Wednesday, December 29, 2004

Mobile Python under Mac OS X

Well I've been playing with Python for Series 60 for a couple of hours now, and I'm in love. It is so much easier to use than Java ME to do the stuff I want to do that I've already hacked up a couple of quick useful applications, and I can see possibilites for eSTAR as well. After using the application I'm guessing the reason that Nokia went with Python rather than Perl is the console, it seems to make sense to have it around. Apart from PDL Perl doesn't really have a console, although it'd be pretty trivial implement one...

For those of you who might be interested in developing on Mac OS X, it looks like the only useful thing we're missing is the py2sis application from the SDK which is a Windows only tool.

If you're interesting in using the Bluetooth console facility to develop directly on the phone under Mac OS X, this is fairly trivially done. First of all you need to create a serial port that the script on the phone can connect to using the Bluetooth Serial Utility which is found in Applications > Utilities. Then you need to run a terminal emulation program on you Mac to monitor your newly created port, I'm using ZTerm to do that. Then just run the script on your phone as normal.

Update: I've now got and PyBlogger working under Python for Series 60. Wasn't particulaarly hard to do either, which suprised me...

Update: A version of Perl for Symbian/Series 60 has now been released...

Apple and the headless iMac

ThinkSecret is peddling the rumour of a headless iMac soon to be available for less than $500, I'm less excited about this than I could be because it'll ship in the UK at just under £500 despite an exchange rate of almost two dollars to the pound.

Apple has stayed clear of the sub-$500 PC market for very obvious reasons, so I'm going to be suprised if such a thing as a headless iMac comes to pass, but there are lines of arguement not mentioned by ThinkSecret that do make it seem like an attractive proposition to Apple this time around. I'm going to be watching this one closely, even if it turns out that in the end I can fly to New York, buy a headless iMac, and fly back to the UK for less than the price of an "native" UK iMac. So much for the global economy.

Python for Series 60

I've known that there has been a "secret" developer release of Python available for a while now from Nokia for Series 60 phones, such as my Nokia 3650 and 6600. However what I'd totally missed, was the public release of Python for Series 60.

I'd tried developing applications for the Series 60 using Java ME and found it hard going, despite having a fair bit of "normal" Java experience. It was especially hard going since I don't use MS Windows and couldn't take advantage of most of the infrastructre that has been written specifically to make things easier for the developer.

While I was struggling with Java the rumours of Perl for Series 60 leaked out of Nokia, I more or less stopped working on trying to get my head round Java ME at this stage, despite actually having a working toolchain and some tinkertoy level applications running. I was vastly disappointed to latter learn that Nokia was going to go with Python not Perl as their choice of high level language.

I've played with Python in past and I definitely prefer writing Perl. But even Python is a vast step forward over the intricacies of Java ME, so I was still somewhat cheered by the prospect of any decent high level language appearing for the Series 60 platform.

I'm currently downloading the SDK. I'm discouraged that it seems to be shipped in zip format, as I'm hoping the toolchain isn't platform specific. However, I've seen other people are talking about developing from Linux so I'm hoping I'll be able to do so from Mac OS X which is my prefered platform these days.

More information about the release can be found on the Python for Series 60 Wiki site.

Update: I managed to get Python working and developing under Mac OS X is quite simple. I even managed to get it to do something useful, well sort of, in a fairly short time scale.

Update: A version of Perl for Symbian/Series 60 has now been released...

Tuesday, December 28, 2004

The what foundation?

No sooner had I finished bemoaning the fact that we're not doing anything about the threat possed by NEOs, than Slashdot reports that the B612 Foundation has been established to do something about just that.

Of course the foundation was established in late 2002 and seems to consist of paper studies and a website, unlike the Planetary Society they don't seem to have hardware ready to launch. They also don't seem to have any money, at least at the moment. Time will tell...

Thursday, December 23, 2004

Don't Panic...

