Re: [9fans] c++

[tlaronde@polynum.com]

On Thu, Nov 22, 2012 at 11:00:51AM -0500, erik quanstrom wrote:
>
> putting aside that i don't believe that the big problems like war and hunger
> have anything to do with programming errors, ....

There have been already numerous hundreds of millions if not billions of
money losses by financial teams due to software bugs. And since, more
and more, programs make "decisions" based only on superficiality (no
semantics), software acts as an amplifier.

There is the example of big mirrors that are hand polished, not because
a human is less error prone : on the contrary, but because a machine
makes few errors but always the sames, in this case always in the same
area, while a human will make far more errors but random ones, not
localized leading to a more even surface.

>
> and even that aside, can you cite studies that show that the choice of programming
> language is the dominant term in determining the error rate of the resulting
> code?

No, i have none since I'm already too busy programming so I have time to
gain not to loose...

But I will make an assumption: one will find that the most desastrous
software bugs are from the softwares the most widely deployed and
the more "high level" ones; that these are put in the wild because they
are easy and need to catch an audience in the hurry; and that these
have been wanted by people with the most "high level" view of
programming that is the more farther, enforcing the use of the
language "du jour". So the results are biaised: the languages are
not bad by themselves, but because some people use them not by need
and not by understanding the needs, but because this is the mandatory
language of the trend...

--
        Thierry Laronde <tlaronde +AT+ polynum +dot+ com>
                      http://www.kergis.com/
Key fingerprint = 0FF7 E906 FBAF FE95 FD89  250D 52B1 AE95 6006 F40C