Rowan Sylvester-Bradley wrote:
> Thanks for your reply and thoughts.
>
> "singhals" wrote in message
> news:mailman.9.133337...@rootsweb.com...
>
>> The answers for the web-display and for the print-offs are different.
>
> Of course they are - but related, surely? I want to solve both problems, and
> maybe there is some software that does both. If not, fine - I'll have to
> have two solutions.
>
Not related in my experience. On paper, your size is
limited by the dimensions of the available paper. A4 or
8.5x11 or multiples thereof. Digital, you can specify the
size of the chart and if you end up scrolling 8 miles, you
scroll 8miles.
>> A "tree" is by its very definition a display of ANCESTORS. Whether in print
>> or on-screen.
>
> Whose definition? Why shouldn't descendants be a tree - it's a perfectly
> good tree structure with a single root, plus branches, sub branches to
> whatever depth and then leaves. It just has the root at the top, not the
> bottom.
I don't /make/ the rules, I just report 'em. I was told
that back in 1971 by an 80-year-old Genealogy instructor who
had been told the same back when he started out, well before
I was born. I quite agree that logically roots are below
ground and should go DOWN from oneself; however, that's not
how trees are constructed.
>
> But OK - I used the term "tree" loosely, and maybe "chart" would be a better
> word. But it's too generic for what I meant. I meant in particular a diagram
> showing how many people are related by birth and by marriage.
>
At a certain point, which I find to be somewhere around 500
persons, such a chart or diagram or tree or what-the-flip
becomes too large to be useful. A map of the United States
or even England which is large enough to show the entire
country down to legible small-town names is going to be
cumbersome to use...perhaps one-eighth life-size.
>> It will however, be happy to print CHARTS for you on paper. You can always
>> print those to file, open them in WORD, and then save-as html.
>
> This must be about the worst possible way of creating a web site. The
> resultant site can surely then make no use of web specific features which
> ought to make it easier to navigate around a large diagram. Also Word seems
> to create a huge amount of "clutter" in the HTML that seems to be designed
> to reproduce precisely what was there in the printed Word document - which
> is the opposite of what you want in a good web site.
>
I didn't say it was the BEST way. The BEST way to do what
you want is to produce the html with some sort of generator
and then hand-edit it.
>> The easy way to do that is to create a GED that includes ONLY the items you
>> wish to include. Import that GED into a new database and push PUBLISH WEB.
>
> This seems a rather cumbersome way of customising a chart. Possible if there
> are no better facilities though.
>
May be cumbersome, again depending on the number of people
you're dealing with, but it works easily and is available
whatever program you use.
>>> 4. Intelligent chart layout to minimise the length of long links, and put
>>> closely related people close together, and make the chart easy to
>>> navigate,
>>> and to minimise the number of pages...
>
>> I've used several programs and none create exceptionally long links. Most
>> are of the ".../filename.htm" order.
>
> GenesReunited does. You often find you are following the "third line from
> the left" over pages and pages and pages until you find the person it came
> from, by which time you are never sure whether you are still following the
> right line, or have slipped one to the left or right. There must be a better
> way.
>
Ahhh. You're not talking html links, you're thinking of
lines on the screen. Can't help you there. On the wall
charts I print to paper, I generally use Magic Markers (tm)
or HiLiters (tm) to follow the major lines. I hand-edit to
use different font-faces if I'm going digital.
>> You minimise the number of pages by putting more on each page. That's an
>> option in most generators.
>
> I was thinking of trying to arrange the people on the chart in the way in
> which minimises the space taken, and the lengths of links between them. By
> analogy with routing of printed circuit boards, if you put the chips down on
> the board in a non-optimum order, it requires more tracks, and more track
> crossings, so it all takes more space and performs less well. Auto-routing
> programs therefore spend some effort optimising the placement of the chips
> before they ever try to connect them up. By analogy the same thing should
> help with a chart. It's not obvious, it seems to me, how best to arrange the
> chart, especially since each spouse that marries into the family potentially
> requires another ancestor tree to be fitted in, and those ancestors may
> themselves have descendants etc. etc. I think there is scope for some
> software to help optimise this arrangement.
>
I know people who use org-chart generators for that.
Personally, I think it's more bother than it's worth and the
chances of transcription error somewhere are up around 99%.
FTM does a box chart or drop chart and yes, you may
rearrange the boxes.
But, any way you go, you'll end up with: the more info in
the box the bigger the box needs to be to hold it; the
bigger the box the more space it'll take on-screen or
on-paper; the more space each takes the more pages or
screens it will take to hold all of them; and the more
screens it takes, the easier it is to get lost.
There's a reason those charts you see in biographies rarely
follow more than one child for more than 4 or 5 generations.
>>> 7. You can manually adjust the automatic layout, and such adjustments can
>>> optionally be saved and reapplied even if the automatic layout process is
>>> re-done, including if there are changes to the chart. Obviously the
>>> changes
>>> may be such as to make the manual adjustments meaningless, in which case
>>> they are abandoned.
>
>> There's a special software for this -- TreeDraw.
>
> Thanks for pointing this out. I'll have a look at this package. It sounds as
> if it does a number of the things I was aiming for.
>
>> Someone is bound to mention PhP -- you might try it.
>
> Do you mean PHP the web programming language , or something else?
>
Yes, but it seems to be being discussed as if it were a
program ... my reading of its benefits is that fans think it
will solve everything but World Hunger.
>> In general though, I think you'll find that panning and zooming are mortal
>> enemies of not-lost.
>
> So without pan and zoom, how do you navigate around a chart that has 2000
> people on it and if printed out would cover a large wall, so you can find
> any individual, see how they relate to people you know in the chart, easily
> get back from this person to your familiar area of the chart etc.? This is,
Far's I've ever seen, you don't. Several programs probably
make it possible to DO, but it won't be easy to create and
it won't be easy to "navigate" and most folks will get good
and bored before they find the hang of it. That's
EXPERIENCE speaking, now. Most non-genealogists can't even
follow a simple one-page descendancy chart. It isn't
possible to make it easy enough for them. And most
genealogists won't want to do the "how are these two people
related" drill, when they can just ask the program.
> it seems to me, a difficult problem which needs very good tools to solve.
> I'm interested in any good tools that may exist in current software
> packages, and in ideas for new ways of doing this.
It's a difficult problem, yes.
Case Study Follows feel free to quit reading already:
One of my family associations wanted a Descendancy chart for
their reunion. I sent the planners a sample, and we reduced
the project to include only the lines of those who had
already registered to attend. The resulting drop-chart took
100 sheets of paper, which I glued together. We had to
pleat it up to fit the only wall without a door. Of the 80
or attendees, our 6 officers looked at it.
Couple years later, they wanted family group sheets; again,
75 attendees, our 6 officers looked at the sheets.
And 3 years later, I took a draft of the mss of the
genealogy, complete with index (3 reams of paper) and
absolutely no one looked at _their_ entry. Several picked
it up, read the top page, and put it back down.
It was a waste of my time and their resources.
Not a sermon, just a thought, as a local ad claims.
Cheryl