Home > Meetings > Weekly Meeting



BetterGEDCOM Weekly Meetings: Mondays 10:00 AM - 10:30 AM Pacific Time

Meetings are usually held on in a GoToMeeting setting. Prior to each weekly meeting, the GoToMeeting connection is posted below and forwarded to members of the BetterGEDCOM Wiki via WikiSpaces email. Be sure to review the GoToMeeting User Guide

No Meetings are presently scheduled


NOTES FROM PREVIOUS MEETING
Times Around the World

Comments

DearMYRTLE 2010-12-27T18:23:10-08:00
Proposed Agenda 3 Jan 2011 Developers' Meeting
PLEASE feel free to add items to the agenda, by continuing this discussion in the wiki.

After contributing to the Google Docs "Comparison Worksheet", participants will better be able to consider:

1. Immediate vs. long term goals.

2. Timeline for completing first BetterGEDCOM model.
DearMYRTLE 2010-12-31T10:54:34-08:00
3. How to arrive at a consensus of opinion.

4. Which to settle first?
Families
Sources
Places
???
???

(If we can agree on one to discuss for weekly has-out sessions.)
AdrianB38 2010-12-31T11:53:32-08:00
I hesitate to add this to the Agenda but...
If we are to get our heads round the time-line for completing first BetterGEDCOM model, we need to have some specific objectives to get us there, maybe bit at a time. And it may be useful to bear in mind the difference between goals - which I interpret as often fairly loose aims - and objectives.

Objectives need to be (using management-speak) SMART, i.e.
Specific
Measurable
Agreed
Realistic
Timed

OK - you can have your own variants on that lot. But the crucial thing is that at some point we need to move from the loose aims / goals to specific objectives - otherwise we'll never know where we should be going or when we've got there. (NB - that does NOT mean we forget the goals)

OK - I'll take my Project Management hat off now.
DearMYRTLE 2010-12-31T11:59:02-08:00
Thanks for posting, Adrian.

The time-line is essential. Determining specific objectives to position on that time-line will be our challenge.

Let's keep talking...

What do you think the time line should include.
AdrianB38 2011-01-01T03:26:30-08:00
What do I think the time line should include?

1. An agreement on the basic aims of this "project" (sorry - can't think of a better term, even though that's a bit formal)

2. An agreement on the methodology to be followed (e.g. do we keep to the Wiki idea - which is bringing up some interesting stuff, like a slow brain-storm - or do we at some point say, OK, let's freeze that bit, work out a specific objective relating to it and do some specific work on that topic?)

3. A decent set of requirements for BetterGEDCOM - e.g. what are the shortcomings of GEDCOM? What do we want in our future? (Ah - we have a topic for that but is it detailed enough to draw up specific objectives?)
This will include an understanding about what physically we want to end up with. (And _I_ don't understand enough about XML to know what the options are there...)

4. An agreement about whether we go for everything in one go (not a good idea, it seems to me) or, if we phase things, what do we put into the first phase? (E.g. phase 1 could be a minimum change to GEDCOM equivalent to what FamilySearch proposed themselves. Phase 2 might then be to expand the data model to include new entity types, etc.) Another decision would be whether or not to release the Phase 1 output to developers? Possibly not if we needed to revamp it for Phase 2...
Note here that I am talking about putting a decision about phasing onto a time-line, not about putting the phases on - too early for that.

My personal problem here is that I was always used to big, disciplined (allegedly) projects with long phases. What you really need here is someone used to working iteratively in short bursts. No offence to anyone but I can see getting consensus and agreement will be like herding cats...
louiskessler 2011-01-01T11:39:08-08:00

I agree here with Adrian on all points.

Seems like the discussion on thi wiki is going off in all directions and we need some cat chow to urge the cats to come together.

Louis
Kin_Khronicler 2011-01-03T06:36:37-08:00
I agree with Louis on this as well, on all points.

This is a big project. We may need to break some of the bigger topics (sources/citations, location formats, research-layer, etc.) up into sub-projects, with sub-teams assigned to them. We may also need an official project manager to keep everything on-track.
Kin_Khronicler 2011-01-03T07:26:28-08:00
One more thing we need to consider for BetterGEDCOM is national vs. international "scope". I would think we'd want a world-wide scope, but that really opens the project up... we would need to support different languages and analyze different source material from other countries. If there's an international scope, then perhaps it needs to be implemented in phases.
DearMYRTLE 2011-01-03T08:31:43-08:00
OK, everyone, I've gathered your comments and here we go with the official agenda for our meeting:

1. Agreement on how to vote.
2. Agreement on basic aims of this "project”.
3. Agreement on methodology.
4. Agreement on national vs. international.
5. Agreement on initial time line.
6. Agreement on meeting frequency.

Respectfully submitted,
gthorud 2011-01-03T09:51:48-08:00
National vs International

Those who wan't to make sure their culture is covered will have to contribute. It is as simple as that. It will be very difficult for those outside a culture to try to cover it.
theKiwi 2011-01-03T11:21:29-08:00
I tried to join the meeting about 1:15PM EST and was told the meeting was full. What is the capacity of such a meeting?

Just curious <g>
GeneJ 2011-01-03T15:07:51-08:00
@theKiwi:

We had a 15 person limit for today's meeting. We hope that will be expanded in time for the next meeting.

Hope this helps.
DearMYRTLE 2011-01-03T15:39:02-08:00
The limit on the room was attendees and one presenter.

At the meetings, I explained I would canvas all our genealogy friends who host large webinars and ask that we can use their room on a rotating basis with other corporate entities.

We don't want to meet only in one corporate-donated space, as that might be construed to showing preference for that genealogy entity over another.

At the end of the meeting I stated my resolve to personally schedule space in time for our twice monthly meetings.

Thank-you theKiwi, for taking the time to make your posting.
GeneJ 2011-01-14T07:33:50-08:00
Developing Agenda Items for 17 Jan 2011 Developers Meeting
I'm posting a section on the PAGE for us to provide input for the 17 Jan 2011 Developers Meeting.

Please edit the PAGE to add agenda items.

This discussion topic supplements input on the PAGE.
AdrianB38 2011-01-14T13:05:19-08:00
See my proposals for Goals 2 & 3.

Somebody else please try Goal 4!!

I have tentatively added an item for the BetterGEDCOM "Hook", a.k.a. the Unique Selling Point / the "Sizzle", etc. The concerns about it are so very important but we don't seem to have got much there yet. Maybe we won't, until we know what BG is, but maybe thinking about what it might be is important right now. All of this is frightfully interesting to those of us with an IT background and an inner Spock - but what about human beings? <PS - In case of doubt: I'm a Brit, this is my attempt to be mildly humorous and encouraging!>
DearMYRTLE 2011-01-16T13:01:06-08:00
Thank-you, Gene J for writing this up. I am just returning from the Salt Lake Institute of Genealogy, have one more event to attend regarding Winter Retreat, and then I can devote my attention to the usual items on my plate.
theKiwi 2011-01-31T10:04:41-08:00
31 Jan Meeting
I'm trying to get into the meeting now, but am told "waiting for an organiser" - presumably I'm not the only one?
theKiwi 2011-01-31T10:06:33-08:00
Well of course posting this message caused the meeting to get started :-)
todrobbins 2011-02-07T10:19:21-08:00
Just saw the notice
I should have checked the Developers Meeting page earlier, I was wondering what happened. Sorry to hear about the router issue. I'll be at RootsTech, hoping to run into some of you there.

Tod
DearMYRTLE 2011-02-07T10:53:34-08:00
Apparently it is the ISP blocking GoToMeeting/webinar... isn't that odd. My router is fine.
SandyRumble 2011-02-21T10:01:20-08:00
Custom GEDCOM tags
I've attached a PDF file with the list of GEDCOM (standard) tags as well as the custom tags that I have identified across multiple products. Abbreviations are sometimes used for products as well as version number. For example, RM3 = Roots Magic v3, and FTM2010 = Family Tree Maker v 2010.
AdrianB38 2011-02-23T09:55:27-08:00
I'll buy Tom's 4th approach.

Not sure we'll be in any position to _require_ anything but we could at least set up a "GEDCOM extension documentation" site / Wiki / GoogleDoc quite easily and do something useful as a quick-win (ouch - at least the 2nd time I've used management speak - apologies)

I'll bet some app providers will think their stuff is secret so won't play, but at the very least, the software provided extensions to events & PFACTs should be do-able easily for a number of apps.
louiskessler 2011-02-23T11:31:41-08:00
gthorud 2011-02-23T16:03:49-08:00
I think there should be a registration process so that as many as possible event types are listed in a central list, but additional user defined events should be allowed, possibly with a definition in the header. It should be considered bad practice to distribute a program with predefined defined events that are not in the central list.

I am not sure I yet see all criteria that a registration must be checked for, but
- unnecessary redundancy should be avoided,
- codes that are merely a translation of another, and
- perhaps events that duplicate (in a slightly different way) other data carried by data in the standard (could perhaps be used to claim standard conformance for things that are not part of the standard), and
- events that would better be served by extensions to the standard (but could get a temporary tag)
- other criteria?

The criteria must be very clear so that we don’t end up in long discussions because some vendors will try to block the tags of others.

Registration candidates should be published for public comment before inclusion in the central list. There will be a balance between a complex evaluation process and a quick process. Note that such a list could be a separate part of a standard that will be updated eg. twice a year, but the rest of the standard must be more stable.

There could PERHAPS be problems needing further consideration if BG decides to have classes or subtypes of events, eg, birth class or Birth event with sub types. Similar problem with attributes contra events (are there anyone that wants to retain attributes?).

I prefer the term Event type since TAGs occurs everywhere in Gedcom.
AdrianB38 2011-02-24T09:39:17-08:00
In the list of custom GEDCOM tags for events or attributes, much as I hate possibly making people go back over stuff, do we need to note how the custom tag is "encoded" in the GEDCOM?

If I understand this correctly, I see (at least) 3 ways of an application program encoding a custom tag. Let's just look at custom _events_ first.

