|
|
|
| Authored by: kuroshima on Tuesday, February 12 2013 @ 08:21 AM EST |
They remind me of the origins of a certain Redmond based
company... [ Reply to This | Parent | # ]
|
| |
| Authored by: Anonymous on Tuesday, February 12 2013 @ 08:38 AM EST |
[Replying to my own post]
If anyone is still wondering "what's the problem?": just imagine if
real engineers did the same thing. Grab a bunch of girders and bolt them
together anyhow, copy part of someone else's arch bridge for their suspension
bridge, and then are amazed when their "Frankenbridge" actually stays
up while one car drives across. That's the testing done, now lets open it to the
public![ Reply to This | Parent | # ]
|
| |
| Authored by: designerfx on Tuesday, February 12 2013 @ 12:42 PM EST |
I see this as a great thing. Do you think people were even
willing to *look* at code before? It was a voodoo magick thing
before. This is a significant improvement and the (was it
MIT?) sketch program or whatever it's called has helped this
move forward as well.[ Reply to This | Parent | # ]
|
| |
| Authored by: artp on Tuesday, February 12 2013 @ 02:08 PM EST |
Only partly kidding.
The major implication is that nobody will be able to write
software for FDA applications. You need to be able to
demonstrate that your program has been tested, documented,
and can be rebuilt from scratch and produce the same results.
Of course, this implies that the FDA is actually enforcing
its regulations, a situation that is not always true, in the
FDA or in other government agencies.
---
Userfriendly on WGA server outage:
When you're chained to an oar you don't think you should go down when the galley
sinks ?[ Reply to This | Parent | # ]
|
| |
| Authored by: Anonymous on Tuesday, February 12 2013 @ 02:23 PM EST |
Just to mention some extremes:
1) Control the space shuttle
2) My first 'hello world' program
For 1) 'the methodology' needs to be incomparably more
strictly applied than for 2). Ignoring exceptions, that is
what will happen. So long as the 'grab - tweak - run' method
is not applied to 1) i'm fine with it. In the process, the
FrankenProgrammer will -over time- need to study, understand
and 'get better'.
And even if a given FrankenCoder never gets to the level of
producing anything that would qualify as a 'real' program,
he/she will never again look upon software as 'black magic'.
If anyone remembers the 'grins' 'here' when a judge 'the
other day' mentioned that he had learned to code in Java, i
hope you'll agree that even if the same judge would never
produce any code worth a dime, the 'win' was immeasurable.
Just compare his attitude towards software ('I could write
RangeCheck in ten minutes..') to the one of the folk that
with a straight face come to present in court that a
software-implemented invention the equivalent of adding one
to an integer is worth being 'protected' for the next 20 or
'life + 70' years.
[ Reply to This | Parent | # ]
|
|
|
|
|