Copyright © 2004 Buena Vista Pictures Distribution

A teaser trailer has just been released for the forthcoming Hitchhiker's Guide to the Galaxy movie. But it's just not long enough to get excited about, and we don't get to see any of the acting. I'm worried that the Hollywood interpretation just won't be as good as the orignal BBC radio or television series of the book. It's very easy to spoil a classic, especially something as off the wall as the Guide...

Google is as Google does...

An recent post to the Google Blog talks about something that interested me for a while. If you put my name into Google it returns a whole bunch of results and you'll find that I take up most of the first few pages, although you get the odd hit for the SNP politician of the same name. On the other hand, if you put my name into Google Groups you have to filter the returned hits a lot to figure out which posts are mine. Quite a few of the returned hits, pointing to Usenet posts which aren't mine, I wouldn't want to be associated with...

We rely on Google a lot, but where should this stop? The first port of call for someone who'd just recieved my CV will probably be Google. Over the years I've posted a good deal on Usenet, but I've used a lot of different routes to do so, and posted using several different email addresses. It might be hard for someone who doesn't know about my non-work interests to figure out which posts are mine, and which posts belong to someone else. Would they even go to the trouble of doing this? I could end up not getting a job because of what someone else with the same name said five years ago. That's pretty scary!

But what's more scary? What happens when you insurance company starts googling you before giving you a policy, you really shouldn't have mentioned your love of hamburgers on your blog. Or, maybe it isn't your blog at all? Want to try and convince your insurance company?

The Register had an article earlier in the year about the dangers of exposing material you didn't mean to be seen to the web at large, but what happens if it isn't you that's doing it? Identity theft is pretty scary, but in a world where your reputation is coming to mean more and more what happens if someone who isn't you can damage your reputation just by the accident of birth, by having the same name...

Close call...

I was expecting more reaction to my recent post about global warming and extinction level events, but I got much less mail about that than I did regarding my post talking about the recent Apple iPhone rumours.

Then you get a close call, and an asteroid passes near by and you end up with a bucket load of mail in your inbox.

While it was pretty small at only 5 meters, and would probably have entirely burned up in the atmosphere depending on the composition, 2004 YD5 passed below geostationary orbit and was the second closest asteroid approach ever observed. The closest involved 2004 FU162 that flew by last March which we didn't even spot until August.

2004 YD5 was discovered on December 21st by Stan Pope of the FMO project, but it was later calculated that closest approach had been two days earlier on December the 19th. So, like the other one we only saw this one after it had gone by as there is a significant blind spot for people trying to detect NEOs. As, just like all those old war films with aerial dog fights, they come at you out of the sun. This isn't very reassuring...

Like all the other near misses, this one illustrates that we really should be doing more to detect near-Earth asteroids before they become a threat. It's always suprised me that the Shoemaker-Levy 9 impact with Jupiter in 1994 didn't cause more comment. After all, what sensible species would watch a rock the size of a large mountain strike a nearby planet and then do absolutely nothing about it?

Tuesday, December 21, 2004

More AG rollout

Looks like I wasn't the only person blogging the AstroGrid rollout meeting at the NeSC. There were a lot of WiFi enabled laptops in the audience, so I'm not that suprised. Maybe I should wear my "I'm blogging this..." t-shirt next time?

Thursday, December 16, 2004

Back home...

Another day, another Dash 8, and another commuter flight home to Exeter. I fly enough that the whole experience of airports and airlines leaves me feeling a vague sense of horror, but it's certainly a lot better than the alternatives. I hate to think how long it'd take to get from Edinburgh to Exeter on the train...

Another communter flight, this one running somewhat late

Christmas in Edinburgh

I wasn't flying back directly after the close of the rollout meeting. Instead I was catching an early flight the next day, so I had an evening in Edinburgh to myself. I therefore did what any sensible person in my situation would do, I went Christmas shopping...

The Scott Monument on Princes Street

This was a mistake, I don't think I'm cut out for big city Christmas shopping. In the end I retreated to the hotel bar to lick my wounds, without having bought any presents. I've got a bad feeling that I'm going to end up doing the bulk of my Christmas shopping on Christmas eve (again) this year

Wednesday, December 15, 2004

The AstroGrid architecture

Next up was Guy Rixon who talked about the AstroGrid architecture. He pointed us to the architecture document which describes the architecture of the current iteration 6 release.

