Google Wave is/was an interesting product. It is nothing less than an attempt to oust email as our primary communications medium and therein lays its story. It is basically Instant Messaging with two additional functions - the ability to automate a response so the user doesn't really see a difference between a human and a program (or Robot). It also supports mini-applications (or Gadgets) in a similar way to iGoogle and Facebook, blurring the lines between conversations and documents. Each individual exchange (or Wave) is logged and can be added to at anytime. It is better than email because it is real-time and richer exchanges can be made. It is worse than email because using this functionality is confusing for all but the tech savvy (having multiple Robots, Gadgets and humans all involved in the same Wave is rife with issues of ownership, progression and timing). Also the success of email is due to its ubiquity - everyone can use it. Hardly anyone knows what a - Wave is or what to do with it.
It was un-ceremoniously released to the general public a couple months ago (although some developers have been using it by invitation since its tech-celebrated debut last Summer) and pronounced dead by Google this week.
In Google's own words - "Wave has not seen the user adoption we would have liked". After less than fifty working days? For such a new and cart-upsetting product? With hardly any permeation of Wave concepts to the general public, next to no marketing and no specific commercial targeting? Of course! It was inevitable that this would be the outcome if an organisation were to review a key product launch after so short a time frame with so little support. This has to fall in the space of Google testing out new concepts/obtaining user feedback, with little fanfare (and so little possibility of failure) with a view to releasing the (inevitable) email replacement again at some point in the future.
Rich real-time communication supported by an easy/interactive interface to a computer (rather than a simple search dialogue) are most definitely the way forward. At the very least, collaborative programming environments benefit from this type of interface, as does trading and social networking proper. Google Wave uses the Extensible Messaging and Presence Protocol (XMPP) protocol which enables the most efficient peer-to-peer communication currently available. These needs are not going away and neither will the constituent parts of Google Wave (in some form).
Wave Robots are perhaps the most interesting component right now. In operation, they are a bit like a Turing Test or a Twitter-bot in that they facilitate a conversational and collaborative approach to establishing (and getting) what you want. Moving past the simple search/response model we have now, it is inevitable that there will need to be some interaction, some toing-and-froing and narrowing-down for something to understand (disambiguously) what it is that you actually want. It happens in real-life and so probably needs to be modelled in virtual-life.
Wave Robots are coded relatively straight-forwardly in Java using a Wave API. This element of Wave remains significantly undersold. As an example, I developed one to allow collaborative SPARQL queries to be made against any open linked data. A few working SPARQL queries have been uploaded to give you an idea.
Queries can be collaboratively changed in real-time and results can be sent out to a named Google Docs account as a spreadsheet. Once in Google Docs, there are several charting options available to make the data more accessible. Its like a hugely more powerful Google search (for those of a technical persuasion). You can add other Robots to the Wave to allow syntax highlighting of the code e.g. Kasyntaxy (although at the moment, this doesn't seem to specifically support SPARQL). As its all Robot based, its server-side and so you can use it on your mobile device - whatever that may be. Just add querytheweb@appspot.com to your Wave contacts to use. Type "cycle" to cycle between one of two endpoints. An endpoint is basically a SPARQL query engine. The two used are both generic meaning that they are not tied to a particular data set. This means your queries will need to use the FROM clause to identify which data you are querying (by URI). Type "help" to get a list of other options.
This Wave Robot took just a day to develop in Java and was deployed to Google App Engine. It is basic but even as it stands, it is still probably the best way to access and present open linked data currently available.
However long Google maintains/runs/supports Wave, its constituent parts will, at some point, be mainstream. The demise of Google Wave is not the demise of the email replacement concept. Get ahead of the game and develop some Wave Robots now. Get used to the concepts and the working environment Google has provided until the end of this year. Your work will be able to re-surface in some new product next year.
UPDATE: Interesting Scoble commentary on Google Wave ending here.
No comments:
Post a Comment