OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: ??? (was RE: A simple guy with a simple problem)



John Cowan wrote:
 
> In general, yes.  But are you willing to bet lives that
> there won't be a bug somewhere in the compiler or support
> libraries that surfaces one fine day?

Yes. Part of the software testing process. Know your code. Know every
bug in 3rd party libraries and code around them. Doable in any
environment - XP or 2167A.
If it is something so obscure that a mountain of testing doesn't find
it, then you have found a rare bug indeed. 

> Maybe true by now, "milspec" probably doesn't mean what it
> once did.  But consider how long you have to recover from
> various disasters:

And, as you will read in _any_ contract, these are typically referred to
as Acts of God, for which neither party is required to conform too. A
power outage is not your fault as a contractor - particularly if you are
providing hardware and software only services. If the purchaser didn't
provide adequate measures for it or didn't require a multi-site system
then that is their problem. Make sure your lawyers read the contracts
properly.
 
> You don't catch me promising 30 sec/yr downtime for
> anything.

Well I would and have done - back with good old Java 1.0 and 1.1
systems, not to mention the much more debugged stuff these days. To me
it seems like you are not one who has lived in that environment. Take a
walk on the wild-side. It's easy and fun to do.
 
> > Depending on well-debugged existing code/gear is one of the
> > the best practices in achieving high reliability. -T
> 
> High reliability, yes.  Extremely high reliability, the kind
> we are talking about here, no.

Extreme reliability is even more dependent on standard libs than normal
programming. There is a much higher probability of bugs in custom
written code than normal stuff - unless we are talking Shuttle software.
From a software development cost standpoint, it is cheaper to find
software with known bugs that to write it yourself and have unknown bugs
or to formally prove that it has no bugs.

If someone requires greater than five nines then they start going the
multiple independent implementations from different vendors running in
parallel route. Mucho dollar, but if the customer wants that, then they
pay for it.

-- 
Justin Couch                                    Author, Java Hacker
http://www.vlc.com.au/~justin/               Java 3D FAQ Maintainer
http://www.j3d.org/              J3D.org The Java 3D Community Site
-------------------------------------------------------------------
"Humanism is dead. Animals think, feel; so do machines now.
Neither man nor woman is the measure of all things. Every organism
processes data according to its domain, its environment; you, with
all your brains, would be useless in a mouse's universe..."
                                              - Greg Bear, Slant
-------------------------------------------------------------------