Scoble’s Link Blog

http://blog.softtechvc.com/2005/04/rojo_the_return.html

Weblog: Software Only
Source: Rojo - The Return - As a Taggregator
Link: http://blog.softtechvc.com/2005/04/rojo_the_return.html

As mentioned by Susan a few days ago, a happy few got a preview of the new UI and functionality of Rojo, which I must say are great improvements from the one I had tested back in Nov 2004. Seeing the UI for the first time, I thought that this looked like a typical Adaptive Path work product (it turns out it is).

So what is Rojo ? Well it is an RSS aggregator, with a few twists (as previously noted, twists are very important). It belongs to the web-based aggregators category, led (until now ) by Bloglines. As you would expect, Rojo integrates RSS/Atom feeds, displays them in your web browser (working great in Firefox PC), display individual posts, search through the feed archive, etc. It also provides a simple directory of blogs, which has been implemented using a limited taxonomy/tagsonomy maintained (as I understand) by an editor.

Like other RSS readers, you can also pre-populate Rojo with pre-defined category (ex: Venture Capital) and sources (CNET, NY Times)… Signing up is actually a breeze and in a few minutes, you are rssreading away. An advice though: be easy on the topics and the sources, because you might end up looking like Scoble very quickly with hundreds (and hundreds) of feeds to read. I actually wish that you could see which blogs are included in the topics in order to choose a subset.

The search works well, too well almost. It would be appropriate to aggregate the different blogs pointing to a given entry (a la Feedster or Technorati) as opposed to having multiple entries at the same level. And I would move the search box above the list of feeds.

Now the interesting stuff:

• Tags: Rojo allows you to tag post AND feeds. And in plain English, unlike del.icio.us. This means that I can organize my feeds much more flexibly than my NewsGator folders. Nice. And I can read all posts related to a given tag, or all posts of all feeds related to a given tag. When available, related tags are also displayed (tags used for similar stories, I assume that there needs to be a minimum occurrence to appear).
  One thing though: it sounds as though - unlike Technorati - that my own tagging, as defined by my TypePad categories, are not used. I know that there is a big debate as to whose tags are to be used, but I do categorize my posts, and would want to have the option to have my categories used as tags.
  But overall, this is great taggregation functionality.
• Social Network: Yep, another one (expect another round of "Connect to me ?" emails - though these can be turned off, to be fair). This functionality was already available in the November release, but I did not go beyond inviting a few contacts.
  Rojo’s implementation is similar to Buzznet’s and Flickr’s: I can add someone to my watchlist to see what he/she is reading, but the relationship does not need to be bi-directional.
  Privacy issue: once I have added someone to my watch list, clicking on Contacts > Manage allows me to tag the person (neat) but also see their email address. This should only happen if the person has accepted me as a contact. Issue solved, see below comment from Michael Eakes.
• Sharing: Once a person has accepted me in their network, I can see their profile, their tags, feeds, etc. And I can see what stories they have shared with their network. Attention(.xml) is creeping up here (as hinted by Steve Gillmor). This is quite interesting, but will only be useful if people tag as well. Check Chris Alden’s shared stories, there are some relevant to me, and some which are not. Had he tagged them as well, I could have used the "Sort by tag" view to go through them (I know I am a pain).
• Recommendation/Discovery: Similar to the Top Links feature of Bloglines, Rojo recommends posts based on the number of incoming links. Recommendations can be produced on a one to seven day period. This is cool, but I would like to be able to filter recommendations by topics/tags. Otherwise many of these recommendations will be totally irrelevant to me.
• Other neat features: list of stories I have read, marked and shared.

Overall this is a great job compared to the initial version that was shown around at Web 2.0. Still not enough to get me out of NewsGator, but certainly quite competitive to Bloglines feature-wise (needs to scale in terms of UI and back-end now). Who said that the Web-based aggregator market was done and dusted ? (Me ? Naann). I can’t wait to see the rumored Yahoo!Mail client resulting of the integration of Bloomba and and Oddpost.
Oh yeah, and it is now officially out of beta. And as usual Om beat me to it. You can take a tour of the product here, and check out their blog here. Interestingly, their corporate site is implemented using a closed Wiki (JotSpot actually).

A few issues now, in no particular order:

• The version I used was really slow, but eh, it is just the beginning and I am sure that more servers will be provisioned.
• At the end of the day, a lot of time is spent by Internet users tagging posts on del.icio.us, furl, etc. I think that these tags ought to be integrated and displayed somehow, and likewise Rojo tags ought to be made available through other tagging services. This raises the issue to del.icio.us single word tag limitation, but I do hope that this will eventually get changed.
• When you follow through a link to the post itself, I would do it in another window as opposed to reusing the Rojo one (or is there an option) ?
• Manage Feeds: In the Tags view, having a maximum of 4 or 5 tags to manage as a set, and then having to click individually on others is a bit of a torture. Why not manage all of them on that one page, or at least multiple ones ?
• As a general point, I think that the UI is nicely designed, but seems to target casual users, as opposed to RSS maniacs. Otherwise there would be more space for viewing stuff and less space taken by UI controls (icons, etc.).
• When you upload an OPML file, a bunch of feeds appear as "Unnamed", as if they had to be processed before the feed title could be displayed.

Related…