 1" instead of "Any Witness" in the first term of the filter. Note that filtering for "Principal 1" will still cause households with a person who has a name variation such as Name-marr that matches your specified name to be included. Typically this would occur when a divorced or widowed woman with matching maiden name is heading a household and therefore listed as Principal 1. In the report options include the Date and Prin1 Last, Given fields as output columns. You can specify the sort order with Date first to have your report completely date ordered, or specify Prin1 Last, Given to be the first sort in order to have it grouped by person's name. On 5/3/2017 12:06 PM, Lee Hoffman/KY wrote: > At 5/3/2017 13:37, nancy wright wrote >> I want a report that shows the census records for each person. I >> want the >> report for all persons surnamed BLACK, but later I would like to get >> this >> report for persons of other surnames. >> If possible I would like the census' to be listed in order. I don't >> know if >> that can be done - I just use Census, not (for example) census1930, etc. > > There are a couple of ways to get this -- > 1. Run a List of Events reports using the filter: > Tag Type ... Label =Equals CENSUS AND > Any Witness... Surname =Equals [?] END > Each time you run this report/filter the [?] will cause TMG to ask for > the search surname. > > > 2. Run a List of Witnesses report using the filter like: > > A> > Witness Is Not a > Principal AND > Tag Type ... Label =Equals CENSUS AND > Any Witness... Surname =Equals [?] END > Each time you run this report/filter the [?] will cause TMG to ask for > the search surname. > > -OR- > > B> > Witness Is a > Principal AND > Tag Type ... Label =Equals CENSUS AND > Any Witness... Surname =Equals [?] END > Each time you run this report/filter the [?] will cause TMG to ask for > the search surname. > > Note that the List of Witnesses report will either give you just > persons that are Principals or that are just Witnesses. The List of > Events report will give both Witnesses and Principals (that is all > witnesses). > > In all cases, set the sort order in the [Options...]=>Output Columns tab. >

I want a report that shows the census records for each person. I want the report for all persons surnamed BLACK, but later I would like to get this report for persons of other surnames. If possible I would like the census' to be listed in order. I don't know if that can be done - I just use Census, not (for example) census1930, etc. Nancy

Couldn't you just use a focus group of people surnamed Black? Then create a journal report including only census records. I use census records as part of my customized journal reports, and they are in order by sort date. Is that what you want, or do you want the censuses, then list all the Blacks in each census (I don't know how to get that)? On Wed, May 3, 2017 at 10:37 AM, nancy wright <wright4766@bellsouth.net> wrote: > I want a report that shows the census records for each person. I want the > report for all persons surnamed BLACK, but later I would like to get this > report for persons of other surnames. > If possible I would like the census' to be listed in order. I don't know > if > that can be done - I just use Census, not (for example) census1930, etc. > > Nancy > > The TMG archive is found here: http://archiver.rootsweb. > ancestry.com/th/index/TMG/ > Instructions on how to subscribe to TMG: http://lists.rootsweb. > ancestry.com/index/other/Software/TMG.html > ------------------------------- > To unsubscribe from the list, please send an email to > TMG-request@rootsweb.com with the word 'unsubscribe' without the quotes > in the subject and the body of the message >

Hi listers, The Southern California TMG users group will meet Sat May 6, 2017 from 2:00 to 4:00 p.m. at the SCGS Library. 417 Irving Drive Burbank, CA Everyone is welcome. -- Patt Ricketts www.jackandpatt.com

I want to be able to find out what I should do in TMG before I transfer it to any/all genealogical programs. One thing that I am doing, and have been doing for quite a long time now, is checking that my TMG sources are fully described as aTMG. Jim Payne New Zealand

Thanks very much Barbara Jim Payne New Zealand From: Barbara Zanzig [mailto:bzanzig@gmail.com] Sent: Saturday, 29 April 2017 11:35 a.m. To: The Master Genealogist Rootsweb Email List <tmg@rootsweb.com> Cc: familytreesearching@gmail.com Subject: Re: [TMG] What did you do when TMG no longer updated regularly? Sorry for the premature send there. James, are you looking for the TMG Refugee materials? Here's the contact information for the maillist and website: TMG refugee list address: http://lists.rootsweb.ancestry.com/index/other/Miscellaneous/TMG-REFUGEES.html TMG refugee list website: https://sites.google.com/site/tmgrefugees On Fri, Apr 28, 2017 at 4:32 PM, Barbara Zanzig <bzanzig@gmail.com <mailto:bzanzig@gmail.com> > wrote: On Fri, Apr 28, 2017 at 3:13 PM, James Payne <familytreesearching@gmail.com <mailto:familytreesearching@gmail.com> > wrote: I used to be able to go to a program/website/link?, that enabled me to find out what I needed to do to do before changing my TMG data to another genealogical program such as Legacy/RootsMagic, etc, etc. This was especially useful to me when TMG more or less shut down and we needed to link to other genealogical programs. (I hope I've described it correctly) Could a list reader suggest where I should go to find it? -- Barbara Zanzig