Category: Thoughts

So many contrasts where to begin… but I have really fallen in love with India, despite the clear poverty visible on almost every street corner, the dynamism of this country is amazing.

India is running an e-GOV programme which makes the efforts of the UK seem rather underwhelming, and although I think I still need convincing on the true value of this, the aim to install 2Mb broadband connections to every local government office in India is breathtaking when you think how many offices there are in the very rural parts of this huge country.

Much on the informed debate at the Graticule conference has been around the upcoming GALILEO programme, the EU programme to launch a Global Navigation Satellite System. It would be fair to say that there was not universal support for this, with many questioning the need for higher accuracy positioning technology when WAAS, EGNOS and the Indian developed GAGAN system can provide improved performance of the current GPS and GLONASS satellites.

I think these questions are missing the important value of quality assurance built into the system which also provides real-time alerts of poor accuracy data, a point made forcefully by Jorn Tjaden of the GALILEO Joint Undertaking, a real requirement for high end navigation requirements such as landing aircraft !

More pressing questions are those I think about the business model for GALILEO, how much funding can really be expected from commercial operators as a “free to air” service is mandated, how many customers will pay for the higher quality service ? I think a Public Private Partnership model is right, the split where the majority of funding comes from the private sector may need to change?

In the location Intelligence arena I was disappointed at the amount of progress in terms of application development, little seems to have been done so far perhaps due to the lack of up to date geodata, although I was impressed by the work of SiRF who are the manufacturing of most OEM GPS chipsets found in phones and pda’s

In my presentation I made my usual point that LBS was never going to be the killer app, to see real value in location, all applications need to be made location aware, so we get away from the current mad situation where I have to type my location into Google Local search even though my wifi enabled PowerBook can be easily located on the network using technology like Skyhook.

And now I’m off to eat another curry !!

Written and submitted from the Taj Place Hotel, Delhi, using the hotels 802.11 network.

This weekend I’m travelling to India to deliver a keynote presentation at Graticule the first International conference on positional technology and Location Intelligence to take place in India. Understandably perhaps the focus on these technologies has been in the west, however this is changing.

India not only offers highly skilled engineers and scientists who are advancing the industry globally, but also has an enormous potential market for the next generation of “location aware” applications. I’ll report back on the conference here next week.

Yesterday, the OS ran the first terrafutures event, a one day workshop to try and bring together scientists of all types, sociologists and IT experts to look into the future and spot the trends in technology and society that will impact on the GI industry.

Some of the key findings included the future pervasivness of network connectivity and the ability of almost everything to be located across these networks in real time. At the same time the focus of economic activity globally is shifting eastward with North America and Europe becoming displaced by India and China, which will result in large new potential markets and well trained, skilled GI professional becoming available at costs a fraction of those today.

How is the traditional GI industry responding to these changes.. well not very fast, for the past 20 years you could argue that the innovation which has taken place has been a result of technology changes in mainstream IT, exploiting developments such as the web and object relational databases. Recent innovations again have come when mainstream IT developments have been adopted for use in geographical applications, Google Maps is the application of AJAX programming to mapping, but this time the traditional GI industry has been bypassed.

The area where GI innovation could make a real difference in the next few years, is also a computer science problem, but one in which GI is playing a leading role – the semantic web.

The semantic web a phrase coined by Tim Berners Lee, the “Father of the Web” relates to an “upgrade” to todays web, where more structured meaningful data is published and subscribed to by web based applications – todays web is human readable and semantic web is machine readable.

We are beginning to see the first semantic web applications which understand geography, recognising place-names in web published documents. This sounds easy but the problem in recognising when ‘Reading” means a place, rather than what you do with a newspaper is a difficult one. For GI this needs the development of more formal ways of describing geographical features and they relationships – the development of geographical ontologies.

There are many challenges ahead for the GI industry, although operationally in many areas the industry has reached maturity, there are still many questions that need to be answered.. yesterdays event was a fantastic opportunity for us GI guys to look out of the silo and take a bearing on the potential direction of our future.

Watch this space over the next week or so, I hope to bring news of our plans to podcast some of the terrafutures presentations, and congratulations to all involved in organising the first of what I hope is a series of similar meetings.

I have just come across geoTorrent.org a website offering links to geodata sets hosted using the bit-torent peer to peer network protocol. The idea of distributing large geodata datasets as small chucks is quite appealing and I have no doubt that when open geodata becomes more mature – this will be the obvious mechanism of supply. One of the aspects of the age of broadband internet often overlooked is that most ISP’s impose some sort of bandwidth limit, which is easily broken hosting large geodata libraries – bit-torent’s architecture neatly gets around this problem.

