« first day (1799 days earlier)      last day (3379 days later) » 

17:00
I SWEAR IF SIMON SAYS "HIS WEEN" I WILL PUNISH HIM
000
000
pony ween
dammit
sometimes i find a ween during testing
it's very confusing
@AviD #fogey
000
000
17:03
#smbc
000
000
#smbc
heh Im already watching that right now
@Ohnana I usually don't test my code all that much. 99% of the time, it works right the first time. :p
Usually there's just a few things to change
hahaha
17:11
I run through everything in my head before moving to the next
testing begins once I've been writing code for 4 hours or something
000
000
@MarkHulkalo pls.
@MarkHulkalo ahh I see you're a proponent of YDD
#Yolo Driven Development...
2
Rofl
Kidding, of course I test it.
@AviD Vaguely reminiscent of...
I believe in polish
@MarkHulkalo lies!
Especially polish sausage
Also, I'm an avid practitioner of WTF-Driven Development
@Iszi very vague
@Amelia hey now, I am the only AviD practitioner here.
Praise agile development
17:16
@MarkHulkalo or, in actuality: kanban with a backlog that ends up sinking after a month
Explain
@MarkHulkalo the inevitable result when you want to do something agile-like and management has no idea what the shit they are doing
lol
Fortunately, all our managers are pretty good here
I've seen it happen 4 times in a row
What is preventing a website from knowing my password if they can tell if it insecure?
17:18
I'm surprised. I thought they'd suck. They
They are pretty good*
@Alizter They are probably checking it by regex.
@Alizter uh, nothing? You just told them?
@Alizter is this on the page where you create it?
Or is it an existing password
While there's nothing stopping them from stealing your passwords if they really wanted to, most websites don't do that. They're usually checking your password lengths + regex to make sure they're valid.
If existing - it's in plaintext
17:19
@MarkHulkalo "most"
If it's on the page, they're checking it via regex in your browser
Because they claim that only I will know the password
@Amelia unlikely
but I don't see anything stoping them
There's nothing stopping them
At all.
You should just use separate passwords for each website. A password manager, perhaps?
17:21
But why do we trust websites at all if this is the case?
@Alizter we don't
000
000
..?
Is there some sort of check that they get to make sure they are not doing anything shady
000
000
No?
Similar to certificates?
000
000
17:21
You can't control a website.
There's a lot of ways to check it
@Alizter the only point of giving your password to a website, is so THAT website can verify you know the password.
you should never EVER reuse passwords.
000
000
^
^^
You can sign up on a website with a fake username and email that you registered somewhere, and a fake password... then see if anything funny happens with the account.
and then you hope they properly protect the password you gave them
17:22
hope being the key word
000
000
They can do what the hell they want with your password; it is your responsibility to do the good practice and use a unique password on every site.
there should be no key, it should be hashed ;-)
For example, the NSA steals your passwords enroute if communicated over http :D
But is it not the case that ~90% of people reuse passwords?
@Alizter It's probably higher than that
000
000
17:23
@Alizter So what? How do you force people to not reuse passwords?
@Alizter I think its less than 90%, but still way too high, yes.
000
000
You'll never know.
Trust me, it's higher...
I used to fix computers for people
Young, old, and in-between, they all re-used their passwords for everything.
000
000
You were fixing computers for people who are clueless about Computer Science.
Yep... 99% of people
17:25
Exactly
Anyway thank you guys for clearing that up for me
I have reason to justify my paranoia with passwords now :)
@Alizter really no reason for paranoia. With passwords, all you need is a healthy does of fatalism.
It's like you guys are experts on Information Security or something. Who knew. :)
quick advice, if you want to be secure with your own passwords:
1. use a password manager for ALL passwords. Make 'em long.
2. Use passphrases (4+ random words, e.g. Diceware style) for anything you absolutely MUST remember (e.g. password manager password, OS password...)
