Custom Flags and Multiple Data Sets

[Nermal 0]

Hi!

 

Total and utter newbie here. I think that I'm finally beginning to get the hang of this, put this problem has me pretty stumped at the moment. Sorry if it's too simple/obvious or if it's already been asked and answered; I looked, but couldn't find anything.

 

Basically, I have a Project with a fair number of Data Sets (I'm engaged in One Surname Research and it seemed a whole lot better - perhaps wrongly - to have each currently-unconnected family in a different Data Set as this would allow better comparisons, easier exports to other researchers, etc.). I imported my data from another program and then spent some time dividing my data into these different Data Sets (which my old program didn't allow).

 

I then realized that I should have included something along the lines of a "Cleaned" custom flag so I could check and clean my data after import.

 

Now that my data is already in multiple Data Sets, do I really have to go and set up a separate custom "Cleaned" flag for each and every one of my Data Sets?? This doesn't seem likely, but I also can't find any way of applying a custom Flag to multiple Data sets.

 

Please help!

 

Thanks!

 

Robert

Greensboro, NC, USA

[Terry Reigel 0]

Welcome to TMG, Robert!

 

Yes, custom flags are specific to data sets. The whole idea of separate data sets to to make them totally independent of each other.

 

I'm not sure that separating the lines was a good idea. Not only custom flags, but customized or custom tags, customized or custom source types, etc., create the same issues. I really believe, absent a really good reason to keep them separate, keeping everything in the same data set is the best plan.

 

If you want to keep track of separate lines within a single data set, Flags are very useful. You can also create Accents based on those flags to provide visual clues about each line.

[patthepiper 0]

Hello Robert,

 

I agree with Terry.

 

I have one Project for my one-name study. The main database contains every name but not full details. I have separate databases for families lines of particular interest and these contain information in greater detail.

 

In effect, I use the main database as an index to the others when new information comes in. Reports from the main database look a rats nest until tidied up; reports from the other databases have evolved into a workable form.

 

I try to cross-reference between databases by ID numbers; #1:12345 would correspond to #5;12345. I blanche at the thought of total ID correspondence, but ensure that at least Progenitors have the same ID. For my own family line I reserved a series of ID numbers - may father is #1, I'm #3 - you can play god in your own little universe!

 

Good luck with the one-name study!

 

Pat

Dunbar One-name Study

Member of the Guild of One-Name Studies

[Nermal 0]

Yes, custom flags are specific to data sets. The whole idea of separate data sets to to make them totally independent of each other.

<snip>

If you want to keep track of separate lines within a single data set, Flags are very useful. You can also create Accents based on those flags to provide visual clues about each line.

 

First, thanks to both of you for responding so quickly.

 

Just to make sure that I understand, though....you would recommend that I keep all of these unconnected families in one large data set and then add a different flag to all members of each family (e.g., Family #1, Family #2, etc) that I could then use to distinguish these lines of people in reports, via accents, etc.

 

I'm pretty sure that I can merge the families back into one data set (if I recall correctly, Terry's site even has a page to help with this). Do I want to create these different flags for each family before or after I merge them?? It would obviously be easier to do it first by simply adding the Flag to every member of a given data set, but will these custom flags then carry over appropriately when I merge them back together?

 

Finally, based on what you're both telling me.....what is the best use of (or reason to have) separate data sets EVER? (apart from Pat's intriguing suggestion regarding using a main database as an index)?

Are they only truly useful for analyzing data from another researcher (or when retrieving an entire family fromany other source), or am I missing something again?

 

Anyway....thanks again for the quick -- and very useful -- help.

 

Any other insights you (or anyone else) can provide will be greatly appreciated!

 

Thanks,

 

Rob

[Terry Reigel 0]

First, thanks to both of you for responding so quickly.

You're welcome, Rob.

Just to make sure that I understand, though....you would recommend that I keep all of these unconnected families in one large data set and then add a different flag to all members of each family (e.g., Family #1, Family #2, etc) that I could then use to distinguish these lines of people in reports, via accents, etc.

Yes, that's my recommendation. You can do just about anything with flags that you can with separate data sets, but you don't have to deal with keeping all the custom stuff, like flags, tag types, source types, styles, etc., current in each one individually.

I'm pretty sure that I can merge the families back into one data set (if I recall correctly, Terry's site even has a page to help with this).

Yes, it's not all that hard if the various lines don't connect with each other.

Do I want to create these different flags for each family before or after I merge them?? It would obviously be easier to do it first by simply adding the Flag to every member of a given data set, but will these custom flags then carry over appropriately when I merge them back together?

Since you've got them separated, adding them now would be simplier.

I'd suggest creating a flag in each Data Set for that line. Initially, set values of Y,N, so everyone in that data set is set to Y. Then, after it's set but before you do the merge, change it to N,Y, so that during the merge everyone else gets set to the default value N.

 

After the merge, you could keep separate flags, or if you would prefer a single flag with different values for each line, you could use the Secondary Output of the List of People report to set all the values of that based on the individual flags. It's not clear to me that separate or a single flag is better than the other.

Finally, based on what you're both telling me.....what is the best use of (or reason to have) separate data sets EVER?

In my view, they are mainly useful as temporary places to hold data received from someone else while you examine it. You can then compare the new data with your own, and copy people into your data, or merge the data sets if you choose. However, I really don't recommend directly importing another person's data, as I explain in the "editorial" in my Merging article.

[Nermal 0]

You can do just about anything with flags that you can with separate data sets, but you don't have to deal with keeping all the custom stuff, like flags, tag types, source types, styles, etc., current in each one individually.

 

Great, thanks for the info. This is exactly what I needed to know. I didn't realize that all of that other "stuff" was "Data Set specific" either; it's a really good thing I asked now before things got even worse.

 

After the merge, you could keep separate flags, or if you would prefer a single flag with different values for each line, you could use the Secondary Output of the List of People report to set all the values of that based on the individual flags. It's not clear to me that separate or a single flag is better than the other.

 

Hmmm......that's an interesting idea. I'll have to think about it some more, but off the top of my head I think I prefer the single flag solution for my particular situation. Somehow, creating 45+ different flags (one per family) in order to keep track of essentially the same piece of info doesn't seem like the right way to go (if for no other reason than that I don't want to go "flag crazy" before I even start experimenting with any other custom flags I may want to try ).

 

Somehow...this program (as great as it's flexibility is, which is why I purchased it) is making me feel very old. I've been out of college way too long to begin studying like this again!

 

Thanks again for the help and suggestions; they're appreciated.

 

Rob

 

 

 

Okay.....I just went to work on this and realized that both the single-flag and the separate-flag solutions seem to have one major flaw:

I thought you couldn't create a Custom Flag with more than one digit!? How can I possibly create a Flag(s) to accommodate more than 45 separate families? Even the numbers & letters combined won't let me go that high and at this point in my research I'm still accumulating new families faster than I can connect them.

 

Please tell me that I'm missing something obvious (or did you just assume that I didn't have that many families?)

 

It seems to me that my only remaining option is to create a custom Tag, in which I simply state which family a given person is in. I know I can't set Accents via Tags, but that's not the end of the world for me for purposes of this identifier. I know I can still Filter Picklist results via a Tag, so it seems like I'll be getting most of what I want.

 

So...is there any way to create a custom Tag and then enter the same information in this field for everyone in a given Data Set?? (he asks hopefully). This way I can still assign each person to the correct family before I merge my data sets back into one.

 

Thanks again,

 

Rob

Edited August 15, 2007 by Nermal