« first day (813 days earlier)      last day (4076 days later) » 
00:00 - 15:0015:00 - 00:00

3:00 PM
@TerryChia I guess... it's the form with helps you get empanelled.
It really reminded me of the Hakin9 pitch.
Thinking of forwarding this to rook...
 
It's a pity @Polynomial's troll piece didn't work out.
 
@asadz btw you never did answer @ScottPack's curiousity question... where are you from?
If you don't mind the ask... you don't have to say if you dont want, considereing it is indexed by google :-)
we have a pretty wide range of locations here, between california and singapore. (the long way around).
 
@asadz No love for Doktor Mayhem?
 
@TerryChia A smartcard is like a small HSM. If you want to put a CA in a smartcard and not trust anything else, then you have to embed the Web server (with SSL !) in the smartcard. I have done that on a HSM, and it is theoretically doable on a smartcard, but SDK for smartcard which are powerful enough for that are not easy to find.
 
well folks, its that time of day for me... @RoryAlsop enjoy Jaffa, I hear it's really romantic at night ;-)
 
3:05 PM
@AviD my name doesn't give me much privacy protection
 
@ThomasPornin Ahh makes sense. Didn't think of that use case.
 
@asadz We do have one semi-regular that is fanatical about protecting his privacy. We don't even rightly know what timezone he's in.
 
Quiet funny you said Singapore thats one place i was planning to go but it all changed just at the last minute
 
Have a great weekend, everybody.
 
@AviD There you go triggering my irrational time-zone jealousy again. Have a good one!
 
3:08 PM
@TildalWave Morning! Yes, I did take a look briefly, and doesn't appear to me that there's there's any vulnerability there. It appears that the architecture is as I suspected much the same as what I'm familiar with in ASP/ASP.Net/JSP, which is that the session is just a dumb container for pseudo-persistent state management.
@TildalWave So, all creating a session is, is adding an entry into the session table with a key (that you get back in a cookie as the session id) and an empty value. For there to be a data structure or data in there, you have to write the code to create the array and populate it with data, which isn't happening. In any case, it's all server side, so the client doesn't have any access.
 
@TerryChia Unfortunately the cool smartcards (with FIPS 140-2 level 3 or EAL 4+) implement VM so you do not implement in assembly or Forth, but with Java bytecode or CLR/.NET. Sooo uncool.
 
@TildalWave So, I do think that from a security perspective, calling create before destroy is exactly the same as just calling destroy. In fact, I don't think it has a performance impact either. Yes, it does cause some additional execution, but it's so minimal as to make no difference in the scope of an entire request.
 
@AviD take care Avid
 
@TildalWave Even if he just creating a session, and didn't delete it, there really isn't a security implication.
@AviD You too!
@TildalWave So, to my understanding, it's unnessary and I think that you're right that it's driving by a cargo-cult programming mentality, but I don't believe that it's harmful in any way.
 
@ThomasPornin You need to realize the fact that most programmers are mere mortals, unlike you. ;)
 
3:25 PM
Wow, lot of chat. After last night becoming somewhat later than expected I took an easy day- 5 mile run followed by a superb massage from an incredibly strong matronly lady. Just about ready for this evening, now.
Just wanted to say lads- spas are great. Worth hanging out in if only for them being the opposite of a sausage fest:)
@AviD - yeah, even when we left and were chatting to the waitress I hadn't realised it was that late. At least I only had a ten minute walk. Thanks for taking the journey in. And Thanks for dinner. Was an excellent evening.
 
3:40 PM
user image
8
 
@TerryChia OI, I'm a Fedora user!
 
@Ninefingers Hey, I'm a Fedora/RHEL fan myself.
 
@TerryChia That dude in the picture looks slightly bored. I love the mint user one though.
 
@Ninefingers heh yeah.
well off to bed. have a good day/night everyone.
 
4:08 PM
@TerryChia But they all equally don't get laid
 
@Adnan Neo does, while everyone else is having a rave in the earth's core.
 
@TerryChia debian user here, your picture saddens me
 
@Polynomial, damn! This is good.
 
@Xander - Cheers for looking into it! So the recommendation would be against him changing the intended order of session handling procedures as he wouldn't benefit anything with it, but from the security point of view it doesn't really change much anything (the function calls order). I still think it's plain wrong to re-create a session that handles user log-ins on an open URI without also re-checking user credentials, but that's from the question on SO. How does this sound?
 
