|
Top 10 Overhyped Stories for 2005
|
|
|
|
|
I am not a cynic by nature, but years of experience in the IT world have compelled me to make sure that everything passes the smell test.
As a result, I can often tell beforehand whether I am hearing marketspeak or the real deal. For instance, 5 minutes after I figured out
what Larry Ellison's network computer was, I knew it would never be successful. And so did Larry Ellison, given that Oracle never actually
built any of them. So with that in mind, let's revisit the top 10 overhyped, overmarketed, overbsed (if there is such a word)
computer industry events of 2005.
10. SOA
I love it when the hype is deafening. It means that it reached the climax and the technology is about to come crashing down.
Open any magazine targeted towards IT management (InfoWorld, Computer World, etc…) and you'll see conferences, workshops and
other get-togethers on SOA (Service Oriented Architecture). All designed to bilk customers out of their money. Btw, do you
notice that when an executive or an analyst speaks about SOA, they all speak in generalities, like they don't exactly know
what it means? That's because they don't exactly know what it means. Let's break it down. SOA is mostly about Web Services.
Ok? Web Service. That's what SOA is all about. If you thought Web Service was an overused $10 word for an RPC call dripped
in XML chocolate, then SOA is a $100 word. IT executives, afflicted with the Airplane Magazine Syndrome©, that read these
publications normally have only a cursory understanding of what it takes to make a system work (and hopefully perform).
Thus companies that rewrite all their perfectly functioning software in SOA fashion are likely to have a rude performance
wake-up. Web Services are SLOW. That's because their payload is XML, which is bloated. And because the nature of web
services is stateless, many times it requires more round trips to get the basic information. So Web Services are fine for
applications that are not performance sensitive, like general office apps (accounting, sales, HR, etc… ) But stick
them on anything that requires scale and performance… That's what Mozilla Foundation learned when they based their
Mozilla Update code on the use of Web Services - they just don't scale. MoFo has since switched to regular HTTP
for the task. Oh and interoperability, yes, interoperability. If the Web Services are interoperable, why in the world are there products to
facilitate communications between a C# client and a Java Web Service? So while SOA (or Web Services) clearly have a certain
place in the enterprise, please go ahead and rewrite all your software in SOA. Why? Because it keeps software developers
perpetually employed. (Hold done, why am I complaining?)
9. VOIP - This is the way we screw the phone company.
The attraction to VOIP is easy to understand. Both data & voice flow over the same data link. And chances are that eventually, among businesses,
it'll be the norm - the PBXes will be integrated into the overall scheme. A massive movement may happen on the consumer front as well, though,
not until home builders install as many CAT 5 jacks into homes as they do phone jacks today. Another problem to overcome is that many VOIP phones
take several minutes to boot up and come to a usable state (ostensibly to load up the various stacks, connect to gateway, obtain an IP address, etc…
However, the thought, that the consumers will be able to somehow use VOIP for free to replace the phone is preposterous. The phone companies
have been at the greed game for way to long to allow something silly like that. But I am already paying for my internet access and so my line is
already paid up, I hear the objections. That doesn't matter. Consider 2 facts: phone companies provide a large chunk of the world's internet
backbone and you want to take your money elsewhere, but still use their infrastructure. So, logically speaking, you already paid for you internet
connection, but you are taking their business away. So either one of 2 things may happen: consumer internet access will become more expensive as
the phone companies lose more PSTN customers or they'll block the likes of Skype and Vonage. Don't take my words for it - read what SBC wants
to do to you (search for Vonage on the page to get to the point quick). I suspect that other phone monopolies will be just as ruthless.
8. Apple switches to Intel
This event was one of the few that actually lived up to its hype. And the reasons for the switch were sound as well.
What did stun me was the turnaround of the legions of fanboys who were previously screaming "megahertz myth". They just ate
the yummy crow and went on bleating the next party line like the sheep in George Orwell's Animal Farm. So now they'll have
to quickly think up of reasons why CISC architecture is better than RISC and why Intel doesn't suck and why MMX/SSE is in fact
pretty good, and maybe even better than the Altivec. It will be interesting to see when AMD becomes the target of their derision,
because it isn't their new sugar daddy - Intel.
7. Firefox will take over the world
When Firefox 1.0 came out, I predicted that by the end of 2005 Firefox would garner at least 15% of the worldwide market share.
I based that on the following thought: probably 5% of the computer users are techies. Then another 10% of the users are in sphere
of influence of the individual techie (i.e. provides family/friends with tech support). Various pundits and zealots even went as
far as proclaiming the end of Internet Explorer. Well neither prediction occurred. You know why?
Because people are mostly driven by inertia. IE's shenanigans are not an annoyance big enough to most people that they will
actively go out and find a replacement (or would even know how). Most people (who aren't computer experts/hobbyists) are just happy that the damn
thing works at all. The whole situation reminds me of an old "In Living Color" sketch, where Damon Wayans is the Apathy Man -
a super hero with all the powers that a superego usually has, but with no motivation to use them. Despite being an unquestionably
better browser, Firefox basically failed to catch on among the people that are just home users and make up the vast majority of browser consumers.
Still, I would like it if Firefox got a larger market share, simply because it is a better browser and better products should win in the marketplace.
6. The case for IPv6. NOT.
5 years ago IPv6 seemed to be a shoe-in. There weren't enough IP addresses to go around and we kept on hearing that China and India need more.
Today, after the NATs blew the roof of the problem, all that the proponents of IPv6 can do is point to the mythical take-up of the technology in Asia.
This is akin to a failed US rock'n'roll band claiming they are huge in Europe. So despite the never-ending proclamations that it's imminent,
IPv6 is a solution in search of a problem. Oh and did I mention that the basic IPv6 packet is much larger than IPv4 one? Yeah, you get a nice bump in
bandwidth requirements. The only parties still interested in it are large corporations with control issues and hardware vendors who want to sell you a new set of routers, hubs,
network cards, etc… Still, I'd like to see IPv6 take hold one day, because, imo, everyone should have a static ip available to them.
Here is a good synopsis of IPv6 mess.
Next Page --->>> To the Top 5.
