On Wednesday night at Manchester's burgeoning Social Media Cafe or #smc_mcr for short I ran a slightly impromptu session on Mashups. The idea was initially that after getting over the initial hurdle of making sure everyone knew what a mashup actually is or could be we would share some ideas and our favourite ones that we had come across. Thing is (and I blame the bar for selling me fizzy lager...) the easiest ways for me to do this was to show off my own offerings and somehow in the ensuing discussion we didn't really get to see much else... BUT there are of course plenty more lovely things out there and good resources to be had and I'll try list a few of them later in the post.
According to Wikipedia a mashup is:
A web application that combines data from more than one source into a single integrated tool. The term Mashup implies easy, fast integration, frequently done by access to open APIs and data sources to produce results data owners had no idea could be produced. An example is the use of cartographic data from Google Maps to add location information to real-estate data, thereby creating a new and distinct web service that was not originally provided by either source.
However my friend @Narrenschiff during a little twitter-heckling from the back of Craig McGinty's session had put me in mind of a more cybernetic viewpoint which I'm going to express thusly:
When dealing with the web on a day to day basis there is a severe overload of information, ever increasing quantities of noise (content, advertising, whatever) hitting your eyeballs, demanding of your mouse-clicks and making the whole thing ever harder to cope with. A _good_ mashup therefore is something that attenuates this noise, boiling the area in question into something that is easier to deal with and ultimately makes more sense. This can be through combining, filtering, attenuating and re-presenting the information ultimately transforming it into a more useful form. For preference it should also be FAST - or perhaps "quick to implement" and "quick to grasp" - all the best mashups are created in an afternoon and refined through community interaction.
There are a few paradigms which seem popular at the moment; "lifestreaming" combining the disparate sources of one's online output into a single coherent whole and one of these that I've been using occasionally is a thing called Swurl
It takes all your stuff and re-works it into either a stream of summaries or (and more visually interestingly I think) into a calendar-like timeline view with each of your entries organised into dinky little boxes. If you produce a lot of visual material it is very compelling to look at - however a heavy text outputter may find it less interesting. Right now I feel my feed is somewhat swamped by a myriad of twitter posts which rather distract from the more interesting flickr stuff. A degree of smartness has clearly gone into this software, hitting a page that isn't very fresh will trigger the server to fetch any new updates from your feeds which pop lazily into place over AJAX a few seconds after the page has loaded. HOWEVER these are clearly turbulent times... it seems the developer has recently stopped new registrations on the site and it sounds like they're not continuing to work on it, a shame as it's really rather delightful especially in its simplicity.
However if you're itching to lifestream there are plenty of other things in this space - Friendfeed has a very similar set of functionality and a pretty big userbase, if you're more of a hacker there is an promising looking PHP application called Sweetcron which you can hoist on your server and tweak to meet your needs.
For myself (and this is something I didn't get into during the session) I've got a limited amount of this kind of thing going on in Samscam and am feeling the need to expose a little more of my stream - however the really important thing is to keep the signal-to-noise ratio balanced. I certainly don't want the front page of my website swamped by ephemeral tweets!
The IEC Statusometer - also known perhaps unless anyone can come up with a better name for it as Spacefac.es is a bit of code I wrote to help with a work situation. The idea was that as we are working in a distributed environment (with the department's staff spread across the country and world) it would be nice to have some kind of at-a-glance view of what everyone was up to. Being the keyed in bunch of people we are we're all already running Skype, Twitter, Calendaring, Blogs and all manner of other stuff - but that's already around 4*n separate places to look for updates.
So I mashed this data in such a way that you can at a glance see a grid of the people in the organisation, their faces glowing according to their current Skype presence status (online is green, offline is grey, do-not-disturb is red, etc) with their latest tweet, last blog post and the items from their calendar for the day. We leave it running on a screen at the end of the office and consult it from time to time to see where everyone is at and it feels good.
There are caveats; The underlying software is at a very early stage - it works but there are some major issues such as security, privacy and inevitably the time I have to work on it which is minimal. If no-one beats me to it I'm going to look at getting a bit of funding to push it into shape and productise it in the not-too-distant future.
Again friendfeed does a lot of similar stuff - but I've not seen anything out there that has quite the same live feel to it.
I've mentioned it before but there is the little Gigometer Dashboard widget I wrote a few months back. It's not really much of a mashup in the traditional sense but it is a handy view on what's going on albeit within the limitations of the data available from last.fm. However this itself seemed to capture some imagination around the group and was where we really started freestyling - can we get information out of Myspace into it? Would it involve screenscraping, what are the useful screenscraping libraries (several developers weighing in here with interesting projects). What about upcoming and other event data sources? And of course the prospect of doing location-aware versions for mobile devices, something that is well up on my list for the app anyway.
On the location-aware side I sense interesting possibilities of Dopplr (a service for travellers) which already has some great stuff mashed into it from other sources as well as Outside.in's Radar application - which just looks wicked! Thanks to Craig for that last one.
Yahoo Pipes is lowering the barriers to making mashups - you don't have to be able to code any more to mash. I've done a few little experiments with it of which my Delicious feederiser is probably the best if a little wonky. It trawls all the websites you've bookmarked in delicious for feeds and aggregates them all...
Tony Hirst from the OU (who is the undisputed king of mashups from my professional community) created a thing that let you take any RSS feed and turn it into a full-screen navigable presentation. Particularly useful when combined with delicious. I was going to use this to drive the presentation but didn't get it in time so just used Omnigraffle and a browser instead! Tony is a true virtuoso when it comes to this stuff and is terribly good at the quick and dirty mash - do check out his blog.
And finally an essential resource if you're in the business of mashing is Programmeableweb's API directory - where they maintain a list of all the cool endpoints you can suck interesting data from. Great just to browse and let your mind wander through the endless possibilities!
I really had an excellent time at the event and enjoyed presenting - and from the tweets it sounds like everyone else had a pretty good time too. My photographs as ever can be found on flickr, suitably tagged.