4:29 PM
@TildalWave Yup, pretty much. I think one thing that's important to understand is that for a typical web app, creating a session doesn't happen when a user logs in, but long before, on first access. Every user, authenticated or anonymous will have a session, it's just that when you log in, some data gets shoved into it to identify you on subsequent page loads. So, creating a session on any page view is pretty standard.
@TildalWave So my recommendation would be just to remove the unnecessary session_start() call.
 
@Xander I'm re-writing my answer and I'll try to include this best that I can, thanks for dissecting it :)
 
4:48 PM
@Xander Updated it under "So how does this differ to session hijacking...." I hope you approve the changes ;)
 
@TildalWave : sorry I had to attend a meeting, going to the datacenter to create a fresh installed replica
will come back tomorrow on this channel and update you on my findings
 
@Sudhi no worries this chat isn't really 'live' anyway with only a few native & true cave dwellers in it :))
@Sudhi (i.e. you do get updates through the SE network if you were mentioned in any of the rooms, so no need to constantly check it)
 
yeah, though I prefer leaving the tab open and hibernate
@TildalWave see ya tomorrow!
 
@Sudhi it's weekend so i'm not sure i'll be here, but others will... or ping with my username like you just did and i'll get updates through a StackApplet ;)
 
5:06 PM
@TildalWave That's one of the cool things about SE chat. In fact, some people over at Role-playing Games use it for asynchronous D&D (or another TTRPG of choice) now and then.
 
6:01 PM
hi friends:)
 
@asadz Hi, welcome back.
 
thanks
 
6:27 PM
@TerryChia Yeah, that looks about right.
 
@ScottPack you look horror movies?
 
Not generally, no.
 
ok
 
I don't much care for gore nor do I particularly enjoy being scared.
 
@Adnan yo will you be accepting any of the answers on 'solar system in a thimble'? it's already gathering trolls with 'not really and answer' answers trying to be amusing :)
 
6:32 PM
lol
what about suspense thrillers
like hanibal (spellings)
 
@ScottPack you realize that's not the point of horror movies, right?
 
hopkins fans anyone
 
@TildalWave Not all, but especially these days most.
I enjoy a good thriller, I've also enjoyed older movies that fall into the horror camp. Event Horizon is the best example that immediately springs to mind.
 
@asadz It's spelled Hannibal but that's the US/UK version of it. He's supposed to be of Polish origins, so the original spelling wouldn't be with a double N
 
okay friend:)
i like old Hollywood actors
 
6:37 PM
@TildalWave Wikipedia says Lithuania.
 
@ScottPack Ah ok, either way they don't use double N's either
 
and @TildalWave my friend said wikipedia is not to be trusted
 
@TildalWave I will defer to your judgement. :)
 
@TildalWave Lithuanian language is one of the two languages which make casing rules language dependent (the other one being Turkish). A real Unicode-related headache.
 
@asadz Studies have shown that Wikipedia is no more or less reliable than the Encyclopedia Britannica.
 
6:39 PM
is this a good thing?
 
@ScottPack I see this often to be the case actually, the Americanization of personal names. Hack, we use so many different characters and variants they surely don't have all of them to include in the end credits LOL
 
@ScottPack Though I noticed the other day that Wikipedia wrongly claims that Genghis Khan had the formal "Kaghan" title. I should correct it.
 
@asadz Good...bad....I'm not here to make such judgments. Only to point out that as a general reference tool it seems to be in the same level of accuracy as previously trusted sources.
 
yeah make sense
it trusts the trusted
 
@TildalWave Also family names. The American version of your last name depended entirely on the literacy, hearing, or penmanship of your immigration officer.
 
6:42 PM
@TildalWave I'm waiting until the monster wakes up
 
@ScottPack I don't actually see that as anything wrong, it's been much the same in any parts of the world throughout history when immigrants were settling in. We also don't have all the character variants in our alphabet and I always wondered how do they register a person that has some that we don't. Turns out there's also some spelling rules effectively changing the original spelling to suit better with the 'new environment'.
 
@TildalWave Some unnamed yahoo even dared to downvote my answer without leaving so much as an explanatory comment.
 
@ThomasPornin where on the thimble answer?
 
@TildalWave I suppose the standardization was more my concern rather than the changes. One example is the Irish name Conley, which was sometimes translated to Connolly during immigration.
 
In my experience wikipedia is pretty reliable for technical topics. For topics where multiple political factions exist it's not so nice.
 
