On Tue, Nov 4, 2008 at 2:42 PM, Petter Urkedal <[EMAIL PROTECTED]> wrote:
> On 2008-11-04, Timothy Normand Miller wrote:
>> Also, there are updates to the tools that Howard has made that need to
>> be merged.  Also, I myself prefer modularity.  Perhaps a compromise
>> with Howard's needs would be to ensure that everything stays within
>> one directory.
>
> How about a small API for accessing the card?  I made a sketch of some
> headers,
>
>    
> http://git.eideticdew.org/cgit/ogp-pu/tree/tools/oga1lib/oga1?h=oga1lib_suggestion
>
> I'm not sure about the top level directory name (tools/oga1lib), and
> other suggestions are welcome too, aesthetic or semantic.

As a software engineer, I love complete, well-documented APIs.  _I_
think you should go for it.


Howard's statement prompts me to wax philosophical for a moment...

You'll encounter with any engineering discipline, science, or art a
set of attitudes and specialized needs that are quite different from
those held by those of us who come from Computer Science.  Howard,
like many people who come from an Electrical Engineering background,
is competent with software coding, but the software (coding and
end-user apps) are merely tools to serve the job off designing
hardware.  Throw a hierarchical chip design at your EE guy, and he'll
think it's great.  Throw a hierarchical software design at the same
guy, and he'll appreciate all the analogies, but he isn't any more
interested in software engineering than mechanical engineering.

We should keep this in mind when we develop tools for different
audiences.  As an open hardware project, we need to make sure that
software libraries are simple, and applications are exactly on point
and intuitive to the specialized user.  Of course, the hardware can be
as complex as we want it.

It should not be inferred that hardware designers aren't interested in
freedom as much as the software engineers.  It's just WHAT they want
to be free is different.  Stallman wants software to be free but isn't
too concerned about the hardware, as long as that doesn't interfere
with Free Software.  I and my friends want hardware to be free, and
we're willing to compromise a bit on the software (i.e. drivers for
Windows and using proprietary tools) for the sake of the hardware.  A
mechanical engineer is going to want blueprints and machining
processes to be well-documented but isn't necessarily so concerned
about the programs he uses to run his CNC router.  These differing
priorities don't really complete, because we're all working to make
different kinds of knowledge freely available.

-- 
Timothy Normand Miller
http://www.cse.ohio-state.edu/~millerti
Open Graphics Project
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)

Reply via email to