Donmai

Danbooru 2 Issues Topic

Posted under General

This topic has been locked.

albert said:

So I should announce that I've been adding more user tracking on the site so I can have better metrics about how people use Danbooru. It's opt-in by default but if you don't want your search habits being sent anywhere, you can go to your user preferences and enable "Opt Out Mixpanel".

It's alright, I don't mind if you know about my fetish for masculine women and homosexual warships. I'm sure all that information is sitting on a government hard drive anyway.

ToastedWaffles said:

Or users could just, you know, use uBlock Origin or some other competent ad/content blocker and have it automatically nerf your metrics tracking. Sorry, but I disapprove of "opt-in by default". Disappointed that you'd sneak this in without giving us a heads up before doing so.

Didn't he just gave us a heads up about it now, otherwise, you would have not known and complained about it. He's nice enough to give us an opt out option, unlike many other sites.

montesat said:

Didn't he just gave us a heads up about it now, otherwise, you would have not known and complained about it. He's nice enough to give us an opt out option, unlike many other sites.

True, it's nice to know that he's running this sort of stuff and offering a opt-out option. But to me the heads-up is a bit too late. Would have been better to say "hey, next patch update of Danbooru I'm gonna introduce this option..." rather than "oh by the way, I've been adding more user tracking that you probably haven't noticed because I want more metrics".

Apart from search queries that go to the Booru servers that I can't do jack all about since they are required for searching the database by character/series/whatever, I've got uBlock Origin that blocks the client-side scripts from loading so I don't have some script tracking my activities.

I'm not complaining, I know albert wouldn't be doing this for malicious purposes. Or is he? jk

I'm trying to run a Python script I wrote 4 years ago to scrape pools, and it keeps giving me 403 errors. Included in the text is "The owner of this website (danbooru.donmai.us) has banned your access based on your browser's signature". While it is reasonable to block access to urllib2 to the main site, is there a way around this to actually use the API?

I don't see anything about this on the API help page.

albert said:

That's probably Cloudflare banning you for not throttling yourself. The simple answer is to rate limit yourself.

It's an old script so I don't remember exactly what it's doing, but I'm pretty sure it is set to rate limit (no more than 2 requests/second) all requests, and I think it was 403'ing on the very first request.

I can at least wait a day (is that long enough?) and try again with it printing every url it accesses to see for certain if I'm able to get even 1 request in.

bigrich said:

It's alright, I don't mind if you know about my fetish for masculine women and homosexual warships. I'm sure all that information is sitting on a government hard drive anyway.

Those are relatively normal so I'm not surprised you don't mind. If the guys that run this site know I like big tiddy preggo girls getting milked that's fine by me. If they look too deeply in to what I like though I'm scared for their mental health

I have a war against personalization, and approve the GDPR etc. etc. ( including Google Analytics, which I removed on all my sites once I realised that one can be seamlessly followed from website to website ad infinitum by one's friendly local Stasi for all time ), but certainly have no problem with analytics nor with being opted in by default on any individually independent site such as this.

If Danbooru belonged to Disney, it would be different. Or any evil-by-default corporation, including Amazon or a newspaper conglomerate. But this place is like home, and security cameras don't matter at home.

albert said:

I've fixed the CORS headers. It looks like some of the webms are missing, I'll look to see if I can restore those.

Does the fix take a while to propagate? I'm still getting the same error as above.