The Side Channel

Q-Club

02:37

Hello bash. youtu.be/3_AOeaGWwpM

forest

04:50

Anyone know how to do \underset and \overset at the same time in MathJax?

I know some symbols like \sum natively support that, but not all do.

In particular, /bigcirc seems not to want to support it correctly.

Ella Rose

05:13

@forest this forum seems to have an answer

the example there is \underset{\sigma_1 \sigma_2}{\overset{\tau_1 \tau_2}{V_{i,j}}}

forest

06:08

Ah thanks, that worked perfectly!

W[K]=\biggl(\underset{1}{\overset{r=R}{\bigcirc}}\rho[K^r]\biggr)\circ \sigma[K^0]

Using that to represent the Whirlpool block cipher.

e-sushi

09:13

Might be interesting to some of you. 🖖twitter.com/originalesushi/status/1087999301404540938

forest

@e-sushi Doesn't it also use no salt or something like that?

e-sushi

Oh, they use "a kind of" salt. It's more of an RNG problem…

forest

Ah. Yeah it seems pretty horrific.

I've never recommended 7zip for serious encryption, thankfully.

Though mostly because it extracts to unencrypted temporary files.

e-sushi

"… it uses PID and time(null) as seed. Yes, this is 7zip's random generation code running on your computers in 2019…"

"Now that was just the Init() of RandomGenerator which happens once, the next random values are generated as shown here. Yep, it mixes in a totally random hardcoded salt to previous SHA256 state. Then returns up to 16 "random bytes". Doesn't check that 'size' argument < 16"

"… AES256-CBC but supplied with IVs half of which are zeroes, and the other generated by a toy RNG. That's just from a 30min review."

TL;DR: 7zip's crypto implementation sucks big time.

forest

Yeah it looks nasty.

e-sushi

09:26

Anyway, just wanted to drop the hint as some do recommend it. Not a good idea if you do. If you don't, great! 🤗🖖

Meta Crypto SE

13:32

0

Q: Community Promotion Ads — 2019

2019 is here! And with the new year, as usual, comes a new iteration of Community Promotion Ads! Let’s refresh these for the coming year :) What are Community Promotion Ads? Community Promotion Ads are community-vetted advertisements that will show up on the main site, in the right sidebar. The...

Future Security

15:08

Gross. Any idea whether it's safe to use it for unencrypted archives or opening (not creating) encrypted ones?

Future Security

18:28

I've mad fun of how inconsistent PHP's standard library was. (If you can even call it a library and not a pile of functions.) But I've neglected to point out some of the absurdities in the C standard library. Like time(*time_t).

"Should the result go in the return value or should we set the result using a non-const pointer?"

"Why not do both?"

That's not too bad. It's actually somewhat convenient in other contexts. (In OOP with references in, same reference out.)

But the combination of ambiguous pointer use, inconsistency, and readability is nearly as messy as PHP. Why t = time(0); or time(ptr)? Why not t = time(); or *ptr = time();?

kelalaka

18:45

Uh, I gave answer and question gone. I think the reason is when he saw the simple answer

Future Security

New user?

kelalaka

Nope. 3 hundred arounf

Ella Rose

Community opinion is requested in this meta q: crypto.meta.stackexchange.com/questions/1169/…

kelalaka

Anyway I've read the linked article a little and followed two see the answer

@EllaRose Just keep it as a piece of evidence for how one answer can go deeper? :P

Or, for people wants to get rid of the downvote tag, They can target it :P

Future Security

@EllaRose Is that something that happens often? If it's rare then it doesn't need addressing. The score makes it very clear that the answer is wrong or at least suspect.

Ella Rose

18:59

@kelalaka I cannot parse the meaning of your statements

The issue I have with the stance on that answer is that, as-is, only the actions of the top 10 users will be accounted for

which is, in my opinion, not fair to the rest of the community

kelalaka

@EllaRose Yes. badly written. Just having fun.

Ella Rose

But please, you don't need to respond to it here; Whatever your feelings are, post them as responses to the meta Q

Future Security

Ah. I remember that answer now. Maybe it's worth deleting for almost being spam.

kelalaka

I'll. don't worry.

can an answer be deleted permanently?

Ella Rose

What do you mean by deleted permanently?

SEJPM

19:04

@kelalaka SE devs can. Whether mods can, maybe if we destroy users

(but destroying is only available for low-rep users)

kelalaka

destroy user?

