Category: Blog

Once again the issue of “free” and Open Source mapping has been raised by the Guardian , as always a number of good points are made along with a few Gilligan-isms..

The main thrust of the piece is about the potential of open source, community produced mapping data with interviews with representatives of the Mappinghacks and Open Street Maps projects, all great stuff about producing mapping data appropriate for the use of the citizen using GPS and web based editing tools. It is very exciting and does offer the potential to develop the UK GI industry in many different ways similar to the impact of the open source movement on commercial software development.

However.. you knew there would be one..

The article compares OS datasets which are more detailed and maintained daily by a team of 300+ surveyors, surveyed to 10cm accuracy with very expensive GPS receivers, to the data collected by volunteers using handheld GPS costing a few hundred pounds from Dixons.

I am not critical of the open source mapping movement here, but the needs of Utility companies, Local and Central Government often can only be met by high accuracy, up to date data, which is very costly to collect and for which they are willing to pay.

We need to try and take the political steam out of this debate, as I have noted before the decision as to who pays for the collection and maintenance of high quality data is purely political, but I don’t see any potential UK administration choosing the funding of the OS (£100m ?) above other more pressing social spending such as Schools and the NHS.

There are many needs however which don’t require such high accuracy, high cost data for which the OS does not have a viable product offering. This is where open source mapping and the potential improvements to it developed by other third parties offers the greatest potential. Here the UK could follow the example of the US where a range of GI companies could flourish adding value to this data.

The Gilligan moments then…

INSPIRE will result in the mapping of “every lamp-post, phone-mast, river, mountain” etc in Europe !

VMap1 (that’s 1:250K Remember) classified by US intelligence as the most detailed map ever drawn !

Well Known in the GML community Schemasoft the Vancouver based software company was acquired by Apple Inc on Wednesday. Although most well known in the GI industry from its GML – SVG tools, Schemasoft has worked behind the scenes with all the major vendors including Microsoft, Correl and Apple.

So no need to get too excited Apple intrest here is in Schemasofts file conversion and data management expertise which we may see appearing in Apple applications such as iWork , I don’t think we will be seeing GML on a iPod just yet.

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

An interesting article in Electronics Weekly reports on the findings of a DTI sponsored study visit to Japan looking at the uptake of Location Based Services (LBS). The headline is that less than 10% of the users of advanced LBS capable phones on the KDDI network actually make any use of the available services.

This would seem to agree with my perception of the market in the UK, if anything perhaps here it is even less than 10%. In this particular case we may be seeing the effect of a particular technology limitation, KDDI use A-GPS on their phones so get high location accuracy but first fix may take more than a minute – Japanese teenagers it appears are at least as impatient as those in the UK!

My guess however is that the problem lies at least as much with the applications that make use of this technology as with the technology itself. In the UK as in Japan I believe, a user must make an explicit decision to use an LBS type function, e.g. find me the closest ATM. Using the best designed WAP interface that will take at least 2-3 mins including time for the user to connect to a portal, the network locating the users phone, the backend GIS analysis and the presentation of results. Time to ask somebody 30 seconds !!

I have argued before to anyone who will listen that for LBS to work the whole service must be transparent to the user, as soon as you switch on your phone in the background various analysis can be taking place so that the most common requests are pre-calculated, and instantly available as contextual information.

When you are roaming the phone and various networks are working to transfer you to the provider with the strongest signal at any point in time, a process invisible to you other than the operator logo changing on the phones screen – this is how LBS should work !!

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

The ongoing and mostly ill-informed debate on the funding of digital geographical information today hit the national press with a article published by Michael Cross in the Guardian . Now I have been a great fan of Cross’s often insightful journalism in the past but today’s article is full of errors and misrepresentation of the facts.

Now before I’m accused of just following the Ordnance Survey party line I must state that Cross does make some valid and interesting points illustrating some of the challenges that a commercially focused OS must must face to compete in an active market place.

Cross seems to see this competition as a bad thing, but if you accept the principle that the “user pays” and that the OS is a trading fund, then the OS must compete to survive. The Ordnance Survey would not last long if every time a new more competitive product came along, the OS gave up and rolled over. I would argue that this cannot be good for the user of geographic information, in every market competition brings improvements both in terms of value and product innovation.

