Sunday, October 31, 2004

Gmail exploits

The Register is reporting that Google's GMail is vulnerable to a security explot that could give an intruder full access to your email. I must admit to being unsuprised. Web applications are notoriously hard to secure, this hole will be fixed, but there will be others.

I keep the bulk of my mail on local disks and will continue to do so, and using IMAPS allows me to access it from a remote site. Why do I need web mail to do this? As I've discussed before I like the GMail interface, a lot. But I'm not going to trust my mail to someone else's server unless I have to...

Rail chaos

Well, it took me six hours to complete the journey home from Heathrow. In theory it can be done in just over three, but that assumes our rail network actually works. I arrived back in Exeter in the pouring rain, exhausted, and without a taxi in sight. It took another three quarters of an hour for a taxi to turn up. Don't even mention the buses...

The wrong sort of engineer on the line...

I wasn't alone in my misery, indeed compared to some I was relatively well off. Storms closed the Dawlish coastal route. Waves crashed over the lines, cutting power to two Virgin trains, stranding them and their passangers who later had to be rescued.

I'm on my way down to Plymouth tommorrow morning so I'm hoping the line is reopened, otherwise I'm in for no end of grief.

The return leg

Unlike the outward journey, where I was flying direct to LAX from Heathrow, I ended up having to go via Chicago on the return leg.

Now normally, I wouldn't mind this, as I know Chicago airport fairly well. But half an hour after our scheduled take off time we were still sitting at the gate in LAX. It didn't bode well for making my connection back to the UK, and since my connection was the last flight of the day I'd probably get stuck in Chicago for an unscheduled layover if I missed it.

Waiting at LAX, I can see the plane...

According to the ground staff at LAX, who were looking fairly harassed by this point, the delay was due to the airspace over Chicago being, well, full. In the end I think they bundled us aboard the 747 just to get rid of us, as we sat on the tarmac for another half an hour after they closed the doors before we got pushback.

In the end we took off from LAX an hour late, considering I had an hour and twenty minutes to change planes in Chicago I wasn't actually in too bad shape. I was definately going to make it, so long as we weren't stacked and racked over Chicago. The chances of my luggage making it were, of course, fairly slim.

A suspicously empty gate for the LHR flight

Ariving into Chicago I got lucky, my gate for the Heathrow flight was only two down from my arriving flight, an easy change considering I had a full 10 minutes before they were going to close the doors to get there.

When I got there I thought I'd got turned around, the gate was suspicously empty for a flight that was scheduled to depart "real soon now", where were the queues, the massed throngs of people? It turns out that 3/4 of the people scheduled to be on the flight were stuck over in a different terminal on another (late) plane. I walked onto the plane, almost one of the first to arrive.

When I made it back into Heathrow, my luggage even turned up, now that's what I call service.

On my way to LAX

Hitched a ride with the JAC gang to LAX, as they were getting a a flight back to Kona only a hour before mine, and they had a free spot in the car. Fortunately, we only got a little bit lost...

On the way to LAX, err, somewhere...

Wednesday, October 27, 2004

It's a wrap...

We've just wrapped up the conference with closing comments from Arnold Rots. Next year's ADASS, the 15th annual, will be held in San Lorenzo de El Escorial close to Madrid in Spain.

The last talk...

Eric Sessoms talked about hardware accelaration for astronomical data analysis. He compared the performance of your GPU to the main CPU of your machine. Typically the performance of a commodity graphics card exceeds by a factor of 3 that of your CPU.

More information about using GPUs for general purpose computation is available at the site.

Lack of instant gratification

The online Apple iPod Store says that "Instant gratification [is] available at an Apple retail store or an iPod reseller near you". Well I'm all for instant gratification so along with the JAC crowd, and purely in the interests of online journalism, I ducked out of the conference and went down to the Apple Store in Pasadena.

We were quite horrified to learn that, not only wouldn't they have the new iPod Photo to sell for a "few" weeks, but they wouldn't even have a demo model till later that afternoon. So much for our instant gratification. We were forced to retreat to Moose McGillycuddy's for lunch and conciliatory beer...

Just noticed that Brad is also blogging ADASS XIV...


Another snippet of news from the plenary talk was that the Cassini probe sucessfully carried out its flyby of Titan last night. Data download continued through the night into the early morning hours of this morning. The flyby was by far the closest any spacecraft has ever come to Titan.

