Posted under General
Implicating
all to animal_costume.
Reason: They are all people in animal costumes.
There are also some other, more obscure animal costume-related tags, but I'm not sure if they're worth implicating.
Updated by jxh2154
I say they are, because they're all that much less likely to get noticed otherwise.
Put them in the animal costume wiki, they get noticed. It has been policy to not implicate things with less than 15 hits for well over a year now, if not two.
The fact that the implication table times out on a regular basis shows me that this was a good policy, as even with it in place the implication table has grown quite large.
How do you know the table times out?
I've gone-ahead and created the wiki, also referencing some of the obscure ones, seeing as there's no reason to exclude them if we'd be implicating them to animal_costume anyway.
Updated
Log said: The fact that the implication table times out on a regular basis shows me that this was a good policy, as even with it in place the implication table has grown quite large.
That's sort of the same logic as saying that we shouldn't put a post into too many pools because it'd clutter-up the UI.
We shouldn't be forced to have less data because of system implementation issues.
Nothing is stopping you from adding a tag manually, it's not forcing you to have less data. If the implication table continues to time out the implications can't be entered and it will have to be done manually forever anyways.
Bapabooiee said:
That's sort of the same logic as saying that we shouldn't put a post into too many pools because it'd clutter-up the UI.
This is true anyways. Have you ever been reading a comic when out of the blue you have a post in 7 pools? It's cluttered as fuck and it's really annoying to find the pool you were navigating again.
Log said:
If the implication table continues to time out the implications can't be entered and it will have to be done manually forever anyways.
Again, what implication table is timing out?
If you can't find it I'm not sure you're qualified to be a mod.
http://danbooru.me/tag_implication unless Mods+ have something different.
Despite jxh2154 mentioning it bombing out on him, though, I haven't noticed it fail until just now. Not that I am in there often, mind you, but I have been a few times in the last few days.
Are we sure these failures are specific to the pages that are failing and not a general slowing down of everything lately?
If the system imposes an upper bound on the number of implications we can have, will we soon reach an upper bound on the number of tags? posts? users? Maybe we already have?
Something sounds wrong with that logic, from a design standpoint at least.
sorry for off topic?
I've been getting "fail" and "time outs" trying to use the searches like the Implication and Tags page.
That's what I was mentioning. Regular user profiles are mostly timing out again too. Maybe the user statistics are killing things again.
It's sort of bad though because not having the user statistics makes moderating users difficult, being unable to access their profiles makes it impossible.
I've been trying to get to it for at least 6 days now to no avail, it was initially to look an implication up but I don't even remember what it was now so I've just been checking to see if it's dead or just a one time thing. It was dead when albert had all the stats off so it's more than likely a size thing rather than something to do with counts.
It's been working again for me since albert did the stats change, though I need a few refreshings sometimes.
Anyway I might be clueless about the db but since we all saw the whole site suddenly slowing down and going failbooru everywhere at the same time, it's unlikely that the db swelled up this much in such a short time to be the cause.
Bapabooiee said:
That's sort of the same logic as saying that we shouldn't put a post into too many pools because it'd clutter-up the UI.We shouldn't be forced to have less data because of system implementation issues.
Um... creating an implication does not increase the amount of data posts have; it decreases it, since implications form restrictions on the possibilities of tags that a post can have. If you implicate A to B and there were any posts that would have been found by a search for a -b you just lost them.
0xCCBA696 said:
If you implicate A to B and there were any posts that would have been found by a search for a -b you just lost them.
What's amusing is that that search actually produces results.
Log said:
If you can't find it I'm not sure you're qualified to be a mod.
Æh, it's very nice you're so snide and all, but for me "table" means the DB table. So if you could magically discern when it times out, you'd have magical superpowers. And I agree with the sentiment of "just because it's broken right now shouldn't mean we don't do the right thing". I'm very annoyed by the speed at which people start making up bad hacks instead of fixing issues properly.
I think the point about numbers was meant more for aliases than implications. Or maybe the idea came from me being presented with a potentially ginormous request of object -> object_class implications one day and asking to be spared and just given the major ones.
And yes, I'm a broken record now, but the implication page is indeed completely unreachable no matter how many times I refresh so I can't do these or the half dozen others I'm waiting on yet. Aliases, however, load fine, and I thought they'd be bigger but who knows.
Does anyone know of a way to, like... submit implications without going to the page? Some kinda funky URL string or something I could do to send the same request without loading that page?
jxh2154 said:
Does anyone know of a way to, like... submit implications without going to the page?
Maybe it's possible through some magical, undocumented feature in the API?
If it's not possible (which looks like the case), that would make for a pretty good feature request, I'd say.
Bapabooiee said: Maybe it's possible through some magical, undocumented feature in the API?
Maybe. Though honestly I don't know much about how the API works and haven't used it as such before.