This was followed by a break away session where we sat down and talked about what we wanted out of AstroGrid-2 and what bits of the architecture needed tweaking. There was general agreement that, amongst other things, the portal is perhaps the real weakness of the current infrastructre and lets down an otherwise fairly good workflow system. A workbench system with an interface similar to Triana, or AVS, was suggested as a portal replacement.

The second day

The second day of the AstroGrid rollout started with Andy Lawrence talking about AstroGrid-2 and Euro-VO, and AstroGrid's relationship with them and also with the AVO.

AstroGrid will be leading the VO Technology Centre (VOTC) which will be a loose confederation to develop new technologies for Euro-VO. VOTech is the first project under the VOTC umbrella funded by EU-FP6 and partner contributions. Formally VOTech is a design study which will output reports, designs, prototypes and proto-standards. In contrast to the engineering focus of AstroGrid-2, VOTech is a research and development project.

Tony talking about the AstroGrid-2 project

This was followed by Tony Linde talking about AstroGrid-2 and what it's going to be doing over the next 3 years.

Tuesday, December 14, 2004

In closing...

We closed with several more talks on the technical details behind the workflow and the scripting infrastructure. The second to last talk by Neil Winstanley should probably have been right up front. We finally got some more details of the reasons behind the choice of Groovy as the workflow scripting language, some explaination of how to hook these scripts into the workflow, and what and when you need to do that.

After hearing this talk I was somewhat happier about the capabilites of the workflow engine. However, I'm still very unsure as to whether the interface being presented should be represented as the "user" level. It seems too low level to be offered directly to an average user.

PAL and Data Centres

Martin Hill talked about the Publisher's Astrogrid Library (PAL) and how to use it to serve your data. Looks interesting, and fairly feature complete. He pointed us all at a PAL wrapped version of 2MASS, amongst others things, as an example.

The Registry

After coffee we moved on from Workflows, and Elizabeth Auden taked to us about the Registry. Elizabeth took us through the registry portal and showed us how it worked.

At the moment it looks like a fairly primitive interface into a the registry, with a lot of raw XML getting displayed along side what look to be raw Java error messages when, for instance, the XML fails to validate. On the other hand it all looks much improved. To be honest the output of the registries was never really supposed to be seen by humans, it was supposed to be consumed at the software level, so maybe the primitive portal interface really isn't a major problem here...

Workflow too complex?

We've had several workflow demos now, and I'm disappointed by how complex it looks to try and use the workflow engine. This seems especially evident from the Solar Movie Maker demo which seems to rely very heavily on fairly extensive user scripting in Groovy.

The post-lunch demo

We've just come back from the lunch break and are we're now into the crucial part of the meeting, the actual demo. Nic Walton did the warm up and talked about the helper applications, such as CDS Aladin and Starlink's TOPCAT and Treeview applications.

The first demo was the extra-galactic use case with Nic talking us through the wokflow, with a live demo of the workflow going on in the background. They seem to have gone with Groovy as a scripting language for workflow, interesting choice...

Nic Walton and the AG extra-galactic workflow demo

...and indeed a good number of the people in the audience seem quite worried by the choice of Groovy as the scripting harness for the workflow engine.

We've slid into a long and detailed discussion about what should be user scripted and what workflow should provide natively. I'm very confused about why aplications which already has a workflow harness needs user scripting to stick them together? Surely that sort of thing should be entirely transparent to the user?

Connection to other projects

Guy Rixon is talking about how AstroGrid connects to other projects. So far he's mentioned the eSTAR-like agent architectures agreed at the VOTech kickoff meeting, and external applications like those produced by the Starlink project.

Next up...

Following a talk about the GridPP project by Tony Doyle, the next talk was about the AstroGrid (AG) infrastructure by Tony Linde. He seems to be trying to emphasize the mix and match nature of the current AG infrastructure, you can take one, some, or all of the various AG software components depending on your needs.