A) The proper(?) way, using the generic EVEN event which looks like this:
1 EVEN
2 TYPE Educated
2 DATE FROM 1958 TO JUL 1965
2 PLAC Crewe, Cheshire, England

B) The use of an underscore to denote an extension
1 _EDUN
2 DATE FROM 1958 TO JUL 1965
2 PLAC Crewe, Cheshire, England

c) Plain simple just stick it in anyway
1 EDUN
2 DATE FROM 1958 TO JUL 1965
2 PLAC Crewe, Cheshire, England

(Please don't start telling me there's an attribute for EDUC (grin) - I know there is, but I didn't want to use it - I just wanted an event, not the attribute, OK? (grin) )

I think if it's important that we have a list, then it's important that the list specifies the method used, otherwise an application won't easily be able to code to recognise it. I also think it's very important to start if the custom thing is an event or an attribute (unless someone can show it's not an issue for custom tags??)

For attributes, things may or may not be complicated by the fact that GEDCOM had no official means of adding custom attributes until GEDCOM 5.5.1 so far as I can see. Therefore ways of adding a custom attribute may be more volatile.

D) The proper(?) way, introduced only in GEDCOM 5.5.1, using the generic FACT attribute which looks like this:
1 FACT Home Guard
2 TYPE Military Unit
2 DATE BET 1939 AND 1945
2 PLAC Cheshire, England

E) The way FamilyHistorian (the software I use) does it with a custom tag vice FACT since it's 5.5 compliant not 5.5.1. (_ATTR can vary between apps, as well as the type)
1 _ATTR Home Guard
2 TYPE Military Unit
2 DATE BET 1939 AND 1945
2 PLAC Cheshire, England

F) The use of an underscore to denote an extension
1 _MILU Home Guard
2 DATE BET 1939 AND 1945
2 PLAC Cheshire, England

G) Plain simple just stick it in anyway
1 MILU Home Guard
2 DATE BET 1939 AND 1945
2 PLAC Cheshire, England

I'm sure there are other flavours... Please tell me if I've got the GEDCOM wrong anywhere...
louiskessler 2011-02-24T22:43:14-08:00
Adrian,

That's a good analysis.

You left out some things that make the analysis even tougher, such as Events also having descriptions, e.g.:

1 EVEN Appointed Zoning Committee Chairperson
2 TYPE Civic Appointments

and that the TYPE event descriptor can be applied to defined events, e.g.:

1 MARR
2 TYPE Common Law

Basically, the TYPE can be any text the user chooses, and GEDCOM states it should be displayed as given.

So that makes it possible to have an infinite combination of posible EVEN/FACT descriptors and classifications.

Add to that any number of possible user defined tags that start with an underscore.

All I can say is good luck trying to make a list of them. Infinity is a pretty large number.

Louis
louiskessler 2011-02-24T22:53:52-08:00

... and since events can have descriptions (i.e. attributes), the presence or absense of an attribute cannot be used to define the difference between events and facts.

The true difference, is that an event denotes a change of something and when that occurs. A fact indicates a truth that exists and the period of time during which it is true. From GEDCOM:

"As a general rule, events are things that happen on a specific date. Use the date form ‘BET date
AND date’ to indicate that an event took place at some time between two dates. Resist the
temptation to use a ‘FROM date TO date’ form in an event structure. If the subject of your
recording occurred over a period of time, then it is probably not an event, but rather an attribute or
fact."
ttwetmore 2011-02-25T02:44:45-08:00
The quote Louis provided about dates in events is a good guideline. However, I believe it is still reasonable to allow events that occur over a range of dates, so don't believe the quote should have been so strongly phrased.

Examples of events that take more than one day would be a trip, eg, an ocean voyage when immigrating. Yes, of course, you can add two events, a departure event followed by an arrival event, which might be the recommended course, but why disallow an event for the voyage as a whole.

How about a removal event? This is an awkward one to express in GEDCOM since it has a starting date and an ending date and there is no support for such an event. In my LifeLines databases I have gone back and forth between adding a FROM tag with a date an a TO tag with a date under the REMV event tag, to use a single "from ... to ..." DATE tag. I don't think GEDCOM 5.5 has the removal event, which is unfortunate, since removal is a very important event with genealogical significance. I think when using a single date for a removal event, one would put in the date when one got where they were going, but this has always seemed a little asymmetric to me.

How about a multi-day ceremony? How about a vacation? GEDCOM should be an important guide for BetterGEDCOM, but all of its assumptions are fair game for reexamination.
gthorud 2011-02-25T06:00:37-08:00
Removal???? event ---- ????

Are there any practical reasons for the time interval or point in time currently stated by Gedcom? Is it impossible to express an event occuring over a period of time in a time sorted list of events in a report? Or ... Is it more than an academic idea?
ttwetmore 2011-02-25T07:10:18-08:00
Gier,

My view is that there is rarely any reason to limit a data value by number or range. For example:

names -- persons should be allowed to have any number of names, including zero.
places -- events should be allowed to occur at more than one place; unusual but why restrict? Special cases always crop up.
dates -- events should be allowed to occur on more than one day and also over a range of days; unusual by why restrict? Special cases always crop up.

A military campaign as an event occurs over a range of dates and over a range of places. Pathological example maybe, but easy to handle with a flexible approach.

I don't see how sorting is much of a problem. I use the convention that the first occurrence is used for the key. If I add many names to a single person, the person is indexed on all of them and sorted on the first of them. If you want an extra sort name value that's fine. If I have many events of the same kind, the first occurrence is the "master" event that shows in displays and in reports. They're all in there though.

Sorting events with multi-date and date ranges isn't a problem. You can sort by the earliest or you can sort by the "middle" and the user will like the results and never think about what's really going on. Sorting rules are easy to establish for these "special" cases, that I don't believe are all that special.

Again substitute my usual tirade about humanistic data. Genealogical data is humanistic data, not the kind of data that yields easily to format restrictions, and therefore often impossible to fit into relational table columns. The automatically assumed implementation of genealogical data as simple records made up of simple columns in simple relational databases has been the cause of so much of the problems we live with with genealogical systems. I believe we should proactively distance ourselves from the idea of restrictions as far as we can. Take the view that we will embrace a policy of no restrictions and then only introduce them if they somehow become mandatory.
AdrianB38 2011-02-25T12:57:23-08:00
Louis - re your statements "Events also having descriptions, e.g.:
"1 EVEN Appointed Zoning Committee Chairperson
"2 TYPE Civic Appointments"
and
"events can have descriptions (i.e. attributes), the presence or absence of an attribute cannot be used to define the difference between events and facts"

That's, ahem, "interesting". I just double checked GEDCOM 5.5 and the INDIVIDUAL_EVENT_STRUCTURE in that copy seems clear to me that it does not allow a description (i.e. attribute) for an event - not even the EVEN generic event that you quote. TYPE, yes, no problem with that.

Do you know if previous (or post 5.5) versions of GEDCOM relaxed this? Or is it simply a case of software suppliers trampling over the standard again? In a sense, it doesn't really matter either way because if there are files out there with that construction, we need to deal with them. But I'd still like to understand what's going on (my pedantic brain again).

I think we do need to understand AT SOME POINT in time, the difference - if any - between an event and a characteristic (attribute in GEDCOM terms), so I will, later on, stick a requirement in that we do so.
ttwetmore 2011-02-25T14:36:49-08:00
Just ole opinionated me again. Adrian's last about the difference between an event and a characteristic brings up an important (IMHO) point that I have tried to cover in the DeadEnds model.

What is an event in GEDCOM? It is a substructure of lines inside a person record (or family record) that describes a date and place and maybe some other information about an event that occurred in the life of the person (let's forget families for awhile). These are SINGLE ROLE events that conveniently forget that a birth event really involves at least three persons! Thinking about events in this trivial way has gotten so commonplace that it has nearly completely hidden from view what events really are. A role player is NOT MENTIONED in these events because the substructure is inside the record of the event's PRIMARY ROLE PLAYER already. These "events" serve PRIMITIVE genealogy well, the simple quest for birth and death dates for direct ancestors, but they are inadequate for serious genealogy. GEDCOM came out of the LDS's goal to perform temple rites on church members' ancestors, not from anyone's goal to have GEDCOM support serious genealogy. We should not be surprised that GEDCOM is only suited for the fairly simple requirements of the church faithful.

In the DeadEnds model I call these events "vital events" and keep them as substructures inside person and other records. Thus converting a file from GEDCOM format to DeadEnds format does not cause an explosion in the number of now independent event records with single role players. In some sense these "vital events" are really like structured PFACTs and I think can be treated as such. You just have to think of a birth date and place as structured but still pretty simple PFACTs about a person.

But then there are "real events." GEDCOM doesn't have them as first class citizens, but some systems now do. These events are full-fledged, top-level, first class citizen, records, with their own record identifiers, indexed as any record would be. These are true multi-role entities that refer of to the person records of the person who play the roles in the event. The DeadEnds model brings this record type front and center right along with person records, and I certainly believe strongly that the Better GEDCOM model should do the same.

So, the bottom line. In thinking about events, one must be pretty careful to know what one is talking about. I hope I have explained the differences between the two ways that the term event is used most commonly in the genealogical context today. When deciding how to include events in the Better GEDCOM model it is important to know which of these concepts is the one one is discussing.

Tom W.
AdrianB38 2011-02-25T15:03:58-08:00
I've just created a discussion "Syntax09 Define Event vs. Attribute" on the Better GEDCOM Requirements Catalog page to cover the topic of "Is it an Event or an Attribute / Characteristic / etc"
AdrianB38 2011-02-22T14:39:27-08:00
Frankly, I think the GEDCOM standard is a bit of a mess re the EVEN / TYPE construct. It offends my sensibility to see events like BIRT and BAPM with their type explicit at level 1 for an individual; but for custom events we have this generic EVEN at level 1 with the type down a level in the TYPE line.

