Re: gEDA-user: gEDA just hit SlashDotOrg

[DJ Delorie <dj@xxxxxxxxxxx>]

> And from here it looks like you don't want to consider my point.

Your point is that gEDA is a powerful flexible collection of tools
that motivated people can (and should) use to do wildly different
things.

My point is that that will never happen if people can't even justify
trying it because the initial learning curve is too high.

I've conceeded your point many times, but that doesn't preclude me
wanting to add some structure to help people get started.

> No it's not. gEDA is a general purpose toolkit, one (and only one) of  
> whose applications is feeding your pcb program. ISE is much, much  
> more specialized.

Given the wide range of things that ISE can do, I think it qualifies
as as much of a general purpose toolkit (in it's realm) as gEDA does
(in its realm).

I mean, if you want to take the "gEDA is a general purpose toolkit"
too far, I say we just ship everyone a C compiler and let them put
together whatever kind of EDA tool they want.  When we create a
package like gEDA, the value in the package is not just the pieces
within, but how they work together, and how well that combination
solves the user's problems.  Perhaps a gschem->pcb flow is sufficient
for most people, perhaps we need to add a gschem->spice flow instead.
Or a gschem->fpga flow.  Maybe we need to focus on a point-n-click
"send this board/chip to fab" database.  But what we have now doesn't
have enough hand-holding guidance to get people started, regardless of
what they want to do.

> I have no objection to wrappers. What I object to is the constant
> demand to fix perceived problems by violating the fairly clean,
> modular nature of the kit. Rather, we need to make things *more*
> modular (e.g. get the hardwired behavior out of the gnetlist front
> end).

I'm not arguing against your desire to keep geda modular, or to demand
that things be fixed one way or another.  This is the "geda just hit
slashdot" thread, and the original problem noted (on /.) was that gEDA
is too hard to get started with *because there was no Windows
installer*.  What gEDA did was irrelevent because it never even made
it to the user's PC.

Expecting a hardware designer to think outside the box is one thing.
Expecting him to become an expert in an unrelated field (cygwin, gcc,
packaging, posix-like build environments, C, where did I put that
guile-on-windows FAQ?) is too much - he'll just download something
else that works out of the box.

> So run the analogy the other way if you wish. Someone who attempts to  
> use a hand tool as if it is a power tool will be frustrated, but  
> turning it into a power tool is not the answer.

It runs either way.  People downloading an EDA package expect it to
run, at least somewhat, like all the other EDA packages.


_______________________________________________
geda-user mailing list
geda-user@xxxxxxxxxxxxxx
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user