« first day (1538 days earlier)      last day (3354 days later) » 

6:10 PM
@Simon I have successfully moved into my first apartment.
but not before totally ruining my back
 
@DavidFreitag achievement get
 
@Ohnana The sofa I ordered wouldn't fit :[
Now instead of a sofa with a sleeper and a love seat, now I only get two love seats
 
@DavidFreitag cut to fit?
seriously
 
@Ohnana There is this monumentally stupid privacy wall by the front door similar to the ones in public bathrooms. If it wasn't there I'd have my sofa
 
damn.
 
6:15 PM
I got 105mb/s internets with a wired modem, not one of those POS modem/router combos
Beats the shit out of the 9mb/s speed I had before
 
@Xander maybe the chat room type?
@DavidFreitag You have earned the achievement "NAKED AT LUNCHTIME"
 
@Flyk I got that one a long time ago.
 
worrying
 
raz
@DavidFreitag @Xander I don't know if you're a beer guy but Untapped added a badge for checking in beers before noon. I'm Lvl3 :D
 
6:22 PM
@raz lol, i don't really drink beer much, and iirc they don't count angry orchard (probably because it isn't beer)
 
raz
@DavidFreitag They do count ciders!
 
@TildalWave my dad's bringing back mutton chops and it's honestly the best facial hair I have ever seen
 
@raz Well even then, I usually have a glass of scotch or tequila
 
raz
@DavidFreitag I don't go for Angry Orchard too often. Crispin is dry and delicious.
@DavidFreitag Single malt's never a bad choice. Tequila is always a bad choice for me.
 
Mah back D:
 
raz
Skirt width...
So as beards have decreased so have skirt widths... skirt length maybe. Not sure about width.
 
well yes you can pack more of less wide skirts on public transports
 
@raz hoop skirts vs pencil skirts
 
Yep, I'm lost.
 
raz
Yeah I don't know about skirts at all
 
6:28 PM
@raz I think they are kilts for women
3
 
raz
I'm not Scottish, but I see your point.
 
@DavidFreitag i have a kilt
it's great
can't wait until summer so i can wear it more
 
@Ohnana Oh yeah? I've got a mustache hat.
 
@DavidFreitag not a hat fan
 
raz
A mustache hat?
 
6:30 PM
but that's cool
 
Yeah let me see if I can dig up the picture
Jan 27 at 0:57, by David Freitag
user image
That one ^
 
@DavidFreitag Well done! How is it like so far?
 
@Flyk please throttle your edits. new questions are getting lost. I made the same mistake when I started making edits.
 
@schroeder @Flyk isn't new to the edit game, but the numbers seem to ramp up when she's bored...
 
@Simon I haven't been able to enjoy a single minute of it. I moved everything in and proceeded with unpacking. 20 minutes into unpacking I felt a pop in my back and I have been in agony since then
 
raz
6:35 PM
@DavidFreitag How to Train Your Dragon merch?
 
@raz Nope, my sister made it for me for christmas
 
@DavidFreitag glorious
 
@Ohnana It's incredibly warm.
 
@DavidFreitag I'm sure
@DavidFreitag also next time you pick heavy things up and put them down, use your knees and get help :)
 
@Ohnana I picked up a keyboard from my bed -.-
Next thing I knew I was writhing in pain on the floor
 
6:38 PM
@DavidFreitag probably not what caused it
 
None of my stuff is particularly heavy
 
still. sucks that you goofed up your back
 
Also, no microwave.
 
@raz Ha, nice. Packy uses Untappd...I have it, but don't really use it.
 
@TildalWave That's great.
 
6:47 PM
I like that they clarified in the graph that frequency of wide skirts is for women and frequency of beards for men :)
 
7:22 PM
@DavidFreitag Oh, you got a pussy back too?
:2034 I wonder what happens if I change the reply ID.
IT DOESN'T WORK LADS, IT DOESN'T WORK
 
@Simon that message isn't in our room
in RPG General Chat, Oct 11 '10 at 13:36, by RMorrisey
GURPS is a fairly realistic system. If you want a system that's more cinematic, and closer to d20 stuff you're used to, I'd strongly recommend Spycraft. It's not very cross-compatible with other d20 games (because the game balance is just so different) but it's a lot of fun and has a lot of flavor that I think d20 modern lacks
@StackExchange This is the first message in our room
 
@RoryAlsop You ever feel compelled to do something, and then afterwards immediately look back at it and think "why did I even bother wasting the energy"?
(RE: My posts on "what does it take to get an answer" and the related meta.u&l post.
Damnit. I hate when I leave a word missing or parenthesis hanging and the comment/chat message editing period is expired.
 