While I agree with Tom's comments in the last paragraph for GEDCOM, I wonder if there is a case for using TYPE even with conventional events like BIRT. (In fact, I think that as it's part of EVENT_DETAIL, it seems to me that it _is_ currently allowed). Then one might have
1 BIRT
2 TYPE Caesarian

This gives us an advantage of using TYPE with normal events - the ability to concoct sub-classes of events that inherit all the programmatic logic of the main class of event.

If we look at the possible XML for events (sorry - can't do JSON!) we could have
1 BIRT
2 TYPE Caesarian
translated to
<BIRTH TYPE=Caesarian> more-stuff </BIRTH>

and a custom event
1 EVEN
2 TYPE Educated
translated to
<EVENT TYPE=Educated> more-stuff </EVENT>

Yet I would have thought that all event tags should have a common structure and that this common structure would be better served by defining a single EVENT tag to cover them all, thus:
<EVENT TYPE=Birth SUBTYPE=Caesarian> more-stuff </EVENT>
<EVENT TYPE=UserDefined SUBTYPE=Educated> more-stuff </EVENT>
Surely this is then easier to produce DTD / schema / whatever definitions for and therefore, at least in the XML case, not a lose-lose?

An alternative to
<EVENT TYPE=UserDefined SUBTYPE=Educated> more-stuff </EVENT>
would be
<EVENT TYPE=_Educated> more-stuff </EVENT>
which is arguably more consistent in levels and the idea of marking the extensions with underscores. Except I do worry about whether people have used long sentences for the TYPE beneath EVEN

Either way, though, it's important to realise that all events needing programmatic logic need to be specified and all events that we want to exchange with certainty need to be specified - whether or not they are typed or sub-typed. As Tom says.
AdrianB38 2011-02-22T14:59:09-08:00
Louis - re your statement
"There is an EVEN (Event tag) which describes a change that happens at some time,
"and a FACT tag which describe something that is true over a time period."

We've probably had this discussion before (grin!) but while your definition _tends_ very much to be true, we can concoct a definition of the difference between event and attribute (to use the GEDCOM terms) that leads to events happening over a long time.

This is even more true if we go for the concept of an event affecting multiple people while an attribute only applies to one person.

Specifically:
- an attribute must have a value (not one of the existing place, date, etc)
- if something has a value then it's an attribute
- an event must not have a value
- if something doesn't have a value then it's an event

Using this definition, "World War One" qualifies as an event and it clearly lasts for several years. It also affects a number of people, so that's another good reason to take it as an event.

Also, Residence qualifies as an event since the so-called value it usually has is PLACE, which is already present, so it doesn't actually need this "value" item. Some GEDCOM type programs get themselves in knots over Residence because they say "Residence is an Attribute - but unlike every other Attribute it doesn't have a value"

I much prefer my definition of the difference between Event and Attribute because it can be precisely described with no exceptions.

But it's also important to realise that many facts can easily be represented in either fashion depending on whether you bring things like Cause-of-event and Responsible-Agency into play.... So it probably shouldn't cause us too much grief too soon.
gthorud 2011-02-22T16:01:08-08:00
Can the list of Custom tags be used to generate requirements?

The list has been published in a spreadsheet on Google Docs, for access contact DearMYRTLE.
SandyRumble 2011-02-23T04:48:20-08:00
I don't see why not. I'm still merging in some of the changes that Tamura pointed me towards, which should result in a more comprehensive list. As a starting point, some of the work has been done, but I'm sure the list isn't perfect :-)
SandyRumble 2011-02-23T05:18:17-08:00
To address the question of why would someone try to identify the various event types?

To meet user expectations! Let me pose a question: If you were to have a report that identified potential military service, how would you go about identifying events that equate to military service? If you want to see what such a report might look like, email me.
Each genealogy software product on the market today identifies military service differently. Some have a single military service event; others have enlistment date, discharge date, promotion date, and some programs do not even offer a military event. Some users (I am one) use custom events, like Revolutionary War, Civil War, WW1, WW2, etc to identify the specific war, with service dates, and I have friends and fellow genealogists who use all of the methods mentioned.
The typical user doesn't care one iota how fragmented the GEDCOM standard has become, they just want their report. They don't care about esoteric discussions of what an ideal data model or standard should look like; they don't care that not all genealogy programs handle things the same way and have diverged from what was once a standard. They want their software to work for them, not dictate to them how they should work!
A report that is based on a concept, Military Service, regardless of how it is managed by any one particular software product, should run, out of the box for the most popular software products on the market today. If the user defines their own military events, a mechanism needs to be provided that allows them to identify those events. However, if the user uses the "standard military events" defined by their popular, commonly used, family tree data management tool, any software program looking to read the users GEDCOM file should have a good idea what those events are. Some programs that read GEDCOM's don't care what the tags are and the meaning of the tags, beyond the basics of birth, death and marriage. Other programs do.
HTH, Sandy
AdrianB38 2011-02-23T07:24:56-08:00
Sandy - while I agree that users want stuff that "just works", I am also concerned that there is a danger of descending into paralysis of analysis as we attempt the impossible and list all non-GEDCOM standard event and attribute tags.

The non-GEDCOM standard event and attribute tags fall into two categories:
- custom events & attributes created by the users;
- custom events & attributes created by the software suppliers;

I believe that any user who creates their own custom events, and their own reports to pick up data in them, is sufficiently adept to be able to deal with _any_ non-GEDCOM standard events & attributes that they receive in a file. So they are not a problem to them.

The issue lies for me in custom events & attributes created by the software suppliers because a file from one of their applications could well go to another user, who has an application using a different set of custom events & attributes created by their own software supplier. Such a person could have no understanding of creating custom tags and wonders why a report doesn't pick up the funeral details that the guy who sent them the file said were on the file. (Answer - because it's coded with _FNRL in the import but the reports looking for _FUN. NB - what an awful tag name for Funeral!)

So - of that list of yours, do you know how many are encoded by the software suppliers, and how many might have been created by users? Is it possible to tell?

(I'm guessing that sampling of GEDCOM files is useless for this exercise since one would not be able to tell the difference between software suppliers' and users' custom events.)
SandyRumble 2011-02-23T07:58:01-08:00
Adrian, most of what is in the file is what is created by the software products themselves. As a methodology, I loaded each of the "popular" desktop products and created a new database. I recorded one person with one of every person events that the software provided. I then created a spouse for the person, and then created each of the "family" or requires two person events to get the full list on a per product basis, and exported the resulting GEDCOM file. In theory there should be one event for each supported event in the software product, regardless of whether it is product specific or GEDCOM specfic. I looked at what was not GEDCOM standard and identified the "custom for product X events". Anything recorded in the spreadsheet, barring mistakes on my part, should list the products that I found the tag in. If multiple products used the same tag, the tag occurs once with the differing products listed in the third column. In the spreadsheet, I believe I removed my own custom user defined tags, such as WW1, WW2, Obituary, Cemetery, (they were in what I cut and pasted into the wiki).

I used a manual process, and has been pointed out, that can be error prone. I have the original GEDCOM files I created, if anyone wants to look at them. In addition, I could only do this exercise with products I found on the market today, or had earlier versions of. I do not have access to products from the not too distant past or ealier version of that adhered to earlier GEDCOM specifications, so those products are not on the list, and in theory they should be; but that is a limitation on finding the software available for download to perform the analysis.
Sandy
brianjd 2011-02-23T08:04:51-08:00
I can see some usefulness in having a list of real world tag usage. What tags are commonly used from GEDCOM?
What tags have been created by software vendors?
Why would software vendors create tags?

In my opinion tags created by vendors are likely due to one of two things.
1) They didn't read the standard and created a redundant tag. we should ignore these. The interpretation of those tags is only relevant for a translation/conversion program. It should have zero impact on our tag creation.

2) The software vendor, or their users, found a weakness in GEDCOM and created a much needed tag. These are the kind of tags our process should be concerned with! After all we do want a BETTER GEDCOM!

In fact, I'd say that BG should aim to: combine GEDCOM with the extensions made by vendors, discarding the stuff in GEDCOM that doesn't work, and adding in those needful things that the software vendors have not yet coded.

My reasoning on that is that any standard we develop has to have some usefulness to the software vendors or the standard is pointless.

While analyzing every tag ever made and throwing it into the standard is the BEST way of scaring off every software vendor out there.
Ease of use for the User is a great goal, but if you develop a standard that is a huge monstrous mess NO software vendor will use it. Then you have failed in developing a standard. THe standard has to be such that if provides as much ease of use for the user, while still being friendly to the idea of being economically implemented by profit-making software vendors. Let's face it software is hard to make a living off of without large numbers of sales. Writing a single commercial software package like FTM easily costs upwards of $100,000. Easily. Probably more in the range $500,000-$5,000,000+.
I routinely write software which sells for $30,000-$300,000. That is for a team of two, yes two, programmers. The standard that we are contemplating by going down this road, I don't even want to think of it's cost.

So the bigger the standard, the harder to code. Harder to code = time + people. People are VERY expensive.

In my opinion, the GB should retain as much of GEDCOM as we can. How many vendors are going to want to do a complete rewrite of their current application? ZERO.

Just saying, we need to keep in mind who our standard is going to be implemented by. There is probably a price limit where people won't buy the new BG based genealogy program.
brianjd 2011-02-23T08:16:50-08:00
Again with the older GEDCOM standards.

While, they may be useful to know about for translation/conversion they are really meaningless to creating the standard.

It would be worthwhile looking at them for our testing suite, but it is only useful if we can expect vendors to write the conversion process for their own software.

Again, software vendors are looking to make a product that they know or suspect will sell X copies at $50, and 0 copies at $500. While it is a noble goal to support every GEDCOM and vendor standard ever written, it simply is not a realistic goal.

