2,755 reputation
1230
bio website
location Argentina
age 37
visits member for 3 years, 9 months
seen yesterday

Software developer and science fiction fan.


Feb
24
comment Do we need Logging when doing TDD?
You can never know your software will never crash EVEN if you have 100% coverage. 100% coverage, while desirable, gives a lot less information about correctness than it would seem.
Feb
20
reviewed Reject suggested edit on Does it make sense to write tests for legacy code when there is no time for a complete refactoring?
Feb
19
comment Indirection: Readability vs Performance
@aragaer Well, if you've found a better way to handle readability vs performance, you are not going to share your code with anyone else, and it isn't going to be used for anything that requires industry best practices, testing or verification, then suit yourself... I guess :)
Feb
19
comment Indirection: Readability vs Performance
@aragaer Arranging your code in a modular, easy to understand way will make it easier for you to play with it. Well written code will benefit even a team of one. Maybe if you just want to write a one-off, try something and discard the code... in that case you probably can disregard all best practices.
Feb
19
comment In functional programming how does one achieve modularity through mathematical laws?
@leeand00 I think you are misusing the term "verification". Modularity and reusability are not directly checked by software verification (though, of course, lack of modularity can make the software harder to maintain and reuse, therefore introducing bugs and failing the verification process).
Feb
19
reviewed Approve suggested edit on Functional Programming vs. OOP
Feb
17
comment OOP in C: which of these two approaches is closer to a cannonical solution?
I took the liberty of editing the title to make it a proper question. Please edit it again if you think I changed the intent of your question!
Feb
17
revised OOP in C: which of these two approaches is closer to a cannonical solution?
changed title to make it into a proper question
Feb
17
comment OOP in C: which of these two approaches is closer to a cannonical solution?
@Oded Fair enough. After his edit, I see he is asking for the "canonical way", which looks like a qualification. Either there is or there isn't a canonical way to do this in C. This should be answerable.
Feb
17
comment OOP in C: which of these two approaches is closer to a cannonical solution?
@Oded It seems to me he is asking about OOP in C, which is a decently narrow scope. My suggestion would be "don't", but what do I know? :P
Feb
13
comment How can I unit test a class which requires a web service call?
@kevincline Right! I misread your comment, sorry!
Feb
13
comment How can I unit test a class which requires a web service call?
@kevincline I fully agree on the necessity of the tests you propose, and indeed I write them in my day job and have proven themselves useful. But they are by definition NOT unit tests, which is what the question was about :) Consider this: if it is a unit test, and the code fails because the webservice was changed, what is the "unit" that you're testing? What exactly failed? You are not testing in isolation, as needed by unit testing.
Feb
10
reviewed Edit suggested edit on Are developers expected to have skills of business analysts?
Feb
10
revised Are developers expected to have skills of business analysts?
fix spelling errors
Feb
7
awarded  Yearling
Feb
5
reviewed Reject suggested edit on Why is an anemic domain model considered bad in C#/OOP, but very important in F#/FP?
Feb
5
reviewed Edit suggested edit on FizzBuzz - really?
Feb
5
revised FizzBuzz - really?
Removed dead link
Jan
31
comment What is the name of the functional argument in fold
@Izkata Are all higher-order functions catamorphisms? If not, then it isn't an "universally used name for catamorphisms", just like jozefg asserted.
Jan
29
comment Why does the US government disallow dynamic languages for secure projects?
@romkyns But that's not what I said. As a fan of static typing, I'm well aware it gets rid of whole categories of bugs, which is what I argued ("another layer that helps the correctness [...]"). I wouldn't use dynamic typing for mission-critical software.