722 reputation
1924
bio website
location St. Petersburg, Russia
age 33
visits member for 4 years, 7 months
seen May 3 at 17:43
mail: mlvl.jr@gmail.com

Apr
13
comment Is directly using Make considered obsolete?
Same goes for Qmake (despite not being the badly documented inconsistently behaving **ap CMake is), btw ;)
Apr
13
comment How do you unit-test code using graphs?
factor the logic either a) changing the state of a graph or b) producing some data given a graph into standalone-ish "methods" -- and test those with properly crafted inputs; this will guard you against the unit tests becoming integrated ones in disguise (where integrated would be multiple invokations of same logic, essentially -- with its correctness obscured by the chain of intermediate states, just like the interaction of multiple components would obscure their individual behaviors with "normal" integration tests)
Mar
29
comment What does it mean to write “good code”?
@BlueTrin Why not both brain-compile those hi-perf source codez, and also document the hell of what's going on there (right there in comments)?
Mar
29
comment What does it mean to write “good code”?
@balajeerc Thanks (I also fixed the link, so it points to a Springer-hosted version of the same pdf) :)
Nov
9
comment What is your opinion of C++ Frequently Questioned Answers?
-1, FQA actually taught me something (would you find any mention of the last issue in yosefk.com/c++fqa/templates.html#fqa-35.11 in a FAQ? -- and the scope of the enlightment on non-accidental problems with a given C++ feature (templates) is really worth a praise in this case)
Aug
8
comment Is partial classes modern Subject-Oriented programming?
The last suggestion was actually quite doc-y ;)
Aug
7
comment Are unit tests really that useful?
-1 as this proves nothing, and also leaves a false feeling of security in the absence of unit-tests ;)
Aug
6
comment Does cohesion really reduce coupling?
and don't believe any of what you're told, btw -- before you have explained it yourself with mathematical strictness :)
Aug
6
comment Does cohesion really reduce coupling?
one interesting discussion is: youtube.com/watch?v=hd0v72pD1MI (some Kent Beck inside!)
Aug
6
comment How do you learn to take a more OO approach to problems?
Knew that! G'luck, mr. Lowe! :)
Aug
6
comment What is the functional-programming alternative to an interface?
Make it simple: a struct containing those method pointers, plus a function to initialize it from an object instance. Why Haskell? ;)
Aug
6
comment First of the The Ten Commandments for C++ Programmers by Steve Oualline correctly transcribed?
Now you have: dsi.fceia.unr.edu.ar/downloads/informatica/info_II/c++.../… (2nd link in google, must be legal, right?)
Aug
6
comment How do you learn to take a more OO approach to problems?
I remember you were also writing one of your own. Right? :)
Oct
19
comment Ruby on rails not popular in big international IT companies?
big international companies stick to java-on-jails, you know :)
Sep
10
comment General programming techniques to speed up coding time
On RAD: Delphi 6 will be the best bet then, possibly.
Sep
10
comment General programming techniques to speed up coding time
Downvoted with an index finger :)
Jul
29
comment Are there any GUI or user interface design patterns?
As often, the closed ones are the most interesting (if not even most helpful sometimes).
Jun
7
comment How can a Right-Brainer Deal with Massive Left-Brainer Code?
@FrankShearar Sometimes it's like surgery done in the toilet ;(
Jun
3
comment What defines an encapsulation properly?
Another thing is, using several objects together is plain easier when their contracts are more compact -- since the (conceptual) complexity of the code is a "multiplication" of the complexities of those contracts. Of course, in each context there is a limit on "encapsulation level" of the objects involved -- you must still be able to fulfill the given task in the client code, or the (perfectly encapsulated) objects/classes used are ..useless.
Jun
3
comment What defines an encapsulation properly?
Strangely enough, this is still downvoted (while being quite correct) ;) To add, one can expose even the (interrelated) internals of a class' instance, if he obliges the user of the class to maintain respective relationships "by hand" -- it's nothing (conceptually or practically) wrong with having such a contract. But if the object can fulfill the same role with less being known about its proper usage (i.e. conforming to a "smaller" contract), that makes it possible to change the internals w/o changing the client code. That's one thing about encapsulation.