On 3/18/2016 7:46 AM, Richard Hipp wrote:
Separately, I have "improved" the graph generator (on a branch) so
that when you have nodes (such as the above) where the node is near
the bottom of the graph but its next child is off the top of the
graph, then the upwards arrow goes up just a little ways, not all the
way to the top of the graph. Example:
Old style:http://www2.fossil-scm.org/fossil/timeline?b=b8c7af5b
New style:http://www.fossil-scm.org/fossil/timeline?b=b8c7af5b
In the old-style example above, you can see that the second and third
"rails" from the left go all the way to the top of the graph. In the
alternative new-style those same rails terminate just above their
top-most node, which allows the cluster of branches near the top of
the graph to be closer to the left margin.
I like the idea a lot, and I was surprised that I don't seem to like it
as much in actual rendered graphs. I'm not sure why, but it's something
subtle.
It reminds me of the judgment calls when drafting a schematic over
whether to draw a signal as a line connecting two pins (with or without
a name) or to just label stubs at each pin. Too many lines produces hard
to follow rat's nests that hide information. Too many labels without
lines just moves the rat's nest inside your head. There's a real art to
string the balance so that the necessary information is displayed
without too much clutter.
I wonder if it would help if the connecting lines coming in from the
bottom were shortened to a comparable length also?
Look at
https://www.fossil-scm.org/index.html/timeline?b=2015-12-23+21:35:43&n=50&y=ci
and notice [2553] near the top of the page, with a line to it from
[9a69] almost at the very bottom of the "older" page. The previous entry
got an arrow, which is nice. But the long line leading down the page on
the last doesn't hint that the last change was longer ago.
What about the occasional line nearly the length of the page, should
that get a north arrow stub at the lower node and a incoming stub at the
higher node?
--
Ross Berteig r...@cheshireeng.com
Cheshire Engineering Corp. http://www.CheshireEng.com/
+1 626 303 1602
_______________________________________________
fossil-dev mailing list
fossil-dev@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/fossil-dev
