The dictatorship of the presentation layer

Bill Thompson is eloquently sceptical about Web 2.0. (I prefer the term techBubble 2.0 btw.) Here’s a sample of his Register blast:

If Web 2.0 is the answer then we are clearly asking the wrong question, and we must not be fooled by the cool sites and apparently open APIs. Most of the effort is – literally – window dressing, designed to attract venture capitalists to poorly-considered startups and get hold of enough first-round funding to build either a respectable user base or enough barely runnable alpha code to provide Google or Yahoo! with yet another tasty snack. We need to take a wider view of what is going on.

Back in the 1870s Karl Marx outlined the steps through which he believed a capitalist society needed to pass before it could reach socialism. After the revolution came the dictatorship of the proletariat, a painful but necessary stage of oppression and correction, during which the organs of the state would whither away as humanity achieved its true potential and coercion became unnecessary.

Web 2.0 marks the dictatorship of the presentation layer, a triumph of appearance over architecture that any good computer scientist should immediately dismiss as unsustainable.

Ajax is touted as the answer for developers who want to offer users a richer client experience without having to go the trouble of writing a real application, but if the long term goal is to turn the network from a series of tubes connecting clients and servers into a distributed computing environment then we cannot rely on Javascript and XML since they do not offer the stability, scalability or effective resource discovery that we need.

There is a massive difference between rewriting Web pages on the fly with Javascript and reengineering the network to support message passing between distributed objects, a difference that too many Web 2.0 advocates seem willing to ignore. It may have been twenty years since Sun Microsystems trademarked the phrase ‘the network is the computer’ but we’re still a decade off delivering, and if we stick with Ajax there is a real danger that we will never get there…M/blockquote>

Owen’s goals

Owen Barder is returning from his sojourn in Berkeley to resume life as a senior civil servant. He’s published My goals as a manager on his Blog. They’re admirable, and sensible. Wonder if HMG will allow him to achieve them. I particularly like Number 9: “I will not take myself too seriously”. Tut, tut. If the lad goes on like this he will blow his chance of a knighthood.

Apple: keep taking the tablet

Hmmm… It’s that time of year again — open season for speculation on what Apple will get up to next. Australian tech guide Smarthouse is claiming that,

Apple researchers have built a full working prototype of a Mac tablet PC and three Companies in Taiwan are now costing a product for a potential launch in mid 2007.

Sources in Taiwan have said that the focus has been more on the home and the education environment than the enterprise marketplace. Several months ago I was told that Apple was exploring a neat new device that is basically a touch screen that links to various source devices including a brand new media centre that Apple is planning to launch next year.The Mac tablet has been designed to handle third party applications such as home automation software that will allow users to control lighting, audio, entertainment devices and security feeds. It also acts as a full blown PC has wireless linking for a new generation of Wireless Hi Fi speakers that are currently being tested by Apple….

The Rise of Freeconomics

Lovely post by Chris Anderson (he of Long Tail fame)…

It’s a big day for Moore’s Law. I’m not sure anyone else has noticed this, but by my calculations we have in the past few months reached the penny-per-MIPS* milestone. Intel’s Core Duo running at 2.13 GHz now costs around $200 at retail (it’s around $180 at volume), but can do about 20,000 MIPS. I remember my first 6 MHz 286 PC in 1982 that did 0.9 MIPS. I have no idea what the CPU cost then, but the PC it came in cost nearly $3,000 so it couldn’t have been cheap. Say it was around $1,000/MIPS back then. Now it’s $0.01/MIPS. I know I shouldn’t be astounded by Moore’s Law anymore, but that really is something.

Good Morning Silicon Valley picked up on this and added an interesting quote from Alec Saunders, who added some extra historical perspective:

  • In 1977, Digital Equipment’s Vax 11/780 was a 1 MIPS minicomputer, and the Cray-1 supercomputer delivered blindingly fast execution at 150 MIPS.
  • By 1982, 5 years later, a 6 Mhz 286 had about the same equivalent processing power as the Vax.
  • Sometime in the mid 1990’s, Cray’s benchmark was finally passed on PowerPC processors, as PowerMac’s emerged benchmarked at 150 to 300 MIPS.
  • A 1999 era Pentium III/500 delivered 800 MIPS of processing power.
  • A year later, in 2000, the Playstation 2 pumped out an astounding 6000 MIPS.
  • My 2002 vintage Athlon XP clocks in at 4200 MIPS.
  • And today, for about $200, you can buy a 20,000 MIPS processor.
  • *Note for non-geeks: MIPS stands for “million instructions per second”, a standard measure of CPU power.