It is about the *seamless* version, not the full remote desktop version.

1. I suggest looking at Xpra. It's married to X, but is far better than X or NX at seamless remote applications. It's also an example of exactly the sort of remote experience that *could* be done within a Wayland context. The problem obviously being I had to point out something married to X as an example, rather than knowing off hand something that pulls off the same in Wayland, but the concepts are pretty sound.

I suggest looking at xpra. X and even NX I didn't have a lot of fun with, but Xpra has treated me well. The kicker is that the Xpra strategy concept translates to an architecture like Wayland (it captures applications via compositor interface and contextual data through window manager calls).

I have to side with those that say X's inherent network protocols are less interesting. The unfortunate fact is that everyone then cites things like SPICE and typical RDP configurations and VNC, rather than something that achieves the same seamless behavior that the fans of X want.

Spice is pretty crappy (after really wanting to like it for a while). I hear rumor that RDP can be decent, but thus far I've not seen examples of it doing remote application access in a seamless way.

My personal example of how seamless remote applications can work well is Xpra. Despite it's current marriage to X, the strategy used does not use the 'remote' capabilities of X and the concept translates directly to something like Wayland. It intercepts things at the compositor layer rather than remote X calls.

I find this interesting. I know of people who strangely put a lot of their 401Ks into the stock of the company they work for, but mostly in my experience 401k investments aren't particularly tied to the company of employment. At least I make sure my investments are not particularly tied to any one company.

Are there any rendering mistakes or quality differences? Are there any issues with stability? Frame rate is not the only metric, it's just the only metric anyone can simply publish.

Intel has had a process lead, but that's not the only reason they make abnormal margins.

- Lock in. AMD's the only viable x86 vendor and they've been off in the weeds after their glory days that culminated in x86-64 and NUMA x86 architecture
-Ecosystem. Intel invests heavily in things like compilers and standard libraries and so forth. They pay to have good software developers enrich an ecosystem that favors their processors
-Microarchitecture. They frankly have very good hardware engineers.

ARM came to prevail because Intel took their eye off the ball in the low power space. Recent Atom family designs have made a strong showing in that space (in part due to software developers creating remarkable ARM emulation, in part due to excellent hardware design, and in part due to the business call to compete at more typical (for the market) margin). The 'ARMs sip power' reality was (unreasonably) extrapolated to "they'd kick ass in the datacenter", but now studies are being done and the ARM server designs are losing both on raw performance and performance per watt to the Intel offering (e.g. one done by CERN did open power 8 v. aarch64 v. two x86 variants (atom and xeon e3) and the x86 variants won most everything (except POWER did SMT better than most, but even that fell apart most of the time).

Also as people who are dying from other stuff decreases over time, cancer gets a crack at people who were otherwise knocked out by something else first.

While this looks like a perfectly reasonable language, I'm a bit weary of this sort of bragging about line counts. I could do the same thing, it would look roughly like:
ssh pi2.local 'while :; do pitemp=`ssh pi.local get_temp`; mytemp=`get_temp`; if [ abs($((mytemp - pitemp)) -gt 2 ]; then echo "Temperature on pi is $mytemp and on pi2 $mytemp. "'
(abs and get_temp are up to the person to have the functions).

Talk about the syntax being nicer, but lines of code is no big deal in this particular case. It has a nice and non-ugly 'run this on another host' syntax and automatically takes care of the communication channels in a reasonable fashion with a low amount of fuss. Leave it at that.

Also, the trap is that any term that gets adopted will *become* yet another idiotic marketing term.

Any phrase attracting buzz is doomed to become a meaningless marketing term abused by companies with their agendas to be relevant.

I particularly dislike the phrase 'internet of things', but I know I'll dislike any term that sees common adoption while the media/marketers have an interest in the field until that interest dies down and it no longer becomes fashionable for companies to shoehorn it into their message.

For contrast, I had a minimum mortgage payment of 850/month on a 3,000 square foot house, had enough extra to pay it all off in 8 years. It's a fair point that $66k/yr in most areas easily beats $100k/year in SV. That's one thing if you really *want* to live in Silicon Valley, but if you move there because of a better job opportunity and didn't particularly care about being in SV specifically, you are probably making the wrong move.

Basically, I've encountered two classes of H1-Bs:
-Folks who are exceedingly good at what they do and are sought out by name. They are by no means cheaper, but a company has to do H1-B to get them.

-Folks who are cheaper and held hostage to their circumstances.

I think across the industry the latter is at least somewhat more common (it's the simplest explanation for the high volume of H1-B requests from specific companies, it's unlikely one company would need the former case by the hundreds). However this situation results in some reactions that are highly offensive to those in the first category.

I was $55k right out of college 13 years ago. Those were, frankly, the easy jobs to get too.

After reading through the paper a bit, it seems interesting, but perhaps a bit overblown. It seemed to have a lot of work to understand the very specific problem domain before this could be applied. It's more like a methodology *enabled* expert engineers to do optimization, not that it did optimization *instead of* expert engineers.

It's also a field with a lot of solid technical high level algorithms, so there was a pretty good space to map things to. Basically it was identifying what inscrutable code was doing as it relates to well known algorithms, enabling them to start fresh to apply the best practice today of said algorithm. If you are not in this sort of space, the strategy doesn't really have a way to help very much.

Incidentally, it's (probably) not in this case that the code was sub-optimal, just mis-optimized. Whether it's something like code written before AVX2 existed, now adding AVX2 codepath, or scenarios with two algorithms that end the same way that use different operations where a choice was made based on chips that could do one of those faster, and newer chips started getting faster the other way.