RootsMagic 4 Misbehaves! - Fri, 18 Nov 2011

About a week ago, I was sent a GEDCOM that Behold crashed on. Behold crashed on a PLAC tag with a value of: “<>” as the place. It was a small bug, easy to fix.

But I noticed that this GEDCOM was from RootsMagic 4. Looking in my sample files, I only had one other RootsMagic 4 file, and I hadn’t looked at it in detail. But once I looked, I wasn’t happy with what I saw.

It wasn’t too long ago that Randy Seaver did some research and found that programs can not read source citations in GEDCOMs created from other programs. Furthermore, only RootsMagic was able to export to GEDCOM and then import its own source citations correctly.

At first glance that sounds good for RootsMagic. They obviously thought out what they were doing. It seems good … until you go and look at the GEDCOM they create. It includes illegal tags, misused tags and unsupported embedded codes. The GEDCOM they produce is illegal. They have their own algorithm to build up their templates from this illegal code, which if any other developer (such as me) want to try to input their data properly, then we have to work hard to reverse engineer all their ideas.

This is not good. If you are a RootsMagic user and use their Citation Templates, then you will be okay as long as you want to use RootsMagic. But the day you want to export all your data to another program, you are stuck, because it’s doubtful that the others will have done the work to figure out how to unmangle the mangled RootsMagic GEDCOM.

You can keep your hopes up that AncestorSync or BetterGEDCOM will one day do the job. They may enable transfer of this data. But until that day comes, you’re either stuck with RootsMagic - or as Randy recommends, you should NOT use their source templates.

For me, I’m now going through various RootsMagic 4 GEDCOMs and I’m at least trying to figure out how to display this stuff nicely. I’ve got GEDCOM code like this to deal with:

0 @S12545@ SOUR
1 ABBR <i>Echoes of the past Volume. II</i>
1 TITL <i>Echoes of the past Volume. II</i>
1 _SUBQ <i>Echoes of the past Volume. II</i>
1 _BIBL <i>Echoes of the past Volume. II</i>.
1 _TMPLT
2 TID 0
2 FIELD
3 NAME Footnote
3 VALUE <i>Echoes of the past Volume. II</i>
2 FIELD
3 NAME ShortFootnote
3 VALUE <i>Echoes of the past Volume. II</i>
2 FIELD
3 NAME Bibliography
3 VALUE <i>Echoes of the past Volume. II</i>.

and the wonderful reference to this source which is:

1 SOUR @S12545@
2 _TMPLT
3 FIELD
4 NAME Page

First you’ve got user-defined tags that begin with the underscore (”_”). Those are at least allowed in GEDCOM, but they require the reading program understand their meaning to process them correctiy. The same can be said for the TID, FIELD and VALUE tags, except they are illegal in GEDCOM. And the NAME tag has a completely different purpose in GEDCOM and should not be used here like this. Add to this the repetition of the same information 7 times, which is not intended to be displayed over and over, and the inclusion of the HTML “<i>” and “</i>” tags to indicate where to use italics, and you can see what a challenge RootsMagic imposes on other programs.

What they have done is broken their GEDCOM export so badly, they are saying that they don’t care if any other program can read their data or not. Imagine if every program decided to do this and do it in their own way. That will eradicate whatever usefulness GEDCOM still has.

That is not in the true spirit of what all other genealogy software programmers want to use GEDCOM for. We want, make that “need” to use it to archive your data and to transfer it between programs. Right now its the only mechanism that comes close. Until there is a BetterGEDCOM or an AncestorSync, GEDCOM is all we have. And it saddens me as a programmer to see the damage RootsMagic has decided to do to it.

*** Five days to Version 1.0 and counting *** :-)