He's also talking about the forthcoming AstroGrid-2 work. During this he mentioned the proposed work on agents which has been recommended by the VOTech working groups at the kickoff meeting.

National Grid Service

Neil Geddes talked, fairly briefly, about the National Grid Service (NGS) which is a UK wide grid of super computing facilities, computer clusters and storage facilities. Interestingly the infrastructure libraires used by the NGS seems to be WSRF::Lite, or at least that was what he seemed to be saying on his inital slide, which means they're Perl people. So that's okay...

AstroGrid Rollout

I'm now at the NeSC listening to inital talks during the AstroGrid rollout meeting. If you want to participate in a the beta test of the AstroGrid infrastructure you need to mail Nic Walton to ask for a login to the system.

Nic Walton talking about science in the VO

Monday, December 13, 2004

In Transit (again)

I caught my commuter flight to Edinburgh mid-afternoon, and I'm now holed up in the Jury's Inn in Edinburgh.

You can tell I'm flying from a regional airport, right?

The hotel is only a couple of streets back from Edinburgh Waverley and Princess Street and therefore rather convenient for, well, virtually everything, including the Edinburgh Christmas Market. Funny thing about that, you wait all year for a German Christmas Market, and two come along at once.

The Edinburgh Christmas Market on Princess Street

This post is brought to you through the magic of Bluetooth and a GPRS connection, as one thing this hotel certainly doesn't have is a broadband connection.

Sunday, December 12, 2004

The London Perl Workshop

I'm back in Exeter from my adventures in London, where I attended, and in fact spoke at, the London Perl Workshop which was organised by

The main hall where the Plenary session and the Learning Perl track was held. I spent most of my time upstairs in the much smaller, and less photogenic, room where the Advanced Track as being held.

I think my talk on the Advanced Track went over fairly well in the end. I got some nice comments and a nobody tried to rugby tackle me over my politically incorrect statement that legacy software was, despite initial appearances, a good thing. For the most part, when not be evangelical, I talked about Inline::* and SOAP::Lite which I was surprised to find weren't already in widespread use amongst this crowd.

The most interesting talk of the day for me had to be Simon Wistow's talk about his Whatif module. Originally in the Acme::* namespace, and for good reason, he moved it out of there when he discovered that people were actually crazy enough to use it.

Eric Saunders was also at the workshop with me, and I think it's a safe bet to say that he found José Alves de Castro's talk on "Obfuscation and Golfing" the most interesting of the day. For those of you not in the know, obfuscation is all about making your code as totally unreadable as possible in cool and interesting ways, while golfing seems to be a uniquely Perl phenomenon where you try to write code in as few characters as possible.

The most interesting talk for everyone else at the conference seemed to be Casey West's talk on "Managing and sending email". The smaller Advanced Track room was (over)filled to capacity and there was still a queue of people outside trying to get in to see this talk. Is email really that interesting to the Perl crowd? The popularity of this one really confused Eric and myself...

The Royal Albert Hall which is just next to Imperial College Union where the London Perl Workshop was being held. London gets all the best architecture...

On the down side I'm officially back in transit. Lunch time tomorrow I'm catching a flybe flight from Exeter to Edinburgh so I can attend the AstroGrid rollout which is on Tuesday and Wednesday at the NeSC. Some time soon I really have to stop moving, sit down, and write some software. But with Christmas coming up, you have to figure it's going to be the new year before I get a decent stretch of "real work" done.

Google Suggest

Okay, this has to be one of the most instantly useful things to come out of Google Labs for a good while. Google Suggest offers keyword suggestions in real time as you type your search terms into the query box. It doesn't seem like an Earth shattering idea, and it isn't. However it's one of those things that, once used, you instantly feel is the way it should have worked all along.

Google Suggest uses XMLHttpRequest to query Google as yout type, giving you a drop-down auto-complete box with the most likely results. The drop-down box is updated in real time as you continue to type.