SEJPM

otherwise 3 20k+ users can come together to soft-delete so that only 10k users can see it

(or one mod)

also I think mods can prevent a post from being undeleted by 20k users (probably using locking)

@kelalaka it's a more drastic variant of deleting a user

I think it removes the account from the system and deletes all their posts from the database

and is supposed to be used for spammer accounts

kelalaka

So at least I'm safe, for now :)

SEJPM

I think destroying is only available to mods for <400 rep users

kelalaka

What is A in the quoted part of this A themselves

Future Security

19:09

A = answer

The thing is... we don't normally delete answers just because they're factually wrong. Regardless of how many down or (undeserved) up votes it gets.

Ella Rose

But if a large number of users request deletion via flags?

The current policy in place means that even if every single user on the site, except the top 10, flag as low-quality and recommend deletion, they will be ignored

SEJPM

this particular answer has 4 VLQ flags, 2 asking for deletion, and a spam flag

(and the primary intent of VLQ is indeed to ask for deletion)

Ella Rose

I appreciate our top users immensely, but I don't feel theirs is the only opinion worth acting on

kelalaka

is the policy coded?

Ella Rose

you'll have to ask @SEJPM that

SEJPM

19:20

13

A: How should conspiracy answers be treated?

Let such answers stay (at the discretion of the poster) if they're not blatantly not answering the question. That is, if an answer even partially answers the question, exclusively using conspiracy theories, it shall stay and be left to down-votes. Not-An-Answer and similar flags on these answe...

that's the reason I see why I won't outright delete the A

Ella Rose

How is the answer in question relying on conspiracy theory?

Meta Crypto SE

1

Q: What should we do about this poor quality answer?

There is an extremely low-quality answer on the question "Largest integer factored by Shor's algorithm" This answer is clearly poorly received by the community: It sits at -9 score and has received many flags indicating it is of low quality and requesting deletion. The decision of the moderator...

discussion

SEJPM

this comment has it all:

Please note that as per our current rules outlined on meta, this answer does answer the question "what is the currently largest integer factored by Shor's algorithm" even though it's based on false facts. It will thus not be removed by moderators. Any flags asking for this will be declined. (Also a simulated quantum computer is still a quantum computer based on duck-typing). — SEJPM ♦ Dec 17 '18 at 13:49

I still stand by this statement

kelalaka

I think this user once is backed by a mod.

Ella Rose

"false facts" (a.k.a. simply being being wrong) is not the same thing as a "conspiracy theory"

SEJPM

19:24

@Maeher you did the community promotion ads for CRYPTO and EUROCRYPT last year, would you mind doing them again for this year?

@kelalaka personally I don't look at the user raising the flag but only at the flagged content when judging flags

(and I most likely handled some of these 1500 flags)

Future Security

In this specific case the action I'd take would be to edit out the vimeo link.

kelalaka

My point is user has only editing points but has marshall flag. I'm happy with my Mods.

Ella Rose

@FutureSecurity I'm not sure that using the edit feature to prune content from a post is a good thing to do. Third party edits are supposed to be used for grammatical changes and whatnot, not modification of content. I believe that there is an exception when the post is spam/abusive, but that is coupled with deletion

SEJPM

@EllaRose spam / abusive should be handled with the appropriate flags

(which will have the community user hide these contents)

Future Security

The reason I would use an edit would be if the linked video was a scam/spam or if it gave some shady people SEO thingies.

Ella Rose

19:31

@FutureSecurity Right, and in that case it's better to flag as spam/abusive and let the community user handle it properly

SEJPM

also there are steep penalties associated with getting one of your posts successfully flagged (and approved) as spam / abusive

220

A: What are the “spam” and “rude or abusive” (offensive) flags, and how do they work?

Note for moderators: Some sites, including Stack Overflow, have more permissive, overriding guidance on the handling of these flags. What makes something spam and when should I flag it? A post should be marked as spam only if it advertises a product, service, or similar and is unsolicited or la...

100 rep loss + lock on the post so it can't be undeleted by non-mods

Future Security

That's not steep. If someone spams it will be on a throwaway account. Losing 100 reputation points on an account with 1 reputation...

But in this case it's not the answer that's spammy. I don't think the author intended to promote a sketchy business. It's the destination of the link that's spam/scam.

