You can also click on an image to open the original in a lightbox. (I should make that more obvious.)
The dithering happens server-side (er, laptop-side). Experimented with a JS filter but it just didn't hit the way I wanted it to. Plus I save a lot of bandwidth this way
There's a few that I think would look better with dropped frames (the cake for instance). These are all the raw output of the script. Consider it maybe a starting point for manual editing :)
The unfortunate reality of accidents, lack of intention~
I've been taking pictures to be more effective wigglegrams now that I know I can auto-GIF them. I'll do a followup post at some point. The big trick is that they look better when the subject stays in the center of the frame, with the camera "orbiting" it to create a sort of parallax effect.
As the others pointed out the script is at the end...
> Would be nice to have... the choice of perceptual hashing algorithm
Funny enough the script does some of this - there's a handful of algorithms the library supports and it computes/saves all of them. "Perceptual" seemed to work the best but if you want to experiment w/ the others they'll be there for you.
It’s tough but one of the tenets of Gemini is that a lone programmer can write their own client in a spirited afternoon/weekend. Markdown is just a little too hard to clear the bar. Already there was much bellyaching on the mailing list about forcing dependence on SSL libraries; suggesting people rely on more libraries would have been a non-starter
Note that the Gemini protocol is just a way of moving bytes around; nothing stops you from sending Markdown if you want (and at least some clients will render it - same with inline images).
Maybe the solution is to ban editing. Or let moderators review + approve edits at least.
The gambling site “Stake” was doing the same thing recently, they’d make posts on financial advice or gaming subreddits and edit in a link (as to be “oh btw I need advice because I made money betting”). Were even using Greek Unicode “a”s and “e”s to hide from the automod filters. Scumbags among scumbags
It's two different problems. People who run review sites and blogs and such care about traffic, and not getting attribution will kill their desire to participate. People who post here and on Reddit etc. care about talking with other human beings, and feeling ignored in a sea of botspam will kill *their* desire to participate.
It’s not “clear communication” though. The prose that comes out of LLMs is awful - long, vapid paragraphs with distracting tropes. You can ask them to be concise but then they file down all the wrong bits of the sentence and lose meaning. There’s a reason people bother clocking it and complaining about it, it’s *bad*
It’s like everything else that AI can do - looks fine at a glance, or to the inexperienced, but collapses under scrutiny. (By your own admission you’re not a great communicator… how can you tell then?)
I'm taking the bait whatever. All those projects are just more fucking AI tools. It's all Claude seems to be good for - writing agents, skills, harnesses. Just a big fat ouroboros.
(Going down the /trending page - 13 of the 14 are some flavor of context manager or agent or smth)
Let me know when someone uses Gas Town or openclaw to write something that isn't "the next Gas Town or openclaw" and then we can talk
IIRC "firmware as hardware" only applies if nobody - not you, not the manufacturer - can update it. I.e. your laptop's BIOS is software, the controller in your washing machine is hardware.
Of course, now washing machines connect to the Internet, so the obvious lines have blurred
If I remember correctly you carried that project!