3. NEVER reuse passwords.
4. Turn on 2FA / 2SA on any site that supports it. Prefer TOTP instead of SMS.
5. Sign up for https://haveibeenpwned.com/
4 random words is bad advice
most people are gonna use the top 1000 most common words for that
@MarkHulkalo okay, go for 5
@MarkHulkalo you dont understand
17:31
@AviD TOTP over SMS? Really? Why not an app?
I'm bashing correct horse battery staple
and its 2000 top common words.
@Iszi yes an app
oh okay I see
@AviD even if it's the top 4k, that's like 6 minutes :D
@MarkHulkalo I dont think you understand how entropy works
@MarkHulkalo 4 RANDOM words from that dictionary
@AviD +1 for HIBP
17:32
@AviD if people are using 4 common words without capitalizing, etc., then it's pretty useless
@Iszi I didnt mean to get the code over sms, I meant prefer A over B
@AviD Yeah, good fix.
@MarkHulkalo if you know correct horse battery staple, I dont understand how you could still say that.
@MarkHulkalo my passwords are all lowercase and 50+ characters long
@MarkHulkalo this is completely wrong.
17:33
Wrong
One second
gtg, I will finish correcting you later.
@AviD May/may not be worth saying prefer 2FA over 2SA. I'm not sure if anyone offers both.
better yet, I'll let @ThomasPornin do that:
283
A: XKCD #936: Short complex password, or long dictionary passphrase?

Thomas PorninEdit: there seems to lack a thorough explanation of the mathematics in this comic (at least not as detailed as it could be), so here it is. The little boxes in the comic represent entropy in a logarithmic scale, i.e. "bits". Each box means one extra bit of entropy. Entropy is a measure of the av...

@Iszi true, but so few even understand the difference.
@MarkHulkalo wrong wrong wrong wrong, wrong wrong wrong wrong, your wrong , your wrong , your wrong </DrCox>
2
THEN TEACH ME
17:38
@MarkHulkalo hokay you have lets say 10k possible words
@RоryMcCune s/your/you're :-)
you put two together, how many combinations do you need to try?
10k x 10k
yeah, but I'm thinking common words
4k * 4k * 4k * 4k
What is this? 256 trillion?
GPU cracking @ 300 billion hashes per second
a reasonble number to try
14.2 minutes?
17:39
@MarkHulkalo if you're storing MD5 maybe
000
000
@MarkHulkalo That'd be a shitty algorithm.
but if you're doing that you're already pwned
try that with scrypt
see how you get on
or bcrypt
or PBKDF2
17:40
scrypt = 348 billion hashes per second with 25 gpus
also to get that far the attacker has already Pwned your databse, most likely, to get the passwords
That's the only thing I'm worried about, rory
the attacker pwning the database
@MarkHulkalo and remember and this is critical, you're assuming the attacker knows all the passwords to put them in a list
ie the user had to be constrained
on their password choise
not to use a single upper case character or number
if they did that the attackers 25GPU attack fails miserably
17:41
but nobody's gonna choose the top 4k. they're gonna do 1-2k max. especially old people
old people are gonna do: cake muffins tuesday grandson
also that's 25GPUs per password
no one does that for one old grannys single credential
7.1 minutes per password
you can buy credit cards with real money on them for less than that
17:42
using top 2k, IF that model is followed
True, but I'd probably do it just because I could ;)
and in the real world if they're just getting creds they ran them through a dictionary got most of them and then walked away
You have a lot of good points, but I still think it's somewhat feasible. A nation-state could do it much easier
any system where a feasible attack involves attackers who can pwn your database then use clustered dedicated password cracking systems, should not be using single factor auth
@MarkHulkalo if you're worried about nation states you don't use passwords
2
end of story
I'm not worried about them :p
it's all about threat models
all security is a trade-off
17:45
But a lot of attackers just copy your database and sent it over the network
Let me show you a common method I've seen a lot
correct horse battery staple style passwords are absolutely fine for systems where passwords are ok
@MarkHulkalo if they've got your whole database from the threat model of your system you're probably fecked already
They literally .rar up your entire database somewhere, and split it into hundreds of files, and place them in random positions, then send them
and I've never seen an IDS pick that up
Granted, might be using shit IDS
@RоryMcCune I miss that show.
