And, of course, an axiom is just a starting point; if the axioms are inconsistent with each other, or with reality, the conclusions drawn from the axioms could be bogus, no matter how air-tight the reasoning.

Because some other type of phone would require you to understand the technicalities of a backup? Sounds like the kind of phone most non-technical consumers wouldn't use.

Or because, with some other type of phone, the technicalities of a backup would be simple enough for non-technical consumers to use?

The article pointed to the very very very very very incomplete release notes for stuff after 3.5.

You wanted to link to the 3.5 release notes.

It's only "safe" if you 1) trust that the machine to which you're connecting restricts the ability to bind to ports under 1024 (not guaranteed), and that the only people running processes on the machine in question are either trustworthy or are restricted from running programs that bind to those ports (not guaranteed), and that the system services you care about have ports under 1024 (not guaranteed).

And what guarantees that a "system service" (whatever that might mean) has a port under 1024? Perhaps a better scheme for determining whether to trust a service is called for here - one that would probably obviate the need for "privileged ports".

One would hope that services that trust untrustworthy guarantees would be in the minority; in the best of all possible worlds, they would be completely non-existent.

Good. Ideally, nobody would trust them, and claims such as "It prevents regular user programs from masquerading as system services, which usually sit below port 1024." would be treated as the uninteresting claims that they are.

"CISC functionality" is the ability to execute a given CISC instruction set with acceptable performance. Transistors can be used in several different ways to achieve that, and you can choose to use fewer transistors in one place on a chip in favor of more transistors in another place, and if that choice means you still get better overall performance executing the same instruction set, that choice is a good one.

Again, as long as the full instruction set can be executed (even if some of it is executed by trap code), you don't have a subset. You may happen to execute some functions slower, and other functions faster, but if the net result is faster execution of the code actually run on the machine, you have a better implementation.

Smart people add registers iff they're sufficiently useful that it's worth either increasing the die size or taking transistors away from other functions.

For existing architectures, the reason they don't do it is that it would require changes to the instruction format, which, for most instruction set architectures, would be a royal pain. For x86, they (AMD, to be specific) could and did add Yet Another Prefix to double the number of registers as the instruction set already had a tradition of adding prefixes. For ARM, they were already introducing a 64-bit variant of the instruction set, and didn't have to maintain binary compatibility. For, for example, System/3x0, you'd have to add prefixes to an instruction set lacking prefixes, or somehow use opcode bits to refer to additional registers. If somebody were to design a brand new CISC architecture (in an era where we're not designing many new instruction set architectures at all), they could design one with 32 GPRs.

With 16 GPRs, it has fewer registers than all the major RISC architectures other than 32-bit ARM, just as the 32-GPR System/3x0 (including its 64-bit z/Architecture version) does. It's less register-limited than x86, but that's not setting the bar very high. (Note that IBM recently added instructions to z/Architecture that do arithmetic on the upper 32 bits of the GPRs; that suggests that there's some register pressure with only 16 GPRs, although if they still have to make use of base registers, even with PC-relative branches, that might add some additional pressure that x86-64 doesn't have.)

No, they didn't; x86-64 has, as noted, only 16.

Limited their what?

There's "the alternatives" and there's "the alternatives worth considering". The latter category excludes the developing countries in question, as well as the US. Hopefully the people in charge of health care in Canadian governments (federal and provincial) are looking at the alternatives in the latter category to see what they can learn from them.

That isn't answering the question I asked.

The question I asked was "You can't trade off, say, transistors used for registers (especially given that the bigger processors do register renaming, so you have more hardware registers than the actual RISC/CISC instruction set provides) for transistors used for some other purpose?"

I said nothing about keeping all the same functionality, if by "functionality" you mean, for example, "on-chip caches of the same size" and "same number of hardware registers including ones used for renaming of the architected registers" or "complexity of the branch prediction hardware" or .... Yes, there may be tradeoffs you have to make in how you use your transistors, but if the benefits of the additional registers outweigh whatever performance benefits you lose by reducing the size of other functional units by however much the additional registers require, that might be the right tradeoff to make.

OK, that's probably using "memory indirect postindexed mode". Addressing modes that complex are something some CISC processors had, but not others; x86 is much less complex (scaling, but no memory-indirect or auto-increment/auto-decrement), and S/3x0 even less complex than that (no scaling, just double-indexing).

How often was that addressing mode used, in practice? Was it used often enough that you saved enough code space that you could make the I cache smaller?

Then presumably, if the anonymous submitter was Canadian, they were one of "most Canadians", and offered his or her hypothesis about the attitudes towards science and the Canadian health care system because either 1) they didn't pay attention to the EU results or 2) they assumed that European countries are like the US in their heath care systems.

Or perhaps they were a typical Amurrican and made the same silly assumptions.

In any of those cases, if the second paragraph of their submission had been eaten by a grue, nothing of value would have been lost - heck, something would have been gained. (I mean, my knee-jerk reaction would have been to blame "We depend too much on science and not enough on faith" on Amurrican religiosity, but, absent any data on whether the more-religious countries in the EU have a sufficiently-high level of agreement with that assertion as to drag the EU average down, even with those countries in the EU that are less religious than Canada, I wouldn't wonder too hard about that one.)

That if you think a packet coming from a privileged port is coming from a program run by a user whom you can trust, that's only true if you can trust everybody who's plugged a personal computer into your network or you can ensure that nobody on any of those computers gets to run programs with sufficient privileges to get to use privileged ports. (Remember that this subthread started with a mention of privileged ports, which I do not consider one of the Great Ideas In Network Security.)

When I gave this Connectathon talk, somebody asked about the

MacOSX automounter daemon
Launched in user’s session by launchd - exits when idle
Runs with user’s credentials

slide, with "ZOMG WHAT IF THE NFS SERVER ONLY SUPPORTS MOUNT REQUESTS FROM A PRIVILEGED PORT!!!!1111ONE!!!!", I forget what I said, but, in retrospect, I wish I'd said "oh, we're going to remove the root-only restriction on privileged ports", and taken delight in any pathetic squeals of terror that resulted.

(We ended up, for other reasons, running automountd as root in a privileged session, so that wasn't an issue.)

Who's "you" in this context? I am an admin user on my Mac; if that means that plugging my machine into your network would terrify you, then you'd better somehow make sure that never happens. Don't expect me to care about your problem.

So? Most of us USans are mostly exposed to the US system as well, but that's not an excuse for being clueless about the rest of the world.

I.e., if "gee, our health care system doesn't let people who aren't well off get no health care" is offered as an explanation for why Canadians are less likely than people in the European Union to say that "We depend too much on science and not enough on faith.", whoever offers that explanation really needs to start looking at European health care systems, or, at least, to get a by-country breakdown of the EU figures and see whether there's any correlation between having a health care system that lets the poor fall through the cracks and believing that "We depend too much on science and not enough on faith."

(And if they weren't even bothering to care about countries other than the US when concocting that hypothesis, they need to get out more.)

Or, to put it another way, I refuse to give a damn whether it's what you're most exposed to, as it's completely irrational to care; if I were Canadian, I'd be embarrassed to see a fellow countryman acting as if the US was the only other significant country on the planet and as if the EU didn't exist.

They say

but it sounds as if they're comparing the Canadian system for paying for health care with the US system, as opposed to the systems used in for example, Western Europe.

Better than, say, the health care systems in the UK, Germany, France, Spain, the Netherlands, Switzerland, Taiwan, etc.?

You can't? You can't trade off, say, transistors used for registers (especially given that the bigger processors do register renaming, so you have more hardware registers than the actual RISC/CISC instruction set provides) for transistors used for some other purpose?