[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: gEDA-user: lost newbie



> On Saturday 07 January 2006 03:08 pm, Stuart Brorson wrote:
> > [1] =A0To the well-known simulation developer who e-mailed me
> > recently about LTSpice: =A0Sorry, but LTSpice is a wonderful
> > simulator. =A0 And it's free (as in beer) too, which is
> > probably all that Marc cares about.
> 
> I meant what I said.  Every word of it.=20
> For those who missed it, here it is again:
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
> Promoting the competition again.....
> 
> Closed-source freeware is our most threatening competition.
> 
> Another possibility is Oregano and gnucap.  This is a more=20
> integrated environment, and is still GPL.
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
> 
> If we can't beat LTSpice in the eyes of a newbie, we might as=20
> well shut the whole gEDA project down.  We need to beat them on=20
> their terms.
> 
> I believe we can, but we need to make some changes.  We need to=20
> recognize our weaknesses and deal with them.  In some cases,=20
> major surgery is required.  We need to take the lead.
> 
> We should own the beginner market.  Not just be a player in it,=20
> we should OWN it.  I believe we can do it.  All of "free=20
> technology" in electronics is dependent on us.  We are the=20
> ambassadors of free/open-source to EE's.
> 
> When we tell newcomers to go away, to such a competitor, we=20
> admit failure.  When we tell them to go to a competitor that=20
> laughs at us, we really admit failure.
> 
> Oregano is a friend.  It is GPL, etc...
> 
> Gnucap is here.  The developers are on this list, unlike=20
> NG-Spice...
> 
> But LT-Spice, Eagle, the PSPICE demo, the Multi-sim demo .... =20
> sorry.

Al,

Your large stature in the open-source EDA community deserves a better
reply from me.  Unfortunately, I don't have the necessary time to
formulate and write all my thoughts elegantly enough to do you
justice.  Some day I will write and post an essay about the economics
(in the broadest sense) of free/open-source software.  There is a lot
to say about it.  

As for your points about commercial vs. open-source:  Unfortunately,
LTSpice is simply better for certain newbies than gEDA.  It's a
drop-dead Windoze SPICE simulator with integrated schematic capture.
It's an all-in-one package simple enough for a total fool to use.
Doing SPICE or Gnucap using gEDA, on the other hand, involves chaining
at least three programs together: gschem -> gnetlist -> [ngspice,
Gnucap].  Each program is run from the command line.  Each program has
a learning curve.  Getting results from the simulators requires
learning the CLI langurage.  The learning curve for each isn't bad,
really, if you have a little experience.  However, for the total
newbie who doesn't understand the first thing about electronics or EDA
it can be daunting. 

Therefore, at the present time LTSpice is simply a better alternative
for an impatient, whining, clueless newbie.  It gets him to his goal
faster, and with less grief than trying to learn gEDA (or any other
package, too, I suppose).  In the future, I hope that gEDA progresses
to the point where it is a credible alternative to LTSpice.

> When we tell newcomers to go away, to such a competitor, we=20
> admit failure.  When we tell them to go to a competitor that=20
> laughs at us, we really admit failure.

I have three points about this:

1.  First off, I don't view gEDA as a competitor to anything.  It's an
open-source project which aims to be useful to folks doing
electronics.  If it's not useful to some people, that's fine too.
Open source is about cooperation, not competition.  Folks who want to
compete should go into commercial software.

2.  Some time ago, somebody posted here a link to an article about
"when you should fire your customer".  That is, as people who develop
software, we have customers.  Many of them are happy to use the
software, since it allows them to be productive.  Some of those users
actually contribute back to the project, which is great.  On the other
hand, some customers do nothing other than complain and whine and
demand extra attention to solve their problem.  They don't bother to
read the documentation, but they have no idea about what they are
doing.  And if you don't solve their problem, they say it's because
your software stinks.  Basically, such customers are cranks.  It's
generally good business practice (as well as open-source project
practice) to give cranks the brush-off, IHMO, since they can do damage
to your project.  They can badmouth it publically.  They can cause
dissention amongst developers.  They will waste your development time,
and so on.  YMMV.  

3.  Also, when helping an newbie, I tend to do a cost-benefit type
analysis.  That is, I am happy to answer a few questions, and point
*any* newbie in the direction of some resources.  The cost in my time
is small.  If the newbie seems clueful and might make some kind of
contribution back to the gEDA project (i.e. code patches), then I'm
happy to spend *lots* of time helping him, since investing time
into somebody intelligent can bring greater rewards down the road.  On
the other hand, some newbies clearly haven't read any documentation 
or even done any Googling about gEDA, SPICE, or much else.  If such a
newbie also seems to require lots of hand-holding, but then
won't do anything more than get his circuit working and disappear,
then he's not a good investment for my time.  There's no reason 
such a person shouldn't be directed to LTSpice, since the learning
curve is (somewhat) shallower, and the folks on the Yahoo group
can waste their time on him, rather than us.  YMMV, of course.  

Respectfully,

Stuart