We should be concerned with what versions of the standard the current vendors support. We should be concerned with making our standard in such a way that the major vendors can easily upgrade the users themselves. If they don't do it, who will? Us? I doubt I have that kind of time to devote to this.
ttwetmore 2011-02-23T08:25:59-08:00
Collecting a master list of all tags used by GEDCOM, with all known application and user extensions is, IMHO, an important and valuable step. It goes along with what I believe to the be most important goal of Better GEDCOM, "The data model that underlies Better GEDCOM must be a superset of the models used by existing genealogical applications to the fullest extent deemed possible during design."

To follow the military thread for a quick moment. Genealogy is family history. For most users of genealogical applications it is more than finding the birth, death and marriage dates and places for direct ancestors. It is an attempt to discover a rich biographies for the persons they take special interest in. They are interested in all aspects of these persons' lives. A successful genealogical data model would be rich enough to hold the biographical information without having to resort to either general purpose notes or to a set of user or application extensions. Military events are one of many subsets of events that are important in capturing this biographical information. In my opinion the Better GEDCOM events and PFACTs system must support this fully. I'm not an expert on military events, but I have been frustrated in the past in trying to figure out the best way to record military events. I think it would be a great idea if Better GEDCOM could take the advice of a military historian, or family researchers who pay great attention to military records, to have a well designed package of military events and PHACTs to add to the Better GEDCOM vocabulary.

The richer and wider and fuller and more encompassing we can make the Better GEDCOM vocabulary the better we can reach this goal expressed above, the easier it will be for applications to meet users' expectations, and the easier it will be for applications to fully share information.

Think of Better GEDCOM "standard" as a set of entity type definitions, where each entity type has a rich set of PFACTS that can be attached to it. Maintenance of this standard over time will consist of considering changes to both the sets of entity types and to their PFACTS. One would hope that the original entity model would be complete enough to not require much change, but one might expect that there would be some continuing work to extend or adjust the sets of PFACTs. This would effectively boil down to a Better GEDCOM effort to update the standard on an infrequent basis by issuing lists of deletions (hopefully none), modifications (hopefully rare), and additions (also hopefully rare) to lists of tags.

Better GEDCOM must think hard about whether to support the idea of application and user extensions, as we can see how things have gotten out of control with GEDCOM. At the strictest Better GEDCOM could contemplate forbidding extensions. At the most flexible Better GEDCOM could allow the current GEDCOM approach, where vendors and users can add tags willy-nilly without documentation. The best approach, IMHO, is to allow extensions but require them to be documented, maybe even in the transfer files themselves. This was a disaster in the past when GEDCOM tried to implement a complex SCHEMA approach where lots of hard to understand meta-data was allowed in the header sections of GEDCOM files. The intention was good, but as always in GEDCOM, it was hard to understand so rarely and never implemented correctly, and quickly removed from the GEDCOM standard. I bet we're smart enough to create a simple approach to include the necessary meta-data to explain extensions. A fourth approach, maybe even better, would require any extensions to be pre-registered with Better GEDCOM before its use. Descriptions would accompany the registrations. Better GEDCOM could approve or modify before approval such additions. But most importantly, Better GEDCOM could provide the descriptions of the extensions to all vendors and users interested in following the gore.

Tom W.
SandyRumble 2011-02-23T08:37:34-08:00
I would advocate the fourth approach, where vendors can register their tags and others that are interested can find a simple definition. This would provide a single, definitive repository, making all lives easier.
It would certainly be easier for all vendors to maintain compatibility, whatever their resason for reading BetterGEDCOMs. I would hope that it would also generate "good will" with the vendors to find their customizations supported going forward.
Sandy
AdrianB38 2011-02-23T09:48:52-08:00
Sandy - thanks for your explanation of your methodology of tag finding. Seems both sensible and valuable.

Adrian
SandyRumble 2011-02-21T10:04:14-08:00
I have this as file as a PDF, but am having trouble figuring out how to upload the PDF file. Does anyone know how to do this?
thanks, Sandy
SandyRumble 2011-02-21T10:13:16-08:00
Adopted by Father _ADPF Brothers Keeper
Adopted by Mother _ADPM Brothers Keeper
Brit Mila _BRTM Brothers Keeper
Churching of Woman _CHWN Brothers Keeper
Circumcision _CIRC FTM2010, FTM2009
Cause of Death _DCAUSE FTM2010, FTM2009
Degree _DEG FTM2010, FTM2009, FTM2011
DNA Markers _DNA FTM2011
Election _ELEC FTM2009, FTM2011
Employment _EMPLOY FTM2010, FTM2009, FTM2011
Excommunication _EXCM FTM2011
Eye Color _EYEC Brothers Keeper
Funeral _FNRL Brothers Keeper
Funeral _FUN FTM2010, FTM2009
Hair Color _HAIR Brothers Keeper
Verify Home Christening _HCHR Brothers Keeper
Height _HEIG FTM2010, FTM2009, Brothers Keeper, FTM2011
Initiatory LDS _INIT FTM2011
Interred _INTE Brothers Keeper
Medical Condition _MDCL FTM2010, FTM2009, FTM2011
Medical Condition _MEDC Brothers Keeper
Medical _MEDICAL My Heritage FTB
Memorial Service _MEMR Brothers Keeper
Military Service _MILT FTM2010, FTM2009, Brothers Keeper
Military Id _MILTID FTM2010, FTM2009
Mission _MISN FTM2009, FTM2011
Namesake _NAMS FTM2009
Not Living _NLIV Brothers Keeper
Never Married _NMAR Brothers Keeper
Permanent Number _PRMN Brothers Keeper
Separation _SEPR FTM2011
Unique User Id _UID My Heritage FTB
_UPD _UPD My Heritage FTB
Web Address _WEB FTM2009
Weight _WEIG FTM2010, FTM2009, Brothers Keeper, FTM2011
Yartzert _YART Brothers Keeper
Abjuration Abjuration GenBox3
Accomplishment Accomplishment My Heritage FTB
Address ADDR FTM2010, FTM2009, TMG7, FTM2011
Adoption ADOP gedcom
Ancestral File Number AFN gedcom
Alias ALIA gedcom
Query1 21]Feb]2011
Page 2
eventName eventGED fromWhere
Alternate Birthdate Alt. Birth Legacy7
Alternate Burial date Alt. Burial Legacy7
Alternate Christening Alt. Christening Legacy7
Alternate Death Alt. Death Legacy7
Alternate Birth Alternate Birth My Heritage FTB
Alternate Christening Alternate Christenin My Heritage FTB
Alternate Death Alternate Death My Heritage FTB
Ancestor Interest ANCI Brothers Keeper
Anecdote Anecdote TMG7, My Heritage FTB
Annulment ANUL gedcom
Apprenticeship Apprenticeship GenBox3
Arrival Arrival FTM2010, FTM2011
Arrival ARVL gedcom, prior v5.4
Association ASSO TMG7
Association Association My Heritage FTB
Attributes ATTR TMG7
Award Award My Heritage FTB
LDS Baptism BAPL gedcom
Baptism BAPM gedcom
Bar Mitzvah BARM gedcom
Bas Mitzvah BASM gedcom
Birth]Covt BIC TMG7
Birth of Child BIRC RumbleSoft
Birth BIRT gedcom
Blessing BLES gedcom
Burial BURI gedcom
Cancel Seal CANC TMG7
Cartulary Cartulary GenBox3
Caste CAST gedcom
Cemetery CEMETERY RumbleSoft
Census CENS gedcom
Charter Charter GenBox3
Christening CHR gedcom
Adult Christening CHRA gedcom
Church Church RumbleSoft
Brith Milah Circumcision Legacy7, AQ12
Citizenship Citizenship Legacy7
Codicil CODI TMG7
Commemoration Commemoration My Heritage FTB
Communion Communion GenBox3
Confirmation CONF gedcom
Confiscation Confiscation GenBox3
Query1 21]Feb]2011
Page 3
eventName eventGED fromWhere
LDS Confirmation CONL gedcom
Court Court Legacy7, AQ12
Court Civil Court Civil GenBox3
Court Criminal Court Criminal GenBox3
Cremation CREM gedcom
Criminal CRIM TMG7
Crime Crime My Heritage FTB
Death of Child DEAC RumbleSoft
Death of Spouse DEAS RumbleSoft
Death DEAT gedcom
Death Notice Death Notice GenBox3
Deed Deed AQ12, My Heritage FTB
Degree DEGREE RootsMagic4, Legacy7, AQ12, My Heritage FTB
Departure Departure FTM2011
Deportation Deportation GenBox3
Descendant Interest DESI Brothers Keeper
Divorce DIV gedcom
Divorce Filed DIVF gedcom
Separation Divorce Separation GenBox3
DNA Test DNA RootsMagic4
Dowry Dowry GenBox3
Departure DPRT gedcom, prior v5.4
Description DSCR gedcom
Education EDUC gedcom
Election Elected AQ12, RM3
Election Election My Heritage FTB
Email Email FTM2010, FTM2009, FTM2011
Emmigration EMIG gedcom
Employment EMPL TMG7
Employer Employer My Heritage FTB
Employment Employment Legacy7, AQ12
LDS Endowment ENDL gedcom
Engagement ENGA gedcom
Enrollement Ennrollment GenBox3
Custom Event EVEN gedcom
Miscellaneous Event Event]Misc TMG7
Excommunication EXCO TMG7
Excommunication EXCOMM RootsMagic4, RM3
Excommunication Excommunicated AQ12
Family Reunion Family Reunion My Heritage FTB
First Communion FCOM gedcom
Foster Child Foster RumbleSoft
Query1 21]Feb]2011
Page 4
eventName eventGED fromWhere
Funeral Funeral AQ12, My Heritage FTB
Funeral Home Funeral Home GenBox3
Galley Galley GenBox3
Other GEDCOM (unrecogni GEDC TMG7
Godfather Godfather AQ12
Godmother Godmother AQ12
Graduation GRAD gedcom
Hebrew/Yiddish Name Hebrew/Yiddish Nam AQ12
Hobby Hobbies Legacy7, My Heritage FTB
Hobby Hobby AQ12
Honor Honor AQ12
Honors Honors Legacy7
Hospitalization Hospital Legacy7
Hospitalization Hospitalization AQ12, My Heritage FTB
HTML Document HTML TMG7
National Id IDNO gedcom
Illegitimate Birth ILLE TMG7
Illness ILLNESS RM4, Legacy7, RM3, AQ12, TMG7, GenBox3, My
Immigration IMMI gedcom
Interview Interview Legacy7
Introduction Introduction GenBox3
Election ISELECTED FTM2010
Journal Conclusion Journal Conclusion TMG7
Journal Introduction JournalIntro TMG7
Land Land Legacy7, GenBox3
Land Grant Land Grant GenBox3
Land Lease Land Lease GenBox3
Land Purchase Land Purchase GenBox3
Land Sale Land Sale GenBox3
Living LIVING RM4, RM3, TMG7
Marriage Banns MARB gedcom
Marriage Contract MARC gedcom
Marriage License MARL gedcom
Marriage MARR gedcom
Marriage Banns ]] Ann Marriage Banns ]] An GenBox3
Marriage Betrothal Marriage Betrothal GenBox3
Marriage Common Law Marriage Common L GenBox3
Marriage Consanguini Marriage Consanguin GenBox3
Marriage Dimissorial Marriage Dimissorial GenBox3
Marriage Dispensation Marriage Dispensatio GenBox3
Marriage Engagements Marriage Engageme GenBox3
Marriage Intention Marriage Intention GenBox3
Query1 21]Feb]2011
Page 5
eventName eventGED fromWhere
Marriage Rehabilitat Marriage Rehabilitat GenBox3
Not Married Married]]Not GenBox3
Married, unknown Married]]Unknown GenBox3
Marriage Settlement MARS gedcom
Medical Medical Legacy7
Membership Membership Legacy7, My Heritage FTB
Military Service MILITARY RM4, RM3
Military Award Military Award AQ12, My Heritage FTB
Military Discharge Military Discharge AQ12, GenBox3, My Heritage FTB
Military Enlistment Military Enlistment My Heritage FTB, My Heritage FTB
Military Induction Military Induction GenBox3
Military Promotion Military Promotion GenBox3,My Heritage FTB
Military Service Military Servic FTM2011
Military Service Military Service Legacy7, AQ12, GenBox3, My Heritage FTB
Military Service Start Milit]Beg TMG7
Military Service Ends Milit]End TMG7
Miscellaneous MISC RM4, RM3, TMG7
Miscarriage Miscarriage AQ12, My Heritage FTB
Missing Person Missing Person GenBox3
Mission MISSION RM4, RM3, Legacy7, AQ12, My Heritage FTB
Move Move AQ12, My Heritage FTB
MYHERITAGE:BABY_NAM MYHERITAGE:BABY My Heritage FTB
MYHERITAGE:HASSIDUT MYHERITAGE:HASSI My Heritage FTB
MYHERITAGE:LINEAGE MYHERITAGE:LINEA My Heritage FTB
MYHERITAGE:PIDYON_H MYHERITAGE:PIDY My Heritage FTB
MYHERITAGE:ZEVED_HA MYHERITAGE:ZEVE My Heritage FTB
Name Change Name Change GenBox3, My Heritage FTB
Namesake NAMESAKE RM4, RM3, Legacy7, My Heritage FTB
Naming Naming AQ12
Namesake NAMS TMG7
Nationality NATI gedcom
Naturalization NATU gedcom
Number of Children NCHI gedcom
Newspaper Article NEWS RumbleSoft
Number of Marriages NMR gedcom
Note NOTE TMG7
Nullify LDS NULL TMG7
Number of Children Number of Children RumbleSoft
Obituary OBIT RumbleSoft
Obituary obituary RumbleSoft
Occupation OCCU gedcom
Ordiance ORDI gedcom
Query1 21]Feb]2011
Page 6
eventName eventGED fromWhere
Ordination ORDN gedcom
Organization Organization AQ12
Passenger List PASL TMG7
Passenger List Passenger List GenBox3
Passport Passport GenBox3
Pension Pension AQ12
Permanent Record File Nu Permanent Record Fi GenBox3
Personality and Interests Personality and Inter My Heritage FTB
Phone Number PHON FTM2010, FTM2009, TMG7, FTM2011
Police Report Police Reports GenBox3
Poor Law Poor Law GenBox3
Population Register Population Register GenBox3
Probate PROB gedcom
Property PROP gedcom
Ratification RATI TMG7
Rebaptized REBA TMG7
Rebellion Rebellion GenBox3
Reference REFN gedcom
Relationship on Page of Tes Relationship on Page AQ12
Religion RELI gedcom
Reseal RESE TMG7
Residence Family RESF RM4, RM3
Residence RESI gedcom
Residence Residence
Residence before Deporata Residence before De AQ12
Retirement RETI gedcom
Reunion Reunion GenBox3
Record Id Number RIN My Heritage
Roster Roster GenBox3
Sasine Sasine GenBox3
School School Legacy7
Separated SEPARATION RM4, RM3, AQ12
Settlement Settlement My Heritage FTB
Slavery Slavery GenBox3
Sealing of Child SLGC gedcom
Sealing of Spouse SLGS gedcom
Social Security Number Social Security Num Legacy7
Solemn Communion Solemn Communion GenBox3
Social Security Number SSN RM4, RM3, AQ12, TMG7, GenBox3
Stake STAL TMG7
Stillborn STIL TMG7
Stillborn STILLBORN RM4, RM3
Query1 21]Feb]2011
Page 7
eventName eventGED fromWhere
Tax Payer Tax Payer GenBox3
Testament Testament GenBox3
Title TITL gedcom
Travel Travel AQ12
Tribe Tribe AQ12
Twin Twin My Heritage FTB
User Reference Number User Reference Num GenBox3
Voting Registration Voting Registration GenBox3
Vow Vow GenBox3
Web Address WEB ADDRESS FTM2010, FTM2011
Will WILL gedcom
Will Dated Will Dated My Heritage FTB
Will Proved Will Proved My Heritage FTB
SandyRumble 2011-02-21T10:14:23-08:00
Its a little hard to read as cut and pasted text. Its three columns: Label/Desc, Tag, Product(s) that use tag
louiskessler 2011-02-21T13:55:37-08:00

