Overpromise and underdeliver. It's not a way to build credibility. Or Windows Vista.
Yet, Microsoft continually does it. They feel under the gun to do some big things in order to win back some love. "Why do people hate us?"
Microsoft has been far less evil than most of its competitors in recent years, but people still do not trust the software giant. Today's release of Vista Beta 2 now with even less features should be an indication that releasing software when it is "done" is a failed software model.
Done in software is like done in life - if you are done, you are dead.
Windows Vista would have been released and in use by now if they would have focused on its core needs and then added functionality in the future. But Microsoft has an 18 month release cycle for even simple software. Why not strip out the cool new features and build an OS that does what XP does now - and allows easier plug-in extensions? Simply release new extensions once every two months.
- New stuff to play with = new interest.
- Promised stuff not delivered = disappointment.
I don't know about the rest of you, but my only real complaints right now about Windows XP are that it is gigantic (how compatible with Windows 3.1 do I need to be?), it has the worst error messages conceivable, and it is hard to administer (no real automation or scripting). Other than that, it's stable (which beats all Windows before it), relatively secure, and predictable. Oh, I also hate that I can't change the IRQ settings on my laptop.
But Windows Vista has no need to include a bunch of new bloat. The Peer to Peer system referenced in the Microsoft Watch article above would be cool - and would be a nice plug-in. But now, a plug-in that would make people happy in the future is seen as a failure today. All because Microsoft overpromised and underdelivered.
... you reached for the secret too soon / you tried for the moon / shine on you crazy diamond.
Photo: Eduoardo
Jim, well I'm well behind the curve on keeping up with what's happening, but I like the title and content of this posting. But I think that obesity is not just a microsoft problem. Firefox, that love child of the browser-istas, is also a behemoth. What ever happened to Small is Beautiful?
I know, I know it's the way of the world of innovation and the new: try this!, hey man, try this!, no, try THIS! Just "super-size" my plug-ins.
But your thoughts make me think that what would be really useful is a small (there's that idea again) kernel that didn't change much. A core of software that was used over and over and over and that was reliable, and that had error messages that are useful to me. Instead of just crashing, let me know what state my data were in. Give me some hints on how to get back to work, or play.
This might let me then import all the plug-ins I wanted, or maybe only the few. This way developers might be able to make promises I could believe. They could deliver new tools I could try. And I could throw it all away and start over with the small clean kernel. I could take the risks and see what happens. This isn't a new idea. Heck, VMware provides that now. Don't we already architecture for this now? I don't know; I'm just askin'.
Posted by: Bill Anderson | 12 June 2006 at 18:18
Hey Bill,
Well, theoretically Linux is just a kernel with a buncha plug ins. But usually you download it as a package (Red Hat, Umbuntu, etc.). But the difference here is that the package has some inherent value to you.
I really would like to have great P2P stuff in my Windows distribution - but I'd rather that it work initially.
As it is, I never use MSFT stuff until it's been out for at least 12 months. I am not unique. That's a really lousy customer confidence indicator. It's assumed that the MSFT software just won't work right initially.
I really think that's a Waterfall issue.
Posted by: Jim Benson | 13 June 2006 at 11:53