28th August 2007

I have something of a love/hate relationship with PayPal. On one hand, it's undoubtedly one of the most convenient ways of moving money around the Internet, and as I'm usually the person sending the money its essentially free to use. They also have a payment protection policy that (in theory, at least - in practice there is a list of exclusions as long as your arm, and they're not afraid to use it) will refund your money in the event that an eBay seller defaults on a deal or a company disappears before sending the purchased goods.

Unfortunately they do have a number of annoying little quirks, and I ran into one of these when an eBay transaction for a 2nd hand printer fell through. The seller was based in Ireland, and so the auction listing was in Euros, but the miracles of eBay's and PayPal's international outlook means that this is rarely an issue. If the seller is prepared so ship to another country, and the charges are reasonable, it's generally no harder buying from overseas than from the UK.

With this in mind I paid for the item without concerns, and when the seller told me the next day that the printer had developed a fault and that he couldn't in all honesty complete the transaction I didn't expect problems: he hadn't claimed the PayPal payment, so all I had to was cancel it and the money would be automatically re-credited to my card. On checking my online bank statement a few days later, however, I noticed an anomaly: although my PayPal account had been credited with the whole amount paid, €440, instead of the £302.56 that had originally been debited from my card, all I had received back as the refund was £289.67 - a difference of almost £13 or around 4.2%.

I had a similar problem with PayPal a few years ago, and my experience then did not fill me with optimism this time, but the sum involved is far from trivial and I wasn't going to roll over without at least a token protest, so I gritted my teeth and I filled in their annoying little online enquiries form. Their first response was typical boilerplate, going into massive detail about something largely unrelated to my query. All foreign currency conversions attracted a fee, they assured me, and in fact PayPal's exchange rate was very reasonable. As usual for any initial enquiry to PayPal (or eBay, for that matter) this form response didn't answer my question, so I tried again, speaking slowly and using small words. This time the reply was that the exchange rates are prone to "normal market fluctuations" in the time between the conversion from Pounds to Euros and the conversion back from Euros to Pounds, but in fact a quick check at the web site of the Financial Times showed that in fact the exchange rate between these two currencies is extremely stable at present, varying by a mere 0.43% during the period in question.

When I pointed this out (and each time I was dealing with a different operator, first Jayson, then Melody, then Paula, which always makes it hard to achieve any degree of continuity) the response veered back towards the initial defence that there are charges and fees and commissions and in fact they're very competitive in comparison to those elsewhere. While this may well be the case, I am not happy paying even a nominal fee when in fact I have received no goods or services whatsoever to show for it, but I know a lost cause when I see it and in any case with a trip to Plymouth for the bank holiday weekend looming I didn't have the time to pursue the matter further.

I'm not alone in having experienced this kind of rip-off from the company (and the PayPal Sucks web site has more horror stories about dealing with them than you can shake a fat credit card statement at) but unfortunately they're probably still the best of a bad bunch - if only thanks to their widespread acceptance by online sellers worldwide. I mentioned my less than positive first impression of competitor PPPay a few months ago, and that seems fairly typical of the catches and quirks of the other equivalent services. I've said it before and I'll almost certainly say it again: on the Internet, the emptor has to caveat like they've never caveated before.


27th August

Just back from a few days staying with my parents in Plymouth, at the house they share with this:

His name is Ariel, after the delicate, airy spirit in Shakespeare's play The Tempest, and we think he's about fifty years old. This is a fair age for a domestic tortoise, but although my parents worry each year that he won't survive the winter hibernation, right now he's still going strong. He's acquired something of an attitude in his middle age, however, and although he's only seven inches long he's a surprisingly powerful animal and is quite capable of upsetting a foot-tall earthenware flowerpot: plants, soil and all. We know this because he seems to have taken a dislike to the pots that my parents have installed on a paved section of the back garden, recently, and over the last few days I've watched him squaring up to them like a sumo wrestler on a number of occasions - only a timely intervention has prevented another disaster. He's also surprisingly agile and speedy when the mood takes him, and in fact it turns out that he can walk noticeably faster than I can squirm backwards while lying on my stomach for a dramatic low angle - while I was taking the photograph above he came within a few inches of soundly head-butting the lens of my Canon PowerShot G5, an experience which I doubt it would have lived to record. Tortoises haven't changed much since their ancestors were dodging out from under the feet of dinosaurs 200 million years ago, and being nose-to-nose with an example of the species one gains a real sense of that...


