Re: emc really needs a copyleft cad/cam package

[Jeff Barlow <Jeff-at-Barlow.net>]

Matt,

The point that I (and, I think, Anne) are trying to make is just to take
this one step at a time and not get ahead of ourselves. 

Your list of steps, while somewhat incomplete, is in about the right
order. The wish list is definitely a good place to start. It's clearly
about WHAT we are proposing to build. The rest of your list items are
all more about HOW we will build it. 

There is a bit of the WHAT hiding within number 5, though. That is, we
need to have an idea of which of the existing file formats we intend to
support. Our internal data formats are clearly part of the HOW, though,
and that must come later, after we have beaten the WHAT issues to death.

My original background is also in the hardware end of this field. I made
the transition from "hardware guy who has written a few programs" to
"software developer / team leader" over the course of about twenty
years. I have learned, sometimes very reluctantly, that there are
important differences in methodology required for large software
projects versus hardware design.

Large scale software projects tend to be overwhelmingly complex. This is
true not because they are software, but because of their sheer size.
This is hard to fully appreciate until you've "been there" a few times.
I think this is the main reason that many fail. 

I'm not trying to scare anyone away. I believe this can be done. I think
we have the right mix of folks here to do it. I think we can manage the
complexity. The best analogy I can think of right now is as follows.

You can move a wrench from one end of your shop to the other without
having to think too hard. You can also move a two ton machine the same
distance. The latter, however, requires a lot more careful forethought,
lest the thing fall over and crush you. Both are doable, though.

I don't think there is any one right way to do this stuff. I have been
involved in doing it wrong enough times, though, that I think I at least
know a bit about what not to do.

There is a huge temptation, at the start of a project like this, to want
to move quickly from the very abstract to the more concrete. I'm sure we
all feel this. It's who we are; it's what we do. On smaller projects we
can get away with it. On one of the scale that we are contemplating it's
deadly.

I know I'm beating this topic to death, but I really believe it is that
important.

BTW, lets move this discussion to Terry's new list.

Jeff

On Sun, 19 Nov 2000 23:41:34 -0500 (EST), Matt wrote:

>
>Maybe it is premature, but I don't know what else to do! The only things
>that come to mind are:
>
>1. Wish (feature) lists
>
>2. Block diagrams
>
>3. Sketches or actual mockups of user interfaces
>
>4. Investigating possible languages and libraries
>
>5. File format definitions
>
>What should we do? I'll admit that my concept of CAD/CAM programs is
>defined by those programs I have seen or used, and I just expect others
>will have similar concepts (this may be an error). Should we decide to
>clone an existing application and fix the things we perceive to be
>deficiencies? I'm really an electronics technician/machinist by trade
>who has written a couple of programs so I don't know how this is
>supposed to go.
>
>Matt
>