3 ms·
I sincerely hope that the author of this article is not employed in a professional software capacity, or at least not in any software I rely on. As someone who
by mdoms 6y ago
I sincerely hope that the author of this article is not employed in a professional software capacity, or at least not in any software I rely on. As someone who was raised in the peak of the Kent Beck test-first philosophy, reading something like this is shocking to me.
Based on the declining quality of everyday software it's not altogether surprising, but it's disturbing to see it written in black and white so brazenly.
- brohee 6y agoDunno what he does now, but he worked quite a bit in static analysis, which does a much better job at discovering nasty code properties than tests do. He found issues in code you most likely rely on, like the Linux kernel : https://linux.kernel.narkive.com/Cw4EtIti/coverity-untrusted-user-data-in-kernel https://linux.kernel.narkive.com/Cw4EtIti/coverity-untrusted..., and also the pretty important CVE-2010-5298 in OpenSSL.
- UncleEntity 6y ago> I was under the impression that kmalloc has an implicit bounds check as it returns null if attempting to allocate >64kb (or at least it used to). Can someone confirm/disconfirm that? If only it had a unit test...
- rimliu 6y agoDon't you think it is strange that "declining quality of everyday software" is observed at the same time as testing becomes the norm?