@Ohnana Lol and @AviD has been here pretty much the whole time!
 
@RоryMcCune ikr
 
@RоryMcCune Wow, there's another face we don't see much anymore I think - @GrahamLee.
 
7:34 PM
@Iszi yeah I've not seen Graham about in aaages
I think he does more general iOS stuff these days, rather than just security
 
anyone really up to date on session management in ASP.Net? Is it really as bad as it appears?
it blows my mind that by default they reuse session ids and don't provide any mechanism for changing SessionIDs on permission level changes at all
 
@Simon Yes, it runs at the sight of even the slightest bit of work
 
@Ohnana nurd
@DavidFreitag I usually suck it up and keep going.
 
@Simon so nurd=has critical thinking skills?
 
@AJHenderson are you thinking of ASP.NET_SessionId or .ASPXAUTH tokens?
 
7:43 PM
ASP.Net_SessionId
 
@AJHenderson in most cases from what I've seen, both are set but it's .ASPXAUTH that handles session management (you can delete ASP.NET_SessionId and it'll just get recreated)
that's not always the case but most of the time it is
 
@Simon Yeah this wasn't a "suck it up and keep going" it was a "Ha bitch try and move now"
 
IIRC the older ASP.NET apps may have just ASP.NET_SessionId
 
it stems from an attempt being made to re-write our auth so that we can remote kill sessions. Forms auth doesn't provide a way to do that, so they want to move to using the session, which seems like a bad idea
 
@AJHenderson ah yeah by default IIRC ASP.NET doesn't really do session token invalidation
 
7:45 PM
@Ohnana It's not about the critical thinking, it's about how you apply it.
 
it works on a sliding time window
 
@RоryMcCune not only by default does it not, it appears to be impossible to do any kind of change on an active session. You can use the SessionIDManager to kill the current cookie and make a new one, but if you had an existing session, you lose it as it makes a new session rather than changing the id of the current session
basically, whatever ID you provide as the "session id" in the cookie, ASP.Net will happily create a session that corresponds to it
 
@AJHenderson yeah in that, it seems like ruby on rails using cookie store, you can't invalidate session as it's not really managed server-side
good for scaling
not so good for security
 
and I think it hasn't been made a big deal of because most Authentication in .Net is done through Forms auth which is a fair bit more secure and then you can cross check your forms auth with your session to make sure your session access is valid
but then Forms auth can't be invalidated before being applied as a valid state by anything remote because it is literally just an encrypted token that says "I'm this user until time x"
with no way at all to say "no you aren't gtfo"
 
@AJHenderson indeed, you'd need to custom code a check server-side to handle that
so an additional cookie set, that is stored server-side, correlated to the user and can be invalidated
 
7:50 PM
@AviD Is that the Sec.SE chat room Easter egg we've been hunting for?
 
right, my initial idea was tie the Forms auth to the session state via a key that we could invalidate serverside, but session isn't available until way after forms auth is done, so the user will already be authenticated and active on the system before I can check the session to see if they are valid or not :/
 
@AJHenderson Session state is concept that shouldn't exist in an HTTP app.
 
lets just say I've written up "ineffective logout functionality" for ASP.NET apps once or twice
 
@RоryMcCune I have trouble with this finding. It's so incredibly narrow as to be a non-issue for 99% of apps. There are occasions when it can be an issue, but they're infrequent exceptions rather than the rule.
 
@Xander that's a nice concept, but not very practical. You can't effectively push everything to the DB all the time and still have any kind of performance
in certain cases
 
7:53 PM
@Xander if an app provides a log out feature, it should actually invalidate the session server-side, or the developer should be aware that it doesn't
@Xander the finding is a low which says "hey after using the logout function the IDs are still valid for x minutes, you might not want that to be the case"
@Xander so the site realises that (they may already do so, but in my experience many don't)
 
@RоryMcCune That goes back to my argument that an HTTP app should (generally) not have a concept of session side state, but should operate on request/response pairs. And yes, the dev should understand how cookie replay can be an issue in that model.
 
