Say Yes to Everyone
Wednesday, April 16, 1997 by Dave Winer.
It was Netcaster day yesterday on the Net. A press release from Netscape. Sun briefing the analysts. I spent a fair portion of the morning reading press releases and scanning the web for information.
I learned something important -- that push is bigger than it seems. True, when it's reduced to its implementation, it's just another file in a website folder, or a new server that works much like a web server.
But push is bigger because it says that there will be another format after the web. I don't doubt this at all.
If you're building a web presence you want an insurance policy that you can easily make the transition should an attractive proposal come along.
And, judging from the press releases, a lot of content developers want to try new stuff out. So if you make content systems, as we do, you want to offer your users a choice, a way to easily support new formats as they come along.
Yesterday I found my place in the new order of things. Even if we're paying more attention to big corporate names these days, it's possible to cut thru all that and put a system in place for evolving to new protocols.
It's possible to get into agreement with the world. Here's how...
In marriage you can't say yes to everyone, but in technology, one way to win is to get into position where you *can* say yes to everyone.
We said yes to Microsoft. It took a couple of days to adapt our content system to Microsoft's proposed standard format for channel definition. We announced this on the Scripting News page on Monday.
People assume that we're aligning with Microsoft. We are! But don't miss this point -- we also want to align our software with Pointcast, Marimba and Netscape too.
CDF was very easy to support. A new table at the top level of a website contains the addresses of the dynamic pages on the site, with descriptions, and time information for the life and frequency of the dynamic pages.
We built a script that runs at 2AM every night, rebuilding the CDF file for the site. We're on solid ground. Bottom line: DaveNet is a channel now!
The structure our developers create is general. For each format a different script. Not a big deal. The reason it's easy is because we have a framework.
Just like object oriented programming techniques make changing your mind easy for software programs, a web content framework has places to hook in new formats without changing your workflow.
It's nice! More value for our users. And a chance to say yes to every proponent of push, or other new ideas that might come down the road.
Yes, yes and yes.
We've looked at both Pointcast and Marimba's format for push channels, and we can do the same thing for them, we think. We're working on the Pointcast stuff, we still need a hook for Castanet that triggers a publish event on the server, and Netscape's format is still unspecified.
Switching gears, but not really...
Craig Cline, firstname.lastname@example.org, is an editorial exec at Seybold Seminars, and a friend and coach. Craig and I sing a similar tune about where websites are headed. That's the basis of our friendship.
He teaches me about the publishing industry and I try to do new software that fits his vision. It's an iterative process.
I like to build on someone else's vision to scope out a problem, and then refine the implementation until they're happy it meets their needs.
I've done this kind of collaborative development in the past, developing a game with one of my best friends, Bernie DeKoven, email@example.com, who's now a game designer for Mattel, doing new software and hardware and websites around a very popular platform -- Barbie!
I did the same kind of collaboration with Chris Gulker, firstname.lastname@example.org, who was a system developer for the San Francisco Examiner in the early 90s. We worked together to automate print production for his newspaper using Frontier and Quark XPress, which was (and is) deeply compatible with our scripting environment.
Now Craig and I are talking about how the print industry moves onto the web. I believe the efforts of 1996 and early 1997 are on the right path, but more software, publishing system software, is very very important for the next steps.
The print publishing industry went partially electronic in the 80s and early 90s. Macs and desktop publishing software replaced expensive typesetting equipment. But the expensive publishing software still runs on minicomputers, and the presses are still huge machines that distribute paper and ink using trucks and airplanes.
The next step in the process completely replaces the publishing system software and the presses and trucks and airplanes. Production costs will go to virtually zero. Lead-times will go to zero too. Time becomes a much more flexible element.
Some players won't keep up, and new players will enter. It's another publishing revolution, and it's even bigger than the one that happened the last time around.
The web works now but we wonder about the future. Electronic transport mechanisms will evolve and change. If you're in the electronic publishing business, being able to easily try out new distribution mechanisms will make a difference.
When a new transport comes along we all want to be first on the block to flow our content out thru the new protocols. And as a software vendor I want to make that easy.
We've been working with Craig on a new website. Rather than play with theories, we're putting up the system and hope to bring the idea online in the next few months.
Here's the plan. It's going to be an integrated system with editors submitting pieces via email and an editor in chief who owns the front page. Stories are linked in as they become available. The site is automatically rendered using our Frontier-based Content Server software.
The front page is edited with wizzy software that's connected to the Content Server. There's a boiler room (our stuff) and a conference room (the wizzy editor) -- and happy readers (we hope!).
You can preview a prototype of the Seybold news site thru this page:
Please take a moment to read the design guidelines on that page. Thanks!
Yesterday I saw the Gartner Group in the list of content vendors supporting Netcaster and thought of Seybold Pubs and the website we're building.
I wondered how easily Gartner would make the switch. Then I understood my job -- to make it totally easy for Seybold to support any new protocol that comes along.
That's where we fit in in the feeding chain, inbetween the technology leaders and the content leaders.
Let the best format win! I say. And let's have lots of new formats.
PS: If you're in New York next Wednesday, 4/23, come to DaveNet Live, 7:30PM, Room 1E04-5 at Javits.