Slashdot Mirror
Pointless IT Innovations Considered Harmful
Makarand writes "According to a comment column in the guardian innovations in IT are most often simply more trouble than they are worth. Most innovations in IT today are platform specific and are easy to
come up with in the computing fields. Innovating gets easier if the platform sticks around
for a long time. These innovations accrue incompatibilities making it difficult for users to switch platforms and absorb the costs of switching to a new platform. Users will not switch to
a competitor's product if they believe that their platform will be later updated to deliver the same benefits."
innovations in IT are most often simply more trouble than they are worth
I see this everyday. Not just in the areas that they are talking about in the article. I see it most commonly on enterprise applications.
For example a company will have a mainframe based app that they have used for years through a terminal emulator. Everyone knows how to use it and flies through the application often typing several screens in advance. But, some bright spark thinks that green screens are passe and insists on "updating" the application. They spend LOTS of money developing some gui database application or, worse yet, some browser based interface to the application.
Suddenly, the application is slower than molasses, going up hill on a cold day. No one knows how to navigate the new interface and productivity takes a major dive.
Naturally, the bright sparks asssume the problem is old hardware and spend another fortune upgrading equipment to get performance back to where it was before. It's a total waste of time and money, not to mention that it pisses off the user community in a major way.
Actually, though, the big value added with Linux will, for the foreseeable future, be the assurance that you will not get sucked into some long term recurrent license fees.
This shows a remarkable lack of insight into how similar things today are to a few decades ago. A few decades ago we had IBM mainframes and terminals with local blockmode editing; today we have web servers and PC's with web clients with form-filling capability. Are the PC's capable of much more? Yes. Are they often used to do much more? No, not really. The only real difference (ignorning frilly graphics) is that Internet Exploder and Netscrape crash a whole lot more often than a 3270-type mainframe terminal :-)
The article's point, IMHO, is that change for change's sake is not good. Sometimes change is clearly the right thing to do - for example, replacing job-control language with a modern operating system is (usually) the right thing to do, as is replacing assembly language with a high-level language for writing applications. The gains in reliability and maintainability make the effort worthwhile. However, change just for the sake of change is often - usually? - leads to a degredation of reliability and maintainability, rather than the other way around. Companies that pursue a will'o'the'wisp often rush into a bog. The point is, it's not too much to ask managers to perform some basic cost-benefit analyses before they sign onto the latest fad.
[this
It's like: rather than replacing the entire baby, you only change the diaper. However, after 20 years, it's time to rethink that stategy.
Yes, I'd say it is. If you're still changing diapers after twenty years, it may very well be time to think about replacing the baby.
Lost: Sig, white with black letters. No collar. Reward if found!
They obviously have the same playbook, and we all know which one: "#199. I will not make alliances with those more powerful than myself. Such a person would only double-cross me in my moment of glory. I will make alliances with those less powerful than myself. I will then double-cross them in their moment of glory."
One line blog. I hear that they're called Twitters now.
Today's ERP systems (SAP, PeopleSoft, Oracle, etc) suffer from the exact same problem: they promise you the moon (and many times actually deliver it), but once you depend on it you're completely stuck with it. In the case of SAP (and the same case happens to other ERP systems) if later you want to change something it's going to cost you big. Plus you usually pay very high consulting and maintenance fees.
.Net.
The same can be said of other packaged applications which do not make public their data storage formats and/or communication protocols.
This is why I think it is such a big deal to have (1) a true cross-platform executable platform (i.e.: java), (2) a true cross-platform communications protocol and data interchange (i.e.: XML), and whenever possible (3) a comprehensible and standards-compliant-as-possible data repository (i.e.: mySQL, Postgress).
Note that regardless of the article being viased or not 9as some other readers here point out), the reality is that many IT managers are beginning to realize this now. This is why the huge push to Linux, Java, PHP, and XML, and many Open-Source technologies.
It is also why Linux, XML, and J2EE (Java 2 Enterprise Edition) has had such a success, and why many IT managers are thinking twice about Microsoft