Google wave and implications for science

Google Wave (see Google Wave first reactions and links and comments) has clearly wowed a lot of people and this in itself is important. Assuming that a combination of the technology and the excitement turns it into the The Next Big Thing that in itself is world-changing. So I’ll assume that it will happen and will sweep the public face of the internet.

Nothing is really new and we can find these ideas in Ted Nelson, TimBL and others. But the added features are:

• there is an engineering company that the world believes in

• social computing has arrived

• computing is becoming pervasive (except airplanes fly and railways go through tunnels)

  These have all been foreseen but now they are here. So what does it mean for me and for chemistry?

  My understanding of GW is that it provides shareable, realtime, compound XML documents with well engineered APIs for editing, aggregation and publication (including alerts). If I’ve missed anything please say so. I have no doubt that this is revolutionary and I and Henry Rzepa have been trying to do this for 10 years in chemistry. We coined the term “datument” for a composite of text and data (The Next Big Thing: From Hypermedia to Datuments ) which I gave as a 5-minute presentation at Hypertext2003. The central point was that the information could be as large and complex as you like on the basis that it was held in XML. It could be rendered in different styles according to need and taste and interactivity but the central content was unaltered.

  I shall be most interested to see how GW holds its XML. Compound XML documents are possible using multiple namespaces and we’ve been working with Microsoft on this in Chem4Word. There we have three main languages, OOXML (the text), CML (chemistry) and MathML (maths). Add SVG for the graphics and you have the basis of a modern piece of scientific information.

  Our experience of the last 10 years is that much chemistry can be chunked into XML, most commonly molecules, reactions, spectra, crystallography and computational chemistry. CML can hold all of these and there are Open Java libraries which can manage all the basic functionality (display, editing, filtering, searching). This means that we should be able to start to play with GW.

  Note the word Open. GW is based on Open Data, Open Source and – hopefully – Open Standards (these are the least clear). I doubt very much whether the closed source found in much of chemistry can play with GW. In the virtualised or cloud environments you don’t know where you software will end up and if you have to worry about licences and binaries the impedance will kill it. Similarly the legacy formats (CDX, SD/MOL, JCAMP) cannot distribute well in such an environment whereas generic XML technology (XSLT, XML-DBs) can be found everywhere.

  The chemistry community is well-placed to take this up as long as it adopts all of the Opens. That is, however, a culture-changing requirement. Currently many players make money by creating toll gateways for data, code and integration. GW blows that away.

  So you need to start mouthing the Blue Obelisk Mantra – ODOSOS.

  What’s it going to impact most? I’ll be thinking and writing about this. But any publisher who ignores GW and does not let it into their soul will not be able to survive. Any data aggregator whose data are not truly Open will struggle against the new ways of publishing. And anyone whose code is closed will not be able to play in GW.