Sandy,

I appreciate the work that you have done to try to identify GEDCOM tags that various programs use.

But it looks like you have been manually putting this together, which is inefficient and could possibly include typos. Better is to use purely mechanical means for this.

For example, I've uploaded the first part of the GEDCOM tag display from my program Behold at: http://bettergedcom.wikispaces.com/file/detail/GEDCOM_tag_list_in_Behold.GIF

This is the first part of a list from a GEDCOM made by RootsMagic 4 that I imported. It also shows tags Behold recognizes but are not in that particular GEDCOM. Tags in bold are not valid GEDCOM tags but are found in some programs.

The log file produces a list of the tags from this file as follows:

Summary of GEDCOM Tags

Records: 11293
FAM 2696
HEAD 1
INDI 7821
SOUR 706
SUBM 1
TRLR 1
_EVDEF 67
Pointers: 27356
CHIL 5257
FAMC 5718
FAMS 5369
HEAD.SUBM 1
HUSB 2682
REPO 704
SOUR 4938
WIFE 2687
Data tags: 111990
ABBR 706
ADDR 4
BAPL 666
BAPM 164
BIRT 4514
BURI 395
CAUS 7
CHAN 7821
CHR 11
CONC 1112
CONL 1
CONT 449
DATE 17308
DEAT 1982
DIV 26
DSCR 1
EMAIL 2
ENDL 661
EVEN 2
FCOM 89
FIELD 7056
FILE 6
FORM 6
GIVN 7102
HEAD.ADDR 1
HEAD.CHAR 1
HEAD.CONT 2
HEAD.CORP 1
HEAD.DATE 1
HEAD.DEST 1
HEAD.FILE 1
HEAD.FORM 1
HEAD.GEDC 1
HEAD.NAME 1
HEAD.PHON 1
HEAD.SOUR 1
HEAD.VERS 2
HEAD.WWW 1
MARR 2696
NAME 14880
NICK 5
NOTE 1632
NSFX 97
OBJE 6
PAGE 179
PHON 1
PLAC 4832
QUAY 1
REFN 26
SEX 7581
SLGC 461
SLGS 117
SSN 1
STAT 1466
SURN 7265
TEMP 397
TID 706
TITL 712
TYPE 69
VALUE 2294
_BIBL 706
_EMAIL 1
_EVDEF.ABBR 67
_EVDEF.DATE 67
_EVDEF.DESC 67
_EVDEF.PLAC 67
_EVDEF.SENT 67
_EVDEF.TITL 67
_EVID 1
_FSFTID 1155
_INFO 1
_NAME 2
_PRIM 6
_QUAL 1
_SCBK 6
_SOUR 1
_SUBQ 706
_TMPLT 5644
_TYPE 6
_UID 7821

Total number of tagged lines: 150639

I've got about 220 log files and it technically wouldn't be difficult for me program something to compile this all together to make a list.

But I have two points I have to make about all this:

1. Using sample files does not mean you'll catch all tags and what programs use them. It depends on what data people entered and whether or not they used certain features of the program or not as to whether or not those tags would be output in the GEDCOM. So no matter what you do, you can only catch some this way.

But my more important question is:

2. Why bother to do this at all????? In reality, who the heck cares what tags a program uses. A tag is just the meta-information for the data. GEDCOM allows user-defined tags starting with a "_" and so what. It takes one word description to describe that tag. That is simply a table that we can build here, but it has nothing whatsoever to do with the underlying BetterGEDCOM we want to build. BetterGEDCOM is the data structure we want to built. Tags are just the data.

What is important are the tags that certain programs misuse, or use for other purposes - not in a trivial data way, but in a data structure way. For example, I've had to program Behold to handle Legacy's _PLAC_DEFN record, which is basically a level 0 record for places which GEDCOM doesn't have. Or the various improper ways programs do not start the NOTE text on the NOTE line, or include CR's and Tabs in NOTEs, or don't handle Concatenation of lines correctly. Those are the real issues.

