Crash-Free Software

Crash-Free Software

[Larry Hazel]

There is a short article in the paper today that NASA and Carnegie Mellon
University are forming a High Dependability Computing Consortium.  Systems to be
targeted are air trafic control, the space program, health care, banking ... 
Twelve companies will be involved including IBM, Sun, and Microsoft.  No mention
was made about language, but this might be a good place for Ada advocates to
target their efforts.  The first planning meeting is in January. 
-- 
Larry Hazel

Re: Crash-Free Software

[Petra Lynn Hofman]

I heard a short sound bite from the project leader.  His example for
research was "truly plug and play device driver".  I am not much
impressed by this statement.

Larry Hazel wrote:
> 
> There is a short article in the paper today that NASA and Carnegie Mellon
> University are forming a High Dependability Computing Consortium.  Systems to be
> targeted are air trafic control, the space program, health care, banking ...
> Twelve companies will be involved including IBM, Sun, and Microsoft.  No mention
> was made about language, but this might be a good place for Ada advocates to
> target their efforts.  The first planning meeting is in January.
> --
> Larry Hazel

RE: Crash-Free Software

[Beard, Frank]

> I heard a short sound bite from the project leader.  His example for
> research was "truly plug and play device driver".  I am not much
> impressed by this statement.

Being one who is subject to having things taken out of the context in
which I meant them (and I accept that part of the fault lies with me),
I'll take a stab at what he might have meant.

Maybe he was saying, "Unlike Microsoft Plug-n-play (and play and play)
devices that can crash the operating system, we will have devices on
a robust OS, with robust drivers, that will handle problems, report them,
recover, and continue working".

Of course, that's just my opinion.  I could be wrong.

Re: Crash-Free Software

[Ken Garlington]

"Beard, Frank" <beardf@spawar.navy.mil> wrote in message
news:B6A1A9B09E52D31183ED00A0C9E0888C469953@nctswashxchg.nctswash.navy.mil
: > I heard a short sound bite from the project leader.  His example for
: > research was "truly plug and play device driver".  I am not much
: > impressed by this statement.
:
: Being one who is subject to having things taken out of the context in
: which I meant them (and I accept that part of the fault lies with me),
: I'll take a stab at what he might have meant.
:
: Maybe he was saying, "Unlike Microsoft Plug-n-play (and play and play)
: devices that can crash the operating system, we will have devices on
: a robust OS, with robust drivers, that will handle problems, report them,
: recover, and continue working".
:
: Of course, that's just my opinion.  I could be wrong.

No, I think that's correct. It appears that this work is going to be an
outgrowth of the CMU EDCS program, particularly the INSERT project. See

http://www.cs.cmu.edu/Groups/real-time/insert/

http://www.sei.cmu.edu/community/edcs/CLUSTERS/HA/index.html

Re: Crash-Free Software

[Petra Lynn Hofman]

Perhaps I read more into the statement than was there.  I think that
happens when one has used MS and then moves on to more stable and robust
systems.

Ken Garlington wrote:
> 
> "Beard, Frank" <beardf@spawar.navy.mil> wrote in message
> news:B6A1A9B09E52D31183ED00A0C9E0888C469953@nctswashxchg.nctswash.navy.mil
> : > I heard a short sound bite from the project leader.  His example for
> : > research was "truly plug and play device driver".  I am not much
> : > impressed by this statement.
> :
> : Being one who is subject to having things taken out of the context in
> : which I meant them (and I accept that part of the fault lies with me),
> : I'll take a stab at what he might have meant.
> :
> : Maybe he was saying, "Unlike Microsoft Plug-n-play (and play and play)
> : devices that can crash the operating system, we will have devices on
> : a robust OS, with robust drivers, that will handle problems, report them,
> : recover, and continue working".
> :
> : Of course, that's just my opinion.  I could be wrong.
> 
> No, I think that's correct. It appears that this work is going to be an
> outgrowth of the CMU EDCS program, particularly the INSERT project. See
> 
> http://www.cs.cmu.edu/Groups/real-time/insert/
> 
> http://www.sei.cmu.edu/community/edcs/CLUSTERS/HA/index.html