6:47 PM
thanks god it wasn't me lol
@CodesInChaos thanks
 
@ScottPack Heh with Irish names I always imagined there must have been some funny incident how the O'Reilly was formed :)
 
thanks for saying something in wiki defense
 
@TildalWave I had not considered that, but now I am greatly enjoying it.
 
My biggest practical beef is that many mathematical articles are hard to understand if you don't already understand the concept in question before.
 
@ScottPack oh really? :)
 
6:48 PM
Generally speaking the more popular a topic the more likely those pages are to be correct.
 
@CodesInChaos Yes, I noticed that, too. Articles about mathematics are all about formalism and not at all about actually explaining things in non-self-referential terminology. Strangely enough, articles about nuclear weapons are quite the opposite.
 
Tried to figure out what finite fields etc. are from wikipedia
I still don't get non-prime fields
 
@ThomasPornin isn't maths 'non-self-referential' by definition?
 
@ScottPack if there is ever a movie made on this forum or site; i like to vote you for the role of hmmmm you suggest?
 
The problem with math is if you get a definition+proof that's not really useful
you need to understand the concept behind it
 
6:51 PM
you can fit in many roles where looks count..
ghost town
 
 
1 hour later…
8:04 PM
I'd vote scott for the great lebowski
@WesleyDavid Ohai there kitty ^^
 
@LucasKauffman mew
 
@WesleyDavid pet pet pet
long time no speak @WesleyDavid what you been up to lately?
 
@WesleyDavid Hey big boy
 
8:20 PM
Yo @TerryChia for your collection:
http://www.agnitas.de/uploads/pics/000003771889_01.jpg
(OS X user?)
 
/me is an OSX user
 
I daily use Linux, Windows and OS X
 
@ScottPack so you're saying it's not nearly as immersive to not notice the bar you were sitting in moved somewhere else in the mean time?
 
@TildalWave Meh, it's just a tool yo.
 
@ScottPack who is? :P
 
8:27 PM
It should be pointed that I'm sitting in my workplace office web browsing on my Linux desktop and doing training/email on my 10.8 laptop.
 
none of them are so good you wouldn't notice on which one you're on anyway, but if it gets the job done...
i guess they're good enough
 
I think it's kind of funny. As long as I'm switching between desktop and laptop I make the transition completely fine. The minute I sit down at an OSX desktop? I keep trying to use it like my linux box.
Like trying to use Gnome3's magic hot corner to bring up the dock+Expose+notification bar thing.
 
'user experience' kinda ruined it for me
everything seems to be designed for idiots nowadays
 
8:45 PM
in The Comms Room, 16 mins ago, by Michael Hampton
@ChrisS Oh god. Go ask a neckbeard.
 
@ThomasPornin On the basis you use OS X, have you an answer to this ?
@ScottPack I love that hot corner. Every time I go to work, I keep trying to do it on Windows.
C'mon this should show me all my apps!
 
9:00 PM
@Ninefingers With the laptop I can at least 3 finger swipe. Sitting down at an actual keyboard and mouse? I get frustrated.
 
@LucasKauffman Been busy thinking and working and thinking. Re-evaluating work and life.
 
9:16 PM
@David Hahaha
 
@David Cool, a whole day since :) BTW Mozilla fixed that in under 24h (mozilla.org/security/announce/2013/mfsa2013-29.html)
 
I tested the January exploits versus my Windows VMs and a Linux one. Didn't work versus my open source java runtime, sadly.
 
This one is also hilarious: istherejava0day.com
 
the Windows XP one worked, scarily.
 
@Ninefingers No, sorry, I only have one OS X system, and I haven't bought Mountain Lion (yet).
 
9:20 PM
@ThomasPornin Eurgh. Well I decided I really ought to investigate OS X and seeing as Apple hate people virtualizing OS X or installing it on anything but a smooth silver finish box, actual hardware was needed. Now I have said hardware, I can't build install media!
Haha on that java-0day website there's a great line at the bottom:
> Java isn't the only thing you should be worried about. There are others.
With links to flash and adobe reader...
 
@Ninefingers seeing where most exploits are coming from, having a security model based on inability to purchase necessary equipment to later exploit it doesn't sound like such a bad idea
 
@Ninefingers I think the "Apple way" is that if you render your machine unusable, you just have to go to the nearest "genius bar" to get it fixed.
 
@ThomasPornin Am I allowed to mug the genius for his installer usb that he undoubtedly has?
 
@Ninefingers Could be firewire.
 