The bottom line is that we need in BetterGEDCOM to define the data structures that will in turn define the Record Tags and the Link Tags that will be needed. The data tags as far as I'm concerned can be open ended and maybe even defined in the Header section for other programs to interpret, e.g.:

1 _TAGDEF AKA=Also known as

which incidentally is a tag missing from your list, presumably because none of the GEDCOMs you looked through used it.

The bottom line is: Don't get hung up on Tags. They are data descriptors and are relatively unimportant to what BetterGEDCOM needs to do.
gthorud 2011-02-21T15:23:59-08:00
The following rephrases some of what Louis has written, but I disagree on one issue – lists of EVENT tags.

There are several classes of tags. Two classes are those that would describe an event, and the rest are for example things that is currently used to transfer info that would rather fit in a structure defined by BG. (And there are certainly other ways to classify them.)

Some event type tags are special, and very important, since they are not just data – as seen by the program – for example BIRT which the program must know because it carries info about relations between persons. Other are just “data” and the knowledge a program has about the tag is stored in sentence templates (possibly with roles and more).

I disagree with Louis on one point: I think that it would be useful if BG could contain a long list of these “just data” tags, and it should be done in a way so that it can be updated – say once a year. The list could classify the events into groups – e.g. birth group, and could list roles that apply and more. The advantage of having such a list is that programs can be pre-configured with sentence templates and other info about the event types so that users need to define such info for fewer event types when importing. Also, setting up a process for update of the list may also discover proposed tags that should give rise to changes in the structures in the standard.

A standard list will also be useful for those creating translations of event tags, but it will probably be to much work – and too complex – to try to harmonize all these tags across borders (thus we will see a lot of duplicates of what is essentially the same event, but with different tags.)

It should of course be possible to have user defined event tags in addition to those in BG.

But I agree with Louis in that priority must be given to the tags that may affect the structures in BG.

The issue has been discussed here :

http://bettergedcom.wikispaces.com/message/view/Event+entity/30805871

These discussions may also be of interest in classifying tags
http://bettergedcom.wikispaces.com/message/view/Individual+Data+Elements+Discussions/30806561#30806779
http://bettergedcom.wikispaces.com/message/view/Individual+Data+Elements+Discussions/30806561#30806779
louiskessler 2011-02-21T16:40:31-08:00
Also, just to complete:

There is an EVEN (Event tag) which describes a change that happens at some time,

and a FACT tag which describe something that is true over a time period.

Most other tags are simply descriptions of one of these, and will be the data for the TYPE tag under the EVEN or FACT, e.g.

1 EVEN
2 TYPE Graduation

Except that there are a few important Events that need to be distinguished:

BIRT and DEAT that are physical indications of the start and the end of life.

And some of the most important social events, e.g. MARR, DIV, BAPT, BURI, ADOP but these should be kept to a minimum.

Let it be known (from other threads) that Tom and I disagree on this. He'd like to see more tags. I'd like to see fewer and use the TYPE tag to handle the rest.
gthorud 2011-02-21T17:09:01-08:00
Wrt to event types, I think the question is, how should a solution look like in order to create less problems for users? We all know that the situation today is not satisfying, and it will not be better with fewer standardized event types.
ttwetmore 2011-02-21T22:33:25-08:00
Concerning events. Applications need to know the kind of an events for various reasons, especially the events that have special genealogical significance, those events used in internal computations by the software, and those events used to generate specific output in reports and so on. The applications CLEARLY HAVE TO KNOW THE EVENT TYPES. Whenever software needs to know something there has to be an unambiguous way to signify it. Whether this is done like this:

1 BIRT

or this

1 EVEN
2 TYPE birth

is semantically equivalent. But the TYPE advocates would have it that using the TYPE approach is somehow simpler or more elegant or requires "fewer tags". That's ridiculous. In the TYPE method you need TWO tags per event, AND THE VALUES OF THE TYPE TAGS STILL HAVE TO COME FROM A LARGE, FIXED, AGREED UPON SET, SO THE SOFTWARE CAN UNDERSTAND THEM. Using the TYPE approach you DON'T GET RID OF THE NEED FOR A LARGE SET OF FIXED TAGS, YOU JUST PUSH THE NEED FOR THOSE TAGS INTO BEING THE VALUES OF THE TYPE TAGS. There is ABSOLUTELY NO ADVANTAGE FROM USING THE TYPE APPROACH and there IS A DEFINITE DISADVANTAGE. The TYPE approach REQUIRES JUST AS MANY TAG EQUIVALENTS AND ALSO REQUIRES TWICE AS MANY TAGS IN THE FILES. This is a loss loss situation. I can't understand why anyone sees any advantage in forcing the TYPE approach.

The EVEN tag should be kept, the TYPE tag should be kept, but only used in special cases when the fixed set of event-specific tags does not have an event type that covers the need. And of course when I say the EVEN and TYPE tags I mean whatever their Better GEDCOM equivalent turns out to be.
gthorud 2011-02-22T09:57:20-08:00
My reading of Gedcom tells me that the EVEN – TYPE construct is an attempt to classify the event as “a just data” event when the tag is not standardized, something you can not assume any such thing about the data followed by an “underscore tag” – you can not even know that it is an event. Am I wrong?

(I could not care less about the number of tags in a file – computers do not have problems with files unless they are in the order of Gigabytes. Counting tags is a science of the 1980s. (HTML/XML are very inefficient encoding compared to some other syntaxes, but they are still used.))

As stated above, I agree in principle with Toms last paragraph.

The fact that someone bothers to collect the above lists indicates to me that there is a need to standardize/harmonize a long list of tags – what is the purpose of the list if not to identify a (long list of) problem(s).

Except that this standardization will require some work, I have not understood WHY it would be a problem to standardize a long list.
gthorud 2011-02-22T10:06:17-08:00
Strike the word "something" in the first paragraph.

Is one of the next steps to classify the tags according to what has been discussed above, i.e. the first question is - is this the data associated with a tag something that should be in a structure, not just an event type?
GeneJ 2011-02-28T10:09:17-08:00
Unable to enter meeting as registrant.
Hi all. But for repeated attempts to relog, I've been in GTM since well before the Webinar began. Each time, I'm being admitted as guest, listen only--despite having registered properly and having rec'd the registration confirmation.

The link I've been using to try to connect correctly IS the link in the registration confirmation e-mail.
Christine_E 2011-12-12T22:55:14-08:00
Excuse me if this is not the best place to post this, but I want to let Andy know that I can't enter the meetings when he hosts them. This was happening when he filled in for Myrtle, as well as at the Developer's meeting today. I attempted to enter one minute before the scheduled start time, backed out about 5 minutes later, and attempted to enter again, then gave up at the half-way mark. The message said something like it was waiting for the moderator.

Is there something I might be doing wrong or was the room full or??? (for today I may not have been signed in as I recently cleared all my browsing history, but usually stay always "signed in"). This would not be the case when he filled in for Myrtle.

Since I'll be attending RootsTech, I want to know what's happening. Andy, can you e-mail me so I can Reply to you? Christine
theKiwi 2011-12-13T04:32:51-08:00
Christine - I don't know if this situation applied to you?

I couldn't get into the meeting yesterday from the Meetings page either, and after 20 minutes or so looked a bit closer and found that the Meetings page actually hadn't been updated with yesterday's meeting details - I was trying (like you twice) to log in to the meeting from 5 December. Once I got the correct link in an eMail from GeneJ I was able to get in no problem.

During the meeting the Meetings page got updated to include the link for the meeting in progress on 12th December.

Does this sound like it fits your situation too?

Roger
DearMYRTLE 2011-03-07T12:45:05-08:00
Should we NOT record Developers Meetings?
While on the road, I've received a request from wiki members to consider the following:

PROPOSAL
Discontinue the practice of recording the Developers Meetings on the following grounds:

1. Discussion should be free-flowing. If you are being recorded, you may not be willing to say what you otherwise would say.

2. Though the meetings are held in English, for many participants, English is a second language.

3. There are timely invitations to attend the meeting. If you don't come, for whatever reason, you'll have the meeting notes.

What do you all think? Should we rely on the "Notes" typically typed up during the meeting itself?

Respectfully submitted.
theKiwi 2011-03-07T12:59:22-08:00
I think the notes are adequate, and the recording, at least at this stage is un-necessary since it's hard to imagine a "he said - she said" situation that would require review of the tape in what has happened so far.

Roger
ttwetmore 2011-03-07T14:52:07-08:00
I don't feel that the meetings should be recorded.

Tom W
gthorud 2011-03-07T15:41:26-08:00
I tend to agree that we don't need to publish the recording.

Since it is very difficult to record the minutes of the meting, currently done by the moderator in the meeting, in other ways than using "keywords" and short sentences that are in many cases meaningful only to the attendants, I would rather prefer a short summary of just the agreements/conclusions/ action items/VERY important info presented in the meeting, that would be understandable to people that do not attend the meeting - starting with the agenda.

It should be very brief. Maybe 1-3 sentences on each agenda item, although this is not a hard restriction. Two persons could cooperate to write this.

This would be much faster to read than listening to the recording.

The recording could be helpful to someone writing such minutes, but it should be deleted after that.

Geir
DearMYRTLE 2011-03-08T15:53:20-08:00
I think we are then agreeing that the meetings shouldn't be recorded.

I am not available to write the more detailed minutes that Geir is suggesting by listening to a recording of the meeting. We are running short of volunteers.

Perhaps more detailed notes can be taken during the meeting by the moderator?
benfsayer 2011-03-19T15:29:55-07:00
Effective and efficient meeting minutes in my experience list only three discussion items:

Action Items (with assignee and due date)
Decisions
Issues

Anything more (the he said, she said) takes too much time and isn't the most important part.
GeneJ 2011-04-10T09:20:28-07:00
Being logged out of wiki
I'm being logged out of the wiki at the most inopportune of times.

