touch_abs

Im not sure you actually achieve any of this with the SLS, and NASA in general. 1 and 2 are irrelevant if the cost is too high for anything that isnt a political stunt, to be a real alternative it needs to be something that a commercial payload might plausibly choose. 3 and 4, are compromised by the political nature of NASA spending, I dont believe the current institution is capable of innovation. The talent of the people is uncertain, not due to the individuals, but the way work can become visible (also the hiring, promotion, and leadership determination is based on politics not merit) Apart from long range experiments, nasa is really the cutting edge at the moment. 5 is just another way tax money evaporates.

It does, thats the point. A consumer then sees similar final costs when compared to greener products.

We are trading things and systems for social interaction; its a qualitative thing not really equivalent to how tightly coupled we are. We no longer need people at all, because we dont need (or even particularly want) to interact with a person in order to solve our needs.

Im not really sure what your point is; If there were a similar tool for software dev im sure it would gut the industry in the same way. The thing about these AI art tools is they emulate the normal commission process for a client; You describe what you want with example pictures and a short statement and you get roughly what you wanted. I dont think there is an equivalent for SW work yet, everything I have seen is aimed at accelerating an expert.

The thing about computers/computing is that being better at a task usually gives someone a commercial advantage; finding them and exchanging money for the implementation seems fairly straightforward...

So Im not an American, and live somewhere with proportional voting; however from what ive read FPP is a fairly easy to justify system if you take into account two ideas:

First, that the group represented is as small as practically possible, so small numbers of people, small cultural subgroups, and a limited social distance between the elected official and the constituents.

Second, that the rules governing a group are as local as possible, large disagreements should be unusual, and ideally lead to splits in region. Issues on a state or federal level should be limited to interactions with other states or nations; weather or not you want to live in a group with legal abortion for example should be determined at the the smallest size that can make that decision; probably city by city, what people in other cities think should have no bearing.

Unfortunately the nature of government is to centralize and accumulate powers, things that were once the sole preview of a town or county are now federal issues. It seems impossible find a "fair" voting method, to reconcile how to weight two distant strangers opinion on what should be a local matter.

I think it might go further than that, a whole lot of the commission and entry level illustration work is pitched as "draw this scene in a style similar to x" or "draw my character doing x thing in y style". AI has the potential to completely gut this area (and I suspect that the number of artists employed doing this type of work is substantially higher than the few pushing in novel directions).

This changes the design and availability of the software tools, the willingness for educational institutions to engage in these topics, and may even reduce the idea of a professional artist back to high art only (and we only need like 50 artists a year thanks).

There is a whole lot of artistic work that is done in fairly mechanical environments, game dev in particular is what Im thinking of, where you need a couple of lead charicters and then a bunch of extras. If you can generate the extras with AI then suddenly you have a substantially reduced need for artists.

I wonder how much this applies to other fields, corporate art and imagery for sure, but also a whole bunch of the low cost illustration and commission work I can see getting completely gutted, with the existing space for an entry level human to build from is no longer worthwile.

There isnt really a way to gather the data free from a manufacturer, every current self-drive mode has limited operating scenarios, if you want to record an incident you first need the manufacturer to confirm that it was operating, and that it was one of the intended use cases.

Companies are less interested in safety vs a human to begin with, they have a emphasis on not causing any accident they are culpable for.

The big limits of LiDAR are cost, more than anything. There have been dozens of public driving trials where from a functionality level the answer has been positive (apart from traffic lights, the bastards), but nobody wants to buy a solution with a six figure BOM, before integration.

Its interesting, that kind of object level fusion is a fairly different problem to training visual perception, following some of the less in fashion robotics techniques. I wonder if its a case of the Tesla engineers focusing on the fad technologies (or just their strengths) more than its a hardware cost thing.

Command line and config files are really not a great way to learn a system, once you know what you are doing its faster, but arranging the actions you can take in space with something that represents the current state of the system is by far superior for anything new. There are bad GUIs but fundamentally I believe a system where you can see what is going on will always be easier to learn.

M$ forgot whats actually good about their OS, its been slowly metastasizing, but even so I dont think desktop linux is the answer. There are some real advantages to the windows paradigm, the idea that the program and the window are the same thing is important; the application should always show what its doing and what you can do. Settings are selected from all the possible options (you dont need to know whats available in advance)

I have always found that personally as I become more capable I get stranger and more obscure problems with computers, but never less problems. The edge cases and weirdness that you can get help with in an active forum is greater for sure.

When talking about common problems, I mean truly common, the things an average high-school student runs into like joining an online classroom or editing a document (or even playing a game!). These things do fail on linux, and without being a power user that knows where to start and some keywords to search getting past square one is not so easy.

This is an advanced task, and if you want to focus on games, this so wrong it hurts. Most new games (and especially AAA titles) need a substantial amount of work to run well on the part of the proton (and sometimes game) developer, very few have good support, less still without some performance penalty, and even less within a few months of release. Thats before you run into driver oddities, and good luck with multiplayer.

The idea that you can slap wine on the few things that you need is a weirdly repeated fiction, it wont make it true by repeating it, stop doing it.

Because even with the low bar set by windows, desktop linux is genuinely less usable. If you need anything outside of a browser and encounter a problem you need substantially more understanding of how the OS works get it fixed, not to mention many common use-cases are either not supported or have an inferior less stable version on linux.

For common problems in windows you can find a current guide with pictures in seconds, on linux you are more likely to find a slightly out of date forum. And thats ignoring the complexities of different desktop environments and hardware compatibility.

Ideally a wired camera would record up until its destroyed, even warning the owner of motion. In that case even wearing mask still gives away height/build clothes, direction of approach, numbers etc. If its jammed first then you dont have that potential information, and you are relying on a loss of connection to be recognized as an attack.

It used to be that you could assume that you would need a PC of some kind, for work/email/homework whatever, so upgrading the pc so it could do games made sense.

Roughly the same time as the 360 era I built a PC for uni, and the extra cost to get console-beating parts (and adding in that PC games were cheaper or pirated) was way better value.

Now Im not sure what I would do in the same position, a laptop and console costs significantly less than a PC that even matches it, and unis mostly stopped having any focus on CAD

Many of the things used to balance the scales in the past are no longer possible, historically if you owned a business, the workers (who lived in the same area, spoke the same language, and drank at the same pub) could unionize, or decide that they were going to compete with the original business. Now you have a generic international labor force, who dont trust each other, and have large legal infrastructures preventing them from directly forming competing businesses.

Not true, if you can compress a real refrigerant slower you get better performance; modern control systems that can ramp up and down are significantly more efficient than older on-off styles.

Old ones did a few things really well; they typically have the hot radiator coils on top, or better separated from the internals (at the cost of space) better insulation (at the cost of internal space), and didnt have built in defrosters (or air circulation). They also used a somewhat more effective (but worse for the environment) refrigerant, freon.

After the 50s, we demanded a larger volume, not just being larger but also sacrificing insulation and ideal hot-side placement. And we had to change the refrigerant used (and get rid of latching doors). Several features that were poor from an energy standpoint became standard (auto-defrost, air circulation).

They exist, many new lines or expansions are highly automated. Retrofits are hard for two sets of reasons:

Political: union strikes (meaning if you want to automate a line you may have to accept it being closed for the duration) and public outcry over a reduction in jobs.

Cost: You often only get the benefit of the simpler domain if you can implement the whole system from scratch, if you have to deal with the existing signalling, engines, (and human-driven trains) that's a complex problem, with massive capital outlay.