21 Sep
2011
21 Sep
'11
7:42 a.m.
At 03:13 PM 9/20/2011, Fred Cisin wrote:
I concede that my attitude can be a "bit"
cynical.
But do you contest the points that I attempted to make?
Of course not. And there are many more points to be made about error
checking. Personally, for the era you speak of, I blame Charles Petzold
for writing code examples that didn't error-check, then putting them in a book.
At 05:41 PM 9/20/2011, Ian King wrote:
We did test, and test and test and test. We had
databases filled with bug reports, thousands upon thousands of them for any given product.
It's not that the test process doesn't exist or that it's not effective.
It's the priority given to quality vs. ship dates and new feature sets. -- Ian
I'm glad you mentioned that. I don't have any doubt that the software
industry has better tools, more experience, and more expertise for software
testing than ever before. Whether companies choose to use it, and how they
interpret the results and assess the risk of shipping - that's another story.
Software delivery over the Internet has certainly changed the landscape, too,
making it very easy to patch your program every few days once it is in
the field (cough Adobe Acrobat cough).
The days of putting disks in boxes are long gone.
As for the connection between consuming disk space and software testing,
I'm routinely appalled at the way programs leave megs and megs of temp
files, past install patches, etc. all over users' drives without regard.
Or why an HP printer driver or a .NET upgrade takes 20 minutes to install.
Cynically, I think it's all working well towards Windows eventually
having the more Unix-like separation of user data and program files.
The lesson will be re-learned.
- John