Friendfeed, lifestreaming, and workstreaming

As I mentioned a couple of weeks or so ago I’ve been playing around with Friendfeed. This is a ‘lifestreaming’ web service which allows you to aggregate ‘all’ of the content you are generating on the web into one place (see here for mine). This is interesting from my perspective because it maps well onto our ideas about generating multiple data streams from a research lab. This raw data then needs to be pulled together and turned into some sort of narrative description of what happened.

Friendfeed does quite a good job of the first bit. I can aggregate content generated on this blog, photos I have uploaded to Flickr, as well as relevant streams from my LaBLog, papers I have looked at, and other material that I have shared through Google Reader. Some of these are generated by feeds that I create using Yahoo Pipes to filter my posts from a larger stream. As an aside, the one thing I haven’t been able to do in a sensible fashion is to aggregate all the comments I have made on other people’s blogs. Fundamentally this is because I can’t do a semantic search on ‘Items that are comments, written by me’ and then convert that to an RSS feed. But that’s a separate issue. I can subscribe to other people’s feeds and they can subscribe to mine. Friendfeed is quite good on showing you stuff from Friends of Friends as well which is a good discovery mechanism. I have seen several papers of interest in Neil Saunder’s feed for instance.

Where Friendfeed isn’t so good is on post processing. For instance, it isn’t straightforward to define the relationships between things. I have been taking notes on other people’s response to my Blog posts, sharing their posts through Google Reader and then annotating them in Friendfeed as a response or vice versa but this is less than ideal. Also Friendfeed currently doesn’t allow any flexibility in what you want to see. I could probably manage without seeing anyone else’s Twitter posts for instance (not that they’re not fun to read guys, just that they end up taking up more of my time!). Maybe I don’t want to see Dan Brickley’s shared videos but I do want to his shared items in Google Reader. I could in principle do some of this in Yahoo Pipes – I haven’t tried as yet but I should give it a go.

The next step beyond this would be much deeper integration and then actually acting on items. Converting an incoming result from an instrument into a meeting (or a booking on the next instrument). Acting on the flow to notify a specific person of something important, and again, starting to connect it all together into a web of interconnected information that can then be subject to machine processing. This is much more than just a lab book, or just a data feed, or even just laboratory management environment. Its about controlling all the data streams in your life and adding the metadata that will help you to make sense of them.

In this context there is a lot of interest in Workstreamr a new initiative that looks to be part lifestream, part integration with existing management tools like email and calendars, and ties this up with the kind of social tools that enable communication, discovery, and sometimes just a conversation about the weather. The idea seems to be, like FriendFeed that you are involved with generating and viewing the feeds for yourself and others but beyond this you can process and act on these feeds, generating calendar items, send stuff on, embedding the whole thing into a planning process. This will be very interesting to see in action. My concern is that it will be too heavyweight for the kind of things I would like to do and that it may end up focussed on business applications leading to a poor fit with a research environment. However, I am happy to be proved wrong and I have requested an invite to the closed beta.

Data portability remains fundamental to all of this and interoperation between all these different tools and services is a must. Easy conversion of an incoming data file into an annotated entry in a lab book. The entry in a lab book kicking off a conversation, leading to a calendar item for a meeting (with any associated paperwork). An awful lot of this could be done today, by hand, using a mixture of RSS, markup, and open and agreed data formats. The challenge lies in building to tools that will allow us to manage and process this automatically.

p.s. Zemanta is doing much better now on suggested Links and Tags, much less on volkswagen’s coming through. Pictures not so good so far. Will continue to update on progress.