FR: A command to check for a previous occurrence of a link in Metasmoke. For example, !!/history example.com could output 8 previous reports: 6 tp, 1 fp, 1 naa. Could also be extended to more content, such as !!/history yet-another-muscle-supplement
@ArtOfCode looking through ChatExchange trying to reproduce it in Node.js ... I have it authenticating w/ SE but I am not following how the fkey or cookies are being applied to other SE sites ... you got a min?
@ArtOfCode ok, so once you have obtained fkey and or cookie ... how is that being used against chat.stackeschange.com ... it looks like your going through http://%s/users/login but I cant seem to get that to work
@CaffeineAddiction Looks to me like ChatExchange (a) logs into Stack Exchange's OpenID provider, openid.stackexchange.com (b) logs into a Stack Exchange site
Check for the presence of the usr cookie as a way of making sure login to openid.stackexchange.com has succeeded. Apart from that, you can discard all cookies and fkeys from openid.
yep, i have the usr cookie and the fkey ... and the openid site has something that says """If you want to use your own URL to log in, add the following two tags to the head of the HTML:"""
but once I have the fkey how do I use it on a diff subdomain of SE
is it like chat.stackexchange.com/usr/login/<fkey>
oauth_version: "",
oauth_server: "",
oauth_identifier: "https://openid.stackexchange.com",
fkey: "the value you stored from the last GET request you did"
What other operations are there that privileged users can perform that result in data being written to metasmoke? That's other than feedback and reports.
Who're our good Pythoneers? @Andy, @Undo, @hichris perhaps? Can someone implement the Smokey end of the MS API report post route? Essentially, a message comes down the websocket containing in the format of {report: {user: "username", post_link: "link"}}, and we just need to put that data through the same process as the chat-report function. I haven't been able to do it because circular imports played havoc with it.
I had a terrible time getting GitHub to believe I was in school so I could have the student pack. Finally had to get someone from Google to send them a nice email.
I have a private repo under my own account that I need to pull down to EC2, but Smokey's logged in on EC2 and I can't figure out how to log out or invalidate the credential cache or whatever (and that would break Smokey, anyway). Solution: give Smokey access to it, so that I can pull it while Smokey's logged in.
@Undo it has potential but Atom doesnt have full multi-cursor support that sublime does ... close but not quite. And I use the multi-cursor in sublime way way too much to give it up.
Atom is an awesome text editor and on-the-fly small-changes dev environment. It's not so great for big changes.
'Specially with Rails. Comparing my development with Atom to that with RubyMine... the number of bugs that RM has caught for me today that would have turned into debugging is quite large.
While announcing the second iteration of the Stack Exchange quality project, I not-so-briefly alluded to a collaboration we're kicking off with the University Of Melbourne.
The project presents some very interesting possibilities for us if their model validates as well as is hoped:
Knowing ve...
Dang. I don't have close vote privs on any of those sites
> Our results suggest that we can increase the number of duplicates by around 45%, by annotating only 0.0003% of all the question pairs in the data set.
Bold claim. Let's see how the rest of the paper is.
Lol, the user puts as the Face icon an advertisement to the thing they are spamming! stackoverflow.com/users/7076726/mabel-gerald?tab=profile That way even if you delete their spam answer, there's a user in the system with a Quickbooks icon as the profile image.
Spammy profile icons can't necessarily be caught by the system because that would require image recognition.
Well I guess if the profile is an unused one, its image will be minimally received. However, it's the principle of the matter because a spammer left an artifact of spam living on the stackoverflow servers.
@ArtOfCode No, and that's the problem. The spammer has won, even if by a tiny amount, in principle. They left an artifact of spam up, and we can't downvote his profile image.
So it's the problem of spammers using the 'about me' and profile aspects as the place for spam. I guess the strategy of dealing with spam there is old fashioned moderator whack-a-mole?
Yeah, those user profiles are ignored by everyone and Google doesn't even give them any Google juice for the links. Pretty pointless to search them out.
Well there's the first hole in Stackoverflow armor I've seen, thousands of profiles out there that are the definition of spam hanging out. I wonder if you could sneak in some hard core pornography though there and have it ignored. lol. data.stackexchange.com/stackoverflow/query/516380/…
@EricLeschinski Good old fashioned whack-a-mole, except it's kinda unfair because I can go through the last 48 hours of their spam and kill all the accounts in a couple minutes.
So like whack-a-mole, except with RPGs.these things.
@ArtOfCode do we call unofficial sites distributing software drivers (while using official brand name logos, etc on the site - basically something that would be considered 'scammy' or 'potentially malicious') as fitting in that category? Gotta ask because I"ve seen this one guy post two links already for "Canon Drivers" that don't come from the official source, and could be considered dangerous
(didn't catch in Smokey, but caught community attention including a couple spam flags)
they weren't deleted as spam, hence why i'm asking (a couple spam flags, but mod deletion, at two separate times, so it didn't trigger the spam system)
in this case i'll just keep an eye on it, was generally wondering thoug hwhat the guidelines were for blacklisting it :)
returns to churning through the flags at Ask Ubuntu
While announcing the second iteration of the Stack Exchange quality project, I not-so-briefly alluded to a collaboration we're kicking off with the University Of Melbourne.
The project presents some very interesting possibilities for us if their model validates as well as is hoped:
Knowing ve...
