July 29, 2004

Jon Udell on Bloglines

Since last fall, I've been recommending Bloglines to first-timers as the fastest and easiest introduction to the subscription side of the blogosphere. Remarkably, this same application also meets the needs of some of the most advancedusers. I've now added myself to that list. Hats off to Mark Fletcher for putting all the pieces together in such a masterful way.

What goes around comes around. Five years ago, centralized feed aggregators -- my.netscape.com and my.userland.com -- were the only game in town. Fat-client feedreaders only arrived on the scene later. Because of the well-known rich-versus-reach tradeoffs, I never really settled in with one of those. Most of the time I've used the Radio UserLand reader. It is browser-based, and it normally points to localhost, but I've been parking Radio UserLand on a secure server so that I can read the feeds it aggregates for me from anywhere.

Bloglines takes that idea and runs with it. Like the Radio UserLand reader, it supports the all-important (to me) consolidated view of new items. But its two-pane interface also shows me the list of feeds, highlighting those with new entries, so you can switch between a linear of scan of all new items and random access to particular feeds. Once you've read an item it vanishes, but you can recall already-read items like so:

Display items within the last

If a month's worth of some blog's entries produces too much stuff to easily scan, you can switch that blog to a titles-only view. The titles expand to reveal all the content transmitted in the feed for that item.

I haven't gotten around to organizing my feeds into folders, the way otherusers of Bloglines do, but I've poked around enough to see that Bloglines, like Zope, handles foldering about as well as you can in a Web UI -- which is to say, well enough. With an intelligent local cache it could be really good; more on that later.

Bloglines does two kinds of data mining that are especially noteworthy. First, it counts and reports the number of Bloglines users subscribed to each blog. In the case of Jonathan Schwartz's weblog, for example, there are (as of this moment) 253 subscribers.

Second, Bloglines is currently managing references to items more effectively than the competition. I was curious, for example, to gauge the reaction to the latest salvo in Schwartz's ongoing campaign to turn up the heat on Red Hat. Bloglines reports 10 References. In this case, the comparable query on Feedster yields a comparable result, but on the whole I'm finding Bloglines' assembly of conversations to be more reliable than Feedster's (which, however, is still marked as 'beta'). Meanwhile Technorati, though it casts a much wider net than either, is currently struggling with conversation assembly.

I love how Bloglines weaves everything together to create a dense web of information. For example, the list of subscribers to the Schwartz blog includes: judell - subscribed since July 23, 2004. Click that link and you'll see my Bloglines subscriptions. Which you can export and then -- if you'd like to see the world through my filter -- turn around and import.

Moving my 265 subscriptions into Bloglines wasn't a complete no-brainer. I imported my Radio UserLand-generated OPML file without any trouble, but catching up on unread items -- that is, marking all of each feed's sometimes lengthy history of items as having been read -- was painful. In theory you can do that by clicking once on the top-level folder containing all the feeds, which generates the consolidated view of unread items. In practice, that kept timing out. I finally had to touch a number of the larger feeds, one after another, in order to get everything caught up. A Catch Up All Feeds feature would solve this problem.

[Update: The feature, of course, exists. Thanks to David Ron for pointing this out. The reason I didn't find it: the Mark All Read link is right-aligned at the top of the left pane, and not bound to the other controls found there. Since I have some feeds with very long titles, it's necessary to scroll rightward in the left pane to find the Mark All Read control. Operator error on my part, but I'm sure I'm not the only one.]

Another feature I'd love to see is Move To Next Unread Item -- wired to a link in the HTML UI, or to a keystroke, or ideally both.

Finally, I'd love it if Bloglines cached everything in a local database, not only for offline reading but also to make the UI more responsive and to accelerate queries that reach back into the archive.

