TMG-L Archives

Archiver > TMG > 2002-02 > 1013873784


From: Walt Flory <>
Subject: RE: [TMG] Bob V., one man conspiracy? (was: Death to...)
Date: Sat, 16 Feb 2002 10:36:24 -0500
References: <05ee01c1b6a9$c8bd9340$6501a8c0@office1>
In-Reply-To: <MABBJHBCCCAFJHJGELJDKENOEGAA.holly.hendricks@att.net>


Holly

This was extremely interesting.

Thank you,

Walt Flory



At 07:54 AM 2/16/02 -0500, you wrote:
>The titles of the conversations keep getting more outrageous.
>
>I haven't been participating much in the various permutations of this one,
>but wanted to add some insights I gained at Gentech last month. One of
>the best sessions I went to was given by Robert Charles Anderson, the
>compiler/author of The Great Migration Begins. If you want to meet someone
>who thinks about longevity of data, the nature of structured data, how to
>build databases to support the kinds of linkages we might have in the
>future, and the right balance between content and presentation, attend one
>of Anderson's sessions. He claims not to be a techie, but has surely
>grappled with the data structure issues like a pro. His goal is developing
>data bases for multiple media: book, CD and web.
>
>He recently decided that his ancient word processing program of 12 years or
>so, DOS based XYWrite, would no longer serve in creation of this work. He
>told session participants that he decided to find the best people in the
>industry who were grappling successfuly with cutting edge data issues to
>develop a long term strategy. After considering MS Word, he realized
>that he could move directly to an XML-based solution, and he told us that he
>hired the best consultant he could find to help develop the XML strategy and
>templates: Bob Velke.
>
>My notes from the session include the following:
>An XML solution for TGMB
>- is simple and elegant
>- supports putting elements of a data base into prose
>- allows one content file to generate multiple outputs and different
>versions
>- supports separating evidence from conclusions and assertions (and should
>be integrated with genealogical data models)
>- must be able to support new projects and yet unknown types of output
>simply by writing a style sheet with no change to the underlying data
>- should allow you to check validity against the underlying schema
>
>Other notes
>- Because HTML is fixed and can only be changed by a committee, it is too
>limited for his purposes; XML allows user defined elements
>- Check out the XML spy tool for authoring (I think he might have said this
>was a Bob V suggestion)
>- A future direction for his project will be doing more with data attributes
>- RCA put in the links between the sketches by hand for the web-based
>version, and it took 120 hours
>
>I think we will look back on Anderson's work someday as we do that of Donald
>L Jacobus in a previous generation because he is breaking new ground and
>setting new standards in support of a very high standard of data integrity.
>I marvel at Bob's restraint in these discussions when he could easily engage
>in name-dropping about some of the consulting projects he is doing outside
>of TMG to break ground in areas some people here are complaining that he
>does not pay enough attention to ;-) All of this will probably cycle back
>into TMG at an appropriate time. I appreciate more than ever that Bob does
>not rush to market with the latest bells and whistles until the underlying
>design work can be done right, and developed and QA'd appropriately.
>
>Anderson's session was early at Gentech. Afterwards, I volunteered at the
>TMG booth regularly for the remainder of the two days. It was very hard to
>get Bob's attention for sales questions because of the steady parade of
>conference speakers and presenters and other senior people in the field
>coming into the booth to consult with him at length. Luckily, Lissa was a
>great resource for such things.
>
>Holly
>
>PS, for those who like code, a sample from Anderson's presentation showing
>user-defined XML tags
>The simplicity of his model in xml was interesting. Within each
>category, evidence is separated from conclusions and assertions, and that is
>not shown in the simple example below.
>-----------------------
>[headers, version, style sheet, parse info]
><sketch [URL of schema and namespace]>
>
><sketchee><name></name></sketchee>
><migrate>
> <origin></origin>
> <migration></migration>
> <first_residence></first_residence>
> <removes></removes>
> <return_trips></return_trips>
></migrate>
><bio>
> <occupation></occupation>
> <church></church>
> same format continued: freeman, occupation, education, estates
></bio>
><gen>
> <birth></birth>
> same format continued: birth, death, "spice"[=spouses], assoc
></gen>
><comment></comment>
><biblio></biblio>
>
></sketch>
>---------------------
>
>
>
>
>==== TMG Mailing List ====
>Please remove any "Reply-To:" address in your e-mail program. Including a
>"Reply-To:" prevents others from benefiting from replies to your questions.




This thread: