Re: CNC Control software recommendations

mfea_2000 wrote:

> Matt,
>
> As for who I work for, as with most companies, I am bound by company
> policy to keep that information confidential.
>
> I can tell you that we produce CNC controls and hardware to retrofit
> to the mills. We sell through distribution and directly to end users.
>
> I don't want this to start the guess who Alex works for game. If
> anything pans out I will fill in the blanks at a later date.

OK, fair enough. I had to try though, ;) .

If a linux based solution would be acceptable, I can recommend the EMC
system. Details can be found at:

http://www.linuxcnc.org

I have thought for some time that the _FIRST_ CNC control manufacturer
that adopts the EMC will gain a huge competitive advantage if (and ONLY
if) they use the publically available version that everyone can access
through the CVS at Sourceforge. Here's why:

1. Most shops that buy and use CNC equipment can relate their horror
stories of poor or no support when problems arise and pricing for parts
and service that border on extortion. In a nutshell, once burnt they are
twice shy.

2. Some CNC control company needs to offer a modular, open archetecture
system that:
a. would always be repairable at a reasonable price and...
b: could be upgraded piecemeal in the future as a hedge against
obsolescence.

3. Considering the widespread reality of #1, selling #2 should be
reletively easy because the potential customers already understand
(really well...) the problems with closed, proprietary technology.

4. HOWEVER, the benefits described in #2 HAVE to be GENUINE, not just
unfulfilled promises. That's why the control has to use the publically
available code with NO proprietary extensions. ALSO, that part about
"modular, open archetecture" means the system has to be understandable
enough for even third parties to effect repairs and to enhance and
extend the system's functionality.

5. If they read this far, astute managers will raise some version of
this objection:

"If we don't keep some of our system proprietary, competitor X will just
download the publically available code and start selling knock-off
systems in direct competition with us!"

This is much less of a threat than it appears because if your system is
attractive enough, in form, function, and price, it will be hard for a
competitor to develop an alternative product with all the benefits of #2:

a. Since you'd both be using the publically available code, the
competitor's product would look almost identical to yours and you could
counter their advertising with "why buy an imitation, when you can have
the original", or something to that effect. Machinists have incredible
brand loyalty, IMHO.

b. If the competitor tries to gain some advantage by developing a
proprietary extension, you can easily counter by accusing them of trying
to do the things described in #1.

These are the reasons it's important to be the _FIRST_ to adopt the EMC...

6. If you stay true to the principles in #2, you'll probably have the
support of the EMC user/developer community. You'll have help in
troubleshooting problems and adding features as long as all the
improvements are plowed back into the publically available codebase.

7. At the end you might ask, "Well Matt, if your so smart and all fired
confident, why don't you start up a company and do this yourself?".
There are two reasons:

a. I lack sufficient capitol to pull off something this ambitious.
Think about it; inventory, 1-800 tech support, etc. This is way more
than a one man job.

b. It would be really helpful if there were existing name recognition,
i.e. if Centroid or Southwestern Industries or Fadal or Haas or Fanuc
(yeah, right) or Hurco (even less likely than Fanuc) did this, they
would have an advantage over an unknown company. They might want to say
something like, "We've been listening to our customers comcerns about
proprietary cnc technology and we agree! That's why we've decided to go
in a bold new direction with our CNC control software, a direction that
the rest of the industry will soon follow...".

These are the reasons I asked the name of your employer. Without knowing
their identity, I can't guage whether this idea would work for them or
not. If any of this resonates, let me know.

Matt