Jump to content


Photo

multiple data sets (or Projects) and flags problem printing accents in charts


  • Please log in to reply
4 replies to this topic

#1 pete run

pete run
  • Members
  • 18 posts
  • Location:Sheffield, South Yorkshire, UK
  • Interests:Fell (and Hill, off road) running.<br />Many many other things. Science, politics, etc etc You name it, I've got an opinion, even if they do get Grayer and Grayer.

Posted 01 November 2007 - 04:40 PM

I have a problem with multiple data sets and flags.

I've read through some past forum entries and it seems there are problems with anything customised inc flags, sources etc, when using across different data sets.

I would have thought you would be able to export these customisations from 1 data set or project and import to another. Is that not possible?

Despite what "the experts" tell us, I can see many reasons to have separate datasets, or even projects for different groups of people. For instance if you are a professional doing it for different families, or in my case doing it for unrelated friends and none blood related family. We are recommended to use flags, which I agree are wonderful things, and TMG very good, but TMG is not the fastest software on the block, and separate datasets or projects for large amounts of distinct individuals can make it easier to manage.

I've not tried it yet, but the only solution I can think of is to have a "master" project or data set, and use that as a "pro forma" for any other set of individuals. Then regularly copy it, delete individuals and import individuals. Very messy, but is it any messier than using many sets of flags and filters.

Anyway, enough of my grumpy old man bit, here's the problem.


I've created some flags to help me show individuals who link and hence appear in more than one descendant chart. i.e. (via report option Boxes tab, and define the Accent colours) if they have flag LINK1 they print with a background colour of say pink, LINK2 would be beige, LINK3 etc. So that when they are printed on different trees it is easy to see the link between the trees. Works a treat and makes those vital "inter family" connections stand out, double cousins, 2 brothers marry 2 sisters, widows marrying other family members etc. I find as little as 5 colours enough, reusing the links for different unconnected branches.

Over time I have created other data sets and deleted some. And at present have just 2 data sets. These links seem to be repeated such that I now have
LINK1#1, LINK1#2, LINK1#2 (yes 2 of LINK1#2) and the same set of 3 for all but 1 of my customised flags. (There is just 1 for SEX, LIVING etc).

If I then try to run a chart report with data set 2 individuals, I get the following message for each LINK* flag

======
"The selected chart will include people from data set #2 but one or more Accent conditions are based on the LINKl#l flag which is exclusive to data set #1.
The chart may therefore not reflect the intended Accent colors.
Do you want to generate the chart anyway?"
=======

I then proceed and produce a chart without any accents at all, even for SEX which did not get an error message for which I normally have as a blue or red box.

So, in summary my main questions are:

1. Is there a way to export and import flags and other customisations between data sets or projects. It sounds like a basic requirement to me.

2. How have I got 3 copies of each flag, and why won't the report recognise them?

Or is it simply that the dataset element of TMG is really only of use as a staging post for new data from the like of GEDCOMs.

#2 RobinL

RobinL
  • Senior Members
  • 1,356 posts
  • Gender:Male
  • Location:Adelaide, Australia
  • Interests:charting, managing cultural differences in genealogical information, analysis for community and one-name studies

Posted 01 November 2007 - 05:05 PM

I know that this seems confusing.

The inbuilt standard flags (LIVING, SEX, etc) have the same name and meaning over all data sets (as you can't edit their definitions).

If you create _custom_ flag then it gets the name that you give it _plus_ a #N where N is the data set in which you defined it, hence LINK1#2. This is because you could have defined a flag with the same name in one data set as one with a different meaning in another data set. TMG builds and maintains a "firewall" between data sets.

Report configurations that use filters or accents on custom flags are particular to that data set as they contain the flag name including the #N suffix. So that is why you appear to have a number of different flags and why such a report created in one data set will not work as expected with another data set.

BTW: the duplication of name with same suffix should not occur and suggests that you have found some sequence that has led to some confusion. If it still persists after an Optimise, VFI sequence then you should report how you got it to Support.
Robin Lamacraft
Adelaide, Australia

#3 pete run

pete run
  • Members
  • 18 posts
  • Location:Sheffield, South Yorkshire, UK
  • Interests:Fell (and Hill, off road) running.<br />Many many other things. Science, politics, etc etc You name it, I've got an opinion, even if they do get Grayer and Grayer.

Posted 02 November 2007 - 02:26 AM

Thanks for the info. My fears confirmed. Although I still can't understand with a database why flags and other customisations can't be exported and imported.

Back to your answers: I've done "Optimise, VFI sequence" etc but still have the multiple flags. They all show in the full "Flag Manager" list, but the extras only show for individuals from data set 2. As to how I got it I've no idea. I think I will have to try merging the 2nd data set with first, then deleting data set 2, see if that tidies it up. Before that though I suppose I could try deleting the extra "data set 2" flags via "Flag Manager".

#4 Terry Reigel

Terry Reigel
  • Moderators
  • 4,643 posts
  • Gender:Male
  • Location:Boone, NC

Posted 02 November 2007 - 05:15 AM

I have a problem with multiple data sets and flags.

I would have thought you would be able to export these customisations from 1 data set or project and import to another. Is that not possible?

You want to be able to export the Flag definitions to another Data Set or Project? While I've seen a number of requests to export Source Types and Tag Types, I've never before seen a request to do the same with Flags. I suppose because generally Flag definitions are so simple that re-creating them is simpler than an export/import process would be.