I accidentally joined The Bridge - It was horrible!!
 
9:25 PM
@ThomasPornin oh ALRIGHT, can I mug said genius for his peripheral-based installation media?
 
@ScottPack Hold me.
 
I suppose you can use violence on the genius as long as you used an iHammer.
 
Oh, and I changed by gravatar to a bear, so am I cool now?
 
@WesleyDavid Come here sweetness.
 
@ScottPack (>^-^)>
 
9:26 PM
@WesleyDavid There's a cache delay -- I still see a drunken cat face.
 
@ThomasPornin ...whose name is "Honeybear" incidentally.
 
@WesleyDavid You can vastly improve your experience by adding users to your ignore list.
@ThomasPornin Ok. I'll try to iNegotiate first, I think.
 
@Ninefingers I vastly improved my Stack Exchange experience by blocking myself from joining The Bridge.
 
@WesleyDavid That also works.
 
Gah! Taking so long to refresh my gravatar
 
9:34 PM
@WesleyDavid the internets have decided you don't get to be a bear.
 
                __
     __/~~\-''- _ |
__- - {            \
     /             \
    /       ;o    o }
    |              ;
                   '
       \_       (..)
         ''-_ _ _ /
           /
          /
RAWR! I'm a bear now.
At least, Gravatar updated for me.
Thunder, lightening, rain, and hail all night and day. Who says the desert is always hot and dry.
 
a small victory for our bear brothers from my parts of the woods:
http://translate.google.com/translate?hl=en&sl=de&tl=en&u=http%3A%2F%2Fwww.vier-pfoten.de%2Fnews-press%2Fnews%2Fslowenien-verbietet-wildtiere-im-zirkus%2F
(only 23 EU countries left to go LOL)
@WesleyDavid you made that gravatar from 'The Edge' movie?
 
9:54 PM
I made that gravatar from the entrails of freshly killed salmon
 
you are still a cat for me
 
@David Try reloading the DMZ page.
 
Ctrl+F5 did it for me
 
@ThomasPornin you cant make me
 
The wait for Gravatar to update is unbearable.
Strips my patience bear.
Makes me want to go to their offices and rip them apart with my bear hands.
But that would be rather grizzly.
So I'd better grin and bear it.
Put my plans for destruction on paws.
Clearly I'm not koalafied to make rational decisions in this frame of mind.
 
10:01 PM
@WesleyDavid there you go
 
@TildalWave It was so snowy that day, I lost my bearings.
 
@WesleyDavid heh pretty fierce fella you chose there
 
Only when he's doing coke. Otherwise he's chill.
 
@WesleyDavid oh that explains it :P
 
Shog9 on March 08, 2013

It’s time once again to cast your vote for the next Stack Overflow moderators. The primaries have just ended, and the top ten candidates can be found here: http://stackoverflow.com/election.

We’re running the election now (rather than a year from the last election in June) because veteran moderator Tim Post is stepping down in order to work with us as a Community Manager! While we’re extremely lucky to have his hard-working brilliance brought to bear on the problems we face managing all these sites, his transition does create an immediate need for a replacement on the SO mod team. …

 
10:05 PM
@TildalWave He was on an eight-ball bender in that photo, so is totally agro.
 
did that already... how come there's only 3 you can vote for but 4 positions open?
 
So there's no chance of a 4-way tie? =)
 
@WesleyDavid Hey ! Koalas are not bears ! Even sheep are genetically closer to bears than koalas are.
 
@WesleyDavid touché
 
@ThomasPornin What about drop bears?
 
10:08 PM
bears would make for good astronauts
 
10:26 PM
What's the advice on our non-question related comments to questions/answers (that might have helped establish what's being asked/answered) that are no longer really needed there? Should we delete them or leave that to mods to decide?
I mean those comments that pointed to some issues that were later addressed in answers/question... if they don't have any replies in the comments, then there's no point in keeping them there, right?
 
10:39 PM
@TildalWave Sure, updated answer looks fine to me. Thanks for the discussion!
 
@TildalWave depends. Sometimes comments that prompted the asker to clarify the question are good to keep around because they illustrate what kind of information someone asking a similar question should include
 
@Xander was a pleasure, thanks :) mind me asking, as a software consultant, which web framework is least frowned upon? I don't need a long answer, I'll take your word on it
 
If there are a lot of comments, delete the ones that aren't relevant any more
if a comment has no value to future askers, delete it
 
theres still something i dont get about system calls. if the cpu is in ring 2, it only has access to ring 2 instructions - the same instructions that any userland program has. so how, using just those instructions, can it switch to ring 0 in such a way that a userland program couldn't replicate itself?
 