This image is one of the closest ever taken of Saturn's hazy moon Titan. It was captured by Cassini's imaging science subsystem on Oct. 26, 2004, as the spacecraft flew by Titan. At its closest, Cassini was 745 miles above the moon, 300 times closer than during its first flyby on Jul. 3, 2004.

The Huygens probe, built and operated by the ESA, is still attached to Cassini and is scheduled for release on Christmas Eve. It will then attempt a touch down on Titan on Jan. 14 next year.

Update: The Huygens probe sucessfully landed on the surface of Titan at the start of 2005.

The Genesis probe

Listening to the plenary speech this morning one snippet of interest is that, despite the bad landing, it looks like the samples from the Genesis probe remained intact enough to get some science output from the mission. Apparently there will be "almost no data loss", which is good news considering how badly beaten up the probe must have been after the crash.

Conference Banquet

Last night was the conference banquet, weirdly enough it was held at a Santa Anita Park which would have made a lot more sense if there was a race for us to watch during the meal.

The after dinner speaker was Raymond Arvidson, from JPL, who talked about the Spirit and Opportunity Mars rover missions.

Spirit and Opportunity, or the Red Sox playing in the World Series?

We decided on Spirit and Opportunity...

Starlink for the VO

The next talk in the afternoon VO session was given by my boss Dave Giaretta who talked about the Starlink project, and how we've gone about VO enabling the Starlink legacy classic software, along with developing complimentary "next generation" Java software.

Open SkyQuery

This afternoon Will O'Mullane did a live demo of the NVO Open SkyQuery portal. During the demo he showed how you could use it to upload and then cross match your own data against one of the "standard" catalogues (such as 2MASS or SDSS) served by an existing SkyNode. I must admit to being much impressed by the progress the NVO guys are making...

Tuesday, October 26, 2004

An afternoon of VO

This afternoon we had Keith Noddle and Nic Walton talking about AstroGrid. Disappointingly the URL which Nic advertised where we could find the AstroGrid release currently gives a 404 error, presumably something will turn up before the NeSC meeting in December.

The iPod Photo

Apple has just released the iPod Photo. Rumours about this product have been spreading for several weeks. With the new iPod also comes a new version of iTunes (version 4.7).

The new iPod Photo

The new iPod Photo comes in 40GB and 60GB versions. More information about the new iPod is available at the Apple iPod Store.

The VOSpec tool

Pedro Osuna from ESA/ESAC talked about VOSpec. While there was a demo of this tool yesterday in one of the focus sessions, I can't find anything on the web about it. It looks like it hasn't actually been officially released, so despite the pretty pictures, it's pretty hard to get a handle on what it can actually do.

The new ESA VOSpec tool

To be honest, it looks a lot less capable than the Starlink SPLAT tool.

The Aladin experience

In the first of the Vitual Observatory (VO) sessions François Ochsenbein talked about Aladin, the CDS killer application.

Aladin development started 7 to 8 years ago and has evolved into a tool which includes a number of services, fufilling the need to put astronomical data onto the astronomer's desktop. The goals of Aladin development is interactiveity, performance, ease of installation and connectivity.

Aladin in written in Java and one of the strengths of Java is the ease of interactivity, but it has notoriously slow. Aladin has always pushed the limits of Java performance as it has improved (e.g. JIT compiling). However it also provides good ease of installation, although the applet concept is now dying.

The real power of Aladin is in it's connectivity to a large range of different data sources, and it's ability to interoperate with other Java applications like VOPlot and VOSpec.

The next step is to extend Aladin to access the emerging web service infrastructure.

Morning poster session

Either there isn't enough poster space, or there are a lot more posters than normal at this ADASS because posters are only up for one day rather than the full run of the conference. My FROG poster was up yesterday, and today Malcolm and Norman got their turn.

The Tuesday poster session

Norman was talking about the new Starlink build system, while Malcolm's poster compared JAC/Starlink ORAC-DR to the ESO Eclipse environment for ISAAC reduction.

I'm back...

I've got IMAP access again, so my mail "situation" is back to normal.

Breakfast meetings

I skipped out of the conference to go and see an old friend for breakfast, all very Hollywood...

Warren and I did out PhD's together at Keele, but we've only run into each other once since then. However he's out here working for Caltech, doing site testing for for CELT, so despite hectic schedules we sort of had to arrange to meet.

You've got mail...

