Yoshi MCF

So in case anyone needs it:

"Do you want X or Y?" "Yes."

Hey folks! Google is fucking you via sneaky enshittification again!

Want the shit in your google drive to load instantly again, instead of taking for-fucking-ever?

Open your gdrive (web OR app) Settings > Manage Apps

Gemini was checked "use as default" (and i sure the fuck didn't set it that way, this was a silent push)

Mozilla, in its finite wisdom, embedded LLM bots into recent versions of Firefox for the vitally-important purpose of… naming tab groups. Now, some users are noticing CPU and power usage spikes caused by a background process called Inference.

Ugh. Reminder again for Firefox users to visit your about:config page, search for the browser.ml.chat.enabled key, and set that to false:

If yours says true then double-click it until it reads false.

Fair warning though, if you've been on the platform for a long time this archive is likely to be quite hefty in file size. This blog I have had for 5 years with 22k posts, and the export from tumblr came to be 48GB. My previous blog I made in 2011 and has 95k posts, so needless to say I did not use tumblr's built in export to back that one up.

If you want more control over exactly what you back up from your blog, I recommend that you use tumblr-utils instead. It allows you to backup specific tags, post types, and to ignore posts that you did not create (reblogs where you've added a comment count as a post you created, to be clear).

To use it:

download and install python 2.7

download tumblr-utils

create an application on tumblr to get an api key

extract the tumblr-utils zip and open the tumblr_backup.py file in notepad

search for "API" and paste in your OAuth consumer key

then go back to the folder where you extracted tumblr-utils and choose open folder in terminal / command prompt, or type cmd.exe in the address bar while inside the folder

now in the window that opens it should show the current path, and you can type "py -2 .\tumblr_backup.py [your options] [your blog name]" and hit enter. Example:

Backing up just my original posts from this blog with this command came to 632MB rather than 48GB, and also gave me the option to save my posts in JSON format which will be useful for converting my posts to a new format for self hosting.

On that note I'm currently looking into figuring out a simple (and ideally free) way of self hosting a static site blog that utilises activitypub, and also converting my old posts to re-host on said blog.

This post series by maho.dev on implementing activitypub with any static site is my primary source of guidance atm if you also want to try figure that out yourself, as well as having an explanation for why you'd even want to do this if you don't already know

but if tumblr goes down before I get things sorted and write up a post about it then i'll be reporting back on it via my bsky, mastodon, and toyhouse accounts

if you dont have an account on any of these I'll also be sharing an update via my personal site's RSS feed, link of which includes an explanation of what RSS is and some feed readers you can use, I highly recommend checking it out as getting a feed reader is going to be the best way you can stay connected with people if they scatter across the internet!

When social media was getting big, you could just NOT get a Facebook or a Twitter. It was simple as that. You can't do that with AI.

Trying to find a job? It goes straight to an AI filter. Trying to look at art? Here's 100+ ai shit. Trying to look up who was in what movie from the 1980s? GOOGLE GEMINI IS HERE WITH THE COMPLETELY WRONG ANSWER!

ShutUp10 lets you turn off annoying and invasive Windows 10 and 11 features.

Lycos is weirdly good at bringing useful search results to the top of the list. No ads, no AI garbage directly in the results page, and less SEO-maximizing nonsense in the results.

If Lycos is having a moment (as it does from time to time) and you still need to use Google, you can at least skip the garbage in Google's results page using UDM14.