It's happened before when I've wanted to post a discussion, but it's really set me back the last couple of weeks when I've been trying to set up the pages for EE & GPS support.

I was logged out again this morning a little after 6--and I have three new unsaved pages (all with table data) and one existing unsaved page.

I've got some of copied over to a text file now, but don't think that will save the table data.

Does anyone know what causes us to be logged out?
gthorud 2011-04-10T09:48:00-07:00
I have experienced the same problem when posting a message a few days ago, that's the only time - and I don't remember that it has happened before. I lost some text. It is reasonable to be logged out after some timeout period, but don't know how long that period is. Or there could be a bug somewhere.

I think we will have to see if others experience the problem.
GeneJ 2011-04-16T11:00:12-07:00
What will it take to get the BetterGEDCOM E&C; part of the model defined?
What will it take to get the BetterGEDCOM E&C part of the model defined?

Have we made any progress this week?
GeneJ 2011-04-16T15:50:04-07:00
http://bettergedcom.wikispaces.com/message/view/Research+Process%2C+Evidence+%26+GPS/37632782?o=20#37776776

Mike writes, "On Friday 8:26am (Mountain Time), Tom presented a concrete solution that elegantly solves the problem Adrian introduced. Does anyone have a reason why this model does not solve the problem? I would like to see more concrete examples like this because it helps me better understand EXACTLY what the person is trying to say. Otherwise, as others have previously stated, it's easy to misunderstand what they mean."

I don't know that all of us have agree to have the source system as the clearing house for the record capture.

Tom's example was only enabling a citation to be "created"--I think in his example he still had the information snippet and logic and reasoning being recorded in the database proper.

If the source system is not the clearing house for these snippets and logic/reasoning, I object.

You wanted examples.

(1) I do not believe information in the source is more important than information about the source. How can you separate a snippet from it's authors identity as though that snippet has some separate value. Ditto, how can you separate a snippet from the identity of its "source of the source" and believe that snippet has some separate value. How can you separate the "snippet" from a delayed birth record from the information that is was delayed. We know that sources come in all flavors. It seems to me you're trying to carefully package all the fruit in boxes and then strip the labels off.
Please let me know if you need a list of authorities for the above.

(2) The "evidence person" concept is not vetted from my perspective. Surely we have to vet the evidence person concept before we make it the clearing house for anything.


P.S. logged out again
gthorud 2011-04-16T19:32:30-07:00
I think we need a page to capture the discussion. As I have said before, that page should contain references to previous pages and discussions.

If we can't agree to have a place to discuss it, and can't organize the work, we might just go back to doing whatever we were doing before the Developers meeting.
GeneJ 2011-04-16T21:24:42-07:00
GeneJ 2011-04-16T17:34:17-07:00
Snap shots of pages ...
Question:

I don't think we can take a snapshot of a discussion page, or can we?

I tried it a few minutes ago with page 1 or Adrian's "missing link" discussion. I failed, but it was the first time I'd tried using Geir's helpful tips. (What I got was a pdf of the page, not the discussion.)
gthorud 2011-04-16T17:52:27-07:00
I don't know how to create a PDF of a discussion. Would have been nice to be able to have a way to print without the "menu" thing to the left.

Did my instructions for creation of a snapshot work - have you tried to link to it from a page?
gthorud 2011-04-16T17:53:30-07:00
Also, be aware, I have seen some funny looking PDFs.
ttwetmore 2011-04-17T19:01:01-07:00
Agenda for April 18th
Agenda Item --

From Second Life Meg Madrigal: "In my opinion it is a mistake to focus on this as a *genealogical* question. The issue is to identify a way to transfer information from one relational database to another. Spending effort convincing people about evidence and citation models is not productive, al though interesting, because those models will most likely change or, as you mentioned are already different for different segments of the community. This issue can probably EASILY be addressed by a information science/database architect expert."

In my opinion we must focus on this as a genealogical question. The issue is to identify a way to transfer information between genealogical systems. Models based on evidence and conclusions reflect the true nature of the genealogical universe so are an absolute requirement in defining a transport format. Relational databases are nearly useless as an interchange standard and introduce syntax and formatting restrictions that cannot be tolerated in genealogical data. A relational model is as subject to change as any other. A relational model is as subject to problems caused by a segmented community as any other. This issue has been dealt with by information science and database architect experts for years.

Tell you what. Let's just throw everything away and start over. And then reinvent GenTech.

Tom Wetmore
AdrianB38 2011-04-18T02:27:06-07:00
This is me, speaking as an ex-programmer, etc.

"The issue is to identify a way to transfer information from one relational database to another"
Sorry, but this is simply not true. If it were, (simplistically) there would be no issue - write some SQL queries to dump the contents of one SQL-compatible database into another. (Mind you, that's spoken like a true non-SQL-programmer!)

The major error in Meg's statement is to assume that all "databases" in all family history programs are relational databases. They are not, nor is there any likelihood that they will become so. There are apps that don't use databases at all but store the data in flat files - e.g. Family Historian stores its data in a GEDCOM file, i.e. a text file that can be read in something like Notepad). FH isn't the only one to do so.

PAF - for which I still have a great affection - doesn't use any database management system (DBMS) so far as I know (non-IT-geeks can read DBMS as just meaning "database")

Even if we look at apps using a DBMS, there are other sorts of database management systems than relational. These, you may have noticed, have their passionate devotees, with whom I have sympathy.

Even when one starts referring to apps based on RDBMS (Relational DBMS), life for a data exchanger remains non-trivial. I said above "(simplistically) there would be no issue". Reality bites when one realises that each app developer does their best to make their app stand out, and this could include adding new sets of data to their database. Hence, their database probably differs from the standard model because they want - oh, pointers from a Person data table to a separate Location table instead of Location names. (Trust me, it's different!) Hence, they'd have to supply the SQL to extract stuff as they're the only people to really understand their own database. How would they know what to extract? We could define a standard for Genealogical Data Exchange to tell them what to extract. Or a standard for GEnealogical Data COMmunication. We could call it GEDCOM. Oops - we're back where we started!

So, Myrt, when you reasonably ask "MY QUESTION IS - are we forgetting our requirement not to favor ... a BetterGEDCOM that favors one type of genealogy research over another?" then the answer is that Meg's suggestion is the one that favors one type of genealogy research over another as it applies only to RDBMS databases, not others, and it also favours RDBMS databases that conform to a standard that - seemingly - that view doesn't see the point of.

Adrian Bruce
AdrianB38 2011-04-18T02:45:00-07:00
The one bit of Meg's suggestion that has merit is "Spending effort convincing people about ... citation models is not productive, although interesting, because those models will most likely change" - notice I snipped out the reference to evidence models.

Let's _assume_ that by "citation model" she means "citation template" - the pattern a printed citation conforms to.

We know that citation patterns / templates change. We've already remarked in the Wiki that there are already several base patterns for templates and that implementing ESM's Evidence Explained models in those exact patterns would be wrong, so the Wiki would agree with that _abbreviated_ statement. What we need to do, re citations, is (as has been mentioned in the Wiki) capture and label the underlying components that go into a citation, so that the receiving application can splatter them out on the printed page in whatever order the current citation model wants.

Note that the underlying data for a citation is a different thing from the citation template. We need the underlying data and that is less likely to change than the format of the template.

Similarly, we need the underlying data for evidence - or is it being suggested that concepts like source, citation, proof argument might change and be repalced? No, I think not.

Adrian
hrworth 2011-04-18T04:05:16-07:00
Dear Myrtle,

I would like you to please add an agenda item about the tone of some of the comments that is used on this wiki.

Example:

This was posted this morning:

"You and GeneJ are opposed to the idea. You two are also the least computer savvy of the people actively participating here. I believe the main problems with the effort are trying to run it with a wiki and having it lead by people with no technical background."

I am sorry, this is NOT the first time that this type of language has been used. The author of this comment has no idea about the experience that GeneJ and I have, nor our background.

I clearly am NOT a programmer, but I am an End User.

I do not think that this type of dialog is helpful to this project.

Thank you,

Russ
gthorud 2011-04-18T06:05:26-07:00
The last sentence in the statement that started this topic tells me that whoever made it does not have a good understanding of what this is about. Take for example the issue of exchanging data between E/C-models using different number of levels (DeadEnd vs Gentech) and the simple one level Conclusion model used by all? personal programs today. I am not convinced we have solved that issue, and it is not solved by one person.

The statement starts with ba reference to "this" which is not defined, so what is realy the issue?

Re. relational databases. The discussion about the various ways to implement databases is about 40 years old. I note that relational databases has a substatial user base, and is used by almost every major genealogy program. Someone may wish that the world did not look like it does, but this is the reality. The discussion does not interest me as long as someone does not try to design BG a certain way because it fits their database or because they use Gedcom internally.

The work we are doing on sources and citations has for at least the last moths taken as a basis that there will be many citation models wich will emerge ower time, so there is nothing new in that. (But this is not solved by simply having different templates.)

Re. Russ comment. One issue is the tone, which I will not comment - I don't know what to do with it.

It is obvious that the E&C-model needs some education material. If users does not understand it, they will not be able to use it. You can not simply expect the user interface to solve that problem. There must be a set of rules developed that users must follow, and they can't be too complex. But until that set of rules is writen down, all users can not expect to understand every detail in the discussions - they are simply complex.

Some tutorial material and a draft set of rules (in terms of the data model) should be writen ASAP, so we can have as many as possible in on the discussion.