21st August

A few days ago I mentioned the problems transferring Audible audiobooks to my new TomTom GO 520, and having established contact with a fellow sufferer he forwarded me the latest response from Audible, dated 14th August:

It appears that our compatibility list is a little ahead of our software as the plug-in for the 720 should be available through a software update later this week. I should be able to email you when this update is ready to installed.


Paul Cook
Audible UK

So all-in-all it seems that the entire TomTom 520/720 launch was a month or so premature, necessitating multiple firmware and application software updates from the manufacturer in the first few weeks alone, and an unspecified wait (Audible have already missed their target of "later this week") for a high profile feature from a major partner. It's reassuring to see that manufacturers of allegedly ready-to-use appliances based on contemporary computer technology are adopting all those old habits we know and love from the IT industry itself.

And talking of which, consumer VoIP provider Skype has blamed a major outage last week on Microsoft's "Patch Tuesday", which they claim caused an abnormal load on their internal network because of all the Skype clients reconnecting after the reboots demanded by installation of the Windows updates. Given that Microsoft have been releasing updates on this schedule for several years, and that this month's patches were nothing out of the ordinary, this seems somewhat implausible - especially given that Skype released a new version of their client software "to address the issue" on Friday. The UK ISP group 186K have also blamed Microsoft's scheduled updates for their recent network congestion, and given that Tiscali and Carphone Warehouse have been agonising over the bandwidth problems they expect when the BBC's iPlayer distributed video player is finally launched it's becoming clear that these days ISPs would rather not be bothered with such difficult, expensive chores as actually carrying data...

Elsewhere, as hoped and expected, the RIAA's most famous victim, Tanya Anderson, has filed a lawsuit against the Association, and if the suit is certified for class action it will pave the way for all the other individuals that have been threatened and bullied on behalf of the media industry to fight back at
last. The charges certainly make an impressive list: negligence, fraud and misrepresentation, racketeering and corruption, abuse of the legal process, malicious prosecution, outrage and intention to inflict emotional distress, computer fraud and abuse, trespass, invasion of privacy, libel and slander, deceptive business practices, misuse of copyright laws, and civil conspiracy. As one of the comments in the thread at The Register notes, that list "would make Al Capone envious". Indeed!


20th August

I used to think that the human brain was the most fascinating part of the body - and then I realised, "well, look what's telling me this"...

 - Emo Phillips

So I was browsing for information on the gospel singer Shirley Caesar (it was for someone else, Ok? Not my bag at all!), and came across her own web site. It has the usual pretty front page, with her photo and a brief news clipping, and along the top are the expected links for her biography, discography, photos, tour details, fan club etc etc. Except that clicking on any of them takes you to a "login here" page, restricting access only to people who are prepared to register and so effectively deterring the casual browser or someone with a healthy distrust of web sites with no published privacy policy. I can't for the life of me imagine why someone ever thought that this was a good idea.

Meanwhile, Apple are obviously a popular target for legal action, right now, and in spite of the fact that last month's iPhone lawsuit is almost certainly too frivolous to survive, two more almost equally frivolous (to my untrained eye, at least) complaints were filed last week. The first claims that Apple is violating the Fair Credit Reporting Act by including certain confidential information in electronic receipts produced by their online store, and the second alleges that Apple is infringing patents that cover the use of lights on power supply units. Given that the law cited in the first suit refers to printed till receipts and not electronic order confirmations, and that the patents referenced in the second were only filed in 2001 (and I know I used laptop PSUs with a power LED built into the connector long before that!) it seems likely that neither case will amount to anything, but it makes a change for the lawsuits to be aimed at someone other than the long-suffering Microsoft.

