Re: WSJ article on software liability

From: DHOLLINGSWORTH2 (DHOLLINGSWORTH2_at_cox.net)
Date: 02/27/05 

Date: Sun, 27 Feb 2005 01:08:45 -0600

"Traveler" <traveler@nospam.com> wrote in message
news:6p41215gsroedu63llt4ucluu2s399716r@4ax.com...
> In article <421FF9CA.A0C3ADC1@wecs.com>, Bryan Hoover
> <bhoover@wecs.com> wrote:
>
>>Traveler wrote:
>>>
>>> In article <9dru11tai75juq68cm27nqekocn8144n1p@4ax.com>, Programmer
>>> Dude <Chris@Sonnack.com> wrote:
>>>
>>> >Traveler writes:
>>> >
>>> >> The Silver Bullet or How to Solve the Software Crisis:
>>> >
>>> >Silver Bullet?
>>> >
>>> >From that *alone* I know you're selling snake oil.
>>>
>>> And I know you can take your opinion and pack it up your ass, for all
>>> I care.
>>
>>Oh the humanity :). I was so interestedly following this thread, and
>>then, it turned ugly. Hate it when that happens. I know what you're
>>saying though -- the snake oil comment was a bit insensitive I'd say --
>>and I do.
>
> I have very little patience for assholes. Sorry about the outburst.
>
>>Seems to me what you're advocating here is a natural progression, and
>>one that's been on the tip of my toungue for a long time. Hope it
>>doesn't get *too* much like hardware though
>
> It will be better than hardware because of the flexibility.
>
>> -- rather, I fancy something
>>more along the lines of "It's alive! It's alive!!"
>
> You mean you want something tangible? Well, somebody has to build it
> first. What I am proposing is a revolution, a radical approach to
> software construction that will require new CPUs optimized for a
> radically new software model. This kind of stuff does not happen
> overnight.
>
>>Rock steady big guy.
>
> I have been saying essentially the same thing for close to twenty
> years: there is a fundamental flaw in the way we program our
> computers. It's only now that the software reliability crisis is
> regularly making the evening news that Project COSA is beginning to
> attract attention from the big software companies. Reliability
> engineers have failed to solve the problem and they have failed
> miserably. It's time for a change.
>
> Louis Savain
>
> The Silver Bullet: Why Software Is Bad and What We Can Do to Fix it
> http://users.adelphia.net/~lilavois/Cosas/Reliability.htm

I think what your after is a marketed aplication of technology that is 15
year old.
you shouldn't act like software can be changed faster than hardware. Have
you ever heard of Programmable Gate arrays? In fact the intire CPU can be
changed while running a future version of Enternet Explorer. The technology
is not new, the technology is not implemented.

The problem is everyone has there own standards of acceptable practices.
Which in themselves work fine. Take for example MS Word used to write
information to the unused portion of a block of Hard Drive space. Great way
to hide stuff, except that disk compressors started writing the next bit of
information right after the end of file marker to capitolize on that same
unused space. Now we have Problems with two Software products, That in
themselves worked perfectly.

This is the cause of 90 percent of all software related problems, or bugs as
you call them. The Fact that neither Software package would adhere to the
standard meant that they could not be used in conjunction with each other.
Now consider the same irrisponsible pracrtices applied to any new CPU. You
will find that the 10% of design & programming flaws youv'e fixed was
nothing compared to the 90% that was due to the lack of standard
practices;or the lack of a standard intirely.

The Amiga computer originating in europe was by far a better Proccessor than
anything Intel put out at the time. It possessed several SubProccessors
that did multitask Completely, efficiently and not Psuedo Sequentially.
The Reason Intel took off was becuase of "Open Architecture". This meant
anyone with a compiler could develope software, and anyone with an eprom
burner could develope hardware.

I'm sure youve heard of VESA, the Video Electronics STANDARDS association.
They are the reason everyone could write one chunk of code to put out
computer Graphics. however, there was never any Audio Electronics Standard
Association. And thus we have a program with one video output unit, and 100
Audio output units.

I believe We are in need of new CPU architecture. It looks like Several
pentiums on one board and each with a host of Programmable logic output
gates. or one more, a Programmable Gate Array CPU that can rewire itself on
the fly. Imagine turning your computer off in the middle of a game, coming
back 3 days later, flipping the power on, and continuing your game where you
left off.

That IS the future! I just hope somebody pays for it before I Die.

Dan

Loading