nathanlied

I feel like I need to nitpick a little: Models don't need source code available to reproduce software. See all the "full decompilation" projects cropping up. There's no putting this genie back in the box, because LLMs can also "refurbish" a project enough that it ceases to look like the original. People don't bother now because they don't have to, but in a world where they'd get hit with copyright notices, they would.

I can live with the tears of Google and Elon, frankly.

The adtech industry has, time and again, proven they cannot self-regulate to any decent capacity. At this point, the only reasonable course of action is to shackle them down with such heavy legislative burdens they're rendered de facto extinct.

I will not mourn their loss.

China is being very careful to provide enough support not to be seen as abandoning their trading partners/allies, while keeping the support at a low enough level to not get entangled in conflict or create expectations for future conflicts. They want to be able to paint this as "just business", in spite of any rhetoric they may publicly have. In some cases they'll help more in covert ways (Russia), while others they'll do the bare minimum (Venezuela).

So yes, China did give (note: sell) Iran some hardware, but it's not the most cutting edge tech China has, and it's not in sufficient quantity to make much of a difference.

The US is still ahead of China in a lot of military tech, even if the gap keeps getting narrower.

Some laptops had them, and came with IR remotes. Some of the marketing was around using those laptops as "media centres", and you could control them from the sofa while it was plugged into a TV.

Thank you for your openness here - and yes, it would be nice to see this kind of reasoning in the changelog, even if it's tucked a little out of the way! Those of us who care will read it.

Also very welcome is to separate it into a small blogpost providing details, if the situation warrants a longer, more detailed format.

Your comment highlights some tensions in deterrence theory, but it also oversimplifies over a few things.

If you notice, most countries with nuclear weapons also have published and publicized nuclear use policies. These documents usually highlight lines and conditions under which they will consider the use of nuclear weapons. This is by design. Ambiguity in nuclear policy invites miscalculation. Of course, you don't want complete certainty, lest you risk your enemy push right up to your line and no further; you want your lines defined, but a little blurry, so that the enemy is afraid to approach, much less cross. This is called strategic ambiguity. This is why Russia has been criticized a lot by policy experts for their repeated nuclear saber-rattling. They're making the line too blurry, and so Ukraine and their allies risk crossing that line accidentally, triggering something nobody truly wants to trigger.

In the case of a nuclear-armed Ukraine, given Russia's tendency to like to take over neighboring countries, they could include "threats to territorial integrity" as a threshold for going nuclear. They could also be a little more 'reasonable' and include "existential threat to the state" - which the initial 2022 invasion very much would fit.

What this looks like in practice is that Russia, in their calculations, would factor in the risk of triggering a nuclear response if they tried to take Ukrainian territory. Now, they may believe, as you seem to, that Ukraine would not risk the annihilation of its people over Crimea/Donbas. At which point, Russia would invade, and then Ukraine would have to decide. If Ukraine does not escalate, then they will lose deterrence and credibility for any future conflicts, assuming they survive as a state. If Ukraine does escalate, announces to Russia they will launch a nuclear attack to establish deterrence (reducing ambiguity that this is a full nuclear exchange), and then launches a single low-yield nuke at Russian invading troops, they place the ball back in Russia's court: Ukraine is clearly willing to employ nukes in this war - do you believe they won't escalate further, or do you believe they will launch their full arsenal if you continue?

This is essentially a simplified version of deterrence theory. The idea is to give the other side all possible opportunities to de-escalate and prevent a full nuclear exchange. If you do not back up your policy with actual teeth - by using nukes when you said you would - you're signalling something very dangerous.

This is also why nuclear-armed states do not tend to rely solely on their nuclear deterrence. They want a solid layer of conventional capabilities before they have to resort to their proverbial nuclear button. A strong conventional force keeps conflicts below the nuclear threshold, where deterrence theory tends to get very dangerous, very fast.

As someone who suffered from deep depression, but never alcoholism - the way alcoholism is described by alcoholics always rings true with how I experience (and hear described by others) depression. I am no longer suffering from depression actively; the symptoms of it are essentially gone. But there's life events, certain situations, certain moments of deeper vulnerability, that feel like I might slip back into it.

Surprisingly enough, although there seem to be parallels with how people experience 'life after' both things, I find it curious that alcoholics I talk to often use the "caged animal" metaphor, whereas depressives tend to describe it more as walking "a tight rope" or "at the edge of an abyss" metaphor.

