Really old CPUs (486 or earlier) lack the computing power to do SSL at any decent speed. So you're limited to serving http, gopher, nntp, or telnet. Maybe IRC.
You could have a modern computer handle the SSL connection and proxy requests through to the antique but some of the purity of the thing is lost by doing that.
To clarify - we've had that setting for a long time and it won't be automatically changed on existing instances unless an admin chooses to. I've just changed the default of it for new instances (of which there are much less, lately).
It's the easiest way to stop scrapers, way easier than getting Anubis or fail2ban working properly. Unfortunately it comes at the cost of walling off an instance but most instances are for individual / small group use so it's an ok default to have.
Activitypub could be used to harvest content on a ongoing basis but to get all the historical data, which is the stuff they want, they can't use activitypub. Lemmy only has the last 50 posts in each community's outbox.
That is the convention. It is also common for devs to make a new branch for each release so you can issue updates for that version by adding a commit to that branch.
He seemed pretty coherent to me. At times clearly thinking faster than his mouth can keep up but really no big deal.
Pretty interesting how he's fully accepted that there will be "a wealth tax" as a result of a backlash against AI. And how the CEOs he knows are "livid" about how totally useless AI is even after they jumped in boots and all, lol
If you are running public-facing code, security is going to become a huge deal in coming months. So use AI to find all the security holes in your stuff before the hackers do.
If you have lots of databases and APIs for internal use, create a MCP server so agents can do anything with your data and APIs.
If your data is even a little sensitive and you don't want to send it all to the USA (read up on the CLOUD act), look into running models locally and the hardware investments you'll need to make.
The other day I made a machine learning model that classifies images as either 'a certain type of undesirable image' (no, not porn) or 'any other image'. It is 96.4% accurate and takes 14 ms to classify one image (using CPU only - with a GPU it could be 5x - 10x faster).
I plan to offer this as an API service that social media networks can use to filter posts.
whaaat https://en.wikipedia.org/wiki/Alcohol_enema#Native_American_ritual_usages