To recap,

XML usage is a dead subject as far as the geda project is concearned.


For the following reasons

1) xml tends to be bloated file size wise.
2) incorporating xml into geda would distract developers.

My belief, is that the files are what the files are because no developer
has taken an interest in making them something else.

Do they work. Sure for flat designs. But, not for complex hierarchical
designs.

My reality is that I forked from geda back in 2003. I made a hap hazzard
attempt to merge my code with the 2005 geda version. Then I decided that
libgeda data structures as is just don't support complex designs. This
is not saying thet they are trash they have a lot of important concepts
with in them. But as they stand they are not adequate.

If geda is to be relevent in the future it needs to address:

1) hierarchical data and file structures (xml or other)
2) an integrated hierarchical netlister
3) use of a better supported scripting language
4) a method of dynamically partioning devices into meaningfull symbols
5) a method of communicating between the schematic design and the layout
program on issues of slot swapping, i/o pin swapping, differential i/o
pin swapping, bank swapping
6) strong drc
7) quality libraries of symbols, land patterns and simulation models
8) a clear deffinition of what is GPL and what is available for
commercial use. I have even dumped the busripper.sym because I don't
know if it would contaminate my commercial projects


Thanks,

Steve Meier


ps geda is now available through sourceforge and 7 developers are
registered. I have in the past for other projects made donations to
support development. It looks like of the seven developers only one is
interested in such donations.


_______________________________________________
geda-dev mailing list
[email protected]
http://www.seul.org/cgi-bin/mailman/listinfo/geda-dev

Reply via email to