Re: Languages don't matter. A mathematical refutation

[Jeffrey Carter <spam.jrcarter.not@spam.not.acm.org>]

On 04/02/2015 05:59 AM, brbarkstrom@gmail.com wrote:
> 
> To change the subject a bit, is there any possibility of using
> the probabilistic approach to estimating the long-term cost of
> maintenance given a history of errors with a particular language?
> I'm thinking here of the statistical record of revisions to TeX
> that Knuth maintained [see Knuth, D. E., 1999: Digital Typography,
> Center for the Study of Language and Information, Stanford University,
> pp. 655-662.]
> 
> It seems reasonable to suppose that there are two costs to long-term
> maintenance once a particular project has completed development:
> correcting errors and adding features.  Part of the cost is due to
> having maintainers climb the learning curve.
> 
> So, do we have a way of developing a model for the cost of dealing
> with errors that would show an advantage to Ada in long-term maintenance,
> particularly for long-term information preservation?  I suppose it would
> be helpful to have the AdaCore history of error discovery and correction
> to compare with Knuth's documentation.

The Rational study (http://archive.adaic.com/intro/ada-vs-c/cada_art.html)
compared developing an Ada compiler in Ada to the same project in C. It found ≈4
times as many post-delivery errors in the C compiler, and the average effort to
correct a C error was ≈10 times that to correct an Ada error.

-- 
Jeff Carter
"He didn't get that nose from playing ping-pong."
Never Give a Sucker an Even Break
110