Progress: n. Steady improvement, as of a society or civilization; a believer in human progress. American Heritage Dictionary
I'm told that August 12, 2006, was the 25th anniversary of the launch of the IBM 5150 better known to you and me as the "personal computer." Happy birthday, PC.
The original PC had a 4.77MHz Intel 8088 CPU with 16 kBytes of RAM. There wasn't a hard disk, just a 5.25-inch floppy, and we had to wait for the PC/XT before we had access to the truly awesome 5-MByte "Winchester" disk.
Goodness, we've come a long way since then. Or have we? My desktop at home has a 3GHz CPU, which is a long way from being state of the art. I've got 1 GByte of RAM, which is fairly typical these days, and a 500-GByte hard drive, which is probably on the large side for a home PC.
So compared to the 1981 standard:
- I have over 60,000 times more RAM
- I have over 50,000 times more disk space (compared to the XT)
- My CPU is clocked at just under 1,000 times faster
- My system cost me about £450 (US$852) (self-build), compared to the £4,000 ($7,578) that IBM was charging in the U.K. in 1983 (the year the PC was launched here)
- The 1-GByte SD and CF cards I use in my digital cameras have around 7,000 times more capacity than the 5.25-inch floppy on my old XT.
The CPU statistic isn't really that meaningful on its own. The original 8088 was an "8/16"-bit CPU. In other words, it had 16-bit internal registers, but only an 8-bit bus. IBM had saved a bit of money by using this chip, instead of the rather better 8086, which was "16/16."
The Norton System Index (SI) benchmark for PCs of that era set the original XT at 1.0, and everything else was measured against it (because everything else was faster). The latest number I can find for an SI benchmark is 890, for a Pentium 266MHz MMX CPU (circa 1997). Goodness knows how a modern CPU would really compare against an XT! If you find a comparison, please let me know.
So the bottom line is that over the next few days you'll probably see plenty of news items on this silver anniversary, and lots of people will be pretty smug about how much progress we've made.
But I can't help thinking how little progress we've made, or in fact how much we've regressed, in terms of the security and useability of PCs. Back in 1986 when I was using my first networked system (a clone powered by an 8MHz over-clocked 8086) my PC booted, logged onto the network, and was ready to use in about 10 seconds, with another 10 seconds to crank up Banyan Mail. Today it takes between 2 minutes and 5 minutes for my laptop (a modest 1.6GHz Pentium M, only 200 times faster clock speed and 1,000 times more RAM than that PC clone) to finally become useable, after the Windows boot sequence, loading all the AV and anti-spyware scanner stuff, and the disk activity light finally winking off.
I know a lot of people blame the bloatware phenomenon for the tedious delay in Windows boot times, and the reduction in the useability of PCs. But here's a fascinating article by Joel Spolsky that argues that major applications are actually getting smaller in real terms. I agree with Joel's logic, because what should be happening is that CPU, memory, and disk drive performance increase faster than software developers can use them up with bloatware. But I'll leave it to you to decide if you agree with the old cliché, "What Intel giveth, Microsoft taketh away."
And in terms of security it's clear that the situation today is worse than its ever been. More computer users get infected than ever before, both in the enterprise and in residential networks.
So, any suggestions for a new definition of progress?