Show Navigation
Notices by Verius (verius@community.highlandarrow.com)
-
@maiyannah To be fair Nginx Plus was already killing a major part of its appeal. On the plus side I hear lighttpd is active again, though I have no personal experience with how good it is.
-
This is quite nicely done for teaching people about the hazards of deadlocks: https://deadlockempire .github.io
-
This brings back some childhood memories: http://www.imagisoft. com/games.htm
-
Despite not liking brexit I have to say I look forward to it finally happening, if only to get rid of the constant whining about it.
-
Sjeesh, I haven't been working with containers for a few years and it seems half the stack has changed. That stuff is approaching JavaScript iteration velocity.
-
CERN scientist gets kicked out for being sexist, the usual "there are differences between men and women" stuff, and for cartoon images mocking women. Said images mock SJWs. Sigh.
-
I guess the lesson is that I'm more suited to deterministic than non-deterministic puzzle solving. Though I have a way to convert the latter into the former: writing a prolog / finite domain constraint logic program to solve the puzzle for me. Seriously, Prolog is like the best language ever when you need lots of backtracking, it beats even politicians.
-
Playing an old adventure game. Logic gate puzzle, expert level. Four binary input wires, construct a chip to handle them. I notice pretty straight away that the D line goes 0 1 0 1 0 1 except that it switches in the middle of the logic table. So I look and it's pretty obvious that only the A and D lines are in any way significant and that if A = 0 the result is equal to D and that if A = 1 the result is equal to not(D). So I throw in an XOR and yup, the puzzle is solved. Define expert level. (The odd thing is, the game game gives you actually an actually difficult puzzle in the math section where you have to fill a grid of numbers to get specific sums (a variation on magic squares).)
-
@maiyannah Customer is king, owner is emperor.
-
@maiyannah Not what I mean. I mean the function is considered to have a precondition that the memory is sufficient and that the contents of the memory encode a string correctly (i.e. current position + string length < size of memory). Because that's a precondition I don't consider _this particular function_ to be buggy if it crashes due to a segfault.
-
@maiyannah Ok, maybe I should have said that you can assume memory allocation will succeed, that the buffer has infinite size (i.e. no reading out of bounds) and that the strings are always smaller than 2^15-1 bytes. But if there's more bugs here than just the endianness conversion going dumb on bytes above 0x7F (unsigned) yeah, I'm curious.
-
What could possibly go wrong? ```std::string read_string(const char* mem, int pos) {
short len = mem[pos] + (mem[pos+1] << 8);
return std::string(mem+pos+sizeof(short), len);
}```
-
Interesting. That Win 10 calculator that MS open sourced is the first C++/CLI app I've seen in the wild. I wonder why they went for that and not C#. I mean, it's not exactly a very complicated thing and it doesn't involve interoperability with native code.
-
Ah, Microsoft, where even the calculator has telemetry. :p
-
@maiyannah Ok, so more corrupt than our news reports, less in trouble than they report.
-
NSA releases open source reverse engineering tool. I'm torn about this. On one hand it's the NSA so one has to expect some kind of creepy function somewhere. On the other hand it's the NSA, if there's ever an organisation that has expertise in these things it's them and since nobody trusts them it's unlikely that a backdoor would be very effective (it would need A) an interesting target to use the software even though they know it's literally NSA supplied software and B) said target not to do thorough analysis of the source code first).
-
So news in NL is that Trudeau is in trouble. Is this our news blowing stuff out of proportion or does pretty boy actually have a real problem?
-
@maiyannah The problem with the EU is that to make it really work well you need the member states to give up full sovereignty so you can set things like tax rates and fiscal policy at the EU (or at least Eurozone) level. But member states and the politicians leading them (not to mention, you know, the people) don't want that. At the same time the benefits of economic cooperation are too big to ignore and people don't want to give up on those either. But it's a messy system with lots of stakeholders pulling in different directions and although I understand the attractiveness of doing it over I don't believe we'd end up with something more pleasing (obvious software development parallels here). I do get the impression that the eurocrats are trying to understand however, but there's both a conceptual gap (eurocrats think at such a different level and in different contexts) and a practical gap (what people want and what's possible in the real world are two different things) that it's inevitably going to be an unsatisfactory process.
-
@maiyannah @mangeurdenuage @lnxw48a1 @lnxw37a2 The League of Nations can't be compared to the EU though. The League was an - even weaker - predecessor of the UN. The primary characteristic of the EU is the common market, a strong economical pact which the League did not have.
-
@maiyannah I agree, to the extend that I don't believe the current European Union will be able to survive long (I'm assuming that you'll live a long life of course ;)). Rather I see it either fracture into a few blocs (north/west vs east/south) and/or become a full political union where the primary decisions are made at the EU level rather than at the national level. As that requires leaving the principle of subsidiarity though I'd say such an entity is no longer the EU as we understand it.
Verius
All Jonkman Microblog content and data are available under the