The heat generated about national address databases continues, and one day this sorry story will become public, Cross is right to complain, but on behalf of the tax payer not commercial interests. Acacia which Cross mistakenly calls a company was actually a government funded project which failed to make much progress in developing a method to combine the existing databases to produce the much needed definite national database.

The “Open Access” debate is a good one, and I can see the benefits of providing data a no cost to end users of course, the more people who get access to digital data the better!! But there is one fact here we often lose sight of, there is no such thing as free data.
It costs somebody to collect, manipulate and manage the data collection process, and to maintain the detailed data we are used to here in Great Britain is very expensive. The “open access” advocates argue that this funding is the role of government or of course the tax-payer really, as takes place in the U.S.A.

But is Government really willing to fund the activities of mapping agencies in this way ?

In the US, the equivalent of the OS the USGS is so poorly funded that it has not yet completed mapping the whole country!! and those maps which do exist are often decades out of date ! Because of the wonderful “Open Access” policy the wealthiest country on the planet will never be completely mapped to a consistent standard!

I personally wish there was a way to make digital data available more cheaply, but detailed accurate information about the every changing world around us is expensive to collect.

I hope that the various “Open Source” GI database projects such as Mappinghacks and Open Street Maps are successful in providing free mapping data and these initiatives are very exciting however they will never be able to do what the OS does.

As a politician what would you argue to fund using you hared earned tax payers pounds or dollars? Building more schools or decreasing hospital waiting lists or funding the creation of digital geographic databases an activity which can cost you nothing and indeed generates you some income if done well!

Written and uploaded from the Airport Lounge at Leeds/Bradford Airport using a 3G data connection.

Now back from the GITA show I am reflecting on one of the few events, which caused much excitement at the show – or actually did not occur at the show. Much of the GIS press both print and online flew into a rage on Wednesday on the announcement in a number of blogs of the release of the next version of AutoCAD, yes you guessed it AutoCAD 2006. Why the fuss – well these guys had been working to a press embargo for the official launch, which was to have been today!!

This was particularly embarrassing for my ex-employer Autodesk as the bloggers had been given permission from marketing to “leak” information and included notable AEC figures such as Lynn Allen Autodesk’s Technical Evangelist.

Red faces all round but it is interesting that Autodesk’s marketeers are recognising the importance of the blogging community as a influencer in addition to the traditional press.

A strong set of presentations today from Autodesk, ESRI and Oracle all touched on the hot topic of the moment, interoperability what does it mean and where do you do it ?

The Autodesk (and I guess MapInfo, Intergraph, LaserScan etc) view as expressed by my old friend Geoff Zeiss was that this is a database issue using an open but proprietary interface on top of spatially enabled RDMS (e.g. Oracle) multiple users using different vendors are able read data from a single repository. Geoff to his credit pointed out that it is still not possible to have consistent write access or manage anything like a long transaction between vendors.

The ESRI view is that there is a place for this database level interoperability in some cases but more often than not interoperability would be at the application level both to other GI based applications and other corporate enterprise applications such as CRM or ERP. This I think is a development of the previous ESRI message which was solely focused at application level integration.

What I found most interesting however was David Maguires ability not to mention Oracle once during his presentation but mentioned IBM websphere and DB2 on numerous occasions – is there more to this I wonder ?

Written and submitted from my hotel room using the in-room high speed internet connection

A good presentation by Peter Batty of Ten Sails which introduced the concept of Sentient Computing, I guess an extension of using Geographic Information to provide context to other systems. For example the flow of a patient moving through a hospital along with their records could be tracked, if the two were separated an alarm would be activated. Another example could be the reconfiguration of a meeting room to suit the needs of its occupants. These application rely on accurate short range location determination technology based on things like RFID tags but as Peter pointed out the technology is not there yet.

My worry is that when the technology becomes mature as it has with mainstream LBS application developers will still take the conservative user activated view of services, so that rather that automatically offering services tailored for you based on your location you must ask for them through some clunky interface.

Written and submitted from my hotel room using the in-room high speed internet connection