I hardly ever see IDS decently configured anyway :)
and I'm a security tester
@Iszi me too :/
And you're right, fecked already... but there's a lot of information that a re-used password might get you. Apparently even John Brennan was reusing passwords
s/command/function
17:48
@MarkHulkalo sure, for the user it's a problem, but as they have no control over the security of the systems they use, for their threat model the only sane answer is a password safe
But isn't that a single point of failure?
@MarkHulkalo sure, but what are the alternatives?
good point
000
000
@RоryMcCune SQRL BRAH
any hardware tokens are rarely cross-platform, so no dice with todays multi-device world
17:50
Personally I would just silently modify the login command to send passwords over the network, encrypted somehow
@MarkHulkalo sure client-side JavaScript encryption, keybase use it
@RоryMcCune Unless you go for a dedicated device, e.g.: RSA token.
@000 I've just eaten!!
000
000
lololol
Yep, you'd have to install a program to get the key from the hardware, but even that could be intercepted/changed
17:51
@Iszi yeah those are good but rely (apart from perhaps yubikey) on server-support
@RоryMcCune Doesn't all 2FA?
@Iszi IIRC yubikey can pretend to be a keyboard and "type" a randomly generated password in
so it works with ordinary browsers
but provides hardware storage
but I may have mis-remembered that
@RоryMcCune But how does the server know that's supposed to be your password?
@Iszi you use the yubikey during registration
to set the password
again IIRC, it's been a while since I read the docs
@RоryMcCune And it never changes? Aren't we just talking about a different form of password manager now?
17:53
@RоryMcCune I agree wholeheartedly
@Iszi yar hardware backed one :)
That's another way of looking at it
@Iszi yubico.com/products/services-software/personalization-tools/… so it's only one part of the keys operation and they don't recommend it if proper 2FA is available
@RоryMcCune Of course.
But when we start talking about proper 2FA, then we're back to requiring server-side support.
@Iszi aye, more people are getting there, but still some notable absences...
18:00
@RоryMcCune I'd call the latter part of that sentence a gross understatement.
@Iszi well yeah
jrg
jrg
@RоryMcCune @Iszi So you can abuse a Yubikey as a poor mans 2FA - password you type, plus the static stored password on the key.
I do that for a few different places (Root passwords, mostly)
@jrg But that's not 2FA. Both factors are still "something you know".
@Iszi well you could argue you don't know what the yubikey "types" if it's starred out :op
jrg
jrg
@Iszi hence "poor mans". One is a hardware token... makes the password database useless if stolen, and vice versa.
It's hardly ideal.
18:02
@RоryMcCune Then we start to approach the question of "When is 'something you are/something you have' equivalent to 'something you know'?".
Here, I think it comes damn close.
jrg
jrg
So the new FIDO stuff that Yubico is supporting is pretty interesting... it actually does a handshake.
@Iszi true true, he did say "poor mans" :op
@RоryMcCune If you don't mind me asking, what kind of security testing do you do?
Bummer that YubiKey's top-end device has neither NFC nor FIPS certification.
@MarkHulkalo mainly web app. , some inf. I work remotely so internals/database stuff etc comes up for me less often
jrg
jrg
18:04
@Iszi yubico.com/products/yubikey-hardware - NFC is on some.
Just the NEO full-size, I guess.
@jrg Yeah, but not on the "YubiKey 4" series, which is supposed to be their latest and greatest.
jrg
jrg
@Iszi Yeah, I think they are trying to move towards the consumer market with that.
And the only one that's FIPS certified is the "Standard" (or its Nano variant), which can only be bought in bulk.
jrg
jrg
Well, now it can only be bought in bulk. Wasn't always.
A few months ago I picked up one from Amazon.
@RоryMcCune Inf?