In fact I've got quite a lot of it. While I've managed to access my inbox, for the first time since late Saturday night, my home machine is now dead. I'm having to wade through all the spam by hand that is normally automatically pre-filtered using SpamAssassin. I hadn't actually realised how much spam I was getting these days, it looks to be several hundred messages a day. In any case, this means that while I do have access to email, I'm going to be a lot slower to respond than normal.

The afternoon demo

I spent the afternoon manning the Starlink demo booth. Most of my time was spent burning DVD's of the current Mac OS X build of the Starlink classic software which seemed to be in very high demand...


Over the lunch hour Mark Talylor presented the TOPCAT, the Starlink interactive editor for table data, and the underlying STIL infrastructure library in a focus session. It was well attended and we got a lot of interest afterwards at the demo stand.

Mark talking about TOPCAT and STIL

Ground based observatories

After coffee was the first of the ground based observatories session. To kick off the session Eugene Magnier of the University of Hawaii talked at length about the Pan-STARRS image processing pipeline (IPP) architecture.

Following this Roc Cutri from IPAC talked about the 2MASS Extended Mission and ancillary data products. There were three main data products which come out of the 2MASS survey,However there are ancillary data products now being created, while there is no new data acquisition, this data will push fainter but with lower reliability and non-uniform sky coverage. Will mostly be available by the end of this year.

Super-resoultion for Spitzer

Charles Backus talked about super-resolution imaging with the Spitzer Space Telescope.

Spitzer provides good S/N, critically sampled with a relatively small telescope, and their observation strategies provide redundant coverage. They do their super-resolution, using the Richardson-Lucy algorithim, assuming a spatially invarient point-spread function (PSF) and uniform pixel noise. With Spitzer data you typically see a 3:1 reduction in the PSF width using the technique.

Monday, October 25, 2004

Visual data mining and VR spaces

Julio Valdes talked about data mining. Which is, apparently, "the non-trival process of identifying valid, novel, potentially useful, and ultimately understandable". The goal of his research is to combine the best results from the human and the machine in a visual tool.

Why use a visual representation at all? He argued that the amount of information in relational structures is large, humans are well qualified to understand complex geometric structures. Additionally, when you present results visually you need little mathematical knowledge to understand the results.

He talked at length about the formal mathematics behind representing data, and then demonstrated some "virtual spaces" for data representation.

Blind Deconvolution

After the obligatory speeches of the plenary session, Tony Chan of UCLA talked about blind deconvolution. It's called "blind" deconvolution because you don't necessarily know the point spread function convolved into your image, you still want to recover the original image. Tony talked about using total variation (TV) regularisation technique. Results looks fairly impressive.

He also talked about ways to distinguish the correct from alternative non-correct deconvolution solutions. For instance by using the "sharpness" of the recovered image.

Mail problems...

For anyone trying to reach me right now, my Exeter email address is out of action, along with my hard drive and my entire desktop system back home. It always waits until you're on the other side of the planet to fail catastrophically doesn't it?

You should be able to reach me on my address, however the Apple is acting up since my Exeter IMAP account isn't there and the message filters are trying to put messages into non-existant folders.

If you really need to reach me, I think you more or less have to resort to the phone at the moment...

Free as in beer

Well it seems like there really isn't such a thing as free beer. The beer remained stubbornly non-free at the ADASS conference reception. Normally there is at least a couple of free beers to get you in the mood before they start charging people for the stuff, but this time the cash register was ringing right from the start of the reception.

In the end I folded early under the weight of jet lag and my early start. I went to bed well ahead of the crowd, maybe I'm just getting old...

The Apple Store

While I was in town today with the JAC gang we wandered across, quite by accident, an Apple Store in Old Pasadena.

The Apple Store in Old Pasadena

There were, predictably, a lot of wonderful toys in the store, but the two things I was drooling over the most were the new iMac G5, and the amazing 30 inch Cinema HD Flat Panel Displays. I mostly resisted the urge to spend a lot money...

The Apple 30 inch Cinema HD Flat Panel Display with an iSight, attached to a Power Mac G5

The new Apple iMac G5

However, I couldn't resist one purchase to simplify my life a little. I picked up an Apple World Traveller Adapter Kit which will allow me to get rid of the tottering tower of adaptors I'm currently using to plug my PowerBook into the mains while I'm in the States.

