@Makyen Since you're around, but whenever you have the time, could you look at this? You made this watch herbal[\W_]*+native(?:[\W_]*+(?:oil|cbd|\d++|[\da-f]{4,}+))* which catches 6 TP. There's a recent post with grown instead of native, i.e. herbal-grown-cbd-oil. As usual, I expect a couple more of these so I was going to watch it with the usual pattern.
I'm wondering if it makes sense to merge them, so something like herbal[\W_]*+(?:grown|native)(?:[\W_]*+(?:cbd|oil|\d++|[\da-f]{5,}+)s?)*, which catches the previous 6 and the new one. Does this make sense, or should I just watch the new grown pattern separately? Also, the TP on the previous watch are all a year old, and I don't know if that's something to be taken into account as well. There's no FP on any of these watches of course, merged or otherwise.
@Yatin It's a reasonable answer though, and the link is legit. It's link-only, but it does address the question sort of. Are you basing this on something other than the username?
Not sure if the username is sufficient to addblu them. Maybe they thought the username was funny. Still, no harm I guess. Next post that's FP will sort it out.
@cigien I'm not Makyen, but probably try to merge everything into one pattern -- unwatch the old and watch the new, or do a PR to update the existing pattern
@Yatin Yeah, I would agree with that. Blacklisting is making a fairly strong claim, and should be backed by more evidence. btw, if you're sus about the user, I think you can get hal to notify you personally the next time that user posts. I'm not sure exactly how it works though, triplee would be able to give you clearer advice.
@cigien actually Hal only reacts on things Smokey reports, so it can't watch things on its own; a Smoke Detector watch would be the way to go if you want to get an alert when they have any activity related to posts (still no watch on comments)
two apparent FPs on Stack Overflow but those are the only hits across all of Stack Exchange so feeling that we can watch this at least for a while stackexchange.com/search?q=netrack
the FPs are on a Github organization with the same name who publish a Keras add-on
several in fact, but looks like a one-man operation actually
@SurajRao I would probably just flag that as no longer needed, but I'm unsure if self-promotion in comments under their own posts necessitates any other type of flag.
Hey, I found a user who left at least 2 answers with the same youtube link: https://stackoverflow.com/a/65733150/845414 (deleted) https://stackoverflow.com/a/65733771/845414 Still alive.
@JohannesKuhn If the youtube channel is theirs and it's relatively easy to see that, then it's considered undisclosed affiliation and an excessive amount of that is spam-flag worthy. I don't immediately see that the user owns that account, but I imagine they have a higher chance than normal of owning it considering the post date and that they're mentioning it in that context.
Well, I think it is strange that one video should be able to answer two completely different questions. Posted by a new user. To a video with 22 views.
@Yatin That pattern looks like it's already caught by Potentially bad keyword in answer and Potentially bad keyword in body; append -force if you really want to do that.
@Mast Done. Is there context for your current request on the Team which I need to take a look at, or are you just mentioning the Team as a reference to as to why you're asking an admin? I'd appreciate a link, so I don't have to go searching all of the Team to try to figure out what your mentioning the Team might be about. [I'm not at all upset, just a bit confused as to the reason for mentioning it and not wanting to go searching without knowing what I'm looking for.]
@Mast Thanks for the link. Yeah, I could see how that might be a bit confusing if you're not familiar with that general pattern of using an API. As with the SE API, the MS API requires all accessed have a key. The key identifies the application which is making the request. That key isn't considered secret, as it only identifies the application. For most applications, the key is something which is stored in the application's code, which may be publicly available.
The key can be used by the API provider to track the application's accesses or control access by the application. For example, if it's determined that a particular application has gone rogue and is making enough requests to cause disruption (e.g. the app has a bug that causes huge numbers of accesses), then disabling the key can allow the API provider to block that application until the issue is resolved by the owner of the app.
In order to associate a request with a specific user, a token is required. For the MS API, a token is necessary for all write accesses, because write accesses are done by specific accounts. All tokens are considered secret and should not be displayed or stored in a manner which would disclose them, other than to the user who is the owner of the token. There's an authorization flow which allows the user to either confirm or deny granting a token to the application.
The SE API uses tokens both to allow access to user-private data and for write operations. It has an OAuth2 request flow that requires specific indication by the application and authorization from the user for the type of access which will be granted to the application.
@Makyen Apparently there's an API key, an app key, a token and a code. I think the token and code are what we use to authenticate userscripts, the app key I got, now I just need to figure out how to feed it correctly to MS so it gives me my data.
@Mast The API key which is used by the app is the app's key. A token identifies a specific instance the the app and is associated with the user owning that specific instance of the application. I'm not sure where "code" is mentioned. It could mean either of the above depending on how it's used. If it's in our documentation, it should probably be changed such that it's clearer as to what it's referring to.