@Xander apps (as users experience them) to me need to have some way of recording when a user logs in and out and once they log out, from a security perspective, they shouldn't be able to access content only available to logged in users
 
@Xander in our case for example, we actually had to deploy a specialized session management solution for storage because we had been storing down to the DB and it was killing our firewall on the information we had to pull back about user every single request, even at the AJAX level
 
@Xander so HTTP as a protocol might not want/need a concept of session, but apps as most users experience them do, and for better or worse, most are written over HTTP these days
 
@AJHenderson You don't have to push to the database all the time. I'm not arguing for persisting all the things to the database, I'm arguing for not shoving a bunch of data loosely associated with some request/response pairs into semi-volatile storage and calling it "stateful."
@RоryMcCune That's fair.
 
7:57 PM
@Xander ah, well, that's just abuse of session state
 
@RоryMcCune Yes, I do understand that this is far too infrequently noticed or understood.
 
not that session state itself is inherently bad
 
raz
@Xander It's been nice since I've switched phones 3 times in the last couple of weeks. All my beers are all centralized. As I often forget if I've had a beer.
 
@AJHenderson Well, it's certainly useful in many cases, and I wouldn't go so far as to say it should be actively avoided if it does in fact serve a useful purpose for an app, but I would call it an anti-pattern. It's duct tape on top of a architecture that wasn't designed to support that sort of mechanism.
@raz Yeah, would be super handy for me for that as well, though at this point, it's a bit of a lost cause. I'm just trying to implement better search on my blog instead so I can use that as my library, since it at least covers most (but alas, still not all) of the beers I've had in the last four or five years.
 
@Xander yeah, on that we agree
I think HTTP as an application platform in general is boneheaded
since it started as a way to get text files that linked to other text files
it's the epitome of a chain of hacks spanning decades
 
8:04 PM
@AJHenderson Ha ha, yes, that is very true.
 
(which would be why I normally hide out in back end land)
 
raz
@Xander You should be able to script and create your own beer database. Have a column for the blog post link. Then create a full beer list webpage with links included. That'd be kinda sweet. Then search the DB.
 
@AJHenderson So simple answer? Add a piece of data to your session IsAuthenticated and set it to true when the user authenticates, and false when they log off. All new sessions should have it false, and once the user is logged out, the old session will too. You don't need to worry about messing around with sessionids at that point, because the sessions themselves accurately track authentication status.
 
well, the sessionid's issue has to do with fixation issues
ie, because .Net's sessionID by default stays the same, if I were to, say, change my password, it doesn't do shit to lock out an attacker who was already logged in
because my session and their session are still linked
 
@AJHenderson Ah, gotcha, yeah, reading up further now.
 
8:13 PM
any permission level change should result in a sessionID change, but I can't do that
it seems to not be made a big deal because the Forms auth cookie is more secure and can be checked before allowing access to the session (making the session untrusted)
but I can't use Forms auth due to it not being able to be invalidated remotely (in the case of, say, a rogue employee)
it gets set too early in the stack and the user would be valid until session state can be loaded
which is a bit risky, though possibly the best we can do
but I could put a server side value that we can invalidate that will cause the application to say the valid forms auth ticket is not actually valid
but only after I get to the point where session state is loaded
 
@schroeder #careface
 
lol
 
8:31 PM
@AJHenderson Sure, stick something on the request or in ThreadLocal storage. That should work. It's been so long since I've worked on an app with session state that I've forgotten where in the pipeline it get loaded.
 
Excel is really starting to piss me off
Half the time the sum function keeps returning 0, the other half it just enters the formula I write as text
So infuriating
 
raz
Software is only as smart as their operators.
 
I think the massive team of developers had something to do with how smart the software is too...
 
8:47 PM
@Ohnana A nurd has nurd skills.
 
@Simon yes. a badge of honor
 
@DavidFreitag Yeah, I've had those a few times, had to stay in bed for ~3 days.
 
9:12 PM
freakattack.com <-- another named SSL bug
'cause we've not got enough of those in our test reports already
 
Don't roll your own crypto, they say.
 
@Xander thread local won't work in any distributed environment
because you have no guarentee that the same system will serve both requests
 
If I look at all these beasts and freaks and the crime rate, how can I not roll my own crypto?
 
@AJHenderson Ah, using out of proc sessions?
 