@Gilles the up-votes on comments have no real value in this case, right? I mean, they're only there to show approval and they're shown higher up if the list is long, or is there any other point to it i'm missing?
 
10:43 PM
@TildalWave agreed
 
@Gilles cool I'll do some clean-up then :)
 
i just finished reading an extensive chapter on windows system calls and x86 memory management, but I don't feel this was addressed.
 
@lynks the general idea in architecture design (x86 may have additional quirks) is that the only ways to elevate privileges jump to an address that can only be changed with elevated privileges
e.g. unprivileged code can trigger an interrupt, which causes a switch to interrupt mode (ring 0) and a branch to the interrupt vector
in other words, any program can jump to kernel mode (it does that to issue a system call), but that always invokes the syscall processing code, not some other code that the program could choose
 
@Gilles hmm ok. so executing the 'system call instruction' - something like INT 0x80, actually looks something like JMP [<privileged register>], RING 0 (in pseudoish code), where <priviledged register> is something like CR1 ?
and RING 0 is an instruction to put the CPU into ring 0
 
@lynks yes. I can't vouch for the specifics (I only know them on arm, not on x86), but that's the general idea
 
10:53 PM
@Gilles thanks, that appears to make sense
 
1
A: Dangerous interrupts

GillesThe details depend on the processor architecture, but generally speaking, when an interrupt is raised: The CPU saves some registers (at least the current program counter and an indication of the current processor mode). The CPU switches to a privileged mode. The CPU reads a pointer value from v...

2
A: What mechanisms prevent a process from taking over the processor forever?

GillesVarious peripherals are connected to the main processor. When an event happens in a peripheral, the peripheral sets an electric signal which causes the processor to take some action. The act of setting this signal is called an interrupt request (IRQ for short), and the aftermath of the interrupt ...

@lynks Two somewhat related answers ^^^^
 
@Gilles From the book, i had understood that memory management is the major player, in that kernel code is stored in pages marked as 'ring 0 execute', which led me to ask why you couldn't then build arbitrary kernel execution out of ROP gadgets that live in kernel pages...but I guess where the code lies it not sufficient to make it privileged code...
@Gilles thanks :)
 
@TildalWave I don't believe on frowning upon frameworks. :-) In the enterprise space, you've going to find a vast preference for .Net or Jave. They'
re pretty solidly entrenched.
@TildalWave Ruby (particularly using Rails) is the hot new thing, but the push for ease of use has led to some pretty serious security vulnerabilities, though I believe all the known issues have been patched at this point.
PHP is meh. There's a massive amount of it out there, it can get the job done, and nobody takes it particularly seriously.
@TildalWave I'll give you the Official Approved Consultant Answer. It depends. The best framework for you depends on what you want to do and who you're doing it for. There aren't any I would say "don't use." Some make thing easier than others in various scenarios.
Personally, I've been doing ASP.Net for the last 10 years, so that's what I like. It's probably not the best option if your goal is to minimize up front costs though. Rails or PHP is a better option in that situation.
 
@Xander I'm one of those rare eccentrics that roll their own, but is always good to know what to follow and compare to. Potential vulnerabilities might be completely different due to the nature of my framework being limited to an extremely tiny subset of features big guns would have to support (or that it's completely pre-compiled), but I'm following new ideas in this field and try to match those I like.
 
@TildalWave And if you go to an Oracle or WebSphere shop, you use Java, end of story.
@TildalWave Probably the ones to look at for feature comparisons would be ASP.Net MVC 3/4, and Ruby on Rails. Those are the latest generation of fairly mature frameworks.
 
11:09 PM
@Xander I'm mostly following ASP.net MVC so far as it's similar to my style. Also following with great interest server-side JavaScript based frameworks that are just starting to trickle out of the woods slowly. Some neat ideas there :)
 
Ah, yeah, Node.js and the like?
Definitely.
 
@Xander Yup. Cool to follow as it's merely at the development stage (library-wise) and some really excellent coders working on it.
@Xander But I will definitely need to look more into Ruby on Rails. And Equinox/Jetti
 
@TildalWave Yeah. Just moving from the threaded model to the event-driven model has some really incredible potential.
 
@ThomasPornin would you mind if I sent you an email for some feedback on something?
 