There was a lot of inter-community conflict in the years (decades) preceding the formation of Israel, so it wasn't exactly peaceful. That there were some groups (on both sides, though the Jewish ones were far more effective, well-trained, and well-funded) that exploited those conflicts for escalation does not deny that the conflict already existed.

I would also argue that imposing the jizya/dhimmi status, creating "second class citizen status" for non-Muslims was, in and of itself, a form of Muslim-supremacist society in Palestine before Israel existed. Either convert to being a Muslim, or be stuck as a second-class citizen.

When Iran directly, materially, and openly, supports groups or organizations that have as an overt stated goal to destroy Israel, and actively work towards it (both with indiscriminate attacks against civilians, and building infrastructure for future invasions/attacks), I don't think the war is necessarily 'unprovoked'.

We may say that it was unproductive, badly conducted, or a lot of other things, but saying it was unprovoked is like saying that Ukraine has no reasons to attack Iran and/or Belarus. They do have those reasons, because both of those countries directly and materially support their attackers. It just might not be productive to do so (and indeed, Ukraine seems to believe it isn't).

I'm by no means defending throwing the baby out with the bathwater - which is what's happening when someone abandons a less-aligned company for a completely unaligned one - but I have a somewhat different perspective on what, exactly, ticks these people off so much with Framework but not Dell, even though Dell is ostensibly worse (from their perspective), and it's not all that unreasonable emotionally, but it leads to bad outcomes, and it is very much not rational.

For them, it's a problem of (perceived) hypocrisy. You see, Dell never claimed to be good. Nor did HP. They're big corporations, they've got contracts with the military, IDF, what have you. Their appeal, as it were, is the product/service itself. Their only ideal is the Capital, and they never pretended otherwise.

In comes Framework; claiming to be sustainable, different from the others, caring about society/the world/etc., instead of just in it for the Capital, like all the others - regardless of whether they really claimed this or not, it is how they're perceived by these people - and then they go and "do something like that", so they go back to Dell/HP, because at least those didn't lie about who they were. This is exactly what happens with Mozilla vs Google/Microsoft.

This is very much a reflection of a fair few Leftist political spaces. Two people may agree on pretty much everything in how a society should be ran, but one of them believes that private property is inherently theft, and another one would like to maintain private property. That singular difference, one that could be set aside until all other goals are achieved - if ever - will cause endless debate, drama, and ultimately a schism which will leave both sides weaker.

This has been my experience, effectively.

Sometimes I don't care for things to be done in a very specific way. For those cases, LLMs are acceptable-to-good. Example: I had a networked device that exposes a proprietary protocol on a specific port. I needed a simple UI tool to control it; think toggles/labels/timed switches. With a couple of iterations, the LLM produced something good enough for my purposes, even if it wasn't particularly doted with the best UX practices.

Other times, I very much care for things to be done in a very specific way. Sometimes due to regulatory constraints, others because of visual/code consistency, or some other reasons. In those cases, getting the AI to produce what I need specifically feels like an exercise in herding incredibly stubborn cats. It will get done faster (and better) if I do it myself.

Frankly? I think this is a whole mix of things. There isn't a "real" reason, there's a smorgasbord of them.

Why do "FAANG"s RTO? Because they're massive people-movers, and cities that host them likely hold C-level meetings to pressure RTO so that people spend more money. More on transport, more on food, more on coffee, more consumption = more taxes = more movement = growing value to office spaces = win for the cities. Not to mention that managers at these corporations are pretty wealthy themselves, and likely hold investments that would depreciate were WFH to continue in any great scale.

Why do smaller companies RTO? Because what works for FAANGs surely works for them, too. Literally. I've seen multiple managers push for RTO because the big tech leaders are doing it. Add that a certain 'magical' belief that RTO means more productivity and an enriching 'office culture' where new profitable ideas brew - they're all only human, after all, and are as prone to magical thinking without any concrete evidence as we all are - and you've got perfectly good reasons. And mostly irrational from a business PoV.

Is this the case for literally everyone pushing for RTO? Of course not, I'm sure there are legitimate reasons there, but most of the justifications I've heard, as a huge advocate for WFH who always seeks to understand pro-RTO management, have little basis on evidence that it is something good for the business.