Olives @olives@qoto.org

One of the blogging systems I'm looking at doesn't even have a drafts feature... Or a scheduling system (less important).

Drafts are important.

I see the guy in the Medium article used a mutex. You have to be a bit careful with these because only one thread can hold a mutex at once (until the unlock method is called). Other threads have to wait until it's released.

If you're doing something that takes a long time, that might be painful.

https://en.wikipedia.org/wiki/Test-and-set The mutex is likely implemented with some sort of atomic instruction like this.

I see he used a mutex. You have to be a bit careful with these because only one thread can hold a mutex at once (until the unlock method is called). Other threads have to wait until it's released.

If you're doing something that takes a long time, that might be painful.

https://en.wikipedia.org/wiki/Test-and-set The mutex is likely implemented with some sort of atomic instruction like this.

Mastodon's ID format appears to be a custom kind of snowflake (the huge number which appears in the URL for posts).

https://en.wikipedia.org/wiki/Snowflake_ID
https://medium.com/@jitenderkmr/demystifying-snowflake-ids-a-unique-identifier-in-distributed-computing-72796a827c9d
To get an idea of what a snowflake is.

One of the good things about having a proper blog is that the URL would presumably be human readable, rather than just being a huge number on it's own.

There are thorny questions, of course, like if they engage in a potentially unwise decision to engage in sexting with peers, is that the same as other groups sharing such imagery? Strong language like "CSAM" probably doesn't fit there.

While there are some who write "NCII or CSAM", the way that I treat it is that NCII (non consensual intimate imagery) is a superset which covers both. That is to say that CSAM is a subset of NCII (as it's accepted that minors can't give meaningful consent).

That said, being very specific can be useful to make things clearer for readers.

The interesting thing is that I don't think I have yet heard anyone mention the term overfitting (a potential design flaw). Depending on how an "AI" model is designed, it can regurgitate it's training material.

https://en.wikipedia.org/wiki/Law_of_the_instrument Over reliance on one particular tool.

Every country has flaws. There is no such thing as a mythical country.

It feels like Norway has found itself onto a list of "mythical" countries, much like the Netherlands and New Zealand were for a while, and someone just assumes that everything is great there.

https://www.theguardian.com/world/2024/oct/23/norway-to-increase-minimum-age-limit-on-social-media-to-15-to-protect-children
Since a few people seem to treat the Scandinavian countries as a unit, it is worth remembering that Norway (from what I've seen) can be worse on speech / Internet issues.

I've even seen people credit Norway with something that a different country did.

Maybe, formal is a better term than professional.

I'm getting a bit tired of these bugs, now I like a lot of things like the functionality on this server, and the admin is pretty chill, but I am going to play around with an alt to try to be more productive.

I see someone is calling someone out for running sensationalism based on Google's internal bug tracker (which was promptly fixed).

See, I was right about these bloggers.

I tried to make the new preamble flow better (although, at the cost of being longer). The previous one was a bit more bam bam bam, which is probably my social media habit to save space.

Of course, it isn't anywhere near as bam bam bam as one of the older science posts.

Since I am trying to be a bit more professional in these science posts, than I might be in casual social media chatter, I don't do things like putting quotes around the word AI.

This was a surprisingly fast update. Part of the reason is that this one was less about reading and more about changing text than the previous ones. Of course, another factor is that I had more free time to do this one in.