Elsewhere, five servers belonging to the organisation behind the popular Ubuntu Linux distribution were shut down last week after the admins realised that they had become so badly compromised that they were being used to launch attacks against other systems. For some bizarre reason they were running an old and unsupported version of the OS (remember, children, Linux never needs patching!) and this is being offered as an excuse to explain the apparent ease with which the servers were 0wned. I find this odd, as we've always been assured that Linux is thoroughly hack-proof, so even an old version should be just as tight as a the current release one, surely? I remember something similar happening at Debian and Gentoo a few years ago, and if the Linux houses can't keep their own servers in order, why on earth should we believe their incessant claims about the higher reliability and security their OS is alleged to provide?

And, finally, I mentioned last month that if Apple's licensing tiff with the Universal Music Group was not settled amicably, the DRM embedded in iTunes tracks could conceivably allow the media giant to insist that Apple remotely disables any and all songs downloaded by Apple's customers. Although several people subsequently told me that was a) technically impossible, and b) would never happen, I stuck to my guns in both cases and feel thoroughly vindicated having heard that exactly that has happened to customers of Google's commercial video download service. Google has decided to abandon the service and so, after only a few days notice, in the middle of last week all purchased videos expired and became unwatchable. This is the sharp end of DRM, and far more significant than the copy protection techniques that are all that most end users experience. No matter how the seller describes the transaction, and Google used the words "purchase" and "own" liberally on their site, in fact you are only buying a license - and the terms of this license can be changed, or the license revoked completely, at the will of either the seller or the media's owner. Rarely has the time-honoured phrase "caveat emptor" been so applicable...


19th August

Oh, when will the madness end...? The Violent Crime Reduction Act which (pointlessly) restricts the sale and purchase of realistic imitation firearms has been law since last autumn, but the specific clause that covers this does not actually come into affect until October, and although I hadn't intended to buy any more airsoft replicas some irresistible new models have come onto the market this summer and I'm taking advantage of the brief window for a final fling.

Last month I picked up Star's interpretation of an ISSF competition air pistol, the PSS-300, and as I write this I'm waiting for a shipping quote for two very similar models from KSC and Maruzen. Sat in front of me right now, however, is an absolute beast of a replica, and even after a weekend of fondling it I'm still slightly boggled at how darn big it is!

This is another model from Taiwanese manufacturer Star, the AW-338, notorious in the past for the dubious build quality of their AEGs but who are starting to regain their reputation with some innovative and unusual gas replicas. It's based on the L96 Artic Warfare Folding sniper rifle made by UK manufacturer Accuracy International, and from what I can see is an extremely faithful reproduction of the original. As the name suggests the thumbhole stock folds neatly in half just behind the pistol grip, reducing the 123cm overall length to a still impressive 102cm. The stock itself is made from an extremely dense polymer, in a flat colour somewhere between olive drab and black (the camera flash has lightened it considerably, as usual), and although like the original it is actually a hollow casting it certainly doesn't lack heft or rigidity - which is just as well as, also like the original, it weighs in at more than 6½kg.

As befits a sniper rifle the replica comes complete with a 3.5-10 x 40mm telescopic scope and a quick-deploy folding bipod, and the adjustable cheek piece and butt plate, together with the unusual extending monopod below them, have been reproduced perfectly. As usual the large metal magazine holds both the gas and twenty three 6mm BBs, but unusually the ammunition feed mechanism can be popped out of the magazine body and replaced with a fully loaded one - and considering how little gas even a powerful bolt action replica uses in comparison with a gas blowback pistol, that could be very useful during an airsoft skirmish.

