Slashdot Mirror
Hailstorm: Open Web Services Controlled by Microsoft
richard writes: "Clay Shirky has an interesting article on Hailstorm on OpenP2P.com. He looks especially at how MS mixes decentralization with strong control of third party development and user data. Think of it as an authentication-centric, rather than hardware-centric system." A very nice analysis, neatly mapping out Microsoft's plans and how they intend to control the system. Well worth the read.
I don't know if Hailstorm will work. It seems like a longshot to me, but I'm a pauper and Bill Gates has an impressive track record with world domination.
.NET and Hailstorm don't fly, they're screwed. Of course they're so big it would take them decades to waste away. The plan offers vast rewards, potentially, but it has enormous risks, as well.
But this does illustrate one of the big differneces between Linux and MS. MS has a master plan. They're building server software, server farms, development tools, business alliances and strategic partnerships. They have a business plan and a technological plan, and they both seem to fit together. Even though the word is going to come off as a joke after all that's happened, this stuff is innovative, in kind of an Orwellian sense. Especially as a business plan.
They looked at the future and decided it was going to be objects running out in the cloud, talking to each other in complicated ways, and they tried to figure out the best places to build the toll booths.
We don't have a plan. We've yet to come up with a really good business model. We've been making incremental improvements to a 70's operating system. Individuals or small groups have ideas and they make it better in a small way. The result is a lot better than anything they had in the 70's. But it's a gradual process of accumulation. No one comes down from the mountain with the new direction.
The first time I realized that Linux had super powers was when SLS dropped the ball. They were an old distribution. For whatever reason they just stopped doing it. And Slackware stepped up to the plate and took over. If Linux had been commercial, SLS would have killed it in its cradle. But you can't kill Linux. Debian will be moving along long after VA Linux and Red Hat have succomed to financial reality.
MS has a plan. Gates says he's "betting the company" on it. I don't think he's kidding, or that he's wrong. If
This is not a clash between rival technical systems. It's about world views.
I've got to be honest, I love the megalomaniacal scope of MS's plan. They're thinking the way the the guys who built the pyramids thought. Part of me wants to email Bill and say, "God speed, you magnificent bastard!"
But ultimately, I think he's going to fall on his ass.
Why?
The OS monopoly was achieved in an environment when no one understood the dynamics of the business. There's the famous story about Bill trying to sell out to IBM for a relative song, and IBM turning him down. That suggests that neither side knew what MS had.
Translation: the lucky SOB *stumbled* into it. And he was helped along by the fact that no one else understood how big the prize was either, or even that it existed at all.
There's another famous story about Lotus dissing Bill, rudely pointing up the difference in the bottom lines. People didn't understand the dynamics of lock in back then, that the person who controlled the OS had leverage over the application market. These were smart guys, the best and the brightest in the industry.
The article at the top of this thread is first class. People are thinking like chess players when they look at the business now. Which squares on the board do you need to control if you want to win? The word is out, the guard towers are fully manned, and no one is going to be stumbling into anything this time around.
No one is going to create a strategic dependence on MS if they can help it. Especially now, when the XP license server shock waves are about to hit. These guys are lining up their ducks to do the same to thing to their customers that OPEC did to the West in the 70's.
It's going to be an intersting thing to watch, though.
It's amazing how the overwhelming impression someone would get from all these slashot discussions is that your typical /.er is a technophobe.
This kind of thing is *cool*.
Should Microsoft control it? Of course not. But there should be a lot more enthusiasm on these boards for the capabilities these things represent; it's this sort of universal capability which is the future.
Is there a security risk? Of course... but you could say the same thing about the postal service, the telephone, credit cards, etc etc. It's *going* to happen.
But the OSS has to stop saying "Boo. Stop Microsoft. They're evil." and start saying "What Microsoft is trying to do is cool, but what we can do is *better*."
That's (partly, anyway) why *nix dominates the web... Apache (and its predecessors) used a protocol (http) which was just plain better than all the commercial alternatives for information dissemmination, and when the commercial companies turned around and came to see how great the whole web concept was, the OSS community was already there.
--
Convictions are more dangerous enemies of truth than lies.
Convictions are more dangerous enemies of truth than lies.
- Nietzsche
You could emulate the interface, but you wouldn't have the keys to validate your authentication.
.NET XML Web Service spec, which is fine, I suppose, unless you like IBM's or Sun's Web Service ideas better. Without the common Passport authentication, Hailstorm is no longer Hailstorm.
So your gnuPassport couldn't be used on sites which only recognize Microsoft's Passport.
And if you can't link up with other sites, you lose a lot of the functionality of Hailstorm and are just left with the
((And if you did get your hands on the keys, it would mean the collapse of the entire Passport security scheme.)
--
Convictions are more dangerous enemies of truth than lies.
Convictions are more dangerous enemies of truth than lies.
- Nietzsche