18:07
@MarkHulkalo infrastructure, so traditional network testing where you start with port scanning /Vulnerability assessment and move on from there
Ah
I mainly try things like hacking myself
Setting up systems and trying to secure/break them
I rarely get the opportunity to test it on someone else.
you should look at CTFs
if you don't already
I live by the creed, "do unto others as you would unto yourself."
jrg
jrg
@RоryMcCune +1 to the CTF recommendation.
Capture the Flag?
18:10
@MarkHulkalo yep there's loads of competitions which are essentially hacking and puzzles
Ah, nice
you can compete either against the computer or in teams against other teams
@MarkHulkalo Ok. Remind me to stay far away from you.
@Iszi rofl
No, not like that. Not at all... I mean, hacking myself is learning stuff... hacking others is wrong, generally. Unless they're haxoring you first, then it's fair game.
000
000
You won't learn much from "hacking" yourself.
18:12
Not necessarily true
000
000
You're not necessarily true.
@MarkHulkalo Very obtusely reminded me of a bash.org quote that I can't seem to find right now.
Something to do with baiting the master?
@TildalWave Awwwwwh yuh, ATHF all the way.
@DavidFreitag I don't know if I want to use UE4
18:15
@MarkHulkalo Why?
Greedy, I guess.
000
000
OVID
@AviD
OVID AVID OVID I CAN'T ACTUALLY CHANGE MY DISPLAY NAME HULP
No. Something on the lines of:

1 - What would you do if the world was going to end in 5 minutes.
2 - I'd screw anything that moves! What would you do?
1 - I'd stand very, very still.
@MarkHulkalo Really?
Splitting the project cash 50/50 for example. It would be 50/50 of what we have left after giving to Epic.
18:16
@AviD Wait, haveibeenpwned.com sounds a bit suspicious? Does that not just give them free usernames to try out?
No, mainly I don't know if we would need it for a 2D game
@MarkHulkalo Epic only takes 5% bro
@Alizter Nothing they don't already have. Data is pulled from public sources.
And that's only after we hit the $3k/4mo mark
Yeah, good point
18:17
That's only $600/yr
I'd love to do an open world game like skyrim, but that isn't too feasible
Would need a team for it
I only program, can't really make models/textures, can you?
Not really, lol
The marketplace doesn't have the kind of assets I'd want for that kind of game :<
@Alizter also the guy that runs it is Troy Hunt, a very well known security guy.
@AviD A very awesome security guy.
18:22
indeed.
000
000
OVID MY DISPLAY NAME
is anybody else keep getting logged out of the site periodically? like, all day long....
000
000
n
No, but the chat keeps acting weird
@AviD Did you accidently ban yourself? Again?
jrg
jrg
18:24
Looks like level3 is having some issues...
@AviD I hear @Simon has some problem with his display name, but it may be unrelated.
@MarkHulkalo thats not the chat, thats just simon.
@ThomasPornin lol
000
000
It may be.
18:31
still not sure at what point you started thinking this was my problem.
000
000
To be fair, whose problem could it be then?
Mine? pls
@MarkHulkalo Graphene is fucking cool.
Yeah man
I can't wait to see things built out of this
Formula 1 cars made out of carbon nano tubes
18:48
@MarkHulkalo Is that really a breakthrough tho? They've basically discovered that you can get a square meter sheet of copper foil that does the job for $1 instead of $115. That's just the substrate, chemical vapor deposition has other costs and it's hardly a clean method for large scale production. Exfoliation with basically sellotape is still cheaper and much cleaner.
I honestly don't know enough about it to give you an informed response
I just see some sciency stuff progressing and go, "yeah!"
000
000
You probably do the same when you drive past a McDonald's too, right?
Silly Americans.
They also give a pretty bad example what graphene is good for. Molybdenum disulfide is much better for desalinization, and now there's even cheaper and faster method employing electric shockwave
McDonalds is pretty nasty. I rarely eat there. :p
@TildalWave I understood some of those words.
@MarkHulkalo heh fair enough :)
18:59
@DavidFreitag If we can prove we can make a decent game, we might be able to convince others to fund a bigger, badder adventure.

« first day (1799 days earlier)      last day (3379 days later) »