TMG-L Archives

Archiver > TMG > 2011-03 > 1299694731

From: "Michael J. Hannah" <>
Subject: Re: [TMG] design question
Date: Wed, 09 Mar 2011 11:18:51 -0700
References: <>
In-Reply-To: <>

On 3/9/2011 Teresa Elliott commented:
> ... But few of us believe GEDCOM is bad, just that
> it hasn't been updated in over a decade...
> That's a lot of being behind the times.
> For what it was designed to do, yes, it still does that well.

Having been following, and to a minor extent involved with, the GEDCOM
specifications since at least 1995 (based on my e-mails I can still
find), I know that there have been a number of attempts to update
GEDCOM, or create a new specification. I also know that Bob Velke has
been very significantly and actively supporting many of these efforts
(much more than my mainly passive lurking).

In my opinion, the problem has not been a lack of attempts to update
GEDCOM or create something else with more features. The problem has
repeatedly been a lack of sufficient agreement by enough players to gain
wide-spread acceptance of any one of these multiple attempts (e.g.
GenBridge). As others have noted, you design the specifications based
on what you want to do. Different groups of people have different
things they want to do, and can't seem to agree on something that might
accomodate most or all of their different goals.

For example, Thomas Wetmore in a message posted _today_ on the GEDCOM
ListServ mailing list, gives his opinion of the status of four current
such activities. (Thomas has always been very active in GEDCOM, and
back in 1991-1994 was the original author of the genealogy software
program called LifeLines which is based solely on the GEDCOM
specification.) Unfortunately, all four of the current efforts he
mentions seem to continue to have the same problem above, each designed
to do something different with minimal wide-spread support.

Until this changes, I agree with others that either a report, or web
pages like SecondSite, are my preferred way to share my data.


For Thomas' current message, see:

This thread: