Thursday, January 03, 2008

Annoying the freetards

Jaron LanierJaron Lanier: gets his hair done in Brixton.

Fake Steve Jobs points me to this article by Jaron Lanier, in which he puts forward his opinion, which seems to make sense to me, that Open Source development is not the saviour of the world.
Open wisdom-of-crowds software movements have become influential, but they haven’t promoted the kind of radical creativity I love most in computer science. If anything, they’ve been hindrances. Some of the youngest, brightest minds have been trapped in a 1970s intellectual framework because they are hypnotized into accepting old software designs as if they were facts of nature. Linux is a superbly polished copy of an antique, shinier than the original, perhaps, but still defined by it.

Before you write me that angry e-mail, please know I’m not anti–open source. I frequently argue for it in various specific projects. But a politically correct dogma holds that open source is automatically the best path to creativity and innovation, and that claim is not borne out by the facts.

Why are so many of the more sophisticated examples of code in the online world—like the page-rank algorithms in the top search engines or like Adobe’s Flash—the results of proprietary development? Why did the adored iPhone come out of what many regard as the most closed, tyrannically managed software-development shop on Earth? An honest empiricist must conclude that while the open approach has been able to create lovely, polished copies, it hasn’t been so good at creating notable originals.

Lanier is not simply some ignorant technology hack either.
Jaron Zepel Lanier (born May 3, 1960 in New York City) is a virtual reality developer. He was a pioneer in, and popularized the term "Virtual Reality" (VR) in the early 1980s. At that time, he founded VPL Research, the first company to sell VR products.

It seems to me that Lanier is correct in his assertions: whilst Linux has become increasingly stable (and is thus perfect for servers), it has failed to produce any really innovative technologies.

And by "innovative technologies", I mean two separate things here.
  1. Making existing technologies easy for the layman to use.
    Whilst many of the innovations in Apple Mac OS X (which is, essentially, Linux at core) have existed on Linux for years, they have not been accessible to anyone who is not a natural computer user. Whilst Linux Graphical User Interfaces have become better, especially in recent years, you are still going to have to find yourself using the command line fairly frequently.

    I occasionally use the Terminal (command line) on Mac OS X in order to enhance certain things, it is not necessary for the layman ever to use Terminal. And, for all of my years of day-in day-out use of computers, I am far from being a command line expert: it is not an environment that I feel comfortable in.

    Thus the innovation of people like Apple and, yes, Microsoft, is to use innovation to allow users access to powerful technologies without them actually realising what this is that they are doing and, crucially, without them needing to realise it.

  2. Developing new technologies to push the boundaries of what is possible.
    Open Source technologies simply haven't done this to any meaningful degree. Even where there is progress, it is more often than not built upon innovative technologies that others have developed.

    For instance, Dizzy showed me a simply dazzling window controller the other day, called Compiz. In terms of eye candy, it is almost unbelievable—here's a video of it in action. Now Compiz is (at least partially) Open Source and one could argue that it is innovative.

    However, Compiz could not exist without a graphics API known as OpenGL which was developed, in a closed model, by Silicon Graphics Inc in 1992.

I'll admit that my knowledge of computing interfaces and coding is very far from comprehensive, but I cannot think of any truly innovative Open Source product (not entirely surprisingly, since the Open Source model is a socialist one and socialism has never been renowned for its technological leaps forward). Can anyone suggest one, and one that does not rely on technologies originally developed under a closed system?

As Lanier says, that does not mean that the Open Source model is not useful in some areas—consolidation, for instance, or incremental steps in a technology—but only that it is conservative. As Fake Steve says,
Amen and much love, Jaron Lanier. You big fat Whoopie Goldberg looking freak.

I shall now sit back and await my flaming from what FSJ calls the "freetards"...


Ross said...

"Whilst Linux Graphical User Interfaces have become better, especially in recent years, you are still going to have to find yourself using the command line fairly frequently."

True - but which of these is easier: telling a computer illiterate that they need to type in "sudo grep whatever" and hit enter to fix their system, or telling them to click the settings menu, select add/remove programs, scroll down the list, select the package they want, click OK in the dependency window...

Anonymous said...

One example of how socalism does not always lead to evil though.

..apart from leading to OS X of course.

John Trenchard said...

the open source model is more anarchist than socialist.

there's agreed standards to make the various open source projects interoperate, but there is no politburo nor is there a central controlling structure.

if anything, the Microsoft way of doing things is vastly more "socialist" in its operation, in terms of its "5 year plans" and all that - just look at how crappy Vista was developed.

