I notice that leading research group, IDC, has come up with a statement of the blindingly obvious. It has suggested that network bandwidth is more important to users than processor power. Gosh, I'd never have thought of that.
In fact, of course, it is something I have thought of many times. Indeed, I have written many times, one way and another. I would take IDC's suggestion one step further, and suggest that it is the industry's (and the users') lust for processor power - and the sloppy applications code that this brings with it as a by-product - which is the direct progenitor of the desperate need for more bandwidth.
We have word processors which are worthy of half a Meg of disk space consuming 10 or more Megs. We have applications which miraculously turn 10Kb of valuable text into a couple of Megs of fart and flair artwork all designed to project a corporate image. All that these do is project an image of companies that value their own egos more than the resources they are consuming to force those egos onto unwilling (or bored) recipients over networks.
Yet here we go with yet more power to the processor. Intel, not content with launching the MMX in January - prompting the inevitable rush from the PC makers to be first on the block with the machines - has now done the same trick again with Pentium II. For any user looking to have the latest technology, it means buying a new machine every three months.
That is barely enough time to get all the bloatware applications loaded and running before they have to be loaded onto the new machine.
What is more, the inevitable seems to have already happened. No sooner does Intel launch a new processor than someone points out that it has (or at least appears to have) a bug in it. And, as fits the tradition, it concerns the chip's ability to do complex maths. This happened with the 286, 386 and 486, so is now hardly newsworthy, except when Intel tries to pretend it hasn't happened.
Then there is the problem that, even if the power is there, the applications and operating systems are not really written to exploit it properly. Microsoft is always having to play catch-up, and the only way it can be done is by adding more bits to systems such as NT. For example, high-end NT systems companies such as Unisys and Data General are already producing symmetrical multiprocessing boxes with more processors than NT can handle. In the case of Unisys, its latest Aquanta system can offer up to 10 Pentiums, but NT can only handle four - at least for now.
The IDC position is that the Wintel platform is in the process of losing much of its market dominance, to be replaced by Java-based systems. While it may be early days for the Big J, it certainly does have some things going for it, not least of which is machine independence through the Java Virtual Machine (JVM). This can run under just about all of the main (and not-so-main) operating systems, and on a growing number of hardware platforms.
Indeed, I suspect it could easily be made into a plug-in for those it eschews.
This raises an interesting side issue which could also be a thorn in Microsoft's side. I noticed the other day a story about, Bournemouth-based Fernlink 2000, which is selling add-in cards for old PCs. The add-ins overcome the problem of getting old PC hardware through to 2001 without dying. For many users with old PCs, this saves the only other option - upgrade.
How long, I wonder, before some enterprising company comes up with the idea of putting the JVM onto an ISA or PCI board, together with a processor most suited to running the system. If the price is right, many PC users could then bypass the arguments of whether the fastest high-powered PC with the latest resource-consumer operating system is better than an NC?
They could have both, and for much less investment.
And, yep, it'll run Android rather than RiscOS
US engineering giant's cost-cutting outsourcing plan is on the rocks, according to insiders
HP Envy X2 laptop only affordable if you've got loadsamoney
Counterfeit code-signing certificates enabling hackers to hide malware being sold by cyber criminals
Certificates can be used as part of layered obfuscation to evade detection by anti-virus software