GDG- Time Lines

[Alan D. Brunelle]

Tom Gilbert wrote:
> Esteemed GDG Member Contributes:
>
>
> Sure, I'll try to do that .. right now they are handwritten in a notebook, probably 100 pages or so .. I just make note of item along with date and/or approx time during my reading, have pages in my little looseleaf notebook according to the pattern I mentioned before .. I also, of course note the reference(s) of my information .. tom g
>   
A slight digression: I actively work in what is called Open Source 
Software (Linux, to be exact), and one of the great aspects of Open 
Source is that the code is available to everyone, and they can modify it 
as they so choose. The code is kept in a central repository, where it is 
maintained by a bunch of different people (who have "control" over 
different aspects of Linux, with Linus Torvalds (the original author) 
having "final say" in the matter). *Anybody* can request a 
modification/addition to Linux (via the notion of posting a "patch" to a 
well-known mailing list). Then there are various people who "agree" with 
the patch, and eventually it gets put into the main tree. [Or not: if 
the Powers That Be disagree, it is held out of the main tree.]

An idea that I've always held in the back of my head was the notion of 
having an Open Source database of Gettysburg facts - a single repository 
where people can request to place in new data and/or modify existing 
data in the database. The goal being that over time the database becomes 
more complete and more accurate as more eyes look at it. Taking Tom's 
data as an example, he could post his data - and then the knowledgeable 
people on the list can agree, disagree, or modify this data as it comes 
in (prior to going into the database). Others could then "fill in 
holes", and as time went on if new ("better") evidence comes in, the 
data can be modified.

The key things are (a) active participation, (b) a goodly sized staff of 
"knowledgeable people" and (c) a controlled database environment. 
Whether or not one needs a "Supreme Overlord" (the role Linus Torvlads 
plays with Linux) is an interesting question for this database, 
certainly if there were a good set of "knowledgeable people" then when a 
large majority of them "agree" with something would be sufficient to 
getting the data change in. [This hearkens back to some of the education 
we on the list have received on "good history" - what to look for when 
evaluating new theories and/or data. Incoming data must undergo a 
certain level of scrutiny, and pass certain tests before being allowed in.]

The community benefits from those that do the work to post information, 
and those that post data benefit by having others fill in data they do 
not have access to.

The key thing towards understanding this is the concept of "multiple 
eyes" and "multiple hands." As people contribute ("multiple hands"), 
more and more data is put into the database - making it more important 
to others, and by word-of-mouth (or word-of-Google) more people start 
using it (and relying upon it). As more people are involved ("more 
eyes") the data becomes better - problems are found quicker, and 
corrections find there way into the database.

Right now, Gettysburg data is spread all over the place - books, 
magazines, WEB sites, first-hand accounts squirreled away in boxes in 
libraries and historical societies, and probably not a small amount just 
sitting in peoples heads... Whenever anybody wants to write a new 
article or book, they have to go to great lengths to find the data - and 
the likelihood of being able to find all the pertinent data is pretty 
small.

Now, having such a database is in itself not an end-all - it would take 
quite a long time (if ever possible) to get /all/ the facts stored in 
one place. But certainly one could imagine that "simple" things could 
get put in quite quickly (orders of battle, for example). Then as time 
went on, more and more complex stuff could be added. One could envision 
that the notion of a time line of events could be added as well - taking 
into account Jim Cameron's noted concerns regarding "accurate time data" 
- one could simple build relational connections between data ("_A_ 
happened before _B_"). Then certain "accurate" times could be added - 
again, taking into account Jim's comments regarding "it happened at 
sunrise" or perhaps using some agreed upon time ("the cannonade started 
at 1:07pm" could be used as a metric when gaging events /before/ and 
/after/ that).

The building up of a more and more complete time line would tremendously 
help narrow down possibilities: /if/ a very accurate time line can be 
crafted at a very fine level, /then/ one can start throwing out invalid 
theories if they don't match the time line. (Someone espousing a theory 
based upon "I believe _A_ happened after _B_" could easily be shot down 
if it were easy to prove that _A_ happened before _B_ given a lot of 
surrounding relationships with _A_ and _B_. "We can prove that _C_ 
happened after _A_, and we can prove that _C_ happened before _B_, so 
therefore _A_ had to happen before _B_." This puts the onus back upon 
the original theorist to now disprove the relations of _A_ & _C_ and _B_ 
& _C_...)

Just tossing this out there, I don't know others have tried this before, 
or if people think it's a dumb idea. In any event, I'd like to hear what 
others have to say.

Alan