@Xander yeah, with a custom provider
which also does mean that our session state really is stable
or as stable as we want it to be
and yeah, that's what I ended up deciding to recommend, I don't like the Identity of the user ever saying it is authenticated when it isn't, but I simply don't see any better option avialable
and everything I find seems to indicate that coupling session to forms auth is the appropriate solution
I could always implement my own secure cookie, but that's dumb when Forms auth is sitting there able to provide it. And if we really wanted to be fancy, we could implement a direct access to our session state service outside of the .Net session manager, but that's probably something for another day
though that would allow preventing the Identity from ever being valid
well, provided I have access to the cookie at that point anyway
 
9:39 PM
@RоryMcCune The dude was so much in a hurry to make the website that he didn't double check his spelling.
"cryptogrpahy"
 
10:11 PM
0
Q: Multiple iptables denied in report from hourly logcheck on Debian

somethingSomethingI get thousands of lines of iptables denied every hour from logcheck, for a week now. What are the steps to troubleshoot this, what are the main causes of this? This is inside an LAN and the addresses denied are local to the LAN. Here is an example of the logcheck report: This email is sent ...

 
5
Q: Why is my debugger detected as a Trojan by anti-virus software?

lyenliangI'm trying to use a deugger (namely OllyDbg) to analyze some exe files. However, all the versions I downloaded from the Internet are considered as Trojans by some anti-virus software. (I use www.virustotal.com to scan the binary I downloaded) Are OllyDbg's main behaviors belong to a kind of Tro...

 
10:29 PM
Future value as in we're now having a LMGTFY 101 session. Once we determine which suitable search keywords would match the image you're searching for, there's no further value to it for our Q&A. I.e. it falls in the unclear what you're asking category, and by the time you edit to clarify what precisely you'd like to find, it will be trivial to do so. How about 1 or 2? — TildalWave ♦ 32 mins ago
This Q&A concept is too hard for most people isn't it?
 
@TildalWave I dunno, seems pretty hard to me.
 
Yes, it's too complicated.
 
@TildalWave sigh, you missed that euphemism completely.
 
I often find myself commenting with stuff like Some conjecture is fine as long as it's sufficiently explained and preferably supported by facts and/or reliable souces of information. Which is just verbal diarrhea isn't it?
@DavidFreitag I'm serious tho
 
@TildalWave I think you have made the mistake of believing that everyone who comes through here actually reads any of the relevant information before posting something.
 
10:36 PM
@DavidFreitag but it shouldn't be this hard to ask for help should it?
 
6 mins ago, by David Freitag
@TildalWave I dunno, seems pretty hard to me.
 
we're a bleeping bureaucratic fortress and I often feel like I'm just walking in circles like that Marvin from HHGTTG just to make the point
And if I let through some simpler (trivial in SE parlance) questions, then you get whole essays in meta asking if that's what we've became
Like this:
2
Q: Is there value in questions about future schedules?

Eric PlatonI came across questions like, most recently, "What is the planned date for the Falcon Heavy Demo launch from Vandenberg AFB?". Such questions are interesting to me, but they also seem very volatile, speculative, and cluttering in the long term. How does SE want to deal with these questions? Jus...

And this is nothing, you should see meta posts on some other sites, e.g. Astro.SE
Anyway, what's new here?
 
@TildalWave They finally killed OpenGL khronos.org/vulkan
 
ah nice
where does it work? :)
and does it run node.js?
 
@TildalWave According to the banner at the bottom it has been adopted by most of the chip manufacturers as well as some large game companies.
 
10:49 PM
ah yes I see logos now ... nice
 
@RоryMcCune Oh well, that's it. I am going to write a SSL library just to show how these things should be done.
 
@ThomasPornin Ah, yes! Make it portable too :D
 
@DavidFreitag The choice of language is still open.
Basically, either something sane like C# or Java, or something portable like C.
 
@ThomasPornin Agda2?
 
@ThomasPornin Heh, I had you pegged for C.
 
10:57 PM
@DavidFreitag I know too much about C to be thrilled about programming in C.
 
@ThomasPornin Since when has C ever been thrilling? I have never found the process of programming to be thrilling, but I have found the result of said programming to be such ;]
 
@DavidFreitag I envision good code as a work of Art. The process of creation is the best part.
3
 
11:31 PM
where do we find these people?
@schroeder Some of my responses are "unethical"? What are "ethics" when we're talking in the context that his school has perpetrated a man-in-the-middle attack which victimized OP and his peers? In this context I think the topic of "ethics" is a red herring -- OP is a victim and he is entitled to self-defense. You may have a point that I don't answer the letter of his question, but I do provide certain hypothetical responses to the grievance OP raised. I strongly feel though that the "ethics" debate flies out the window here. If you're being done wrong, do yourself right, no one else will. — user69557 2 mins ago
 
@schroeder Sigh I don't know, but wherever we find them, we should return them as defective. I hope we kept the receipt.
 

« first day (1538 days earlier)      last day (3354 days later) »