John Trenchard said...

i notice that the "command line" is mentioned in this blogpost as if its somehow backward and regressive.

let me explain - in simple terms.

have you noticed how car stereos are nowadays HARDER to tune into radio stations compared to the old fashioned way of just turning a dial?

its a similar thing with GUI development. the command line is there as a fall back - in case the GUI developers fuckup and make things harder to do.

as for "innovation", i suggest you read up on KDE 4 development.

Shug Niggurath said...


I'd argue that the latter option is easier simply because it's mouse led, which is 90% of how modern users use a machine, it's by the numbers and it's logical.

My father in law as an example, is a 66 year old who got a computer a year ago, he goes online to look for books, Google Maps etc. Other than typing a search term or two in, all he needs is a mouse.

knirirr said...

A lot of this was covered recently here.

the Open Source model is ... socialist

That turns out not to be the case (see the link above).

I am far from being a command line expert: it is not an environment that I feel comfortable in.

As it happens, I am a "command line expert" and hardly ever use the finder or spotlight GUI. It is the fact that Mac OS X is based on BSD unix (not Linux, though they are superficially similar) and therefore has such an environment in addition to the GUI that has convinced me to use it. I thought that "classic" Mac OS was crap.
But, just as I don't wish to be forced to use a GUI, I don't expect others to be forced to use the CLI. Isn't it great that we can have both? I can now recommend the same machines to computer illiterate relatives that I would use myself*.
I should also add that Apple have added a few nice utilities to the command line interface, e.g. mdfind, qlmanage &c.

* Caveat - I would use Linux (probably Gentoo or Fedora) on commodity hardware if I require a server. I only access my servers via ssh anyway.

Peter Risdon said...

Wow... where to start? You do know that OS X is built on an open source core (Darwin) that was originally a fork of FreeBSD? So you're a user of open source technology, albeit one who is happy to pay for a really superb graphical interface that sits on top of it. In this, you're similar to those who pay for SUSE or Red Hat Linux.

The idea that because UNIX began in the 1970s, modern Unixen are old fashioned is just... bizarre. It's like saying that cars are old fashioned because the first ones appeared a century ago. Wacky. Modern Unixen have been written since the 1990s and are entirely new.

You use the terminal in OS X. The Command Line in Windows has been massively enhanced, and has had the Windows Scripting Host added to it because it is also essential in Windows. An unskilled user of Linux, Windows or Mac will never use any of these, skilled users will. What point are you making?

TCP/IP networking came from the open source world, in fact Microsoft initially used FreeBSD's stack (and command line ftp client) for Windows... the DND system came from the same place - University of California, Berkeley - that gave us... FreeBSD (BSD stands for Berkeley Software Development). In fact, the fucking internet came from the open source world, as did email and the web. But I guess these just aren't innovative enough.

I can't be arsed to look it up, but there's an equivalent of Compiz for Windows - these distinctions between proprietary and open source are delicate things to analyse.

Now, here's the biggest error, and it's a whopping great one. Open source software is not socialist, though one of its most influential early figures was, in fact I believe Richard Stallman is a communist. But so what? He wasn't producing a communist system but rather one based on academic computing. Before the mid 1970s, ALL software was open source. If you bought a printer, you got the source code for the control software and could hack it. It was when a printer arrived in Harvard without the source code that Stallman began the process that would lead to GNU and the open source movement.

Stallman's point concerned ownership. If he bought something, he wanted to own it. He also felt that the peer review and collaboration found in all academic science was useful in computer science too. It is.

So far as the involvement of Silicon Graphics in OpenGL is concerned, again you're just demonstrating that you're unfamiliar with the field. MOST large scale open source software projects have always run in partnership with businesses. Hollywood companies like Dreamworks abd Pixar contributed money and developers to the GIMP image manipulation project, and then when theis went in another (more photoshop) direction, sponsored a fork called Cinepaint that handles 24 bit graphics. Of the last dozen movies you watched, I'll bet ten were mastered using Cinepaint in some capacity.

Yahoo runs entirely on a heavily customised FreeBSD network and the company has always fed money, code and developers back into the project.

OpenOffice began as the (proprietary) Star Office, was bought by Sun and opensourced.

Blender began as a proprietary paskage and was then opensourced.

And so on. Business have always been involved.

People talk about this subject in terms of pure fantasy. It's weird.

Peter Risdon said...

Typo - DND = DNS

A couple more things. Linux has been outstandingly stable as a server for more than a decade. It's as a desktop system that people are debating it now.

Not only has the design of Unixen changed hugely over the years, Windows has increasingly incorporated features of Unix into itself. In fact, to start the Windows NT project, MS hired a bunch of developers who had written VMS - a multi-user, multiprocessing operating system that is not unfamiliar to anyone who knows Unix. VMS is one letter shift from WNT, and so the lineage of current Windows versions was made plain from the start.

