Primordial Ooze

iPodder

Last night it took everything I had not to start writing my own iPodder software. I knew I would fail and I simply wouldn't get anything else done.

The stuff out there for Windows still seems to suck real bad even for a more advanced user like myself. I can't imagine how regular computer enthusiasts could survive.

No offense to  the dev groups working on this stuff, I'm simply speaking relative to iPodderX for the Mac, which rocks.

Here are my criticisms on i-podder.js, iPodder (Python), jPodder, and iPodder.NET.

i-podder.js

The script seems to choke on feeds every now and then. This is an intermittent problem. If you read the code, you see that there isn't a whole lot of error checking. Even with this serious fault, I still use it. The other solutions simply seems to mysteriously stop working after a while.

iPodder (Python)

Where's the development on this? I listenened to the ipodder-dev podcast and Garth Kidd says something about a new release soon since not everyone gets the nightly CVS stuff. Well, I must admit that I did see something new last night and it simply didn't work. I got exceptions when it tried to manipulate iTunes and parts of the interface were disabled. Garth mentioned that they were concentrating on the engine and welcomed UI developers to help out. Hopefully someone will have time for that.

jPodder

Neat proprietary install (is lzPack standard in the Java world?). Very iPodderX-like interface. Neat capability to browse the podcast directory. This one seems actively developed and I am a bit surprised to see how much it has come in the past few days (week) since I got my first negative impression.

It's nice that they allow you to pull down an OPML and use that. However, my OPML is behind www-authenticate at Bloglines.com so I can't specify my OPML.

I may write a follow-up if I plan on using it.

iPodder.NET

Very primitive interface and the development seems kind of slow. Feeds have to be added manually and are maintained in the standard .NET config file. Having the feeds in this file makes it difficult for me to use by Blogroll extractor since I have to insert a fragment into the XML file. Not a terrible technical feat, but more than I am inclined to develop right now. (I actually refactored my extractor last night to generate the proper format via a nice generic command-line parameter, but I still have to write the capability to insert the fragment into the app config file).

To sum it all up

The whole revolution is still in its infancy so it's not surprising that the tools are still broken to a certain degree. The way I see it, these are  the key features for me:

• Browsing of podcast feeds
• Auto-synchronizing with feeds you want if maintained on something like bloglines
• Solid integration with iTunes, Windows Media Player, etc.
• Better overall detection of already-downloaded files (lots of ways to attack this)
• Allow applying rules to Podcasts and possibly auto-insert tags. This is important as long as iTunes doesn't allow me to say "Genre is 'Speech' or 'Podcast' or 'Comedy' AND PlayCount = 0" Currently iTunes only allows all conditions to be AND'd together or OR'd together.

Lots more to discuss on this topic, but I don't have a lot of time to flesh out a whole white paper on the topic.