I am assuming that they are laveraging multiple compression technics like using texture specific compression on the gpu, or sub sampling the resolution directly. 8 bytes per channel seems rather large here as well.

It would be nice to have an easy way to define a multi node proxy. Right now i have several devices in my network which i can't install tailnet directly on. For each, i have to create a seperate container for each (since i tailscale node can't share network namespaces).

I find tailscale to be simpler, more reliable and cover my needs better. I am pretty sure that have i known about tailscale before, i wouldn't have got the unify gateway.

I still love my unify though, dashboard galore

This is still L3 layer though. One the main use case of L2 is proper DHCP propagation and avoid subnet collisions. I do not think that this matters in practices though. Only a limited amount of user facing service require proper L2 emulation (apple TVs ?)

You my friend made my day.

care to share a pr :)

That's a semantic distinction which does not matter in the point OP is making. And contrasting rust static approach to general GC.

Hopefully any of those get some traction

You analogy doesn't not hold, and in manyways point at the root of the issue. The point was to show that harsher punishment did not seems to result in meaningful effect on crime stats. It's fine to want harsher sentencing for some kind of emotional fuffilement, let's just not pretend that is helping in any otherways.

punishment is a prerequisit to rehabilitation

How does that help, if after incarceration that person become a much hardened criminal both because of the lack meaning pathway to integration, and you know spending years locked up with the worst of society.

> but we don't really have a good understanding of what specific people can and cannot be rehabilitated, or how to go about actually effectively doing the rehabilitation.

Then let's work on that.

Weakness on crime is not a left vs right thing.It's mostly about which kind of crime each side is soft on.

Why stop at criminal, this also true for non-criminal; Let's just jail everyone, then we will all be safer.

Looking at the past for wisdom on how to treat young adult/people is very dangerous...

> People who start engaging in criminal conduct at an earlier age more likely to engage in criminal behavior as adults

this is not quite what the citation means. I am also very curious on how these study correct for cofactors .People who start a life of crime early are probably not comming from the best conditions in term of life circonstances. Which in it self is also a strong predictor.

> A more rational and less emotional sentencing system would be harsher on teenage offenders while reducing sentences overall.

No, that would be the fear based emotional response. The ration response would be to simply measure whether or not harsher sentencing have a better outcome. Instead of prehemtively jailing people based on dubious stats.

A lot of software engineer are seeing this as compiler engineer only caring about performance as opposed to other aspect such as debuggability, safety, compile time and productivity etc... I think that's where the "sabotage" comes from. Basically the focus on performance at the detriment of other things.

My 2 cents : The core problem is programmers expecting invariant and properties not defined in the languange standard. The compiler only garanty things as defined in the standard, expecting anything else is problematic.