Re: Clear-Sky/software and motion controllers

[Chris Albertson <chris@topdog.pas1.logicon.com>]

Glenn Gombert wrote:
> 
> Chris et all;
> 
>         Below is an interesting WEB site that I came across that offeres a
> full commercial verison of an Observatory (Camera, Telescope Control
> Positional System,etc) control system that is all setup to run under the
> Linux operating system:
> 
>                         "http://www.ClearSkyInstitute.com/"


Looks to me like you could buy an entire MkIV setup off the self
from these guys.  You could get the 16 bit 2K x 2K cameras, stepping 
motor control, shutter, filter wheel, focus, and shaft encoder 
feedback, dome control... the works, all under remote computer 
control.  They even interface to GPS receivers for portable setups
The software looks first rate too.  I know how hard it is to
write those Motif X Windows interfaces.  That is why I did not
propose doing one for TASS.  It is clear that more than a
couple man years of engineering went into the project.  It is all
good stuff but You'd have to pay $20K per system before you
were done. 

I was wondering how they are ever going to make any money.  
The software looks like it cost a lot to develop.  They claim
200K lines of C code.  Using only low industry cost figures, if
they sell copies at $4000 each (they do) they'd need to sell to 
about 500 automated observatories just to break even.  Let's 
assume they can get a 50% market share (very optimistic) Are 
there really 1000 observatories wanting to automate them selves?

Then I found this notice saying that the system was built with
essentially a combination of slave labor and tax dollars:

> Clear Sky Institute, Inc., was formed in November of 1997 
> to bring the research in robotic telescope systems software
> developed at the University of Iowa since 1993 to a wider 
> audience as a fully supported industrial-strength solution for
> local and remote observatory operation. The principle 
> investigator of that effort was Dr. Robert Mutel. The original
> work was developed with NASA and NSF funding and is freely 
> available at
>      ftp://gastro23.physics.uiowa.edu/pub/telescope

I checked it out.  There is lots of source code there and some
precompiled binaries.  It is a big system.  The design is not
based on any of the standards talked about here but the design
is _very_ conventional.  In fact I'd say "classic textbook UNIX"
Loadable Kernel level drivers, daemon processes and Motif user
interface.  The system is well done and I at least can learn
something by studying it.  If I had a scope with a CCD in a dome 
that I wanted to operate from a warm control room this looks
like the system.

One other thing, they have use some neat motor controllers.  I don't
know what Tom will need to spend to build his but I would guess
custom PCBs are not cheap.  "Off the self" could be cheaper. 
 
See           http://www.omsmotion.com/oms_line.htm



-- 

--Chris Albertson

  chris@topdog.logicon.com                Voice:  818-351-0089  X127
  Logicon RDA, Pasadena California          Fax:  818-351-0699