But what about commercial data providers.. despite all the advantages there is a big problem – peer to peer means piracy in many minds, an unfortunate perception ,but one that needs to be disproved if the GI industry is to really make use of a valuable technology.

Written and submitted from home, using my home 802.11 network.

Back at the end of June, the O’Reilly media Where 2.0 conference took place in San Francisco and saw presentations from Google, Microsoft, and Yahoo amongst others on their perspective of the expanding mainstream GI market. The first audio recording from this conference is now available online at the ITConversations website.

The Past and Future of Mapping is a presentation by David Rumsey a well known map collector and historian who spends 30 minutes looking at the major historical developments in cartography and points out that we are revisiting many of the techniques of the past with the latest generation of web mapping sites. This is a very interest presentation well worth listening to.

I’m just disappointed with a comment David makes at the end of the podcast comparing the situation in the US where USGS makes available geodata at no change, with the OS policy of licensing geodata to fund its operations.

David comments, as do many, that OS geodata should be free – well perhaps he would like to send us all a copy of his book Cartographica Extraordinaire: The Historical Map Transformed” for free rather than charging $80 ? – but then again maybe he needs to recover the publishing costs of the book and make a little profit to expand his collection ?

Unlike the USGS the OS is not funded by the taxpayer, and like David the OS needs to cover its costs.

Written and submitted from home, using my home 802.11 network.

Looks like I’m going to be having a busy September, following my vacation I’m back to talk at the Society of Cartographers Summer School in Cambridge and at a event we are holding at the OS called Terra future. The events although in many ways very different are both focusing in on the increasing use of information, and geographic information in particular in mainstream applications.

Steve Chilton has put together a very interesting agenda for Cambridge, and I’m looking forward to debating public access to mapping data with the evangelists of the “open geodata’ movement including Steve Coast and Jo Walsh co-author of Mapping Hacks.

Terra future, later on in the month looks at the role of location information for all business markets over the next decade, with a eye on developments in information technology and the expected changes in society as these impact on users of GI. Speakers include Peter Cochrane ex-CTO of BT and author of a must read blog on silicon.com, Richard Scase, Prof. of Organisational Behaviour at the University of Kent; and Jayant Sharma – Oracle Corporation’s technical director for spatial products development.

These really are exciting times to be in the GI industry, who would have thought even a few months back, that so much interest would be generated by a few web mapping applications and how much these events would fire peoples imaginations as to what is possible.

Written and submitted from home, using my home 802.11 network.

Tim Martin comments on the hype surrounding Google Earth, asking is google as much a hindrance as help to the GI industry? I think Tim is right to recognise some of the data quality issues apparent in both Google apps, but I think it is a mistake to see Google in any-ways trying to advance the traditional GIS industry.

That is not what they are trying to do, to be honest I don’t think they are very much bothered about the traditional GI industry, it is too small and too inward looking for them to care.

What they have done with great success is make “fit for purpose” geographic information available to the general public in a form that is easy to use and very accessible in the case of Google Earth, and with Google Maps, have made available a platform with which web developers can build their own map based applications complete with geodata – something ESRI has been trying to do for years.

Google are not likely to support spatial analysis in the near future because, it is too specialised and would be of little interest to the mass-market, rather than because it would expose issues with the available data. ESRI seem to have plans to address these more demanding applications across the web using their ArcWeb services project and will I assume understand and respond to the issue Martin identifies.

I thought it was about time I actually tried to use the Google Maps API rather than just talk about it, so I spent a couple of hours over the weekend building a simple google map application. The result Where’s ed? took just an hour or so to put together ( Can’ get it to work with Safari but the bug is a least recognised!), in contrast to build a similar application using Autodesk MapGuide, something I used to be very familiar with, would take a day or two and I would have to acquire the data from somewhere – something which could take weeks!

This is the one place where I think Google does justify the hype, in a few months Google Maps has done more to allow the individual to develop mapping based websites than the traditional GIS industry has done in 10 years. The democratisation of Geographic Information in this way is the result of two things, firstly a simple, slick API for developers and secondly and most importantly of all, the making available of a consistent source of commercial geographic information at no cost to the developer or user.

The advertising based business model of Google has not touched the Google Geography apps yet, if will in time .., but at which point, we the developers and consumers will already be hooked.

Written and submitted from home, using my home 802.11 network.