I've been following all the hullaballoo about the Echo Project. Unfortunately there's been more flaming than at a dragons convention. But one of the few calm voices amongst all the hot air is Jon Udell, who today wrote a brilliant essay that got right to the heart of what RSS means.
Jon reminded us that technologies such as RSS are all about helping ordinary people "communicate more easily and more effectively". That's what web technology can achieve...although perhaps in the Echo debate we're not eating our own dogfood :-)
The challenge for web technologists in 2003 is to develop tools that enable people to write structured content for the Web. Once non-technical people can easily do this, we've taken a major step towards the Semantic Web. As Udell puts it:
"Ideally XML, not raw ASCII text, would be the stuff that was written, and refactored, and then mined to produce coherent views. We have no tools that come close to enabling that to happen.
Such tools, combining the power of XML with the flexibility of freeform text, and operating on a universal canvas, are what will really drive mainstream adoption of a two-way Web."
Dave Winer has done an awful lot of work to get us to the cusp of the two-way web. He created a weblog authoring tool and he co-invented personal publishing standards like RSS and XML-RPC. But most of all, it's Dave's ideas and his vision for a two-way web that I value. He is carrying on what Tim Berners-Lee started. As Dave wrote earlier this year:
"Like a lot of technologies people told big stories about something called hypertext, but until Tim Berners-Lee came along there really wasn't something for ordinary people to use. He pushed aside a lot of hairy technical issues, didn't even try to solve them, and cobbled together something that was brain-dead simple and incredibly ugly, and it worked and it was wonderful."
Tim Berners-Lee always wanted a read/write Web and Dave Winer has done more than most to help make this dream a reality. So I hope people lay off Dave and let him do what he does best - create solutions for real people to use.
Jon Udell is also doing a great job in elucidating what we really need in order to achieve a Semantic, two-way Web. Regular people don't want to hear lots of flap about formats and APIs. What we want are easy-to-use XML-based writing tools, and applications to manage our information and subscriptions. Now that would make Mr Safe real happy!
One thing that is definitely "funky" is the blogroll in Radio Userland. I updated my blogroll.opml file last night but - no matter what trickery I do - the changes won't publish. That is why my external links look a bit odd right now.
John Robb recently announced that Radio will soon release a new version, which is exciting news for us Radio fans. I look forward to the new features. But I also hope they address the little things, like the continuing blogroll funkiness.
Radio is a wonderful product and I like playing around with it. But there are some kinks in the publishing process which need to be ironed out, pleeeease :-)
postscript: ha ha, typically as soon as I published this I discovered how to force my blogroll to publish. In my browser, I browsed to the location of my blogroll: http://radio.weblogs.com/0105304/gems/blogroll.opml. I noticed it was still displaying the old version of my blogroll. Hmmm, so I refreshed the browser and it updated to the new version. I thought well maybe that will finally update my weblog. Nothing else had worked and I had tried everything - including deleting the blogroll.opml file from my gems folder and its reference from my homepage template, clicking 'Radio --> Publish --> Entire Website', adding the opml file back in, re-publishing, etc.
So anyway after refreshing the blogroll.opml page in my browser, I re-published this post and - lo and behold - my blogroll had finally updated on my homepage. Praise be. I can go and enjoy my Sunday now :-)
Jon Udell on RSS: "It's about a new way of communicating, one that's defined by personal publishing and subscribing, and that empowers writers and readers as never before."
Amen to that, brother.
People are trying to change RSS into something called Echo. If you want to know why, then I recommend you check out Jon Udell's conversation with Mr Safe. But also read Dave Winer's post in reply. And if you really have to, browse the Echo Wiki.
I admit it. I regularly check my stats at Technorati and Blogshares, plus I do some search engine checks now and then. Everyone does it. I have to say I'm not exactly setting the world on fire in terms of popularity. I'm probably a 'C List' blogger at most :-)
But I am noticing my Google popularity is increasing for the phrase "Read/Write Web". I'm now at number 2 (as of this writing) and only Dan Gillmor's weblog post titled "The Read-Write Web" is ahead of me. I haven't been officially tracking this, but I do recall I was at number 5 a week or so ago, and buried in the back pages a month ago. So I'm moving up the charts.
I also found a directory of Google API tools. I like GoogleDuel, which pits two words or phrases against each other. I discovered that "two-way web" is 6 times more popular than the phrase "read/write web". That is due to Mr Winer's influence. Also "browser/editor" is 6 times more popular than "universal canvas"...
In my recent articles I've explored the concept of the Universal Canvas, a term made popular by Microsoft when it launched .NET in 2000. But things just got interesting, with the news that Microsoft will phase out its Internet Explorer browser as a standalone product. Internet Explorer will be integrated it into Microsoft's next-generation Operating System codenamed Longhorn.
But what does that mean exactly? How will the Internet Explorer web browser be integrated into the OS and what effect will this have?
In a nutshell, it means IE components will be converted into CLR components. CLR stands for "Common Language Runtime" and it is the engine that drives the .NET platform. The CLR sits on top of the Operating System and provides developers with a set of services.
Ahmet Zorlu speculates that CLR components such as Web Services clients and P2P applications will be introduced into the browser, and current IE components such as the plug-ins and Active-X controls (e.g. Flash Player) will be converted to .NET.
