Via Luke Huttermans blog: on keeping RSS subscriptions synchronized.

A standard like this would not only allow you to for instance sync up different SharpReader instances on different PCs (say one at work and the other at home), but would also work across aggregators so you'd even be able to for instance sync up NetNewsWire at home with RSS Bandit at work.

My initial reaction is that the "right" solution is to use a web-based aggregator (any recommendations) but my own personal preference is for a client-side aggregator. In my case - Luke's Sharpreader. To me, reading feeds feels more efficient and responsive when its done locally using a desktop app. They are asking for comments on their proposed standard SIAM

As I writr this, a thought went off. RSS, and other syndicated content, is in many ways similar to email. Back in the day, I use to get my email via POP and it'd be trapped on the PC where I downloaded it (unless I explicitly left it on the server). Nowadays, I use IMAP to read my email and it stays on one server where I can get to it from any computer. And with IMAP, I have the choice of many web-based or desktop based application to read my email.

A next-step could be an IMAP-like server and client apps for handling syndication feeds. From a number of viewpoints, this could have some important benefits. It could reduce the bandwidth used and frequency of polling subscriptions from within an organization. You could also do a lot of interesting matching and recommendation of feeds between users based on what they already read. Any takers?

Whenever I've read about projects or people advocating the use of a Model-View-Controller framework for a php-driven web site, I've always gotten a hunch that its an overly-big hammer to use for your typical website. While the theory of its use may make sense for a desktop client application, I'm not convinced it fits for a website where what you are doing is sending pages of content to a browser. Other people seem to share this viewpoint. See Web Applications: still no ideal architecture... which links to the more interesting The Mountain of Worthless Information who writes:

At the end of the day, we need to stop pretending that the Web browser is just an extension of the desktop, and recognize it for what it really is: the old mainframe terminal, gussied up by lots of colors, graphics, and better fonts. Treat your web apps as we did terminal-based apps 30 years ago, and you'll find your system behaves a lot better as a result.

On a related note, while the MVC pattern may not work for a php webpage. There's no reason not to use a Frontcontroller. The WACT project's page on FrontControllers is a handy reference for all the different ways you can register such a script on Apache including: Physical mapping, using mod_rewrite, and using the ErrorDocument directive.

Turns out all this time when I've been coding I've been using FrontControllers, I just didn't know it. I see some interesting ways I can automate what I've been writing to create a simple, reusable front-controller script.

I'm taking some time off from work to enjoy the holidays and hopefully work on some personal projects (both code and otherwise). Hope everyone has a happy holiday, whether you're celebrating anything or just enjoying time off from work.