About this site
















RPC over HTTP via XML

Friday, February 27, 1998 by Dave Winer.

Good morning!

Today I'm writing this DaveNet piece with WordPad. It's an interesting experiment. Ever since I changed my primary platform to Windows it's been an adventure. I had already found the best writing tool for me on the Mac, it was MORE. Now I'm a wanderer.

I'm more flexible. WordPad is a nice piece of software. I can hear the people at Microsoft ducking, waiting for me to say, but... But it's fine. It does the job. I'm writing. It's receiving. All is cool. It sure is a heck of a lot nicer than SimpleText. And of course, since it's 1998, not 1984, it should be.

Can Java defy gravity? Permalink to Can Java defy gravity?

Galileo figured it out, why can't we? Only structures stay up (for a while). If it's built on hype, gravity rules, and what goes up must come down. Hype always gives in to substance. Java can't defy this rule, nothing can.

Last April, Sun announced an aggressive plan to put a user interface on Java. Now, ten months later, they're beginning to deliver. That's an incredible accomplishment. It takes a long time to develop good user interface toolkits. They're working quickly, it seems, from my point of view.

Last year, the industry had trouble distinguishing Java's promises from delivery. It was given extra credit, it was assumed it could zoom thru all the problems that previous platforms took years to move thru. We should have been able to neutralize the promises of Sun officials by asking tough questions and demanding straight answers.

Instead the headlines captured the euphoria, Java Defies Gravity! every headline seemed to say. Netscape would do a browser entirely in Java. One version of the source could would run anywhere. That was their answer to Microsoft.

All of the promises of Java are evaporating now. It will take many years to mature all the APIs and interfaces. It's going to be a simple world for a long time. That's the reality of Java. It can't defy gravity.

That doesn't mean it can't take its place among other software that does things it can't do. A negotiation is taking place. Java came in like a bull in a china shop. If they get pragmatic, Sun can develop a business around Java. But they may be the last to see that the bubble is bursting.

Is Push the next big thing? Permalink to Is Push the next big thing?

It never was. It was so transparent. If you just thought it thru, you realized push was just email. Instead of creating new protocols, CDF, DRP and OSD, push could have built on SMTP and POP. It would have been a no-brainer, it would have bought them so much compatibility, and since Push was so closely linked to Java, it could have given Java a great email client and server, which it still lacks.

I remember a panel presentation at InternetWorld last March. Microsoft's Brad Chase on stage, and representatives of Backweb, Pointcast, and other companies camped out in the then-hot push market. It was positioning and nothing more. We now know how it turned out. The protocol they were promoting didn't happen, the users didn't want it, and the hotness of the companies on the stage (and off) evaporated quickly.


It's important to take another look at Push because XML is getting the same rap. If Push is a fad-gone-by, can XML be far behind?

Now I'm hyping XML because I believe it won't go that way. I promise if I'm on a stage with Microsoft it won't be on the same terms that Backweb and Pointcast stood up with them. Right now I'm a believer in XML because I believe it empowers everyone, not just Microsoft. If I think it's going in a different direction, I will tell you.

Right now I'm still a believer.

COM Permalink to COM

There's a technical revolution hidden inside Windows, a far-reaching architecture, called COM. Unlike other buzzwords, COM has substance, a long history, but has suffered from Microsoft's mishandling of the marketing job, and the industry's cynicism about Microsoft.

Even so, COM *is* Windows. It's the glue that connects Windows together, apps and .asps, system software and tools. It's Microsoft's biggest strength, and therefore it's the thing they need to defend the most. COM is their runtime, and runtime is a big thing with Microsoft.

RPC over HTTP via XML Permalink to RPC over HTTP via XML

OK, now look at this situation with different colored glasses. The Internet is a platform. A platform is made up of tools and runtimes. Internet tools run on all kinds of operating systems, as do the runtimes. The beauty of the net is its simplicity, its ubiquity and its lack of a controlling vendor.

TCP is the runtime environment of the Internet. It's deeply competitive with Microsoft, and it's larger than Microsoft. The stronger TCP is, the more outside of Microsoft's control it is, the more powerful everyone else will be.

The Unix people have been in a struggle with everyone else over who owns the Internet. I've been writing about this for a long time. It would be great if they could let go of the struggle and realize that the Internet is owned by the universe, no single operating system or flavor of operating system can contain it.

This is important because there is another layer coming on the Internet, a very simple one, that can build on COM, on Windows and elsewhere, and provide a flat playing field for everyone.

It's RPC over HTTP via XML. I believe it's the next protocol for runtimes. We'll be posting more notes on this on Scripting News over the next few weeks. Please, if you care about the independence of the Internet, tune in. Thanks!

Dave Winer

© Copyright 1994-2004 Dave Winer. Last update: 2/5/07; 10:50:05 AM Pacific. "There's no time like now."