Frans Bouma also thinks the CLR is where IE will end up. He says that HTML or other markup "will be rendered by components embedded in other applications, like helpviewer, blog readers and other tools. Such a component can be embedded in winforms as well, as a control."
But whoa, before we get starry-eyed for the future, let's step back for a moment and review Internet Explorer in its current form. Basically it is made up of a number of components. The two main components are called the WebBrowser control and MSHTML. I won't go into gory details, you can read Microsoft's documentation for that. Suffice to say that Internet Explorer is based on a component architecture - and what's more, as of version 4.0 this became "an integrated part of the Windows shell". Version 5.5 added lots of new functionality, including editing capability using e.g. behaviours and the "contenteditable=true" declaration. IE 6 is the latest version and it looks like we're stuck with it for a while - until Longhorn is released.
.NET is also a component-based architecture, but on a larger scale. Internet Explorer is like a small fish about to be fed to a larger fish called .NET.
This is what it comes down to - IE will become JAFWC (Just Another F*cking Windows Component). It will no longer be a standalone product that can be plugged in to any OS - it will only run on the Windows platform.
So what are the benefits of having Internet Explorer subsumed within the OS? It will have a much cleaner architecture - no more plug-ins and add-ons. And we'll finally get a decent edit control, which will enable the browser to once again be editable as Tim Berners-Lee originally intended it to be. The Universal Canvas may finally become reality, albeit in a Microsoft world.
Mark Pilgrim: "I take in a lot of raw data, synthesize it, and spit it back out in ways that many people can understand."
Mark Pilgrim and Neil Deakin are two very smart web developers, but more importantly they both have the ability to document complex web technology in laymans language - so that wannabes can learn it too. This is different to technical writing, which means documenting a piece of software for its end users.
One of the best ways to learn something is to document it. The beauty of the read/write web is that it makes it easy to do this, and easy for everyone to contribute.
Revision 24/7/03: I struck out the sentence about technical writing, because tech writing as a discipline actually covers writing for both wannabes and end users - and many other types of audiences too.
Dave Winer posts a link to a DaveNet from 2 years ago:
"If it were not possible to read my words without annotation, we'd have to invent a medium that allowed that. But in 2001 we already have such a medium, it's called the Web.
We have tools and servers and all kinds of runtimes on all kinds of operating systems.
We don't need or want another medium. So let's not screw it up.
I think that's what the writers are saying to the geeks."
Those words don't need any annotation ;-)
Micah Alpern asked via my Comments form: "Wasn't this term [universal canvas] first popularized by Apple with their failed OpenDoc program?" Only one way to find out and that's pay a visit to Google. I found a definition of OpenDoc, but I didn't see anything that had OpenDoc and Universal Canvas in the same sentence. Anyone know of a link?
Looking for more info, I took a ride on The Wayback Machine and travelled back to 1996 - the Apple OpenDoc website. This snippet from 1996-era Apple is a good summary of what I discovered about OpenDoc:
"You can combine features from your favorite soft-ware applications - including tools to handle text, graphics, photography, spreadsheets, video - even live data links and Internet connections - and use them in a single, simple work environment. The result: a work environment that's truly integrated."
I have to admit, it does sound similar to the universal canvas concept - in a mid-90's sort of way. OpenDoc was all about bringing different software components together into a single application. The Microsoft vision is about bringing together data from different applications, using XML - very post-2000 :-)
Also check out Apple's former product called Cyberdog - does this blurb remind you of something:
"Because Cyberdog uses OpenDoc component technology, it's completely integrated into the operating system and can be extended with other OpenDoc components."
Integrated into the OS, sounds familiar eh. On that topic, this 3-year old Macworld article also seems very prescient now. It was written by Wes George, soon after Microsoft announced its .NET strategy in 2000. Wes said this:
"The thrust of Microsoft.Net is to Vulcan mindmeld the operating system to centralized Microsoft servers by making the browser and the OS one piece of software. All access to information, services, or other applications are controlled from this "universal canvas." And the universal canvas is directly linked to Microsoft at all times."
I don't share the view that Microsoft wants to control us all via centralized servers. But still it is worth considering both the pros and cons of the "universal canvas" - particularly if it does end up integrated into the OS along with the browser.
Last night I wrote about the Universal Canvas. Today in my RSS newsreader, what should appear but a great post from Steve Gillmor on the same topic. Of course being a pro, Steve made his point way better than me. Microsoft has all the pieces, says Steve, to "create a browser-hosted read-write tool for sharing and routing information."
But the pieces are being fitted together to reveal a jigsaw puzzle that looks suspiciously like the Windows Operating System. As Steve puts it: "We'll get the long-promised Universal Canvas, but sorry folks it will have to be Windows end to end."
Steve also wrote in an earlier post that "Office is now a System, BizTalk is now a System (Jupiter) and IE is part of the Operating System."
All this talk (including from me) about the universal canvas moving away from the browser and into the Office/Operating system, is a little scary. The World Wide Web was originally meant to be a decentralized network of information where people could read and write freely, as in both free beer and free speech.
Sure the browser market has been largely controlled by Microsoft these last few years, but at least browsers run on the World Wide Web - and the Web is as universal as it gets in the digital domain. So where does it leave us if the future canvas for our browsing and creating is embedded in a "system", owned by one company, rather than on a universal network owned by no one? Is the Universal Canvas going to bypass the Web?
I've been Scobleized. Now I really am part of the blogosphere...yay :-)
