Notices by Jens Finkhäuser (jens@social.finkhaeuser.de)

@david_chisnall Cthulhu weeps.

Hey, this is to let you know that I'm going to block fosstodon pretty soon. No exact timeline, but it'll be in January.

If you want to keep following this account/instance, I suggest you use an alt on some other instance I haven't already blocked.

Context: https://social.finkhaeuser.de/@jens/113765756511141919

I've been kind of stuck on "too big to block" thinking for too long. Time's up.

:boostRequest:

@SilverEagle I was reprimanded by the fosstodon mods for calling a legally confirmed Nazi a Nazi, because that four letter word apparently is worse than the policies and value system it represents.

Tells you everything you need to know about the instance, and I've had it limited for a while.

@onepict @SilverEagle I keep getting back to that there are two kinds of liberals: those that primarily want liberties for themselves, and those that want to liberate others.

FOSS is essentially based on libertarian values. We tend to call the first kind of libertarian who is also into FOSS a fossbro.

🤷♂️ I don't think it's much more complicated than that.

@jameschip Forget Legos, caltrops^Wd4 are the real danger!

The perfect mug does not exis...

@aral @spanezz All I have on GH is a landing repo that sends people to @Codeberg .

That seems to be the appropriate use for it.

@liztai By some odd twist of fate, I'm acquainted with a nephew or similarly distant relative of the Dutch translator/author. I didn't know until I watched this.

Yeah, it was pretty good, but it did fall a little flat in places, like you said. I can second everything you wrote here.

It was also a lot more enjoyable because of its flaws than many other shows, I think.

@andre actually, @alcinnz worked on an earlier version of the CRDT! Things have evolved since then, though.

The hard part, as he pointed out, is the ordering. Lamport timestamps work - they're logical clocks, where each edit happens one unit after the other, but the units have no real relationship to wall time.

The problem is: how do distributed nodes agree on what is "one unit" and "after?"

Lamport suggested that every node simply keeps its own timestamp, and every edit happens at plus one..

@andre @alcinnz ... relative to the previous one. To get the entire clock state, you have to synchronize the clocks of all the nodes, and take the most recent timestamp (TL;DR). Vector clocks are a generalization of the concept.

The problem here is that you kind of have to know all the participants in the CRDT and keep a timestamp for each.

A different type of logical clock uses a Merkle tree of edit "events". This works, but has other issues. They sort the most recent edit to the root of...

@andre @alcinnz ... the tree, which is useful, but that means updating a partial tree for each edit - the overhead grows O(log(n)).

What we've done is two tiers: on the one hand, vessel is a container - it doesn't have anything to do with CRDTs as such. But it has a different system for ordering chunks of data, which is sufficiently similar to a Merkle tree to work as a logical clock. More specifically, both are DAGs, and it's the DAG properties that make them work that way.

Which means...

@andre @alcinnz ... if you just put a sequence of edits into those chunks, all edits in one precede all edits in another, logically speaking. That's a lot less per-edit overhead, so less stuff to synchronize!

Unfortunately, it also loses enough precision to be potentially bothersome, so I devised a way to do better.

You can combine the logical, DAG-based ordering with a strictly monotonic wall clock per node. The nodes then attach a monotonic timestamp to each edit. It allows merging based...

@andre @alcinnz ... the actual order of events rather than a logical order, kind of.

This is relying on the fact that siblings nodes in the DAG that share a common ancestor must have synchronized state in order to be able to share an ancestor. And so within each generation you can argue that each node counting monotonically can be considered synchronized from that time point where the ancestor was received.

I wrote this up in a lot more detail: https://interpeer.org/blog/2024/01/comparing-vessel-to-a-merkle-dag/

(First of several posts)

@leah good one. I'm not telling you my CVE collection 😃

@grimalkina @writeblankspace There's a lot of thought gone into the similarities between autism/ADHD and PTSD symptoms, with some folk openly stating that the reason is that living as a neurodivergent person in a neuronormative world causes PTSD.

I really wouldn't know.

What it does do is reflect the experience of living with ADHD and autism. And dyslexia, dyscalculia, etc to varying degrees. I can see the attractiveness of the statement.

The whole point of neurdiversity advocacy, though...

@grimalkina @writeblankspace ... is to demonstrate that these conditions (whatever you put under the umbrella) are part of the "normal" range of brain function variations, and creating a system that traumatizes people (with a subset of characteristics) is the thing that needs to go.

It should follow that putting yourself into a pathologizing box isn't really compatible with such advocacy. It is identity politics more than anything, which can help/hinder advocacy .

So...

@grimalkina @writeblankspace ... with that out of the way (is it?), I think that the nugget of truth in this narrowing of the definition may be like the chicken and egg problem: which came first, the brain shape, or the symptoms?

I can cautiously get behind a definition that, roughly speaking, states that if brain shape comes first, neurodivergence is an accurate term. And if symptoms (e.g. induced by trauma) comes first, you can find a term that suggests more of an illness.

But that's...

Predictions are hard, but right now everything I read screams at me that #LuigiMangione is going to die, or to enter a dark hole he'll never return from.

Why? Because violence is power's answer to threats, and people in power want to prevent something akin to what @pluralistic lays out in https://prospect.org/culture/books/2024-12-09-radicalized-cory-doctorow-story-health-care/

Three things may avert this.

a) he's too white. b) there's going to be a better patsy. c) too many people making this prediction too public.

They're not mutually exclusive.

1/n

This is not a comment on innocence or guilt, or on who is or isn't nice. It's not a comment on right or wrong.

It's just an observation on power/wealth and the dynamics surrounding it. The wealthy and powerful need to show that killing them is not going to pay off, so they'll devote significant resources to producing that show.

Your attention is already on Mangione. That's the show that needs to go their way.

The sad part here is that @pluralistic offers a way out at the end of his story.

Recently, I keep thinking of that guy who unironically expressed the opinion that American English was "correct" English because there are more Americans speaking it than there are English people. "Democratic", he called that.

He didn't extend that same notion to International English, though. https://en.m.wikipedia.org/wiki/International_English

Then again, he was also convinced that the US was the only nation capable of bringing peace to the Middle East.

Anyway, smart and nice guy, otherwise.