The most important innovation has been coming from the Unix world, which is where open source lives, for a long time now and it is Apple and Microsoft who have adapted to it. Microsoft is even open sourcing some of its own code now.

It's also important to remember that open source software predates and is not limited to the stuff people talk about now. It has always existed. At one time nothing else existed. It can be covered by any license you care to issue, including restrictive ones. It is where the overwhelming majority of innovation takes place.

A Microsoft developer gave a talk in Cambridge a few years ago, and outlined the technological base of their far-future research facility there. One of my employees went along and gave a report to colleagues, including me. Microsoft do their serious, innovative development on... FreeBSD.

dizzy said...

I was going to add much of what Peter said. I think the problem is that you're defining innovation in terms of "layman" and the fact is that a "layman" still have to understand what is happening to do something.

At the core of almost all "tick box guis" is merely the editing of a settings file. The problem that the "layman" has is that they find the graphical method of typing "true" or "false" easier, but the technology is still essentially the same.

Yes, closed source has produced many great thing. So has Open Source. I also think, incidnetally, that you are making a flawed merge of Open Source (GNU) with the Free Software Foundation. the two things are very different.

Regarding Linux and the Desktop, there has been talk about desktop readiness for newbies for years. This year, Ubuntu and PCOS (I think) have really reached that point. The problem is that Microsoft have such a donminance over what opeopkle are "used too" now that it will difficult for anyone to break that hold. OSX has that problem too of course.

Ed said...

I would switch to Ubuntu if Apple produced a version of iTunes for it.

dizzy said...

Just run the Windows version on it.

Newmania said...

I understood not a word of that DK, computers pah ! Give me a trusty quill anyday

Roger Thornhill said...

Open Source is not Socialism, as there is mindset for compulsion to join any particular collective.

Peter: In fact, to start the Windows NT project, MS hired a bunch of developers who had written VMS - a multi-user, multiprocessing operating system that is not unfamiliar to anyone who knows Unix. VMS is one letter shift from WNT, and so the lineage of current Windows versions was made plain from the start.

Unix and VMS are VERY different. The CLI/Shell is totally, utterly in fact, the APIs, the scheduling, permissioning, accounts. I suppose they both have virtual memory and allow SMP. VMS is so robust and secure from hacking it makes UNIX look like Windows ME.

WNT was meant to be like VMS, but MS pulled all the decent stuff like 32 levels of priority in scheduling, the file versioning and the way resource permissions in general worked. IIRC some parts were pulled as NT evolved, e.g. NT4 was a pale shadow of the ambitions of the earlier versions.

Peter Risdon said...

Roger, certainly. But I said merely that VMS would be "not unfamiliar" to a Unix user. It's a small point, but I think that's true. It certainly was for me (with OpenVMS). I don't think the same would be true for someone familiar just with Windows or Apple systems.

Since I'm typing again, I'll mention the irony that a lot of beardy Unix types now use OS X as the best Unix OS around, with a nice open source subsystem and the ability to run open source X applications if necessary.

Roger Thornhill said...

Yep, Peter. If some people could shrug off their chippiness about Apple, they might realise it, too.

knirirr said...

I'll mention the irony that a lot of beardy Unix types now use OS X as the best Unix OS around

The best desktop unix OS around, perhaps, but I wouldn't say it was the best overall. I don't have a beard, but in my experience the chaps that do love Debian.

Anonymous said...

OS/2 anyone? Personally I use Windows. I have a dual boot partition for SUSE but I just can't get on with it. My brother (who is a socialist) loves Ubuntu. Make of that what you will.

Anonymous said...

The innovation isn't just in the software itself, but what having access to free software results in.

Google (including that this very site uses) would never have been able to get off the ground without Linux - buying server licenses for Windows would have been out of the question when Google started up, and now that they're running so many machines, I dare say the stock holders wouldn't enjoy paying for licenses now either.

As a single, but extremely important, example of innovation in Open Source, The Apache HTTP server was the first to enable virtual hosts, allowing multiple websites to be hosted on one machine. The fact that ISPs and data centres could then host multiple sites on cheap hardware with no licensing costs is what caused the Internet explosion. This absolutely would not have happened without Linux and Apache.

In any case, the Open Source communities couldn't care less what Lanier has to say. We're quite happy churning out high quality software for anyone (including those who choose to slag it off) to use for nothing. Or don't. It's your call.

Roger Thornhill said...

Google (including that this very site uses) would never have been able to get off the ground without Linux

And AltaVista, you know, the thing that did Google's job before Google was invented, would not have occurred unless we had OpenVMS and its wonderful VMS Clusters running on that excellent, highly advanced Alpha chipset. All proprietary, mind - Digital as then used it as a showcase to show how powerful the system was. Of course, later outsmarted by Google's algorithms and inventive use of existing hardware and novel techniques.