1,100 reputation
622
bio website yourcodeisshowing.wordpress.c…
location Quebec City, Canada
age 37
visits member for 4 years, 1 month
seen Oct 27 at 18:55

Developer with 8 years experience. I work with J, C, C++, C#, SQLite, MySQL, VBA, and a bunch of other odds and ends.

On Twitter: @MPProg


Jul
29
comment How much documentation should a client expect from a one-man team?
@rwong - Thanks. Tweaked it a bit more.
Jul
29
comment How much documentation should a client expect from a one-man team?
The write-up on this current project will be 4 weeks. That's just one project out of several this year... But I see your point.
Jul
29
comment How much documentation should a client expect from a one-man team?
I wish I was getting paid all these different job titles' salaries. Or if I can pick, I'd take architect.
Jul
29
comment How much documentation should a client expect from a one-man team?
Cashier? Now that's seriously wrong.
May
20
comment Why are there multiple Unicode encodings?
See what Spolsky had to say on Unicode and encoding.
Mar
24
comment What is the biggest trouble developers encounter when debugging?
Multithreaded memory access bug in a DLL called with an out-of-process COM. It's a world of pain. It's not made up, I got a solid one!
Feb
25
comment Constructor parameter validation in C# - Best practices
Sorry, this is just too much of an anti-pattern to consider.
Feb
24
comment What did your college do right and what could they improve upon?
@zneak: :) Human communication. I can do P2P from home, thank you very much.
Feb
23
comment Constructor parameter validation in C# - Best practices
@ChaosPandion: Maybe not specific, but definitely strongly biased. Tagged nonetheless as suggested.
Feb
23
comment Constructor parameter validation in C# - Best practices
@S.Lott Updated. I think you've hit something on the head there. The more I think about it, the more that specific context was necessary. Thanks again for clearing things up!
Feb
23
comment Constructor parameter validation in C# - Best practices
@S.Lott What @Chad said. There were two cases of the validation in the calling class, which were identical. Even if they were combined, the rule for the class is that the argument cannot be null. It made sense to suppose that the class itself should manage this. Now it's encapsulated, nearly forgotten.
Feb
23
comment Constructor parameter validation in C# - Best practices
@S.Lott For my defense, the pre-validation bit wasn't my creation :) I just raised an alarm because it was duplicated.
Feb
23
comment Constructor parameter validation in C# - Best practices
@S.Lott I see. Forgive me for not understanding what you meant. The alternative which incited me (I wouldn't say forced) to ask the question is that the parameters were pre-validated, before instanciation. On failure, a message would be shown, and the constructor was never called.
Feb
23
comment Constructor parameter validation in C# - Best practices
@S.Lott For fear of repeating myself, I was open to the possibility of there being some other approach, which I did not know. I believe I'm not all knowing, that much I know for sure.
Feb
23
comment Constructor parameter validation in C# - Best practices
@S.Lott I'm a humble programmer, I know there's always something else to learn, so that I'm open to the idea of some other doodad for this sort of thing, even if seemingly unlikely. I actually read somewhere that constructors should never fail (which is admittedly bull...)
Feb
23
comment Constructor parameter validation in C# - Best practices
+1 for the confession alone :)
Jan
18
comment Project Manager that wants to lock in time estimate with a signed contract
I was thinking of a similar clause: any feature change invalidates all time estimates and require a complete reevaluation. But that would just be more work. In either case, would a certified PMP be stupid enough to sign this? Of course.
Nov
26
comment UI design and confirmation paradigm
Obviously there's degrees here. For example, you absolutely want your "Delete all tables, files, and settings irrevocably" option to have a solid confirmation. Yet cosmetic changes, or ones easy to rollback, could be done silently, even if there's no direct undo for some.
Nov
25
comment UI design and confirmation paradigm
Very good point. Definitely better than a modal window proclaiming "Hey, I just did something!!"
Nov
25
comment UI design and confirmation paradigm
@Niphra: I had not thought of it that way. I grant if there's one window in a system that works differently it's messing with people (and their nerves). I don't know what would be the learning curve on "not doing something", though, and how quick that would set in.