You need to enable both JavaScript and cookies, and be using a fairly recent browser or you'll see nothing, but when it does work, it's impressive. Some more information about the new service can be found in the Google Blog...

Friday, December 10, 2004

In Transit

I'm in transit to the London Perl Workshop which is being held tomorrow at the Imperial College Student Union. I've actually volunteered to give a talk on the Advanced Track about legacy software and I'm got a funny feeling I might not have made it quite advanced enough, this is the crowd after all.

Anyway, the plan is to meet up tonight at around 7pm in The Star for a few pints. So if I realise that I've totally misjudged the audience I can hopefully rewrite the talk before I have to speak tomorrow...

Wednesday, December 08, 2004

Rat brain flies jet

The Register is reporting that Florida scientists have grown a brain in a petri dish and taught it to fly a fighter plane. You wake up in the morning and figure the world just can't get any weirder, and most days you get proved wrong...

The end of the World

As an astronomer you get used to thinking in the long term. A few million years is a relatively short time scale when you reguarly deal in time scales of billions of years, so when someone tells you something bad is going to happen in only a few hundred years, you tend to take notice. Astronomers are also used to thinking about global catastrophes. The impact of Comet Shoemaker-Levy 9 with Jupiter in 1994 made the threat of Near Earth Objects very plain, and since then some money at least has been directed towards the hunt for such objects. After all, what sensible species would watch a rock the size of a large mountain strike a nearby planet and then do absolutely nothing about it?

There are other possible extiction level events. A nearby supernovae could send a wave of lethal radiation towards us, stripping the protective ozone layer from our planet, indeed it has been suggested that this may have happened before. Closer to home, the eruption of a Supervolcano might make all our worrying about mankind's effect on the climate totally irrelevant.

CREDIT: British Antarctic Survey/C Gilbert, PA
Part of a massive tabular iceberg adrift in the Weddell Sea off the Antarctic peninsula. In Antarctica, the recent break-up of ice shelves has precipitated increased streaming of ice from much farther inland, which potentially represents the initiation of a phase of much more serious ice-sheet collapse.

We can do little about the most of the external threats to the continuation of our species, at least for now. Maybe in a few hundred years, if we get lucky enough to make it that far, then we will have the technology to address at least some of them. But Jan Zalasiewicz's recent article in The Guardian suggests we might not have those few hundred years.

I'd recommend people read it, it scared me badly, and like geologists, astronomers don't scare easily. We're used to thinking about the world coming to an end, and maybe you should be thinking about it as well.

Monday, December 06, 2004

A merry virtual Christmas

I'm worried by the growth of online shopping. Not so much by sites like Amazon, which will ship actual physical things to you which you can then wrap and place under a Christmas tree, but by the growth sites selling virtual goods.

You can now go to a web site and spend a considerable amount of money, and all you'll get back in return is a decryption token or a license key for some piece of software. How does one go about wrapping a virtual present?

What is the defining thing about Christmas morning as a child? Surely it must be going down stairs and unwrapping the presents piled under the tree? More and more these days things just aren't like that. Even when I was a child there were gift certificates, funny coloured pieces of paper you could only spend in one shop. But does something like a gift certificate or actual folding money, let alone a virtual present, give the same tactile feelings of warmth and caring as unwrapping a "real" present does?

I do remember that no matter how well meaning, grown ups could never buy exactly what you wanted, indeed it seemed like the more well meaning the grown up, the less relevant the present. You always wished, at least when you got old enough, that they would just give you the money instead. But none the less there was always the special moment when you faced an awkwardly wrapped, oddly shaped, present and wondered whether it would turn out to be something you wanted.

Surely that's the beauty of giving a real physical present to a child, that you'll give them something that may change them. Something they didn't know they wanted till they got it. I can recall my first telescope, it wasn't something that I particuarly wanted, I hadn't asked for it and it was little used. I lived on the west coast of Scotland and the weather was rarely amenable to observing, but I still remember the one clear night when I was bored and dragged the telescope into the garden, because I'm sitting here doing what I do because of it. I've stood on every continent except Antarctica because my parents gave me a gift that, at the time, I didn't really want. It's odd how such a small thing can change your life like that.

