General Tag Discussion

@Exedrus  
Well, I could go through all 400+ names and check.  
But I’m not going to.

@WingbeatPony  
Yes, usefulness is subjective, I just mean in terms of searching, it’s going to give random results that won’t be specific in any way.

@Zeb  
Alright. I maybe a dev could check. DB access might make short work of this sort of question.
 
Another way to look at it: Is there a way to see how often these tags are searched? That might be a more accurate metric of usage than watchers. (It’d account for unregistered users which probably make up a reasonable amount of traffic.)

@Exedrus  
Collecting metrics on each and every search made on the site is the kind of thing I’d expect of Google and the raw computing power it has on tap. Seems like overkill, and precisely the sort of data I’d rather Derpibooru never aggregate on its users…  
The easier way might be to use existing metrics to give a rough definition for “active” users - ie, any logins within 6 months, any blip of activity on their graphs, that sort of thing - but it’s still massively overkill in finding a solution to something I don’t particularly think needs solving. Sure, it’d be nice to know, but not really pertinent to answering the question “do people use this tag,” when even a casual glance tells me the answer is a resounding “yes.”
 
 
@LightningBolt  
But there is a trend to it. People watch that trend. Therefore, they are using it, and therefore, it is useful to them. Maybe there are ways to make it more useful, but that involves getting feedback, not delivering an ivory tower sentence of “useless.”

@WingbeatPony  

 
It would also be the best way to have hard numbers to rely on.
 

 
What would be the harm of (fully-anonymous) usage data?
 

 
I disagree. Last time I asked on IRC, I was told that the number of active users was on the order of thousands. That’s of the +200k registered accounts. That doesn’t paint a compelling picture that a few hundred random accounts would be active. The number of watchers could be completely misleading (especially given that there are undoubtedly many users who don’t register).

@Exedrus  
I admit, I’m historically bad at making a point, so let me try explaining what I meant.  
When I said gathering that data would be overkill, I was thinking about the overhead in terms of dev time, server load, and sheer quantity of data to be processed. Having been on the other side of this, I know how a simple suggestion of “what if we had feature x” can translate into weeks of work. It’d be months before search data could even start being collected, and ages before it added up to something worthwhile - after all, there’s no way to retroactively gather such info.
 
As for the matter of collecting searches in the first place, that’s a matter of ethics. There’s a reason staff can’t peruse your PMs, look at what you’re searching, or talk openly about the content of reports - they respect your privacy, and I think they’d agree that even “anonymous” search history isn’t as anonymous as we’d like to think it is.
 
Lastly, I confess that last line to be hyperbole, but it was intentional - if it’s being actively tagged, if it’s got as many accounts watching it as a typical amateur artist’s tag, then I think it’s detrimental to think about the tag as being dormant, irrelevant, or any other dismissive attitude. Particularly if it sounds like someone’s already made up their mind about what to do on top of that.
 
So what if it turns out the accounts watching the tag are largely inactive? They still watched it, didn’t they? Does the score on an image not matter anymore if the users who upvoted it have long since left the site?

@Exedrus  
Well, if it helps give a clearer picture. I just counted 25 separate people adding the tag to images within the last 10 days.
 
But if you want something more accurate:
 
K. I just checked 212 of the watchers on the “sexy” tag. Of those 212, 149 of them have logged in within the last month. The vast majority of those were within the last week.
 
Would you like me to count the rest?

@WingbeatPony  

 
Perhaps, but since neither of us know for sure, why assume the worst?
 

 
It’s as anonymous as they design it to be? It could be as simple as just keeping a counter for each tag. That seems pretty hard to deanonymize.
 

 
Why would you cater to users who no longer come here? They no longer use the tagging system. How tags are used could’ve radically changed since they were active.
 
@Zeb  
Alright so there are around 70%. Those watchers are probably a pretty reasonable absolute estimate then.
 

No. I honestly didn’t think you’d count them in the first place but thanks for looking into it! Now we know that that estimate is reasonable.

Could I get suggestions on the names for the magical geodes? So far I’ve got AJ’s (Super strength,) Rarity’s (Shielding,) Rainbow’s (Super speed), and Sunset’s (Empathy.) For Flutters I was thinking Geode of Nature/Fauna. Fauna would be more specific because she can’t talk to plants, right? Twilight’s could be Telekinesis I guess? I had Magic as a possibility but “magical geode of magic” sounds redundant. Tuypo1 suggested “explosions” for Pinkie’s, but I’m not sure how to phrase that and imply that it’s related to baking and also not have it be a mouthful.

Do we add the “parent:__” tags for adopted?

Spoilered for Season 7: Episode 13.
 
||Alias Grand Pear into Old Grand Pear ||

@Binkyt11  
Shouldn’t it be the other way around? The credits list him as Grand Pear >>1467545 and I don’t recall him being called “Old Grand Pear” in the episode itself, when he first introduces himself it’s definitely just “Grand Pear”.

@Binkyt11  

 
Why not grandpa pear? ^^

@Mad Black  
Grand Pear is a pun on the French “grand-père” which means “grandfather” or “grandpa”.

@Mad Black  
I just realized that with the few uses it has… maybe It just refers to the current state he’’s in?… I don’t know, but if so than that i’s just unnecessary. Looking back on my post maybe it should be vice versa.

@Ereiam  
I was very disappointed he didn’t have a heavy Cajun accent, I’d been hoping that was the direction they’d take since we first learned his name.

Despite it being aliased, pinkie pies still shows up in the tag list with -1 images.

@Pacific Glow  
Don’t believe I can fix that unless I un-alias the tag, make it re-count itself. Then alias it back in.
 
And I’m not doin that on tags over 3,000 unless I have too.

Is there, by any chance, a tag that denotes someone located/sitting on the character’s head?  
I’m talking about  
>>1467146  
>>1463975  
>>1460436  
>>882056  
>>1462634 (merged)  
>>1089985  
>>1120590  
>>1214765  
>>1274739  
>>1278777  
>>1325199  
>>1400674  
>>1410657  
>>1418237 (.gif size warning)  
>>1418944  
>>1461493  
kind of pictures.

@Deliverer  
Apparently sitting on head Though it’s only tagged on 2 images so far. So.. Feel free to add to it! =)

@Zeb  
Thanks!

I’ve encountered a conundrum.
 
Specifically, three tags: war hammer, warmallet and maul.  
I am currently tagging “warmallet” on the images already tagged “war hammer” that have stereotypical large-chunk-of-metal-on-a-stick warhammers as opposed to combat-pick real ones. However, there exist images where there are literal warmallets, as in, large wooden mallets that are suited for knocking on enemies and not just wood, and as such the tag might exist alone. “Maul” also kinda exists, and might describe the stereotypical warhammers better.  
War hammer, specifically polehammer:  
 
Warmallet:  
>>1194733p (deleted)  
Maul:  
 
Stereotypical warhammer (note how similar it looks to a maul or a sledgehammer):  

 
Should the images with stereotypical warhammers be (re)tagged “maul” with “warmallet” being reserved for actual mallets and “war hammer” for actual hammers?

@Zeb  
Looks like the aliasing didn’t work. It’s been a week and a half and I’ve still got two separate tags.

@Ereiam  
My fault.
 
I had left it to wait a bit and see if anyone felt it shouldn’t be done.
 
Is done now.

@Zeb  
Great, thanks!