I usually fire my replicas down the length of the hallway and into the kitchen, a distance of almost 10m, which is ideal for pistol shooting. It's a touch cramped for a sniper rifle, however, and as it's raining at the moment I'll have to contain my enthusiasm a little longer before actually putting some rounds through the new replica - and so for now I shall settle for occasionally picking it up and marvelling, yet again, at how darn big it is.


18th August

I bought one of the first self-contained consumer in-car GPS systems back in 2003, the Navman iCN630, and although it was eye-wateringly expensive back then (I think mine was well in excess of £600, new!), and although it wasn't without its quirks, it usually managed to navigate me around the home counties fairly successfully as well as guiding my occasion forays further a-field. Recently it has been starting to show its age somewhat in comparison to contemporary units, though, with the lack of an internal battery or touch-screen, greater bulk and weight than its relatively small screen size would suggest, and unsophisticated and feature-poor software. It seemed time for a change, and as the market now seems to have swung away from Navman and towards TomTom I thought I would make the switch.

TomTom's new range has just been announced, and having apparently forgotten the pain of being an early adopter with the iCN630, I bought the high-end GO 520 model (the top-of-the-range 720 is identical except for its maps of Europe rather than just those of the UK) only a few weeks after launch. I was expecting a few teething troubles (it is a computer, after all, and a fairly sophisticated one at that) but nothing like what I and the other users have actually found!

Initial impressions of the hardware are certainly favourable - the screen is large and bright, and although it sometimes needs a fairly firm touch, being able to operate it with a fingertip rather than a menu system is a huge improvement over my Navman. On paper the list of features is long and impressive, but at this stage my own experiences, and those of members of the Pocket GPS World forums, are far from reassuring.

To begin with, the device seems incredibly fussy about the SD memory cards it will accept. I first tested it with regular 512Mb and 1Gb cards from SanDisk, pretty much an industry standard, and although I could transfer content onto them via the TomTom Home application, after disconnecting the device to reboot it into mapping mode it completely refused to start up with a card inserted. Switching to a newer 2Gb SanDisk Ultra II card seems to have solved the problem, but others are experiencing all sorts of similar issues with cards of many brands and varieties. The only advice I have is to check the forums and see what is working for everyone else.

Other problems, which I have yet to experience in my own extremely limited testing but which seem to be plaguing more serious users, include an appalling buzzing noise at the far end of a phone call when the unit is used in its hands-free mode, making the user unintelligible to the person he or she is talking to, and reports that the quality of the unit's own audio output is seriously reduced when the battery reaches around half its charge. Further, the built-in FM transmitter that links the unit's audio output to a car stereo also seems to be at such a low level that one needs to crank the volume all the way up - something that can often produce an unpleasant surprise when switching the stereo over to playing a CD or a normal radio station! All these issues seem to be hardware problems, either from faulty design or poor manufacturing quality (they're made in China, of course, which is certainly no recommendation these days), and it's not clear whether any improvement can be effected simply by upgrading the device's firmware.

Another issue relates to the claimed compatibility with DRM-protected audiobooks from market leader Audible - in spite of many attempts the Audible Manager program completely refuses to see the TomTom, and so can't even activate it let alone transfer content. Unfortunately this was one of the main selling points for me, as on long journeys having to manage and power both a GPS device and my Palm to play audiobooks can be an annoyance, and although I'm not the only person having difficulties no answer has yet been forthcoming from either TomTom or Audible. It has to be said that if the files were just regular MP3s this problem would never have arisen, as they could just be copied across directly via the Windows Explorer, so it's another black mark against the DRM technologies that the media industry still insists are somehow for the benefit of the consumer...

The TomTom Home application also has a "remote desktop" sort of tool, whereby the TomTom can be controlled using the mouse and an on-screen representation of the unit, and although this worked in the Home V1.6 application that shipped with the 520, the update to Home V2 that was released the day after I bought mine seems to have broken the facility completely. In fact, the entire Home application crashes and exits when one tries to connect to the GPS unit in this way, apparently no matter which operating system is in use on the host PC!