Like Gmail, Bloglines is the kind of Web application that surprises you with what it can do, and makes you crave more. Some argue that to satisfy that craving, you'll need to abandon the browser and switch to RIA (rich Internet application) technology -- Flash, Java, Avalon (someday), whatever. Others are concluding that perhaps the 80/20 solution that the browser is today can become a 90/10 or 95/5 solution tomorrow with some incremental changes.

Dare Obasanjo wondered, over the weekend, "What is Google building?" He wrote:

In the past couple of months Google has hired four people who used to work on Internet Explorer in various capacities [especially its XML support] who then moved to BEA; David Bau, Rod Chavez, Gary Burd and most recently Adam Bosworth. A number of my coworkers used to work with these guys since our team, the Microsoft XML team, was once part of the Internet Explorer team. It's been interesting chatting in the hallways with folks contemplating what Google would want to build that requires folks with a background in building XML data access technologies both on the client side, Internet Explorer and on the server, BEA's WebLogic. [Dare Obasanjo]
It seems pretty clear to me. Web applications such as Gmail and Bloglines are already hard to beat. With a touch of alchemy they just might become unstoppable.

[Jon's Radio]
Posted by Stuart at 2:18 PM

July 21, 2004

Sometime ago, Dina Mehta

Sometime ago, Dina Mehta and I had met a large Indian company who wanted to make their Intranet more effective, and making the various teams more productive. These notes, made by Dina, capture what would be the ideal information workspace:


We asked the question - what is the dream system you desire - and that was really interesting. What it revealed is the need is not as much for a Content Management System as much as it is for a system that allows them to dialogue and converse effortlessly and seamlessly, brainstorm on ideas and projects, in a manner that is as 'face-to-face' as possible. Here's the gist of what an employee told us :

I know X is not here in my office (in Mumbai) but in another city. I want to be able to talk to her, as if she's in the same room as me. I want to be able to feel all the nuances in talking with her - its got to be touchy-feely and not a cold email or a phone call where I know the time ticking away means my bottomline suffers.

Underlying what this employee told us is her desire for flow. Easy, hassle-free, inexpensive flow. Flow that allows a dialogue as if the other person is right there with her.

Incorporating :

  • Presence indicators
  • Communications system
  • Collaboration space

    One of the key requirements - really a very simple one to have, but something so sorely left out in the current system, is a presence indicator. Much like in current IM systems - telling us who's available, who's logged in and therefore present in the office, who I can ping for a query, and ensuring that a response is received. In the case of his organisation, currently, they'd send an email, wait for a response, followed by more emails as reminders, and finally in sheer frustration, pick up the phone and make a call - which can be expensive if outstation (as is the case very often in their line of business) and does not always ensure that they will get a response - what if the person is away from the office?

    Tied into this requirement for presence indicators is the need for 'real-time' 'live' communication. This is where voice applications, small cam shows, conferencing facilities would be useful. Skype with its conferencing facilities has really shown its possible to do this with terrific quality. Combine this with some of the 'soft profiles' that make a person far more approachable than just another colleague, like those on Ryze or Orkut.

    And finally the need for collaboration spaces - where one can play around with Wikis and Blogs. Not having to rely on a whole host of asynchronous emails - or bothering to archive them systematically - these tools can do that automatically for you. And more food for thought pinging its way in through RSS in Newsreaders.

    Picture this scenario - you have a project on and are racking your brains about how to approach it - you check your presence indicator - see who's available - ping them with a request for conferencing - hitch up the webcam, enable voice - and bingo - in minutes you have a virtual team ! Record the conversation, take notes on the wiki, synthesize it in a team blog which has comments enabled, feed in current thinking on the topic from your newsaggregator, and you have real flow. And, ridiculously easy group-forming to borrow a wonderful phrase from Clay Shirky.


  • We arenÂ’t there yet, but getting close. A combination of new technologies are coming together to make it easier for us to work together in groups – be it in the workpace or among friends and family.

    Tomorrow: India Action: Create Contentrix

    [Skype]
    Posted by Stuart at 10:25 PM