So I'll be giving virtual presents this Christmas, because I know they'll be appreciated. But I'll also be giving some off the wall, probably unappreciated, presents. In all probability they'll end up unregarded at the back of a closet, but some might not...

Christmas isn't about receiving, or even about giving. As far as I'm concerned it's about the potential for change, it's about how small things can affect large things. This is why for me the presence of wrapped gifts under a tree is so important, and why I get so very stressed about choosing gifts for my nearest and dearest, I might be changing their life for the better if only I can choose the right gift. Isn't that the magic of Christmas is all about?

More on the iPhone?

There are yet more rumours flying about the possibility of an Apple iPhone. I talked about this in a previous post and I still think it's unlikely. Like a lot of people I think the rumours leaking onto the net about an iPhone are distorted versions of Apple's deal with Motorola to ship iTunes on their new handsets.

But then, who knows, Apple are notorious for trying to keep new product lines as quiet as possible before launch...

Sunday, December 05, 2004

Saying stupid things...

Do you remember the dawning sense of horror you had when you first heard about Google Groups? Full USENET archives all the way back to 1981, every stupid thing you said in public preserved for posterity. Well it's worse than that, these days all the stupid things you say in private are being archived. At least, if you're in Korea anyway...

The Korean Times is reporting that the Korean government have requested that wireless operators keep records of all text messages sent over their networks and the operators, depsite privacy fears from their customers, are bowing to government pressure. Of course we all know, or at least should know, that text messages aren't secure from prying eyes. But until now the prospect of widespread archiving of text messages hasn't really been discussed. Unless you were already being investigated by the police, or the security services, what you'd said in the past probably wasn't going to hang around to haunt you. Now, every single drunken text message could be sitting out there, waiting. Maybe the Americans have the right idea after all.

Of course, that's only if you happen to be in Korea? Does anyone know what the policy of the major operators in the UK is on this issue? No, nor me...

The only integrated text message encryption application I can find is Fortress SMS. It claims to implement 128 bit RC4, but the application appears to be closed source so there really isn't any way to check this, you just have to trust the vendor. However the flaws in RC4 are well known, and I'm not sure I trust a vendor who provides a "A Comparison of Fortress Mail Encryption and X.509 based Public Key Encryption for Secure Email Exchange" but doesn't raise or address these concerns.

If I'm going to encrypt something I'd prefer to use a public key encryption scheme over something like RC4. Perhaps it's time I pulled Applied Cryptography off the shelf and sat down and wrote something that isn't to do with astronomy for a change.

Friday, December 03, 2004

Phones that know who you are...

Pantech have just released the first mobile which uses a fingerprint recognition sensor to identify the user. The new Pantech GI100 got a mention over at Gizmodo and Mobile Mag although with the sparse details I guess nobody has actually seen one of these things yet.

From a British perspective there are two glaring problems with the phone's specification. The first is the absense of Bluetooth. Although little thought of over the pond, Bluetooth is everywhere in the UK. My phones, headset, laptop, PDA and GPS unit all talk to each other now, and I wouldn't want to return to the days when they couldn't. Although considering the security flaws that have been discovered in the protocol perhaps this absence is deliberate in what is, after all, supposed to be a "secure" device?

Either way, the crippling flaw is that fact that the phone is only dual band (900 & 1800MHz), making it useless in the States. Nearly every phone sold in the UK these days is tri-band, and most of the higher spec phones are quad band. So why have Pantech taken such an backwards step? The conspiracy theorist in me would attribute this to the fact that, in the current climate, the US would no doubt frown upon such a device. At least one which didn't have a backdoor for law enforcement types. Or perhaps the real reason is that, with all that extra circuitry for the biometrics, there just wasn't room to make the phone quad band?

Thursday, December 02, 2004

More on Ariane 5ECA

The BBC has published an interview with Jean-Yves Le Gall, the chief executive of Arianespace, about Ariane 5ECA and the future of the commerical launch business. Interesting reading...