Yet another failing appears in the new "Help Me" facility, which is intended to direct the user to the nearest doctor, dentist, hospital, police station and other vital services. Unfortunately "nearest" in this case appears to be something of a misnomer, as many users have reported being directed to addresses in Northern Ireland or Europe from locations in the heart of England - which renders the entire feature completely useless at a stroke.

As well as the above, minor problems, quirks and bugs have also been noted in the speech recognition facility, the custom POI warnings, the text-to-speech feature that reads out custom messages, transferring MP3 media to a memory card in the device's SD slot, and unexpected crashes at critical moments in the route... I could go on. However, it is interesting to see that none of the above issues (with the single exception of the low output level on the FM transmitter) were mentioned in the current reviews of the device, all of which gave it the highest marks. Once again, I am reminded that the cosy little world that tech journalists inhabit bears very little relation to the one that most users live in...

I'm sure that many of the problems that plague the TomTom will be fixed soon (although the reports of such widespread hardware faults are certainly disturbing) but right now I can't recommend either the 520 or 720 models. In-car GPS really has to be reliable and foolproof, as fiddling with a recalcitrant unit while driving just isn't sensible, and right now the new models are neither. I'm going to be travelling from Essex to Plymouth next week, and I'm giving serious consideration to playing it safe and sticking with the four year old Navman...


14th August

A few random links, before the scar heals over...

A con by any other name - An allegedly ground-breaking agreement between the EU and the US, which ostensibly curbs demands from the US for excessive personal information concerning air travellers, is in fact anything but. Although EU data protection administrators claim that the number of records that are being transferred with each passenger has been reduced from 34 to 19, in fact they have simply merged separate records into one and the overall dataset is almost unchanged! This has not gone un-noticed, however, with Gus Hosein of the London School of Economics describing the settlement as "a complete con".

The leaky establishment - More than a year after reports of serious security vulnerabilities in the Foreign Office's visa application web site, an investigation has revealed that no progress has been made in securing the facility. Confidential data can still be easily extracted with direct SQL queries, no external penetration testing was ever employed, and maintenance plans by the Indian subcontractor responsible for the site's development, VFS, would actually have decreased security had they ever been carried out! Another chapter in the ever-expanding volume of disastrous UK government IT projects...

Unreasonable demands - The BBC's controversial iPlayer video service has aroused the ire of UK ISPs, who are threatening to throttle the application's bandwidth unless the Beeb contributes towards the cost of the infrastructure upgrades that they claim the distributed service will demand. The increased load on an ISP's internal networks and peering connectivity could certainly be a problem if the iPlayer is as popular as some expect, but the article at Ars Technica makes the valid point that in the long term it is video applications like this that will drive demand for the faster ADSL2+ pipes (and the generation that will follow them), which will help keep the ISPs in business.


9th August

At the perennially wonderful Dan's Data, assistant geek Mark Cocquio has reviewed a cunning little blank disk dispensing gadget, and as soon as I saw the article I knew that I wanted one. Or two, actually, as by chance I'd just ordered a pair of 100 disc spindles, one of CD-R and one of DVD-R disks, and the idea of a matched pair of dispensers was too tempting to resist. A Google search turned up a small handful of UK suppliers, but in fact only veteran electronics and components supplier Maplin Electronics appears to have any in stock at the moment. Maplin aren't necessarily the cheapest source, but they have always had items that are hard to find elsewhere and I've been shopping with them for twenty-five years or more.

There isn't much to say about the Disc Pod that Mark hasn't already covered in the review (and his photographs are several orders of magnitude better than mine, as well!), and I agree with his assessment that it does what it does very well. You squeeze the lever, and a disk pops out of the base ready to grab with thumb and fore-finger. There's video at the site of the manufacturer, Point One Technologies, for the terminally baffled or insatiably curious.

