Posts by Charlie Clark
1688 posts • joined Monday 16th April 2007 14:57 GMT
Page:
RIP
Our thoughts are with the family and friends of this brave and pioneering Playmonaut who has done so much in following the manifest destiny of Plamobil to conquer the sun and the stars, if not, unfortunately the sea.
Re: Doing What Jobs Urged Them To
@Steve Todd - I think you completely misread my post. Check out the following for full details of Apple's perverse approach to openness:
https://docs.google.com/document/pub?id=1GWTMLjqQsQS45FWwqNG9ztQTdGF48hQYpjQHR_d1WsI
Re: Doing What Jobs Urged Them To
This would be the same company that, for whatever reason, decided that the video of the I-Phone 5 was best run by Javascript from a collection of still images? Apple's motivation then and now was about more control over and less competition for the user.
Re: UltraHD?
UHD comes in two flavours: 4k and 8k as featured on El Reg a few months including a video with some bod explaining the difference. SuperHD & UltraHD or similar could have been used but were rejected in favour of the industry tradition of confusing the customer in the hope that will somehow help (HD, HD-Ready, 1080i & 1080p, etc. ad Nauseam)
Not that bad
Many office workers have notebooks that spend most of the time in docking stations and they cost quite a bit more than this. They're seldom in use when not in their docking station except to check for e-mail or look something up on the interwebs or take notes at a meeting.
Something like this would probably be welcomed by a lot of people though I think the price will have to come down - even if the hardware is comparable to notebooks, tablets have got significantly lower price points.
I have to have one for a customer and I do resent carry > 2 kg around between docking stations. Have to see what the hardware is really like when it's available and, assuming I can get permission to use it on the network, then I might get one.
Re: The Other Guys and esolving limits
Nigel's right: we often confuse, and manufacturers and broadcaster peddling new things encourage us in this, resolution with other factors. The difference in codecs between SD and HD is probably what most people notice and think it's the resolution whereas it's largely artefacts caused by the codec and reduced detail down to lower sampling rates. But resolution is much more marketable than boring things like that hence the "retina" fad. Me, I'll take higher contrast and colour range over resolution any time, but maybe I'm just weird.
Re: Not just OS X
Linux distros often have lots of BSD stuff in them including, rather bizarrely, configuration utilities for things like WiFi. I think stuff never gets removed once it lands.
Re: "English and Welsh citizens"
Well, no. Britain's current libel laws actively invite libel tourism: any publication that can be obtained in England and Wales can be tried in an English courts and this applies as much digitally as it does to anything else.
Other countries tend to have significantly better protection of individual privacy thus the no nonsense approach of the French courts in the case of the Duchess of Cambridge's baps. Yes, we know they're still online but a precedent was set as to who would liable. Because we don't really protect individual privacy with a constitution, libel is often the only way making it a bit of sledgehammer for a wallnut. Issues of liability still need addressing but the willingness of companies like Twitter to hand over users' details to the courts upon request is as much about self-preservation as anything else. By complying they are less likely to be held liable themselves and required to act as gatekeepers in the future.
And people do need to adopt some sense of responsibility for their actions. Earlier this year the premature disclosure of the suspect's identity in a rape case in Germany led to a lynch campaign on Facebook, the effects of which were difficult to reverse even after the suspect was released as innocent and someone else charged and convicted for the offence. Our obsession with titillation seems to encourage us to trivialise this kind of thing until children are involved when we suddenly seem to rediscover the moral highground: it's okay to hack celebrities' mailboxes but not those of dead children!
Article has at least one elementary flaw
As much as I like the Akamai IO project the effects of the US skew of the data cannot be understated. For example it credits IE (all versions) with around 50 % market share and it significantly increased at the end of June. Comparable worldwide figures including countries like Korea and China with surprisingly high proportions of IE users give it around 30 % market share. The skew for mobile devices will be even greater because of the greater reach of IOS devices in America.
However, on top of this there is gaping flaw in these stats: Apple is a heavy user of Akamai and Google is not so an awful lot of those users with Android are permanently off Akamai's radar. We can add to this the conflation of IOS with phones. Tablets, where Apple still has a commanding lead worldwide, are used far more for browsing than phones including displacement of PCs and notebooks in the home. Displacement is important as it exaggerates the effect of market share. As long as you don't have to do a lot of form filling then a tablet is definitely the most convenient web device which is why Apple is prepared to go to such lengths to try and preserve its lead.
Nevertheless, when it comes to shopping statistics, Luke Wroblewksi, who writes both more intelligently and coherently than Mr Asay quotes IBM (via Techcrunch) for the Thanksgiving period: "The iPhone was the most popular device driving US retail shopping with 10.5% of visits, the iPad accounted for 10.1% of all visits and Android devices were 7.7%." A much narrower gap and that for the US market.
Re: 350,000 queries per second until...
You might very well be right. And there is nothing stopping you from proposing a fix and submitting a patch. Try that with Oracle.
I found "Learning Python" an excellent primer. Python is probably closer to Pascal than Basic so forget GOTO just give your subroutines names and call them functions and you'll be away.