Wednesday, December 01, 2004

The "Kliper" ship

The Russian Energia Corporation today unveiled a mock up of their potential successor to the venerable Soyuz capsule. The reusable Kliper could deliver cargo, or a crew of six, to orbit and the International Space Station where it could remain docked for up to a year.

The new spacecraft is a lifting body where the body of the craft itself produces the lifting force during reentry into the atmosphere. Unlike a traditional reentry capsules, such as Soyuz, which are limited to a narrow landing corridors along the ground track of their final orbit, the new craft would have significant sideways maneouvering capability.

CREDIT: Channel One TV
The Kliper's unveiling attracted a lot of attention in the Russian media, including television coverage and several articles [1, 2] on various Russian news sites. I'd guess that the vehicle in the background, behind the Kliper mock up, is the Buran OK-KS test vehicle. OK-KS was used for static electrical and integration tests, it never flew, and was not flight capable.

Despite being much more capable the new spacecraft is expected to cost less per launch than Soyuz which itself costs less than a tenth as much, at around $30 million, as space shuttle launch.

While there has been considerable coverage [1, 2, 3] in the Russian media, there has as yet been very little in the Western press. However MSNBC has now published a good article about the new spacecraft, so you don't have to get your Russian dictionaries out quite yet.

However as this MSNBC article points out there are significant problems standing between todays mockup and an actual launch vehicle, and like most things in Russia today, the first is money. A production vehicle, currently scheduled for test flights by 2010, seems to be well beyond the meager funding levels currently allocated by the Russian government.

However, in the long term, this might turn out to be the least of Energia's problems. The new spacecraft was designed to be launched using the Ukranian made Zenit booster. Considering the current political troubles in the Ukraine, which could easily spiral out of control, relying on a Ukrainian built booster to lift the new spacecraft may be unacceptable to both the Russians and the west.

Apple iPhone

Update: iPhone released in Jan. 2007

Russell Beattie has been indulging in some fantasizing about the direction in which Apple might be heading with his blog post about the the "iphone".

I think he's dead wrong. The key thing about the sucess of the iPod, which I'm sure is clear to Steve Jobs, is that Apple broke into (invented) a totally new market and became the dominant player with a high margin product. The mobile phone market is anything but new, in fact, it's totally saturated and margins are vanishing thin. On top of this, his suggestion of "...a small version of OSX" is unfortunately totally off the wall.

CREDIT: iCreate/
The iphone, not an Apple product, or likely to be one? I think this image comes from the UK magazine iCreate which has a habit of dreaming up possible new product lines for Apple.

If anything, I think Ross Mayfield has a better idea. The thought of a WiFi enabled iPod is enough to make most geeks break into a sweat. It might not be as revolutionary, but it's a lot more likely.

But if Apple do bring out an iPhone, I'm first in the queue and unless it's another Newton, they'll pry it from my cold dead hands once I've got one...

GMail via POP3

In a move which almost totally defeats the purpose of GMail, you can now access your GMail account via encrypted POP3. Surely the purpose of GMail wasn't the 1Gb of mailbox space, it was the interesting and novel interface for reading your mail? Doesn't offering POP3 access, which means you don't have to use the novel interface, sort of dilute the point of the service? Presumably enough geeks have been banging on the table asking for direct access that they've finally just given in...

Interestingly, they're also offering authenticated SMTP as well. However, I haven't tested it yet to see whether it rewrites the headers to make it appear that the email was sent from GMail itself. If it does, that's going to make it a lot less useful than it initally appears, although somewhat less prone to misuse.

A second chance for Ariane 5ECA

The BBC is reporting that the second launch of the improved Ariane 5ECA will be in January next year. The first flight of the new Ariane 5 variant, in December 2002, was a failure. Leakage in the cooling system resulted in deformation to the nozzle of the main Vulcain-2 engine, which caused an abort in the first few minutes of the flight.

CREDIT: Arianespace
An Ariane 5 lifting off from Kourou in French Guiana