@Xander I started building my framework with Delphi when Indy came along, so quite some time ago. Of course, I'm a one man band for this, so progress is slow. It's nothing easier than to write 10 new lines of code, but not many things are harder and more time-consuming than to properly test them. Still, I think I'm slowly gaining traction after a few years on 'hiatus' LOL
 
11:21 PM
@TildalWave Yeah, that is understandably a lot of work.
 
BTW when I say Delphi I mean those from the Borland guys, not this drag&drop crap they're selling now :)
 
@TildalWave Yep, I remember. :-)
 
It used to be Delphi & Kylix, but lack of support and new/updated open source libraries meant I had to immigrate... now looking to go back to the comfort of true OOP with Lazarus, but re-writing libraries will be a pain in the butt
 
Anyone here good with c#?
 
@AdamMcKissock lots of us know Java, which is rather similar. try us :P
 
11:30 PM
hehe just stumbled upon my first Andrew's post... twilight zone
"This account is temporarily suspended to cool down. The suspension period ends on Jul 3 '26 at 18:01."
ROFLMAO
 
@AdamMcKissock Not as good as some of the folks at SO, but happy to give it a go.
 
That's gonna be some July 4th fireworks worth watching :)
 
Just trying to wrap my brain around all this OOP stuff
 
@AdamMcKissock pastebin it and we'll have a look at it
 
Doing something like this:
JavaScriptSerializer jsonSerializer = new JavaScriptSerializer();
User user = jsonSerializer.Deserialize<User>(doLogin());
I then need to access user from another class without re-creating it
 
11:41 PM
@AdamMcKissock Can you not just pass the user as a constructor parameter or a method parameter?
 
Not sure what's stopping you from accessing your User directly? Where are you declaring it that it's not visible to your other class?
 
OtherClass.DoSomething(User user) in other words?
 
if you're declaring it private then that's the reason why you can't reference it from another unit/class
 
im confused
 
how about public User user = jsonSerializer.Deserialize<User>(doLogin()); ?
 
11:45 PM
invalid expression term 'public'
 
@AdamMcKissock yeah you can only do that in the class scope, so outside any method
 
so if i did public User user, then do user=json...
 
@AdamMcKissock so at the top of your class you would have User user = null; and then you can do instanceOfClassWithUserInIt.user
@AdamMcKissock add a public in there for good measure
 
boosh!
 
hehe
welcome to OOP!
 
11:49 PM
this is really bad, but I've been using c# for nearly four years...
 
@AdamMcKissock its ok, i was employed as a php developer for 3 years, and never once declared a class in php...
 
never touched C# myself, but this visibility/scope thing is pretty much equal in all of them
 
is there a way to do it so that you dont need class.user, you can just call it as user.?
 
@AdamMcKissock yep, but it's messy (static imports, they are evil)...the point is that you want to know precisely which user object is being accessed just by looking at the code.
 
there will only be one user object
 
11:52 PM
@AdamMcKissock if that seems to not make sense in your context, you probably have a slightly bad class structure
 
if you declare jsonSerializer as a static withing your function, then I guess.... ah @lynks answered
 
User itself is a get/set class
 
a.k.a. a property
 
@TildalWave if you say so
 
@AdamMcKissock then sure, make a class for 'global statics' or similar (believe me you want them all in one place), declare it as public static User user = null;, then you can have an import static my.package.heierarchy.GlobalStatics.* but this is a static import, and may god have mercy on your soul. (it's almost as evil as goto)
 
11:55 PM
but thats bad?
 
that essentially says "add all the static fields from GlobalStatics to the scope of this class, so that I dont have to use GlobalStatics. every time I access them"
the reason this is bad, is that it is no longer obvious where user lives, just by reading the code that accesses it.
and if you ever add 2 or more static imports, things get very messy
 
okaaay
could I not make user a public static class and then just import that?
 
@AdamMcKissock for things like that, where there is only ever one instance of a class, I tend to put them in the entry point class (the one which has the main method), make them static, and initialise them in the main method or asap.
@AdamMcKissock you can't import a Class unless you're doing a static import.
 
@AdamMcKissock No! Don't make it static.
That's not thread-safe.
@AdamMcKissock My wife is dragging me out to dinner, but need more information.
 
@lynks So instead of having a seperate User.cs, put the class inside Program.cs, then initialise User user in program main?
 
00:00 - 15:0015:00 - 00:00

« first day (813 days earlier)      last day (4076 days later) »