metasmoke.posts optimize note Table does not support optimize, doing recreate + analyze instead
metasmoke.posts optimize status OK
metasmoke.feedbacks optimize note Table does not support optimize, doing recreate + analyze instead
metasmoke.feedbacks optimize status OK
I believe it does some space optimization and does magic to make indexes faster.
I believe we should see the largest changes with things like search. Or it might not have done much. We'll find out. The next thing I've been looking into is some of the config vars to make it more memory efficient.
By the way, totally unrelated note, but for a while I thought your domain was erwaysofsoftware not erwaysoftware
So, I've started work on the conflicting feedbacks review queue. But I'm not sure if I'll finish before Sunday because I've got a load in tomorrow and my hands are gonna be very ouch.
@Undo As someone who's relatively new here, it seems like gameification is a big part of it. Trying to increase your % accuracy, decrease TTD, feedback first so it shows up in chat, etc.
@thesecretmaster Could be, I suppose - but we haven't put much conscious effort into gamifying anything. Sometimes effort to not gamify things wrong, but that's about it
@thesecretmaster ah, yep. 2am UTC.
Must have been the too-much-success edge case. FPs recalc immediately.
@Undo I don't think it was intentional, but it seems to have worked out that way. You could think about the different priv levels as "goals"/"leveling up" done by playing well/giving good feedback and appropriately using the tools. And the whole thing about decreasing TTD by using autoflagging. I think it's gameified in a good way, the same way the rest of SE is.
I could also be wrong -- I'm pretty new here, and that was just one of the things I thought about when I came. I don't mean to generalize or insult/minimize the awesome work that y'all do.
@user2476549 Commit messages don't need to be works of art. They should communicate what the commit does. Just describe what you did, and sometimes why you did it, so that 3 years from now someone glancing at the commit message can get a feel for what the commit does, or so someone can search and find it when they are looking for when something happened. If you renamed a bunch of (unspecified) files, just say something like "Renamed a bunch of files to make things more readable."
So: I've done some talking with a few main contributors to the Great Meta Post - Bhargav, terdon, Journeyman. Responses are positive (3/3) to idea of adding an autoflag from SmokeDetector on posts, taking us from 3 to 4.
Probably going to do more talking at a larger scale, but I give it 90% that we can do this.
@quartata I've got a lighting load in tomorrow and since part of the rental includes a dimmer system, there's gonna be a boatload of cables. It also means a long day and lots of ouch for hands, especially beause I'll have to fight my least favorite cable: camlok.
CamLok is one of the reasons it must be nice to live in Europe. Apparently y'all have CEE instead of CamLok so your hands don't die when you deal with 3 phase.
@Undo I'm not done yet :) I'll be back as soon as my hands are back.
If anyone wants to talk about tech theater or hear tech theater stories, feel free to email me. I've got an email listed in my profile. Anyways, sleep.
@Makyen That pattern looks like it's already caught by Bad ns for domain in answer and Bad ns for domain in body; append -force if you really want to do that.
@quartata I don't understand the question, if something is tagged #whoisguard then it already has whois info because that's where you can see that it is registered through WhoisGuard (or Domains by Proxy or PrivacyProtect.org or what have you)
so for a typical pharma domain I would (1) tag as #drugs (2) do a whois lookup (3) see if there are any indicators in the whois data that it was registered through an anonymity service, and, if so, tag appropriately; and (4) similarly check if the name servers in the whois data are interesting, and, if so, tag appropriately
occasionally "interesting" only becomes clear after you see many similar entries but with the SQL data explorer you can now search on the whois data we have stored
so you can go back and check if many domains we have whois data for have the same, or somehow similar, information
also, searching Charcoal Test for Halflife reports for a particular name server can be revealing
so for example chat.stackexchange.com/… shows that there is no single popular NS arrangement involving todd.cs.cloudflare.com
whereas some of the other combinations are very popular in spam
which reminds me, some of the watched NS should probably now be promoted to properly blacklisted
@thesecretmaster the real criterion is "spammers who stopped" but it's really hard to establish anyone's individual effort as particularly making that happen
I think there are like 2 or 3 spam operations which no longer target Stack Exchange, quite possibly because of things we did to alert their upstream hosting etc
but they are small potatoes in terms of spam volume compared to the pharma operation
the "tech training in India" which was prevalent on Stack Overflow a couple of years ago is what got me engaged here in the first place and they are completely gone now
While Stack Overflow employs a good number of anti-spam measures, some spammers are bypassing them on a daily basis. The Smoke Detector seems to be doing a good job of identifying these for manual closing, but the spammer just keeps on hammering the site. As a regular user, what more can I do t...
@gparyani There's a posts review queue now. It should be easily accessible from the review page, which is linked, as t used to be, from the site header. At least it is for me. Does it not appear for you?
@gparyani Interesting. Both links work for me. However, the first one gives that error when I'm not logged into metasmoke. Are you sure you're logged in?
metasmoke.erwaysoftware.com/domains/4986 looks vaguely interesting, is this a separate operation? All the posts have the first word of the body as their title, and the whois registration looks different from the main ones
@Makyen That pattern looks like it's already caught by Pattern-matching website in answer, Pattern-matching website in body, Potentially bad keyword in answer, and Potentially bad keyword in body; append -force if you really want to do that.
It's not critical to do so, but it is nice to have a "standard" format. OTOH, if phone numbers are going to be moved into their own code, then there isn't much point in going through old ones.
@tripleee Yeah, that's going to require searching in metasmoke, or at least testing vs. verification that it's a valid phone number. Although that gets tough when you have to consider worldwide.
Well, it shouldn't hurt much if we convert everything that might be a US phone number into that format. There shouldn't be that many things that are caught in addition to the verbatim copy.
[ SmokeDetector | MS ] Bad NS for domain in body, bad keyword in body, bad pattern in URL body, link at end of body, pattern-matching website in body, +3 more: Thermo Burn is really works? by nandsouza on askubuntu.com
> Pattern-matching website in body, pattern-matching website in title ---------- Title - Position 12-26: phonenumber.co Body - Position 12-26: phonenumber.co
I think the room description does not allow formatting of links.
but let's try
room topic changed to Charcoal HQ: This site is awesome. [best-bad-practices] [corn-for-horse] [crazed-maniacs] [dev-on-prod] [everything-is-broken] [failure-as-a-service] [panic-driven-development] [plastic-knives] [programmatic-crowbars] [roadkill-pasta]
