1,054 reputation
721
bio website accelerando.euweb.cz
location Prague, Czech Republic
age 52
visits member for 2 years, 7 months
seen 11 hours ago

Senior developer, algorithms master, PM, analyst, applied mathematician.

The Three Little Daughters Raiser

Hobbies:
logics, history, psychology, sociology, pedagogics, photo, cycling, hiking.

In past:
space-/astro- geodesist, cartographer, astronomer, teacher, radiometrist on the liquidation of the Chernobyl catastrophe in 1986.


Jul
1
comment What UML diagram should be used for modelling permissions/users/files in Linux?
@YannisRizos I am afraid that the system in this case is extremely far from being simple. And any error in permission politics in one single file will cause all the system to fail. And in your example there is not a word on permissions at all.
Jul
1
revised In what programming language did “let” first appear?
added 4 characters in body
Jul
1
asked What UML diagram should be used for modelling permissions/users/files in Linux?
Apr
2
comment How do I prevent unknowningly duplicating code?
@Brendan You are right. And the artificial de-duplication is an excessive piece of work by itself, too. But... You see by votes, people tend to simply repeat what was told to them when they started to program. Very few ever try to think further. :-(
Mar
9
comment Why are scientific programming languages so weird?
@ThomasEding You have used two logical fallacies in one message. I can't consider this as an error. Rather it looks as trolling. Stop it, or I'll flag you.
Mar
8
comment Why are scientific programming languages so weird?
@ThomasEding "If you need a programming language to help you with your math, then bets are that you are at least at this level of mathematics" If you think you are on level to participy in any discussion, do such work as to read the rules of discussion. In wiki, for example. You are positioning yourself out of any discussion.
Mar
8
revised Why are scientific programming languages so weird?
added 10 characters in body
Feb
26
comment How to address the concerns of the software architect but still maintain collective code ownership?
@MaciejPiechotka I had said several times that we have many programmers and one component for a programmer. So your second part is also for some other case, sorry. I understand, that defining processes for small projects is fine and simple, but we should solve tasks that are set, not other ones that are simple.
Feb
26
comment How to address the concerns of the software architect but still maintain collective code ownership?
@MaciejPiechotka If it is a 4-men project, they all know their scopes of work. I am talking about a large project, with tens of programmers. And not knowing who is better in what could be fatal.
Feb
25
comment How to address the concerns of the software architect but still maintain collective code ownership?
@MaciejPiechotka But that would work only if they already know B a bit, to find that A-B connection while analyzing A, and know that b is the best specialist in B.
Feb
25
comment How to address the concerns of the software architect but still maintain collective code ownership?
@Mr.Mindor I would also add the automated testing for UI. It is very demanding and covers really many problems. But I am afraid that when a group is guided by a person, who obviously tries to move his own guilt to somebody else, it needs special treatment
Feb
25
comment How to address the concerns of the software architect but still maintain collective code ownership?
@Mr.Mindor "...code reviews can only help..." Without any doubt. But they are not enough. And this post insists that they are ALL we need.
Feb
25
comment How to address the concerns of the software architect but still maintain collective code ownership?
I agree this may be enough in a normal project. But if the project already HAS bad architecture so that components are too often dependent on each other? And in this case IMHO, we have to choose carefully who will do which code check.
Feb
25
comment How to address the concerns of the software architect but still maintain collective code ownership?
@Disco3 The connection A-B could be easily found by 'b' only. Who will tell the programmer 'c' that he should check THESE TWO components?
Feb
25
comment How to address the concerns of the software architect but still maintain collective code ownership?
@Plutor You want every piece of code to be checked by every programmer in the group? If yes, it is impossible, if no, where is the profit?
Feb
25
answered How to address the concerns of the software architect but still maintain collective code ownership?
Feb
25
comment How to address the concerns of the software architect but still maintain collective code ownership?
Let's many programmers a,b,c,d,... made components A,B,C,D.... Now the programmer 'a' wants to make a change in his component A that will cause a problem in the component B. Programmer 'a' doesn't know about it. I his code would be checked by the programmer 'b', everything is OK, the problems will be found. But the code checker was chozen arbitrary. So, with the great probability it won't help. How this could help the QA with his problem?
Feb
25
comment Why are scientific programming languages so weird?
@phresnel sorry, look better this one: en.wikipedia.org/wiki/Hoare_logic#Assignment_axiom_schema
Feb
25
comment Why are scientific programming languages so weird?
let us continue this discussion in chat
Feb
25
comment Why are scientific programming languages so weird?
@phresnel Of course, it IS the assignment.