Top critical review
2 people found this helpful
Not very perceptive techie material
on 24 June 2010
Most of the action here, assuming it's not updated, is 1990 to 2000. The end-point seems a bit dated now - Amazon gets a mention, but not Google or Windows XP or e-Bay. Modern solid state disks and I think narrow screens didn't exist. There's no anticipation of the huge crash which happened soon after this book was published - though e-Trade is in - and which can be dated at about 2000. He doesn't predict spam and the Nigerian frauds. On the other hand, Mosaic, now almost forgotten, was at that time one of several browsers, and Windows 98 a new product.
I was interested to try to find how much truth there was in the idea that Berners-Lee 'invented' Internet - any spectacular invention tends to have many inventors, in fact. The first thing to note is that 'internet' already existed - Berners-Lee claims to have invented the 'web'. He worked at CERN where of course there were assorted incompatible computers made by different manufacturers. There were also intranets and email. I think Berners-Lee understates by a huge factor the way that de facto standardisation made things easier. For example, everyone uses 8-bit bytes; Mac hardware now is the same as IBM PC hardware, though just one fixed form of it; and the fact most PCs now use the same chip makes it far easier to write common programs. There are standard connections like USB of various types and 'firewire', and their earlier versions - RS232, Centronics, whatever. One very important hidden aspect which Berners-Lee seems not to have noticed is language: all this work was done using the Latin alphabet, and mostly in English. Japanese, Chinese, and other scripts weren't used; even accents, as in French and Vietnamese, weren't taken seriously.
As far as I can see, Berners-Lee managed to get people to use standard 'protocols' - things like IP and ISP and ways of doing things which are highly technical - how difficult to write on these topics without the in-house 'dedicated' jargon! HTML - hypertext mail link, the plain (but Latin alphabet) text plus commands in <these brackets> - was a relatively small part of the action. Sun's Java language, presumably relying on the standard chip, allowed little programs to be run (and introduced the possibility of 'viruses'). The author always talks in a mystical way of how the web is out there, and everything is accessible anywhere, and yet this can't be true, because (as he points out) to check up on say mysite.com must need some sort of look-up system. He also seems to understate the sheer quantity of cables, wires, satellites, transmission systems, hardware etc etc which must be needed. It's a bit disappointing to find hints of misrepresentation. Another aspect is his rather wounded defence of not making money from it - there's an account of a live TV interview which he clearly hated.
What he doesn't say is that CERN was staggeringly expensive, and in fact may have been a waste of money, like NASA. These people at CERN were in a privileged financial position. In fact it's possible the web will be CERN's only legacy.
In between the techie stuff is the human material, mostly rather affectionate descriptions of assorted hardware and software types, and business people typically at shows trying to make sales. However, in my experience, in real life many of these people are grasping and egocentric, and I suspect his accounts are like actors and 'luvvies' praising each other often through clenched teeth.
He's quite good on historical parallels - e.g. he regards tables of contents and indexes, in books, as hypertexts; and he compares tied-in software with a TV that goes straight to one channel and displays it better than others.
There's also intermediate stuff on e.g. censorship. And on secrecy - he described the public key/ private key system but to be honest I couldn't make sense of it.
So - interesting but with bits missing.