This page has moved to
http://www.superluminal.com/dave/weblog/archives/2003_04_20.html
You will be redirected to that page in 10 seconds.
New RSS feed at
http://www.superluminal.com/dave/weblog/index.xml
Dave Seidel :: Wavicle
Holons, nothing but holons.
Sunday, April 20, 2003
« Spring cleaning »
I've been reasonably satisfied with Radio for quite a while. In fact, I recently renewed my subscription. I've even been dabbling lately in Radio's development environment. But I've been contemplating a change, for various reasons that fit into two broad categories.
Technical
I've been having fun playing with UserTalk, I like writing code in an outliner, and I find the idea of an environment built on a hierarchical database very interesting. But UserTalk and the ODB have no applicability independent of the world Radio/Frontier/Manila. I don't use Manila, and I have to plans to use it, so aside from Radio, there's no way for me to leverage any experience or knowledge I might gain in this environment. Dave talks a lot of talk about being locked in the trunk, but I'm starting to feel like I'm in danger of being locked in his trunk. I'd really rather be working with any one of a number of languages that are non-proprietary and widely deployed. I recently installed Movable Type for my daughter's use, and it's looking very attractive. Perl may not be my favorite language, but I have a huge amount of respect for it, not to mention a certain begrudging affection.
In Radio's defense, I like the fact that the RSS aggregator and the authoring environment are (somewhat) integrated. And I'm addicted to using an aggregator that uses a desktop web server to expose its user interface. I run it on my laptop, which travels with me between home and office, I'm always on a LAN, and I like the flexibility of collecting my news feeds in a central place, but being able to access the data from any machine on the network. But Radio is not the only way to achieve these things. And it would be nice to use some software that makes fewer demands on my disk space and CPU cycles.
Emotional
I have a lot of respect for Dave's work on SOAP, RSS, OPML, etc. He's also pretty good at pissing people off, and he can be very inflexible. Now, he has some very nasty detractors, who in my opinion have far exceeded any of his real or perceived offences, and for whom I have no sympathy whatsoever (I can't even bring myself to link to them, and besides the site is down). But every once in a while, something happens that makes me fed up with the way he communicates. Usually, I cool off after a couple of days. But this time it's enough to push me to a decision point. I'm ready to move on.
Moving On
So I'm going to start making the transition from Radio to Movable Type as my blogging tool. I've already got it running on my hosted site, and I'm looking forward to learning about the plugin architecture and the templating scheme. It will take a while — I need to sort out issues like data import (any good scripts out there to move Radio posts into MT?) and templates (should I port my current templates to MT, or use something else?).
The other decision to make is which aggregator to use. My main requirement is that it must work as a desktop server (i.e., not a native UI, or at least not exclusively), and it would be nice if there was some integration with MT on the authoring side. I'm open to any language, but I'd prefer to have source code, and I prefer an MIT/Apache-style license to GPL. I'm open to suggestions.
•••
Update 1
I downloaded Aggie and AmpetaDesk as the first two candidates for a replacement aggregator.
Update 2 (edited)
I wasn't clear in the "Emotional" section about what aggravated me about the discussion on Sam's blog. It wasn't the CSS argument. It was the incivility. Calling someone an "asshole" is gratuitous and serves no purpose other than to piss people off.
In the meantime, here's my view of CSS: it works quite well if you understand it. There's a learning curve, and I'm certainly pretty low on that curve, but people like Mark are doing a lot to educate us. I believe in the value proposition for CSS, which is to separate content from rendering. The browsers are getting better, and IE is not the only choice. I don't believe that we're locked in the trunk on this one. Sure, we have to deal with browser differences. So what else is new? In the old days, I used to struggle with something like 9 or 10 different DOS/WIndows C compilers to make sure my code would compile on all of them. It was a pain, but doable. These days, I no longer see Windows as my only platform, so I keep my native code very portable and use Java and other languages as well. And I use the browser as my UI platform, so that takes care of the desktop. So I'm doing my best to avoid lock-in.
Holons, nothing but holons.
Sunday, April 20, 2003
« Spring cleaning »
I've been reasonably satisfied with Radio for quite a while. In fact, I recently renewed my subscription. I've even been dabbling lately in Radio's development environment. But I've been contemplating a change, for various reasons that fit into two broad categories.
Technical
I've been having fun playing with UserTalk, I like writing code in an outliner, and I find the idea of an environment built on a hierarchical database very interesting. But UserTalk and the ODB have no applicability independent of the world Radio/Frontier/Manila. I don't use Manila, and I have to plans to use it, so aside from Radio, there's no way for me to leverage any experience or knowledge I might gain in this environment. Dave talks a lot of talk about being locked in the trunk, but I'm starting to feel like I'm in danger of being locked in his trunk. I'd really rather be working with any one of a number of languages that are non-proprietary and widely deployed. I recently installed Movable Type for my daughter's use, and it's looking very attractive. Perl may not be my favorite language, but I have a huge amount of respect for it, not to mention a certain begrudging affection.
In Radio's defense, I like the fact that the RSS aggregator and the authoring environment are (somewhat) integrated. And I'm addicted to using an aggregator that uses a desktop web server to expose its user interface. I run it on my laptop, which travels with me between home and office, I'm always on a LAN, and I like the flexibility of collecting my news feeds in a central place, but being able to access the data from any machine on the network. But Radio is not the only way to achieve these things. And it would be nice to use some software that makes fewer demands on my disk space and CPU cycles.
Emotional
I have a lot of respect for Dave's work on SOAP, RSS, OPML, etc. He's also pretty good at pissing people off, and he can be very inflexible. Now, he has some very nasty detractors, who in my opinion have far exceeded any of his real or perceived offences, and for whom I have no sympathy whatsoever (I can't even bring myself to link to them, and besides the site is down). But every once in a while, something happens that makes me fed up with the way he communicates. Usually, I cool off after a couple of days. But this time it's enough to push me to a decision point. I'm ready to move on.
Moving On
So I'm going to start making the transition from Radio to Movable Type as my blogging tool. I've already got it running on my hosted site, and I'm looking forward to learning about the plugin architecture and the templating scheme. It will take a while — I need to sort out issues like data import (any good scripts out there to move Radio posts into MT?) and templates (should I port my current templates to MT, or use something else?).
The other decision to make is which aggregator to use. My main requirement is that it must work as a desktop server (i.e., not a native UI, or at least not exclusively), and it would be nice if there was some integration with MT on the authoring side. I'm open to any language, but I'd prefer to have source code, and I prefer an MIT/Apache-style license to GPL. I'm open to suggestions.
•••
Update 1
I downloaded Aggie and AmpetaDesk as the first two candidates for a replacement aggregator.
Update 2 (edited)
I wasn't clear in the "Emotional" section about what aggravated me about the discussion on Sam's blog. It wasn't the CSS argument. It was the incivility. Calling someone an "asshole" is gratuitous and serves no purpose other than to piss people off.
In the meantime, here's my view of CSS: it works quite well if you understand it. There's a learning curve, and I'm certainly pretty low on that curve, but people like Mark are doing a lot to educate us. I believe in the value proposition for CSS, which is to separate content from rendering. The browsers are getting better, and IE is not the only choice. I don't believe that we're locked in the trunk on this one. Sure, we have to deal with browser differences. So what else is new? In the old days, I used to struggle with something like 9 or 10 different DOS/WIndows C compilers to make sure my code would compile on all of them. It was a pain, but doable. These days, I no longer see Windows as my only platform, so I keep my native code very portable and use Java and other languages as well. And I use the browser as my UI platform, so that takes care of the desktop. So I'm doing my best to avoid lock-in.