One point that isn't mentioned in the review, but that I think is worth noting, is that the unit doesn't rely on the locking tabs built into the cylindrical cover that comes with a CD spindle to attach it to the unit. This is significant, as not only are there several different layouts around (the handful of spindles I have available to examine right now are evenly split between three and four tabs, for example) but they are not terribly strong and are often damaged. As a case in point, both of the spindles I bought this week arrived with at least one tab broken off, thanks to somewhat inadequate packaging on the part of the supplier, Dabs, and the dedicated and hard working employees of the Amtrak courier service who presumably used my package for an impromptu game of after-hours football.

I was anxious about this, but when the Disc Pods arrived this morning (next day delivery for no additional cost - thanks, Maplin!) it was obvious that it wouldn't be a problem: the unit comes with a tough ABS plastic collar that slides down over the cylindrical cover and jams solidly at the base, and it's that collar that locks onto the body of the unit. This means that even quite extensively damaged covers will work well enough, and the resulting joint is pleasingly solid and stable - quite important when you're using a spindle of 100 disks! I paid £10 each for my Disc Pods from Maplin, and although shopping around could easily turn them up for half that price once the product catches on, they're definitely worth the money and as usual I'm happy to pay the price of being an early adopter.


8th August

Last week ripples appeared in the Windows world following the leak of a pair of updates for the Vista OS, the "Performance And Reliability Pack" and the "Compatibility And Reliability Pack". These have come as somewhat of a surprise, as although reports around the time of the Vista launch suggested that Microsoft was "fast-tracking" the first service pack for release in the second half of this year, in fact the current release date is pegged for 2008 at the earliest - allegedly to dissuade the notoriously conservative corporate sector from waiting until its release before starting the migration to the new OS. The timing disparity didn't stop the pundits from speculating that the leaked updates would form a major part of the service pack, but in this case they were completely wrong - both updates have just been released as stand-alone patches (KB938979 and KB938194), and they are expected to be available via the Windows Update mechanism as part of the "Patch Tuesday" set next week.

Early reports suggest that one of the major performance bugbears of Vista, the surprisingly slow speed obtained when copying files across a network, has been significantly improved - although it may still be worse than that of Windows XP or Windows 2000 before it, and there are suggestions that performance of local moves and copies has not been improved to match. As always, the updates should be treated with a certain caution, as the appropriate discussion thread at Ars Technica already has a complaint from a disgruntled user that they have "completely borked" his laptop.

Elsewhere, The Register reports that a dispute from the dawn of the PC era has finally been settled, when a judge dismissed a lawsuit brought by Tim Paterson against author and journalist Sir Harold Evans over his book "They Made America", one chapter of which discusses Digital Research founder Gary Kildall and the hugely significant but largely unsung contributions he made to the early development of personal computers. Paterson was the creator of the QDOS operating system that was bought by Microsoft and re-badged to before being licensed to IBM for their first PC, and although Microsoft was already successful and influential software house there is little doubt that the MS-DOS deal made their fortunes and set the company on the road towards becoming the apparently unstoppable behemoth that it has become a quarter of a century later.

The lawsuit claimed that Evans' book caused "great pain and mental anguish" by stating that Paterson copied the look and feel of Gary Kildall's pioneering and successful CP/M operating system, which most industry observers found unusual as it has always been widely known and accepted that that QDOS was at the very least strongly derivative of CP/M and, at worst (and according to Kildall himself) a straight-forward rip-off. Judge Thomas Zilly appears to have agreed, and dismissed the case as having little merit, along the way castigating the plaintiffs for failing to provide any real evidence to support their claims and for introducing "irrelevancies" into the proceedings.

Meanwhile, the article's discussion thread at The Register shows that people can make just as much fuss about the operating system wars of the early eighties as about the more contemporary skirmishes, proving once more that, like an elephant, a geek apparently never forgets...


