Monday, March 11, 2002


I'm reading much of Patrick Logan's writing lately. This is especially well done.

11:27:44 PM    comments ()  trackback []  


I like the term full peer, it doesn't have the baggage of the terms client and server. It also doesn't claim the participant is an X and a Y. It just says it is a Z. If you know what I mean.

[Patrick Logan's Radio Weblog]

Yeah, that's the term that Dave uses a lot. I hope it'll catch on, it certainly describes the situation more accurately.

But I DO like the term 'Desktop Website.' Oh well.

10:23:47 PM    comments ()  trackback []  



I can't wait to find out what Dave is teasing us about. OPML is basically a way to capture an outline in XML.[transcendental petroglyphs]

All of the other references he brings up have to do with communication / the elimination of email. I wonder if it has anything to do with the LiveOutline tool, but that's just grasping at straws.

I also wonder if Jabber figures into it, but I'm not sure that the Jabber tool is where it needs to be yet. I can't wait for either the Jabber tool or whatever Dave is teasing.

10:19:18 PM    comments ()  trackback []  



I am SO TIRED of Software Company press releases that use the term 'rich.'

I got enough of that at Microsoft.

7:08:02 PM    comments ()  trackback []  



As I read Michael Fraase and Todd Hoff talk about the problems with a 'desktop server,' I keep thinking that they've gotten hung up on the meanings of the words 'desktop' and 'server.'

When talking about desktops, it sounds like they think the term 'desktop' applies to the semi-bulky machines that sit on/under desks and require you to go to them, rather than taking it with you, like a laptop. To be fair, Todd does seem to understand that you can run Radio on a laptop, but then talks about bad or intermittent connectivity.

The same is true when the term 'server' comes up. When most people think about servers, they think of these big machines humming away in closed rooms. When UserLand talks about a 'server,' they're talking about the ROLE the program is playing, rather than characterizing the hardware the program is running on. Radio is 'serving' the requests that the browser is making, and thus is a 'server.'

A network connection is not completely necessary to use Radio. You can turn off upstreaming and post away to your blog. The next time you have a connection, you turn upstreaming back on and everything flows up to the central server.

Radio is not trying to replace the central server. As long as humans roam freely with no fixed address, there will be a need for centralized servers. Radio just makes the interaction more pleasant. Radio says "I'll worry about all the hard stuff for you. You just write. I'll move the data when I can."

Perhaps the term 'server' and 'desktop' carry too much baggage. Perhaps we could just call Radio an application, and just carry the knowledge that it is in fact a server deep within our geeky hearts.

3:51:50 PM    comments ()  trackback []  



Something to keep in mind when you are seeing strange Radio behavior. I use a proxy at work, so my work copy of Radio has all of the proxy settings defined. Just recently I had to change the settings, but it would only hold on to the settings until the next time I started Radio. I had to check and change the settings every time I started it up. It was getting annoying. After three days of this, I finally realized what was going on.

I had included the original proxy settings in my radioStartupCommands.txt file.

Boy I felt like an idiot after begin annoyed for three days.

The moral of this story -- if you can't understand why certain values in your database are getting reset on startup, check the startup file. (I feel like an idiot just typing that sentence.)

9:51:01 AM    comments ()  trackback []