But there seem to be little interest in discussing how to progress the E&C work. People are just messing around in the details.
mmartineau 2011-04-18T10:16:32-07:00
Has the meeting started? My computer says "Waiting for Organizer". Are we still having the meeting or is my computer just not working right?
theKiwi 2011-04-18T10:18:07-07:00
Same here since about 1255pm - Waiting for Organizer. I'm assuming there will still be a meeting - it's certainly not your computer.
GeneJ 2011-04-18T10:19:28-07:00
I'm going to CHAT.
mmartineau 2011-04-18T10:21:59-07:00
Maybe the organizer is having technical difficulties.
gthorud 2011-04-18T10:24:38-07:00
Waiting for organizer ....
GeneJ 2011-04-18T10:25:11-07:00
To access CHAT go to Meetings > Organizers' Meetings then enter chat
GeneJ 2011-04-18T10:25:50-07:00
I'm in the CHAT LOBBY with Kiwi
testuser42 2011-05-03T15:07:31-07:00
April 18th?
Was there a meeting? Are there any notes?
gthorud 2011-05-04T18:49:23-07:00
There was a meeting but no minutes.
AdrianB38 2011-06-10T05:03:49-07:00
Has BetterGEDCOM addressed "died in Y"?
Gene asks:
Has BetterGEDCOM addressed "Died in Y"--subject of The Ancestry Insider blog post, "The Gretna Green for Death? Why Y?"

GEDCOM is actually innocent here. To save you looking, it appears that thousands of people in Ancestry trees(?) died in the (real) location of "Y, Somme, Picardie, France". Clearly, "Y" in this context meant "Y - Yes, this person has died", not "Died in the place named Y".

There are 3 possible causes of the issue that I can think of:
- people misusing the location for the death by sticking "Y" or "YES" into it to show someone's died - the issue then being compounded by something like an auto-fill to provide the full details of Y's location;
- the software misusing the location in a similar fashion for the same reason;
- (more complex) If software wants to indicate someone is dead, but has no date or place, then GEDCOM 5.5. says to put "Y" next to the DEAT tag. It might then be that an application error resulted in that "Y" being moved into the location;

There is insufficent detail in the blog or linked posts for me to understand which of the above 3 causes is the root cause - or indeed, whether there's a further possibility.

Cause 1 is user error. Causes 2 & 3 are application software error. Nothing GEDCOM (and therefore BG) can do about any of those. One can ask that the design of GEDCOM (and BG) be as robust and clear as possible but in the final analysis, there's nothing that can be done to absolutely stop the possibility of an error located between the keyboard and adjacent chair back.

So - right question, but some analysis reveals GEDCOM is innocent.
GeneJ 2011-06-10T11:30:50-07:00
Thank for posting.

:) I should have included reference to the relevant comments TO the blog article.

See the entry by Brucefuimus.

While Bruce aggrees with your comment that "GEDCOM is innocent here..." I wonder if BetterGEDCOM doesn't need a nice clean little check box option?

See the table in AI's article for the sadistics .. only 8 million deaths at London, 4 million at Los Angeles but *86 million* at "Y."
AdrianB38 2011-06-11T08:56:49-07:00
"Bruce agrees with your comment" - I'd hope I did agree with myself! (Sorry - I might have explained I was quoting myself - must get the citations right! - but when I first posted, Wikispaces had decided to log me out so I lost it all. So the 2nd version was less expansive than the first.

Memo to WL - when you asked me with I wanted to say logged in, which part of "Yes" did you not understand? )

Adrian Bruce
GeneJ 2011-06-11T09:23:23-07:00
Oh my.

Do you think a supplemental check box is a good idea?

http://bettergedcom.wikispaces.com/message/view/Developers+Meeting/40114580#40123682
AdrianB38 2011-06-11T09:30:24-07:00
"I wonder if BetterGEDCOM doesn't need a nice clean little check box option?"

Slightly more seriously... I imagine, Gene, that you're suggesting this because you believe that a check box would be robustly separate and not liable to confusion. Alas...

Let me try to explain - a check box is a feature of the user interface (UI) in the application program and only that. If there is a check box in the UI, it gets stored (a) in a new data item on the database / file and (b) (when checked) as a specific value in that item - usually something like Yes, Y, True, 1, etc, etc. If the value is to go out on a file like a GEDCOM or BG file, then the choices for that value are (roughly) limited to what you see on your keyboard.

So if we turn it the other way round, we would have a specific item on the BG file containing a value like Yes, Y, True, 1, etc, etc to imply "Yes, they are dead". Whether or not the program implements this as a check box is up to the software designers.

Now - the thing is, a specific item on the BG file containing a value like Yes, Y, True, 1, etc, etc to imply "Yes, they are dead" that could be implemented as a check box just happens to be _exactly_ what GEDCOM already has. And yet it still went belly up. So I don't see much that BG can do much to avoid the issue.

And if the issue really is, as I suspect, one which arose in an application error or a deliberate misuse in the application coding of the data, then it's difficult to see what to do to prevent it. Choosing a value that cannot be confused with a place-name sounds a good idea. But somehow Y got extended to Y, Somme, Picardie, France. And guess what? If I enquire on people dying in "N" (exact spelling!), I get 44 people dying in "Øn, Sogn Og Fjordane, Norway". Nope - I dunno either.

PS - Clearly exasperation got to me above - I meant to say above: "Memo to WL - when you asked me if I wanted to stay logged in, which part of "Yes" did you not understand?
AdrianB38 2011-06-11T09:33:53-07:00
Just to be clear - for compatibility, BG needs to _allow_ that "Y" against the DEAT tag or its future equivalent, along with all the other cases where a "Y" can be put against a tag to say, "Yes the event occurred, don't know where, don't know when".

But it won't stop some software company or some users misinterpreting "Y".
ttwetmore 2011-06-11T09:51:28-07:00
My recommendation (and what I do in LifeLines).

Death will either be signified by a separate event record or by a "vital attribute" of a person (as is done in GEDCOM and most other systems).

A death event record or vital attribute has date and place attributes if known. When either is not known it is not given. The logical extension of this is that when neither is know, neither is given. Therefore I find that simply using a 1 DEAT in GEDCOM, with no value and no sub-lines is the proper, logical way to say a person is dead but nothing else is yet known about the death. I know this is not the official GEDCOM way, but it should be, and that's good enough for me and LifeLines.

Better GEDCOM should go the same way. Do NOT use Y or YES, just include an empty record or attribute.
GeneJ 2011-06-11T10:41:53-07:00
TYTY for responses.

Sounds like testing might be needed to know if we have a software issue or if it's all in the user approach.

Tom wrote, "The logical extension of this is that when neither is know, neither is given ... ..no value and no sub-lines is the proper."

In this case, I think function trumps the "proper, logical" way. Say you are adding information from an 1850 census. You'll know the folks in that census are deceased, but you might not know when some of them died. (In some cases, when they died might not even be important in the context of your project.)

TMG has a flag for this purpose (Living = Y/N). The flag setting is actually available during the "add person process

To my regret, I didn't use that flag initially. If I'm remembering correctly, I wasn't aware of how important the setting was until I uploaded my first GEDCOM to WorldConnect. Also from my memory, was able to use filters in TMG to automatically correct most of the entries, then another filter to find those I hadn't corrected.
ttwetmore 2011-06-11T14:32:59-07:00
A living tag works just as well. My only comment is that using a simple 1 DEAT (or <death/>) tag provides the same information without adding a new tag or mechanism. I always opt for smaller models, though I know this is not the only criteria.
GeneJ 2011-08-04T10:42:01-07:00
At least this morning, our "recent changes" results are not updating
...at least in my Firefox and Chrome browsers.
Andy_Hatchett 2011-10-10T11:59:11-07:00
Copy of offer to FTM Beta Testers
Someone asked for a copy of the E-mail from Ancestry offering a free copy of FTM2012 to the beta testers.

The link below will take you to my posting on the Ancestry FTM Message Board. A copy of that e-mail is attached to that post. Just click on the attachment to see an enlarged view of the e-mail.

I don't know how to attach something to this Wiki Post

http://boards.ancestry.com/topics.software.famtreemaker/7378/mb.ashx

Andy Hatchett
GeneJ 2011-10-28T14:54:32-07:00
Did we mean to delete link to the 3 Jan 2011 meeting notes?
If not, the link to the page is below.

http://bettergedcom.wikispaces.com/3+Jan+2011+Dev+Meeting+Notes
GeneJ 2012-01-04T14:59:34-08:00
Woot! The DTO's got BYLAWS
Andy said he wanted them done today and by golly we got 'em!
Andy_Hatchett 2012-01-04T16:18:04-08:00
LOL! (coils whip and hangs it on wall.)

;)
louiskessler 2012-01-27T21:50:51-08:00
Randy Seaver's questions for Ancestry.com
Take a look at Randy's questions at:
http://www.geneamusings.com/2012/01/questions-for-ancestrycom-at-rootstech.html

I was very pleased to note his question: "Are you working with the BetterGEDCOM group or the GEDCOM X group on GEDCOM-like standards?"

And Andy, that's a great comment you added.

Louis
bamcphee 2012-03-22T22:30:44-07:00
Meeting minutes 19 Mar 2012
Are there minutes from Monday 19 Mar 2012 Meeting?
GeneJ 2012-04-23T10:02:13-07:00
"Inactive Plugin"
For some reason, GoToMeeting isn't loading. I receive a message "Inactive Plugin."

Just me, or are others receiving the same notice? --GJ
GeneJ 2012-04-23T10:04:01-07:00
Seems this was a notice that Chrome needed a helping hand. All better and seems to be working now.
louiskessler 2012-07-09T12:19:37-07:00
09 Jul 2012 Meeting
Sorry about dropping my headset and just leaving the meeting without saying anything.

I saw smoke outside, looked out, and on the street, a parked car was on fire.

Fire department came. Car extinguished (but totalled). No one hurt. It will be on our local news tonight.

Always something for excitement. :-)

Louis
GeneJ 2012-07-09T12:26:41-07:00
Oh, goodness. You were missed. Thank you for taking the time today, Louis. Good dialog. --GeneJ
GeneJ 2012-09-17T09:58:51-07:00
GoToMeeting reporting error
GoToMeeting is telling me my meeting ID is not correct. "Enter a valid meeting ID (ie 000-000-000)"
GeneJ 2012-09-17T10:01:00-07:00
I was able to enter with this access code:

345-633-878
theKiwi 2012-09-17T10:03:35-07:00