The Boys of Woodside
Monday, June 7, 1999 by Dave Winer.
Lots of little snippets today!
I just got a Cobalt Qube up and running over the weekend. It's a gorgeous little translucent blue box that says "You want me!" I was over at my buddy Doc Searls house last week, and I saw it sitting on his desk, and asked if I could have it -- to my surprise he said yes! So I took it home, and after a minor configuration hassle, I have it running on my LAN and on the Internet.
It's a delicious little blue box, no monitor or keyboard, just power and a net connection. It's a web, mail and FTP server. It's also a Windows and Apple file sharing server. I've mounted it on both Mac and Windows desktops without problems. The whole thing is configured thru a web browser running on a workstation, so it's easy.
Security may be an issue. It has a little content management system built in, with an Edit this Page function (!) but when I went to view a colleague's page, I clicked on the button, and changed his home page. This is a problem.. (Of course.)
But get this, here's the punchline, the Cobalt Qube is Linux! So, never mind the the rep that Linux has for being hard to use. If you've got the approx $1K that the Qube costs, you can have all the power of Linux and be up and running in 15 minutes. No kidding.
I'm about half-way thru Alan Cooper's excellent "The Inmates are Running the Asylum".
It's such a good book, and very timely for me. I'm deliberately seeking more input on user interface issues, as a refresher, to help my team do better work, and to catch up on what I've missed while I've been exploring web development software. Cooper, along with Tog and Apple's Human Interface Guidelines, have been my teachers.
One of Cooper's theses is that the reason the web is so popular is that software that you run on the web doesn't require installation. Before the web, all software products required complex installation, and users and system managers rightly hated this error-prone and time-consuming process. But Cooper points out that it's a sham. He says "The only reason non-browser programs require installation is that's the way programmers have always done things."
He makes an excellent point! The web is simpler for the user, it gave them a feature that the software industry would never have thought to give them. But what a step backward! I have so little control as a software designer, after five years of working in the web, I'm only now feeling like I'm not stumbling around in the dark, and now that I can see what's going on, it's like GUIs never happened!
I predict a loop-back, that we can bring the simplicity of the web to desktop tools, marry the two, and live happily ever after. (Until the *next* revolution!)
Later today I'll attend the Office 2000 rollout in San Francisco, and meet with Microsoft people on some of the issues above, and perhaps others.
As an introduction, there was an excellent article in Saturday's SJ Merc on the web authoring features of Office 2000. "In any Office 2000 application, you can select Save to the Web from the File menu at the top of the screen. You can then open Web Folders and put the Office 2000 document directly onto the server, with Office 2000 automatically handling the chore of translating the Word document or Excel spreadsheet or PowerPoint presentation into HTML."
Sorry Microsoft, but this is a klooge on top of an old paradigm. The users are going to screw it up and lose documents at an even higher rate than they do now. It can be much simpler for people who choose to write only for the web. Microsoft must (or chooses to) do a balancing act, they serve the installed base, and make it easy for them to write for the web, while not taking away any features that relate to writing for printing.
But Microsoft's disadvantage is everyone else's advantage! We don't have to serve the installed base, we can create a new one. When it's all done, only a complete rewrite of the File menu will work. The more I focus on UI design, and the more we experiment in our own development at UserLand, the more sure I am of this. Only when you forget about the local file system can you make web writing truly easy for writers and system managers.
The Woodside phenomenon keeps building.
Down at Buck's everyone (ie Jamis) has been talking about David Kaplan's upcoming book about the power people of Silicon Valley, The Silicon Boys. The story is centered in Woodside California, where many of the powerful people live, where their kids go to school and where they eat breakfast at Buck's. (Another plug!)
Here come the tourist maps and buses!
Over the weekend we beefed up our content syndicating engine, My.UserLand.Com, and got some of the glitches out, including removal of the cookie requirement to browse individual channels. Now you only need to be a member of UserLand.Com if you want to customize the presentation of the channels.
Also, we came up with a new page that lists the channels as they change. It's updated once per hour, 24 hours a day, 7 days a week. It's the state of the art of human web trawling technology. (Murphy Willing!)
If you run a news-oriented website, jump on the bandwagon, support Netscape's RSS format and register with My.UserLand.Com. If you manage your content with scripting, you can do this in an afternoon, and be a leader, and reap the benefits in the next wave of web growth! Over 153 publications are already on board. Yours could be next.
Finally, in a story in Saturday's New York Times, Jim Breyer of Accel Partners, a leading California venture capital firm, was quoted "We've had far too many low-quality companies go public. Investors can more easily distinguish between companies building a long-term business and those which have been merely opportunistic."
To Jim and other investors, I agree. The quality is out there. Lots of groups of developers working together over the net, stuff is really happening now, but mostly outside of the radar of the Silicon Valley investor community. I want to change that.
There are quality investments, built out of the stuff that happy customers and barriers to entry are made of. Software and people who work together for the love of it. Add support from the investors and more magic will happen. It's getting very ripe!
I've been meeting with VCs steadily for the last month, and plan to keep doing that for the forseeable future. My goal is to get technology and money working together, not just for my company but also for our partners and competitors.
We're changing, and we're looking for change in the investor world. We want to bet on technology, always have, even though it goes in and out of fashion among investors, we never waver. Maybe now, with the rush to Linux, there will be a renewed interest in what new things can be done with better designed software for human communication. I'll let you know, of course.
PS: Luke Tymowski writes: "The Qube is a nice little machine. I've been running one since last July. It hasn't crashed once. I'm using it to host almost a dozen web sites. There is one problem, Cobalt assumes you'll buy one, plug it in, and never upgrade. If you want to upgrade Apache, or recompile it with support for PHP3, for example, you're on your own. I've upgraded most of the software with the exception of SendMail, which I find too too confusing to upgrade with confidence. Apache is pretty simple, however."