Despite what "the experts" tell us, I can see many reasons to have separate datasets, or even projects for different groups of people. For instance if you are a professional doing it for different families, or in my case doing it for unrelated friends and none blood related family.

If you look at the Projects/Data Sets article my website that's the one case in which I suggest multiple Projects, because that's the only reason I could think of. :) But you say there are many reasons - I would be interested in what others might be.

We are recommended to use flags, which I agree are wonderful things, and TMG very good, but TMG is not the fastest software on the block, and separate datasets or projects for large amounts of distinct individuals can make it easier to manage.

Clearly dividing larger Projects will improve performance. Have you found the same to be true when dividing data into separate Data Sets in the same Project?

I've not tried it yet, but the only solution I can think of is to have a "master" project or data set, and use that as a "pro forma" for any other set of individuals. Then regularly copy it, delete individuals and import individuals.

Yes, that will work.

1. Is there a way to export and import flags and other customisations between data sets or projects. It sounds like a basic requirement to me.

Many things can be copied to a new data set when you create it within an existing Project. In 6.12 there is no direct way to copy Flags, Tag Types, Source Types, or Sources between Data Sets. In his comments about Version 7 Bob Velke indicates his plans that Sources and Source Types will be able to be copied in that version. As I said above, I've never heard a request to be able to copy Flags, but maybe others have.

2. How have I got 3 copies of each flag, and why won't the report recognise them?

Robin has already explained that - by doing manually exactly what you say you want to be able to do with a copy function. The report doesn't recognize them because all ancestor/descendant reports are work with only a single Data Set. But some of your Flags are not in the Data Set that you have asked the report to use, so they have no meaning to that report.

I've done "Optimise, VFI sequence" etc but still have the multiple flags. They all show in the full "Flag Manager" list, but the extras only show for individuals from data set 2.

You have multiple copies of the flags because you have flags of the same name in multiple Data Sets, as Robin explained. If you create a Flag named Link in Data Set number 1, it will appear in lists Flag Manger and in the Filters as Link#1. If you create a Flag of the same name in Data Set number 2, it will appear as Link#2. Data Sets are completely separate, so you can have different Flags in each one. Even if they have the same names, TMG doesn't know or care whether they have the same meaning.

You seem to what to keep your data in separate Data Sets, but then treat it as if it was all in a single Data Set. You can't have it both ways. Separate Data Sets are, well, separate. :(
Terry

See my Tutorials and Articles on using TMG at tmg.reigelridge.com

The Second Edition of my book, A Primer for The Master Genealogist, is still available
. For more information see my website.

#5 Mike Talbot

Mike Talbot
  • Validating
  • 362 posts
  • Gender:Male
  • Location:Metairie, Louisiana
  • Interests:Cajun, creole, royal and noble genealogy. Military history and artifacts, old portraits and photos. Chemical Engineer, Tulane U. 1961, retired.

Posted 02 November 2007 - 10:23 AM

Despite what "the experts" tell us, I can see many reasons to have separate datasets, or even projects for different groups of people. For instance if you are a professional doing it for different families, or in my case doing it for unrelated friends and none blood related family. We are recommended to use flags, which I agree are wonderful things, and TMG very good, but TMG is not the fastest software on the block, and separate datasets or projects for large amounts of distinct individuals can make it easier to manage.

I've not tried it yet, but the only solution I can think of is to have a "master" project or data set, and use that as a "pro forma" for any other set of individuals. Then regularly copy it, delete individuals and import individuals. Very messy, but is it any messier than using many sets of flags and filters.

Anyway, enough of my grumpy old man bit, here's the problem.


From one grumpy old man to another (Sophia Loren and Ann Margret are in my single working dataset):

I have never used nor needed custom flags, so can't help and have snipped that portion of your message. Focus groups have always been adequate, where a custom flag would also work.

I Ďm not ďan expert" nor a professional genealogist. For over 20 years, I have done genealogy for relatives, friends, in-laws, friends of in-laws, in-laws of friends, etc., gratis. Name a mistake, Iíve probably made it, including the agony of multiple working datasets.

You never know who is related to whom until you do the genealogy. You don't know who is your in-law's 8th cousin, twice removed until you've researched perhaps a couple of hundred duplicate ancestors before you realize it, or unless you have built a large single dataset and do all work within it.

Having a single working dataset is especially important for those who like to do the genealogy of others, either professionally or just for fun. Never, ever delete anyone from your dataset unless they are incorrect and unfixable (or you are served a court order to do so, impossible, if youíre the least bit discriminate in the US).

As an example, one in-law, "unrelated", married couple turned out to be 7th cousins of each other. Not a close relationship, you say? Alone, she had about 5000 ancestors found, he had about 9000 (Focus Groups can find interesting details like this). Taken together, they only had about 2000 unique ancestors. Furthermore, they were also distant cousins of others in my dataset. Fewer than 400 new people had to be added instead of 14,000. What could have taken several years was accomplished in a few of weeks.

The moral of that story: keep all of your researched genealogy data in one dataset and never discard any correct data from it.

Separate projects and datasets do have an important place. Keep data that you receive from cousins and others or download from the internet in them. Use otherís data as clues and references. Never merge anotherís total dataset with your own.

If you want to provide someone with a GEDCOM of just their own family, it is easy to do by exporting via a focus group from your single dataset. There is no mess. One manís mess can be your treasure.

Best wishes,
Mike Talbot




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users