The Mesh Behind the Mash

Great article by Jack Schofield on mashups. The journalistic detail it brought to this amorphous and currently very trendy Web 2.0 idea helped me understand something that I'd vaguely realised before, but hadn't fully been able to articulate.

The reason that so many mashups use Google Earth (aside from the fact it's a clever application and freely available) is that to bring together information from different sources you need something in common - a kind of peg on which to hang the data. Location is a very natural peg to choose, since everybody carries around in their heads a representation of the physical world, which they use to navigate through it. Moreover, we instinctively use it for our own mashups - the experiences and knowledge of life that are tied to locations. Google Earth therefore provides a convenient and very natural mesh for mashed-up online data.

In fact, it's hard to think of any other mesh that combines such fine granularity with this ease of comprehension. Perhaps something similar could be done with time (which, anyway, is simply the fourth dimension, and very similar to space) or Wikipedia entries (or subsets of them), since the latter are effectively a mesh for the non-physical world of ideas.

Update: I've now come across this interesting matrix of mashups. It shows that Google Maps is indeed the most popular mesh; others include Amazon, Del.icio.us, Flickr and Technorati.

Google: Friend or Foe?

"Don't Be Evil" is the company motto: but is Google for us or against us?

I'm not talking about justifable concerns that it knows far too much about what interests us - both in terms of the searches we carry out and (if we use Gmail) the correspondence we send and receive. This is a larger issue, and relates to all the major online companies - Microsoft, Yahoo, even Amazon - that mediate and hence participate in much of our lives. What concerns me here is whether Google can be considered a friend of openness.

On the one hand, Google is quite simply the biggest open source company. Its fabled server farm consists of 10,000s/100,000s/1,000,000s (delete as applicable) of GNU/Linux boxes; this means that anyone searching with Google is a GNU/Linux user.

It has a growing list of code that it has open-sourced; it has sponsored budding hackers in its Summer of Code programme; and it keeps on acquiring key open source hackers like Guido van Rossum (inventor of Python) and Ben Goodger, (Firefox lead engineer).

On the other hand, Google's software is heavily weighted towards Microsoft Windows. Programs like Google Earth and Picasa are only available under Windows, and its latest, most ambitious foray, the Google Pack, is again only for Microsoft's operating system. This means that every time Google comes out with some really cool software, it is reinforcing Microsoft's hold on the desktop. Indeed, we are fast approaching the point where the absence of GNU/Linux versions of Google's programs are a major disincentive to adopt an open source desktop.

This dilemma is unlikely to be resolved anytime soon, since Google clearly wants to serve the largest desktop market first, while drawing on the amazing price-performance of free software for its own computing platform.

But there is another area where it has the chance to play nice with openness, one that does not require it to come down definitively on one side or the other of the operating system world.

Another Windows-only product, Google Talk, is the subject of a lawsuit alleging patent infringement. However, closer examination of the two patents concerned, Patent Number 5,425,085 - "Least cost routing device for separate connection into phone line" - and Patent Number 5,519,769 - "Method and system for updating a call rating database", suggests that one of the best ways Google could show that it is a friend of both open source and proprietary software is by defending itself vigorously in the hope that the US Patent system might start to be applied as it was originally envisioned, to promote innovation, not as an easy way of extracting money from wealthy companies.

Update 1: Google has come out with a Mac version of Google Earth. It's a start.

Update 2: There are rumours about Google working on its own desktop GNU/Linux. Frankly, I'll believe it when I see it: it's a poor fit with their current portfolio, and the margins are terrible.

Update 3: Comfortingly, these rumours have now been scotched.

Open Data - Good; Open Access - Bad?

Great story in Nature about data mashups - the mixing together of data drawn from disparate sources to create a sum greater than the parts.

This approach is not new: it lies at the heart of open source software - where chunks of code are drawn from the specialised databases known as hackers' brains and stitched together - and open genomics. Indeed, bioinformatics represents a kind of apotheosis of the mashup - see, for example, the way in which data from many researchers is pulled together in a genome browser like Ensembl.

Data mashups are more recent, and have started to gain popularity thanks to Google Earth. This provides a useful and conceptually simple scaffolding for other data to be brought together and displayed - like Nature's own avian flu mashup.

A pity, then, that this paean to the virtues of open data is not itself freely available under an open access licence. (For the benighted, the indispensable Open Access News has a long quotation that conveys the essence.)