The World Travel Adapter Kit

An early morning

I was woken up this morning around 5:30am by the JAC mob who sent me a text message to say they were on their way in from the airport and did I want to meet them in the lobby. Since I was now awake anyway, I thought I might as well and went down stairs for an early breakfast. That'll teach me to leave my phone on...

Sunday, October 24, 2004

On the ground...

On the ground in Los Angeles, fed and watered, but very tired. I'm staying at the conference hotel, the Sheraton Pasadena. It's a nice hotel although it is starting to look a little tired, in need of renovation, but still very comfortable.

Saturday, October 23, 2004

"Company Policy"

Last time I was going through Heathrow I spotted the TomTom Go in-car GPS navigation system in Dixons in the Terminal 3 departure lounge. Well, it's still there, and I still don't have one.

The TomTom Go

Of course I wouldn't buy it in Dixons. Even though it was being discounted, and was "tax free", it's still more than £10 cheaper to buy it online at But I wasn't given the chance...

While I was browsing I took a few pictures of the merchandise with my camera phone. I was only in the store a few minutes however when an assistant approached me and told me that I was not allowed to do so, and that it was against "company policy" to allow me to continue. I was, as they say, politely asked to leave. I did so, scratching my head the entire way out the door.

I guess they don't want people doing comparisons like the one I did above. Dixons is charging £499.99 for the TomTom Go on the high street, so the discounted £425.52 price tag looks a pretty good deal, until you realise that is charging just £412.98. It goes to show that, as always, you have to shop around it you want the best deal.

At Heathrow

Where there is Starbucks, there is civilisation. Even in the Terminal 3 departure lounge at Heathrow...

Friday, October 22, 2004

In transit...

On my way to ADASS which is going to be in Pasadena, CA., this year. I'm currently holed up in the Marriott London Heathrow staying overnight here before flying out to the States in the morning.

Wednesday, October 20, 2004

Frantically doing nothing...

There can surely be nothing as boring as waiting for software to compile. Yet at the same time, it's an activity where your boredom is contaminaed with fear and panic. Fear that the software won't compile, and then panic when it doesn't. Which, of course, is the usual outcome. This is especially true when the software in question is several million lines of ancient Fortran and C held together by a build system so new we're still rounding the corners off, and it all has to work by Friday when the team get on (at last count) four separate planes for the States, which all end up in LAX within a couple of hours of each other.

Watching build messages you've seen a thousand times before scroll up a terminal window, especially when the build cycle is several hours long, can be a really draining experience. The brief happy moments when it passes the point it crashed last time, closely followed by depression when it crashes again somewhere different. Usually to do with the thing that fixed the first crash. It'd make anyone into a philosopher.

So what deep insight into the cosmic all have I come up with? Mostly, that I really should use a bigger font in my terminal window...

Good Bad Attitude

Paul Graham has just published another excerpt from his excellent book Hackers & Painters onto his website. If you're a regular reader you'll know that since I listened to Paul speak at OSCON I've been a big fan of his work, he's fast becoming one of my favourtie dispensers of wisdom.

SOAP::Lite V0.65 Beta

I talked about this in more detail earlier, the 1st Beta release of SOAP::Lite V0.65 has now officially arrived.

Tuesday, October 19, 2004

Mobile phones & Virgin trains

I'm sure everyone who has travelled on them extensively has already figured this out, but Virgin Trains have come clean and admitted that the design of their new Voyager and Pendolino trains cuts off mobile phone reception.

As a regular traveller on the cross-country route it's been very obvious to me that reception in the new trains is a lot poorer than it used to be, with many more patches where there is little or no signal.

However Virgin also said that they were in the process of testing signal boosters onboard carriages to improve the mobile reception, and that if these tests were sucessful the boosters would be rolled out over the entire fleet. But there really isn't such a thing as a signal booster for mobile phones, so I'm guessing that they may actually be experimenting with picocells.

Now this is interesting because it means they might also be experimenting with onboard wireless. GNER has been rolling out wireless on it's Mallard services running along the east coast mainline since July, and as far as I'm concerned, onboard wireless is the holy grail of train travel. I'd be quite willing to pay through the nose for it if Virgin introduced it on the routes I normally travel...

Apple buyer's guide

With the release of the new iBook I thought I'd point people towards the MacRumors Buyer's Guide. The site seems to be making fairly reliable predictions about the Apple product cycle, for instance, it correctly predicted the iBook release.