SEJPM

@FutureSecurity well, it's steeper than the punishment that you get when a mod deletes the post or the post is edited which would be "none" or "a minor question / answer block"

crypto.stackexchange.com/users?tab=moderators

Future Security

You may notice I haven't edited the question or flagged it. (Today, at least. Not sure if I flagged it back when it was posted.) I'm not planning to take any action. Just stating what I think the best action would be if we could ignore policy.

On some websites I've seen naive users copy and paste text that contains sensitive private information. (Like there street address.) Or upload a screenshot that someone could use to hack them or do identity theft. Or a geo-tagged photo.

Mods on other sites have the power to censor that information. What they might do is replace the private part with asterisks and add a "Moderator note" to the bottom of the post. They try to leave as much of the original post they can.

That's really light moderation that doesn't fit into the only fix grammar and formatting idea. Benign censorship I guess.

Similar scenario. Sometimes someone freaks out because they see spam on their website, they try to get help, they copy and paste the spam for others to see, but they don't think to obfuscate or censor the URL. That's bad, of course, because it adds spam to the website they're posting on, it lets spammers get their link on a high reputation website, and they get more eyes on their spam.

Either case, edit and add a note (in line) or edit and comment. The former case isn't something the average user could do because the info remains in the change log. The second it matters a lot less that it's in the log. Removing it from the top level page is good enough to prevent people from accidentally clicking or curiously clicking. (Plus maybe the edit log isn't indexed by search engines.)

So. Ya. I'm not flagging it for very low quality or for spam. And in what-if world I probably wouldn't delete it either.

Regarding the fairness of the policy (that the top 10 are the only people with deciding power), that I too think is an unfair policy. But there are a lot of unfair policies in the world. (Written, spoken, or unofficial) That specific policy on this website just isn't high on my list. (Or any policy for this site.) I assume it would take to much work to fix it with too little benefit, so it's low priority to me.

kelalaka

20:12

This is one of the examples hacks meta.stackexchange.com/questions/156090/… that mod act whenever they see

SEJPM

@FutureSecurity we can also do that

mods can "redact" posts (which needs confirmation by a co-mod) that will remove something even from the edit history

but the guideline there is to restrict this only to things like actual private keys or other stuff

(and I only had to use this once or twice so far)

Future Security

Ah. I should have thought of the private key example.

Of course still tell them to generate a new key. Chances are too high that they probably won't but it is worth trying and it is worth redacting.

Maeher

20:57

@SEJPM I'll see what my gimp skills will allow me to do on Friday.

Future Security

So besides code smell, it looks like the 7z RNG is used in 3 places. To generate IVs, to generate guids (no idea yet what file format its specific to), and to generate salts. @forest @e-sushi

Meta Crypto SE

21:14

1

Q: Adding a Katz-Lindell Tag or Book-Question Tag or None

As many of you already know, questions from the book of Katz-Lindell, Introduction to Modern Cryptography, asked many times in our site. I want to create a tag for these questions as tag:Katz-Lindell to unite them. I've searched one to find since I remember that I've seen it. With the tag, the d...

discussion

b degnan

Does anyone know if there's any work in cryptography over isochronous curves? They have some interesting convergence in the same way that ECC does. Zeroth order, I couldn't find anything; however, it might just be due to lingo. There are some interesting planar properties when you expand to a field with an impulse.

Future Security

21:35

Probably not if you mean work with promising results. If you mean any type of work, then most likely. Lots of people become an expert in a single weekend. Usually inventing some encryption algorithm based on cool mathy stuff.

b degnan

I'm working on something that is completely unrelated. If you see enough math, you start seeing coincidences.

Future Security

Does a discrete version of such a curve even make sense? We can't use continuous math.

I really don't know any of the details behind what makes something difficult or not. Just what some basic EC, DLP, and RSA operations are possible. Plus some hash function based signatures.

Future Security

22:00

That word "difficult" is a bit misleading too. Back when I read Schneier's blog I remember reading something about people in other fields inventing cryptographic algorithms. Something along the lines of: Every scientific and mathematical field has their own hard problems. Usually they assume that no one in their field being able to solve something means that outsiders definitely haven't found an easy solution.

The problem is usually that once they try translating a problem to a computer algorithm they are proposing an algorithm that doesn't have the same "difficulty" as the original problem. It's not that cryptographers magically solve other people's field's problems overnight. Their proposal, if implementable, is just different enough to make things easier. Or they use a subset of the problem space which is a lot easier to solve. It happens with even obscure, pure math.

And Schneier's law: Anyone (regardless of how much they know) can design a cryptosystem that they themselves can't break.

Future Security