XIsProprietary™

Before I go all tl;dr, this is a genuine trademark. Adam has used it, and I’m pretty sure I’ve seen it somewhere else. Probably by a sock-puppet, but there you go.

The thesis is that you cannot blame Linux (the kernel) for the abysmal multiple failures caused by things sitting on top of Linux (the kernel). These things are “third party.” This is a fine and accurate viewpoint.

Unfortunately, it falls foul of the indisputable fact that nobody at all ever uses Linux, the kernel. A kernel is useless without stuff on top of it. This stuff is supposed to be FOSS and many-eyed and so on, and therefore better in some indefinable way. And X is not just “not better,” it is execrable.

Therefore, the argument goes, it isn’t really FOSS. Proprietary software is bad; X is bad; therefore X is proprietary software.

Well, that’s the end of the TM. Everything else below the line is just a discursion on the theme.

————————————————————

To begin with, download the Unix Hater’s Handbook. It has a whole chapter devoted to X.

Surprise! Everything it says about X is still true!

Well, not much of a surprise at all, if you consider the history of X, which I will now rephrase from Wikipedia.

It started in 1984 at MIT and the current protocol¹ dates from October 1987 and the current reference implementation (X.Org Server, which is a lie on its own terms because the server is the client) is provided under the MIT license by the X.Org Foundation.

Loons sure like Foundations.

Now, as to the question of provenance; proprietary or otherwise.

On the face of it, this is ridiculous. X is MIT. X has always been MIT. Where does “proprietary” come into it?

Never underestimate the paranoid ingenuity of the Loon, however. It’s obvious, once you apply the correct warped ideology. And the history of X (being far longer than the history of other failed Loon projects) provides plenty of purchase. Let’s try to follow the chronology:

1984: constructed in an attic (not a basement!) in MIT.

1985: Version 6 (January). Entertainingly, the X-Men foreshadowed the ludicrous versioning of Firefox and the like. Version 6? A normal person who bought into version 4, say, would just ask for their money back. But no matter, DEC thought it was wonderful.

Remember DEC? Before X-mania, DEC was a fairly powerful company.

At this point, a man with the evocative name of Smokey Wallace stepped in. X was not yet 11, and it wasn’t hardware-neutral. So, here we go:

“The release of X11 finally occurred on September 15, 1987.”

Now we’re smokin’, or alternatively we are Wallacin’ (or Gromittin’).

Are you still with me? It’s still non-proprietary. Insanely broken, but still non-proprietary.

“The MIT X Consortium produced several significant revisions to X11, the first (Release 2 – X11R2) in February 1988.”

Jaysus, but these people cannot leave well enough alone. Oh, but there’s a trivial little gap:

“In 1993, the X Consortium, Inc. (a non-profit corporation) formed as the successor to the MIT X Consortium. It released X11R6 on May 16, 1994.”

In other words, everybody in 1988 recognized that the whole thing was an abject failure, but were not prepared to admit the fact.

There are three options open here: a logo contest, an appeal to the masses, or a fork. X chose the fork. And a silly name.

It didn’t really stick:

“The X Consortium dissolved at the end of 1996, producing a final revision, X11R6.3”

Point 3? Things have apparently (although it’s entirely arbitrary) slowed down. And it only took three years to realise that the whole thing is a pathetic waste of time and that graphical interfaces have moved on.

Now, at this point, I could bore for Jesus on the topic of “X.Org and XFree86.” The salient thing here is to remember that X has been broken for fifteen years, and nobody has done a damn thing about it.

——————————————————-

However, the proprietary bit?

At some point in that bit of the past that involves large companies selling their own version of Unix (Loons do not like to be reminded of this), X was touted as a “proprietary” solution. CDE, Motif, etc. Basically, the code was no longer Open, as it had been (for what little it was worth) since 1984.

Evil! Evil! Evil! Proprietary!

Also not relevant in 2012, but such trivialities will not stop the Loons. One last quote, then:

“By 2003, while the popularity of Linux (and hence the installed base of X) surged, X.Org remained inactive,[26] and active development took place largely within XFree86. However, considerable dissent developed within XFree86. The XFree86 project suffered from a perception of a far too cathedral-like development model; developers could not get CVS commit access²⁷[28] and vendors had to maintain extensive patch sets.[29] In March 2003, the XFree86 organization expelled Keith Packard, who had joined XFree86 after the end of the original MIT X Consortium, with considerable ill feeling.”

It’s rather a wordy quote, but it will serve (or, alternatively, client).

I’ve seen a nest of vipers that are more considerate of each other than these useless bastards.

——————————————————-

The following is a list of X principles, drawn up in the early days:

(1) Do not add new functionality unless an implementor cannot complete a real application without it.

Don’t bother your pretty little head with providing a useful API. And completely ignore anything new coming down the turnpike.

(2)It is as important to decide what a system is not as to decide what it is. Do not serve all the world’s needs; rather, make the system extensible so that additional needs can be met in an upwardly compatible fashion.

Nobody ever “extends” X. There are two reasons for this, both equally valid:

(a) If I “extend” X on my workstation (PC) and walk over to my colleague’s workstation (PC), then the extensions are no longer there. Frankly, this alone is a killer.

(b) “Extending” X? Have you looked at the config files? Indeed, can you locate the config files? Are you insane? If so, then extending X would be a nice hobby, along the lines of jigsaw puzzles or crochet or something.

In a Zen sort of way, this makes sense. X is the proverbial tree that falls in the forest, whilst nobody is listening. (Mind you, if they were listening these days, they’d probably be using Pulse Audio and have their ear-drums blasted out.)

(3) The only thing worse than generalizing from one example is generalizing from no examples at all.

I’d love to see what that original example in 1984 was, then. Clearly it wasn’t nicked from Xerox, like the Apple Lisa was.

(4)If a problem is not completely understood, it is probably best to provide no solution at all.

And yet they still came up with a “solution.”

(5) If you can get 90 percent of the desired effect for 10 percent of the work, use the simpler solution. (See also Worse is better.)

Simple? Or just Retarded?

(6) Isolate complexity as much as possible.

Complexity? Isolation?

A challenge for the student here. Download the X package on any distro whatsoever, and use ConfigureMakeMakeInstall™ on it.

Have fun!

(7) Provide mechanism rather than policy. In particular, place user interface policy in the clients’ hands.

“Mechanism not policy” is now a routine Loon chant. Interesting, that. As Kurkos points out, how do I cut and paste between X windows? Oh look, no mechanism.

Cunts.

——————————————————

[1] 99% of computer functionality does not need a “protocol.” I’m being generous about the other 1%. What you actually need, when it comes down to something as basic as a graphics stack, is a bloody API; not a weirdo multicast messaging system. Just a simple API. Is that too much to ask?

——————————————————-

One more irresistible quote from Wikipedia:

“When used across the network, bandwidth and latency can both be significant issues in the usability of certain software models.”

Well, yes. Glad you spotted that.