Interestingly it's currently predicting that a new PowerBook, based on a dual core G4, may be due for release soon.

The new iBook G4

Apple have just announced the new range of G4 iBooks. New features include built-in wireless, a DVD burner and a low end G4 processor. It looks like the highest spec G4 available on the new iBook will be 1.33GHz, this is the minimum spec available for the PowerBook range.

The new iBook

Being Apple the new iBook looks amazing, but we've started to come to expect that these days. Overall it looks to be a solid entry level addition to the current range.

Thursday, October 14, 2004

The Prisoner's Dilemma

The Prisoner's Dilemma is a classic problem in game theory. Wired is covering the sucess of a new stratagem against Tit for Tat, which won the original Axelrod Iterated Prisoner's Dilemma competition.

In this year's recreation of Axelrod's classic competition, the winners, from the University of Southampton's Intelligence, Agents and Multimedia (IAM) Group, put forward a solution which allowed their entries to collude together, rather than compete with each other.

The Southampton press release doesn't cover the topic in much detail, but does they do say that "No outside communication is allowed so the agents have to recognise one another by playing a specific sequence of moves at the start of each game". A lot of people have complained that the Southampton entries into the contest were somehow cheating. This somewhat misses the point of what the Southampton team was trying to accomplish.

Expedition 10

The BBC reports that the Expedition 10 has lifted off from Kazakhstan today on their way to relieve the previous crew. The Expedition 9 crew are scheduled to return home in October onboard another Soyuz craft which is currently docked at the Station. During Expedition 10 an additional two Russian Progress cargo flights are scheduled, carrying supplies and scientific equipment to the station.

Expedtion 10's Soyuz launching from Kazakhstan

Tuesday, October 12, 2004

Global air pollution map

ESA has issued a press release showing a global atmospheric NO2 map built up from the the last 18 months of Envisat measurements.

CREDIT: Institute for Environmental Physics, University of Heidelberg
The image shows the global mean tropospheric nitrogen dioxide (NO2) vertical column density (VCD) between January 2003 and June 2004, as measured by the SCIAMACHY instrument on ESA's Envisat. The scale is in 1015 molecules/cm-2

Usually, despite the pervasive presence of trolls, you can usually rely on Slashdot to turn up some decent comments about technically related news. However the level of ignorance, scientific and political, displayed by some of the posters in response to this story is somewhat scary. So unless you're looking to make your blood boil, I'd steer clear this time. More details about the actual measurements can be found at

Sunday, October 10, 2004

Lifeboat gear stolen

The BBC is reporting that thieves broke into the Norfolk RNLI station and stole equipment including radios, night-vision binoculars and GPS gear. If you're offered anything that you think might have come from this theft you should contact Norfolk Police immediately.

Saturday, October 09, 2004

The coming of SOAP::Lite V0.65

It has arrived! Hallelujah! It has arrived! Or rather, it will shortly at any roads...

It's been over a year since the last release of SOAP::Lite, which is a long time in the web service world. Rumours of the next release started started spreading immediately after Byrne made the announcement on the SOAP::Lite home pages.

Perhaps the most pressing problem with the distribution that the new release should (will?) fix is the documentation. SOAP::Lite has always been powerful, but it was heavily under documented, and most of it's user base had a really hard time getting to grips with it. Randy's new documentation should clear up a lot of problems, for a lot of people. However I'm hoping Byrne has included his own contributions as well.

It's also good to see that DIME support is in the works, although it doesn't look like it's going to make it into the 0.65 release.

I must admit to much shame at this annoucement, as I promised Byrne that I would get round to looking at integrating DIME support during OSCON 2003. Unfortunately I got side tracked into working in Java for Starlink, rather than in Perl for eSTAR, and I never got my code into a state where it could be integrated. My bad.

Multiagent Systems

An Introduction to Multi-agent Systems
by Michael Wooldridge, ISBN 047149691X, paperback, £20.99
Also available in an Adobe e-Book edition for £32.75.

I've just finished reading Michael Wooldridge's book on multi-agent systems. As a physicist I've found a great deal of the literature in this area very hard to penetrate, there is a lot of overlap with sociology and several other of the fuzzier disciplines, which means that the hand waving can get a bit much.

The publisher's blurb says that, "This is the first textbook to be explicitly designed for use as a course text for an undergraduate &graduate courses on multi-agent systems. Assuming only a basic understanding of computer science, this text provides an introduction to all the main issues in the theory and practice of intelligent agents and multi-agent systems."