7th August 2007

Linux! Grrrrrrr! My team has caught the VMware bug, of late, and today while they were busy upgrading the memory in their personal servers to allow them to run multiple virtual machines simultaneously (to four gigabytes! - I used to run whole companies on four megabytes, in the heady days of Netware 2.15!) I thought I'd have a poke around and see what has changed since the last time I looked at virtualisation back when VMware was a mere stripling. From time to time I use CD-based Linux distributions such as Knoppix for ad-hoc security chores such as mounting an NTFS partition to remove some particularly virulent piece of malware from a Windows XCP installation, or scanning for rogue wireless access points etc, so as a test I chose a pre-packaged VM of Ubuntu and Gnome.

The VM's packager obligingly created a default user account with the name "changeme", and as I didn't fancy using something that generic (and in any case would never remember it next time I needed the OS) I obediently navigated to the user and group management tool and changed it to my usual login name. I changed the path to the home directory to match, as well, and this is where the teeth gnashing started. While I was tweaking I also changed the VM's hostname to match our network's naming conventions, and when the system asked me to reboot having done so I obliged - only to find that I couldn't log back in again!

On logging in the system told me that the home directory didn't exist (dammit, Windows would just have created it automatically, and copied the default user profile into it!), and the suggested alternative of using the root directory logged me straight out again after another error message that my %HOME/.dmrc file (whatever that is!) was being ignored - and then it logged me out again automatically! The error message actually suggested that I should use a failsafe session to resolve the problem ("ah-ha", I thought: "safe mode - that's just what I need to change the path back!") and after a surprisingly long search I finally found out that in order to login in this way I simply had to press F10 at the main screen to select a session type. However, this gave me exactly the same result, displaying an error or two and then logging me out yet again! The obvious next step was to login as root and fix my own account, but a further search showed that in Ubuntu the root account is disabled by default, and has to be manually activated after logging in with another ID - and as this was in a thread where the assembled lawn dwarves were arguing bitterly about whether this is a good idea or not I can offer my own opinion that it bloody well isn't!

So, I'm stuck. I can't login as the account I edited, even in a failsafe session, and I can't login as root without logging in as the account I edited! The only solution seems to be to download the VM package again (I'd deleted the RAR file it came in, and at 3Gb it was too big for the Recycle Bin) and start from scratch. I'm sure that there are cunning ways to fix the problem by logging into a command line shell, or administering the system remotely over the network, but they're beyond my limited Linux experience and I don't have time to learn. My opinion of this state of events is pretty poor, of course - there was no warning when I changed the home directory path that the directory didn't exist, or that it had to exist, so I didn't see any risk until it was too late - and although the evangelists would proclaim that this means I'm too stupid/foolhardy/Windows-centric/beardless to be fit to use Linux, I interpret it differently. An OS that allows a veteran sysadmin (if, admittedly, experienced on another family of OSes) to break it immediately after installation, with only a couple of mouse clicks and without any warning at all, is too damn fragile for my taste, and I think I'll stick with Microsoft products in my VMs for the moment.

On a somewhat related note, Daniel Lyons of Forbes magazine (recently revealed as the author of the excellent satirical weblog The Secret Diary Of Steve Jobs) is as unimpressed by the zealots of the free software crusade as I am - but being far better connected he is able to devote an entire weblog, Floating Point, to exposing some of the remarkable machinations and dubious partnerships behind the scenes. Things are especially entertaining right now, it seems, thanks to the recent rift over the latest version three of the GNU Public License that underpins the entire movement, with Open Source Development Labs (current home of Linus Torvalds himself) and most of the small Linux distributions on one side, and Stallman and the FSF, together with giants IBM and Sun, and tech propaganda site Groklaw on the other. The battle is being played out both in the courts (as an off-shoot of the long-running SCO vs. The World case) and on the web, and it's fascinating stuff!


Meanwhile, back at the stats... Although July saw a small climb in the overall totals, some interesting information has emerged about the side-effects of a change to the way Google indexes and returns information. A number of major sites have reported seeing a sudden fall in their numbers after the end of May, the exact phenomenon that has had me scratching my head of late - if on a far smaller scale! As the report at Wired puts it, Google Giveth, Google Taketh Away. Indeed.



Weblog Archive