Yesterday, I completed my second cross-country solo flight. Palo Alto to Sonoma County for lunch with Dad. Then on to Davis, then back to Palo Alto. 3.0 hours on the Hobbs meter. One more step towards my private pilot’s license.

Dave mentions the reasons that Scoble uses an aggregator that’s integrated with his email client. I have wondered why people would do that. With the ever-increasing spam load that everyone has to deal with, it would seem to me that the last thing I would want is more stuff in my Inbox. Scoble lists two things as advantages to this type of set-up: easy ability to email items of interest, and ease of saving items for later. But the thing is that neither of these features are specific to his news aggregator. Bloglines, to use my favorite example, supports both of these features. There are actually a couple of disadvantages to integrating an aggregator with an email client, besides the spam issue mentioned above. Email clients generally take the form of a ‘3-pane’ interface. Folders on the left, message subjects top right, and individual messages bottom right. To view a message, you click on a subject. That works fine for email, where each message is an individual object unto itself. It doesn’t work as well for blog postings. For example, often times Dave will have multiple blog entries for a given day that refer to each other. Perfectly acceptable blogging practice. I want to be able to view all of his posts for that day at once, in a single window, without having to click on each individual blog entry. You can’t do that with an email-client based 3-pane aggregator. This is why Bloglines is a two-pane aggregator. You can view all the new items for a particular blog at once, or you can view all the new items for all the blogs in a particular folder at once. Less time spent clicking. Also, many blog entries don’t have titles, which makes 3-pane aggregators even more difficult. And the big disadvantage to an email-client based aggregator is that it is a desktop app, which means that it polls all your subscriptions all the time. This wastes a lot of bandwidth when you consider all the other desktop-aggregators doing the same thing. Bloglines only polls a feed once per iteration, regardless of how many subscribers there are. This is much more friendly to content providers. And once you have a certain number of subscriptions, the number of which depends on the speed of your Internet connection, your aggregator will spend all its time polling your subscriptions, eventually falling behind. This is a BRICK WALL that all desktop based aggregators face. With Bloglines, because we’re server based, we can throw all sorts of bandwidth and extra machines at this, so that we’re always able to poll feeds on a scheduled, regular basis. Finally, there’s the inherent advantage to server-based applications in general, ease of accessibility. You can access your Bloglines account from any ’net connected machine. You can’t do that with a desktop-based aggregator.

I cranked pretty hard on some new features for Bloglines. We rolled out the Bloglines Notifier, which is a tiny (100kb) Windows application that puts a little icon in your system tray. Whenever you have new items in your Bloglines account, the icon changes and a little window pops up. Great for, well, being notified that you have new things to read. Tomorrow we’ll be rolling out a couple of additional features, as suggested by our great users. I was not one of the ‘blessed’ that was invited to the Foo Camp gathering this weekend. I’d normally be fine with that, as I’d rather be working on features to make my users happier (which is exactly how I spent my weekend). But in this case, I wish I had been there. As Jeremy hints at in his blog, apparently there was discussion among several of the players in RSS-space, and some announcements are forthcoming. I have no idea about what. But as one of the players in said RSS-space, I would have liked to have been included. Bloglines isn’t huge yet, but we are growing fast (we’ll pass Syndic8 in terms of number of feeds in our database in the next couple of days, making us, I think, the largest directory of RSS feeds on the ’net). So anything that was hammered out at the gathering will most likely affect us.

Yesterday, we announced that Bloglines supports blogrolls. See that list of links on the left side of my blog? That’s automatically generated by Bloglines, based on my subscriptions there. We’ve tried to make it very flexible; you can have it show only an individual folder. You can mark specific subscriptions public or private. In other big Bloglines news, on Monday we passed 4 million blog entries in our database. This is a huge resource of knowledge, and it’s completely searchable. Ok, now that that’s done, it’s back to work! We’ve got many more new features to roll out soon.

I’m nose to the grind-stone working on Bloglines, which seems to continue to gain traction. Here’s a nice article on RSS and Newspapers, with a great mention of Bloglines. Thanks! I’ve mainly been working on backend issues with Bloglines, stuff that’s not user visible. It should be done this week, which is good because working on new features is always more fun.