A Discussion of Military Strategy in the 21st Century

Rehashing the "end of Linux" aspirations of systemd is probably not necessary, but another piece of the puzzle falls in place, with ESR's eminently quotable:

If the clang/LLVM people decide they want to eat GCC's lunch, they *will* do it. The reason has nothing to do with any philosophical issue but merely the fact that compiler technology has advanced significantly in ways that GCC is not well positioned to exploit. The clang/LLVM people have both a clean-sheet technology advantage and Apple's money to fund a high-quality implementation with; FSF cannot match either.

This is a long simmering issue, dancing like a rainbow on the surface of Apple's entirely unfounded market valuation : if it's so greatly valuable that it "could buy Russia" and yet "inexplicably" doesn't… then what good is it ? Turns out… it can buy the other half of FOSS, to make a whole with Red Hat's hamhanded assault on "everything south of the kernel itself" (what that move is actually directed at should require little explanation). Or at the very least try to.

Incidentally, all this is not exactly surprising, at least not for the avid log reader :

mircea_popescu: LLVM compiles faster, sucks at encryption (lol) and by degrees at everything else. Sounds like the true C compiler!
asciilifeform: IIRC RMS is concerned about something quite valid – gcc has excellent front-end and ast-level optimizations, while at the same time, a middling-to-poor set of back-ends (depending on cpu arch). Back-end optimizations, that is.

mircea_popescu: Take half and run, yes.
asciilifeform: Aha. Plenty of folks want to take, take, run.

That FOSS is dead needs perhaps no further discussion here, it being one of those things plainly evident to anyone for whom such knowledge makes any difference, and completely hermetic for all others. Nevertheless, it is important to remember that the maggots do not produce any corpses.

The reason FOSS is dead is precisely that it was useful as a paradigm at some point past, but it is useless currently, for two very important, and converging reasons. The first reason is the "thousand fly eyes" problem. It is often said that "open source is secure because a thousand eyes look upon it" only to be discovered later on that nobody actually reads that which "everyone" was going to read. This has been recurring for the entire lifetime of the movement. The abject ridicule of fixing decade old major, security-annihilating bugs on a yearly basis in major distributions is merely a symptom of much deeper rot.

Moreover, in this paradigm the attack against freedom will necessarily follow the path of dumping out tons of code – which is harmful to humanity outright. Each tiny NSA affiliate churning out pointless drivel by the thousands of lines of code,  just so that someone a couple of pay grades higher has a chance at implanting a useful bug is the direct equivalent of a dozen smokestacks pouring out sulphur oxide by the ton per hour just so someone could admire the pretty patterns acid rain leaves behind. It's simply not a sustainable approach to the environment. So in this sense, FOSS actually gives the enemies of freedom a very dangerous incentive structure. (It is always the free that set the rules for the enemies of freedom – never the other way around. Never forget this.)

But even with the horrible pollution it engendered, that first reason was not enough on its own. The second is the clincher, and that is the happenstance that the problem also recently found its solution. The restructuring of thought brought about by a maturing Bitcoin, the colossal ideas of signed code and Web of Trust based packaging as emergent in #bitcoin-assets over the last year neatly do away with all the problems of a temporarily useful, momentarily important but fundamentally broken, philosophically confused and economically unsustainable FOSS. Add to that the backing of the most serene republic, and the results can be nothing short of exemplary.

FOSS was a watershed, for what time it lasted. That time has passed.

4 thoughts on “A Discussion of Military Strategy in the 21st Century

  1. > the colossal ideas of signed code and Web of Trust based packaging

    Please expand. Are you referring to using and running signed executables, or releases of source that have signatures (such as the gitian signatures as described in the Bitcoin release process doc: https://github.com/bitcoin/bitcoin/blob/master/doc/release-process.md )?

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>