Personally I think that the assertion that the book assumes "only a basic understanding of computer science" is a bit on the far fetched side, you'll need a decent grasp of a high level language, several would be better, a nodding acquaintance with distributed systems, some familiarity with the basic concepts behind artificial intelligence and (rather crucially) a familiarlity with set and logic notation which is heavily used throughout the book to explain a lot of the concepts discussed in the text.

However, if do have the background knowledge required I'd certainly recommend An Introduction to Multi-agent Systems as just that, and excellent formal introduction to agent architectures.

After borrowing it from my graduate student's desk, I recently read the book cover to cover during two tedious train journeys. I picked up a couple of new ideas, and got some formal grounding using set and logic notation on a whole bunch more stuff about which I'm already fairly familiar.

If you're moving into this area and need a book to make the rest of the literature a bit more transparent, you could do a lot worse than this one, if a poor beffudled physcist like me can understand it, anyone can.

Friday, October 08, 2004

Breaking Google's DRM in Safari

(Jan 2006): This post is about the pseudo-DRM implemented by Google Book Search, if you're looking for more information about the DRM used by the new Google Video Marketplace the Register has a good article on the issues surrounding its introduction, and of course there is always Slashdot.

Well for Mac OS X users at least Google's DRM doesn't look to be that hard to work around. For instance Google give an example book on the Print Service page, so working from there...

"Mastering Digital Photography" from Google Print

Selecting the image which contains the scan of the book page

Using Safari all you need to do is open the Activity window, and select the image component representing the scanned page, then just double click on it to open it in a new window free of DRM. It's hardly rocket science...

Of course anyone wanting to find out how to bypass the Google DRM will probably have used Google to find this post to tell them how to do it, bit ironic really.

I'm in two minds about the entire Google and DRM thing. The DRM we're all used to hating is normally applied to content we ourselves have bought, and that should we by right have total control over. In that case we are being stopped from using our own possessions in the manner we desire. On the other hand, here we haven't paid for anything. Google is giving access to this content for free. So as they own the content, they should have the right to determine how and when we can access it. But on the gripping hand we must also consider fair use (and fair dealing).

If you go to a library you could photocopy several pages from a book, take them away with you, and indeed republish the content so long as it was properly attributed. The Google DRM denies us those rights of fair use. I'm still undecided whether I think this is an "evil" thing from the corporation whose philosophy is to "do no evil".

Yet More Google

Slashdot has just reported that the new Google Print service is protected by some interesting DRM. Within a few minutes of the story being up onto Slashdot, several means around the copy protection have been suggested. Gervase Markham, who works on Mozilla, found several ways around Google's DRM. He then went and summarised the multiple ways to do it, and even suggested some alternatives in case Google are starting to get worried.

Interesting, Google may indeed be getting worried. The Google Print service is currently down. I'm getting a Server Error (502) when trying to access the site. Perhaps it's all coincidental, but it does look a bit suspicious to my mind...

Google SMS

I do seem to be talking a lot about Google these days, but then so is everyone else as what they're doing right now is interesting, and gives us a good look at where they might be heading as a company in the aftermath of their IPO.

As I'm sure you've seen elsewhere, Google have introduced an SMS service, allowing users to do some limited queries via text message. As a Brit there doesn't seem to be anything innovative here, and the most frustrating thing about the entire businnes is trying to ignore all the Americans talking absolute nonsense about mobile communications when it was covered on Slashdot.

However, the wierd thing about the entire business is that the service is US only, where text messaging has hardly any market penetration, instead of in Europe where this might make sense. Of course we've had services like these, using WAP over GPRS, for a good few years so perhaps they figure there isn't much point in introducing it here?

Thursday, October 07, 2004

And Finally...

In one of those "and finally" stories, the BBC is reporting the replacment of a road sign directing you to the village of Lost in Aberdeenshire. The new sign replaces the one stolen over a year ago, apparently stealing the sign is a popular past time for tourists...

To the village of Lost..

Data Mining the Environment

The BBC is reporting the UK's Environment Agency decision to put its flood data online. Funnily enough the site is now totally choked, presumably with just about everyone in the UK plugging their postcode into the search engine to try and find out whether they're at risk from flooding or not.

Once the inital rush has calmed down the form below should allow you to grab the flood data without having to navigate your way through the Environment Agency's site. All the maps retrieved by this service are subject to Crown Copyright.

