[thingist] 100-150 errors in 1000 lines

[Beatrice Beaubien]

reminds one of DNA:

• the number (2 million) of code mistakes isn't important, it is the 
consequences that are

• only a subset of the mistakes are lethal

• some mistakes will be silenced by other mistakes

• some mistakes will be advantageous

• a proportion of the mistakes can be ascribed to transcription
.
.
.
hmmz. i wish i coded for such large projects. as it is, my coding 
projects are so much smaller
and all my mistakes count bigtime. doesn't give me time to read... well, 
hardly.


Biti


On Wednesday, June 19, 2002, at 06:26  PM, office at plasmastudii.org wrote:

> I'm reading this fantastic article, pertinent to half the discussions 
> on these lists.  http://www.technologyreview.com/articles/mann0702.asp
>
> I'd MUCH rather sit on the couch and read an article in a magazine than 
> look at my monitor anyway.  Page turns beat scrolling hands down.  So I 
> recommend the printed version but sometimes the web is all we've got!
>
> But the article is called
> "Why Is Software so Bad (and how to fix it)"
> The cover story for the July/August Technology Review.
>
> But there are about 16 million lines of code in NT4.
> coders make an average of 100-150 errors in 1000 lines (according to 
> this study)
> therefore, there will be approximately 2 million mistakes.
>
> judson
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: text/enriched
Size: 1347 bytes
Desc: not available
URL: <http://tekspost.no/mailman/private/syndicate/attachments/20020620/2a41bb09/attachment-0001.bin>