Search for:
Enter a postcode or a place name (England only)
© Crown copyright. All rights reserved. Environment Agency, 2004

Reality Mining

Sensors magazine has recently published an interesting article called Browsing Reality with Sensor Networks.

One of the things I'm heavily involved with these days is the eSTAR project, and my work on the agent technologies which have come out of that project has been heavily influenced by the ideas behind pervasive systems.

The ideas that distributed computing power can be integrated into your everyday environment, providing a user with augmented reailty isn't new. However, with the ubiquity of mobile phones (especially in Europe and the UK) it's possible that it is now inevitable, rather than only probable. Like all such major changes in the way people live, it's also pretty probable that we won't notice it arriving either.

Predicting quakes?

Slashdot is currently reporting that the NASA JPL Quakesim project has sucessfully predicted 15 out of the last 16, magnitude 5 and above, earthquakes in California.

However, editorial control at Slashdot is typically poor, if you actually read the linked press release you'll find that Quakesim has predicted the locations of the quakes rather than timing. It's still impressive, but not as impressive as the spin that the story is getting on Slashdot would lead you to believe. The press release also carefully steers round the issue of false positives.

Interesting, but not earth shattering, no pun intended...

Google Print

As reported in several places, Google is now in the process of launching it's new Google Print service.

The new service seems to offer some direct competition to the (oddly) titled A9 search engine, a spin-off from Amazon, which integrates their Search Inside the Book ® service with the results from a normal Google search (amougst other things).

It's going to be interesting to see whether it really enhances the search "experience" or not. I've held off from using A9, as it didn't seem to offer significant enough improvement over plain vanilla Google to justify the (admittedly rather small) effort of moving. Out of habit more than anything else then, I've continued to use Google, despite the possiblity that A9 could offer a better service. Perhaps my assertion yesterday that the Google conspiracy theorists are wrong was further off than I thought...

Wednesday, October 06, 2004

Google AdSense

As you can probably see, unless you're running fairly decent advert blocking software, I run Google AdSense adverts. Just lately I've been getting some rather wierd ads poping up, as the thread of blog postings moves between computing, space and of course scuba diving.

The fallibility of AdSense is actually quite comforting, it means that the fringe conspiracy theorists are probably way off base when they talk about Google having too much power and influence over the internet. Of course, it could mean that Google isn't being evil enough yet. Perhaps if they were much more evil, AdSense would work better...

The nightmare scenario is if Google cross-correlated the keywords from my web searches, my GMail inbox and my blog postings, based on the AdSense keywords, and of course my Blogger login. All of this is possible, even quite practical, for a corporation with the amount of raw compute power available to Google.

But how good a view of me would they get? Amougst other things, one of the wierder ads that have been turning up has been for White Knight tumbler dryers, obviously due to my many recent postings about SpaceShipOne, whose carrier is also called White Knight. Would they think I'm a man obsessed with tumble dryers perhaps?

The Long Tail

Wired has just published an article called The Long Tail. It discusses the so called "shallow end" of the market, and predicts that from now on niche markets will start to dominate.

The article opens by discussing how the publication of Into Thin Air almost a decade after Touching the Void led to the earlier, almost forgotten, book entering the New York Times best seller list. Eventually it overtook the newer book in sales, almost entirely based on it's Amazon recommendations. It's an interesting article, although some of the author's conclusions are, perhaps, a bit on the shallow side.

Monday, October 04, 2004

On This Day: 1957

On this day in 1957, Sputnik I went into orbit, the first man-made object ever to leave the Earth's atmosphere. The BBC has more...

Sputnik I

X-Prize Claimed

I've been stuck on a train all day coming back down to Exeter from Leeds, so I missed the excitement. But the BBC is reporting that SpaceShipOne has sucessfully claimed the X-Prize with a 2nd sub-oribital flight only a few days after it's first sucessful attempt.

Flight X2 of SpaceShipOne

With pilot Brian Binnie at the controls, SpaceShipOne's flight went off without any apparent hitches. It reached 69.7 miles (112.2 kilometers), well above the minimum target. has more information...

Friday, October 01, 2004

SpaceShipOne go for next flight

The BBC is reporting that SpaceShipOne's flight on Wednesday officially qualified as an X-Prize attempt, and that it's second flight will take place at 0700 PDT (1400 GMT) on Monday.