Phorward
You, sir, have just won the coveted Bong! award.
Rory Clellan-Jones trip to coding school was possibly the nadir of an already ignominious career.
Re: Quality? I recommended a colleague buy a Nokia Lumia 710
@JDX
The problem is that such products should be recalled because the problem is known about.
NB. the poster wasn't bitching about the phone just pointing out the less than stellar consumer experience. Not surprisingly this has led to someone jumping ship and while this happens to everyone all the time, it's not the sort of thing that Nokia call really afford.
Thanks for the review
Looking for a Windows 8 Phone for compatibility testing and despite the nice things Andrew has to say about this, it looks like it won't be a Lumia.
Re: Welcome on board!
If Bill didn't want any code review he wouldn't have posted his code.
See post lower down for why global variables are generally not required. The keyword is there so you can use it when you need to but it really is something that you very, very rarely need in Python and has extensive side-effects that you generally don't want. Pointing that out has nothing to do with ivory towers.
FWIW I don't think singletons would be needed here - something for which there isn't a keyword because you don't need them. Moving the functions into the controller class would provide mutable instance variables safely isolated from immutable module constants.
Welcome on board!
Thanks for sharing the code - you might want to put it up on Bitbucket, Github so that we can, er, "fix it", for you.
Code is good for a start - it gets something done - but you'll have to get out of the habit of using "global" variables as that is very much frowned upon. Add in support for dispatching, string-templating and you're almost done. Well, there is more but then there always is!
?
It's arguable that the internet (and it certainly shouldn't be capitalised if it is being used as an "omnibus" term) is displacing other even more energy hungry activities such as physical mail. Be that as it may, growth is now increasingly wireless so it's difficult to see advances in optics making much of a difference there - yes, I know backhaul and data centres are still cable-based, but the article's premise is "growth".
The point about roof-panel solar energy depressing peak electricity prices is spurious: the spot market is irrelevant for data centres. You might want to compare growth in absolute terms of cheap renewable power generation with consumption by data centres and devices, though the increasing fungibility of gas is currently driving prices.
Re: I loved OS/2
Wasn't NT 3.51 the last version that was actually a microkernel and, therefore, uncrashable? But it was slow - this was blamed on context-switching on x86 but OS/2 was managing it fine. NT 4 was was faster but achieved this by putting drivers into the kernel and hence all the BSODs.
Re: I still use ecomstation ...
Your timeline is off. Microsoft started doing the dirty on IBM re. OS/2 towards the end of the 1980s while they were contracted to work on OS/2 v2 & v2.1. Taligent was later and part of OS/2 v3 (Warp) and is the basis for most of our fully object-orientated, widget-based "homescreens". Taligent and OpenDoc promised real productivity ("let me add a spreadsheet to my document…") but most people are happy with widgets.
Hearing you mention MS' version of Unix (Xenix for x86…) makes me feel physically ill. Though, to his credit, Bill Gates was listening to the market. He did employ some great people to work on NT and copied many of the great ideas from OS/2 such as, hardware abstraction, extensible attributes and virtual filesystems. It's just a pity they were doing this a sort of trojan horse while still under contract to work on OS/2. Windows 7 is alarmingly close in many ways to OS/2 3 > service pack 17. And that after only 15 years!
Re: Looking forward to part 2
I agree, great article. OS/2 was/is fantastic - I think UPS handheld terminals are still running it as are numerous other embedded systems and ran in banks and airports for years - which was let down by the less than stellar performance of the PS/2 MCA systems (better than than ISA but in a "malaria is better than typhoid" way). Well, MCA was a major improvement over ISA but didn't bring enough performance and bandwidth to seem worthwhile. So we all got to suffer the abominations of EISA and, shudder, VESA, before MCA's heir PCI was able to triumph. But that was ten lost years. Had MCA succeeded IBM would subsequently have dropped the Neanderthal x86 chips… but they were too busy chasing the hardware monopoly.
A key OS/2 innovation was hardware virtualisation so not only could DOS applications run in complete isolation from each other they also got more memory allocated than they could standalone. Oh, and resources like printers and serial ports could be shared effortlessly across machines. OS/2 also came with a damn good scripting language REXX in which one of the first http servers wasn't written, if memory serves me correctly.
However, there were also downsides: while you couldn't kill the OS, the GUI called somewhat pompously "Presentation Manager" was single-threaded up until v3 service pack 17 (IIRC) which did mean that individual misbehaving apps could ruin everyone's fun. Still crashes came with detailed, numbered errors and stack dumps which I presume trained engineers understood. By far better than NT's BSOD.
Back to Microsoft - I think the article makes a good job of suggesting that the technical wizardry of which IBM was justly proud was, in the 1980s and 1990s matched by a bureaucracy destined to stifle innovation at every turn. The same happened to Apple a few years later.
Re: Opera and OWA
@Pierre you really haven't a clue - if the site was properly written it would "degrade gracefully" either by UA detection, which takes a lot less than 5 seconds, or by client-side feature detection (in action since HTML >= 3.2). Neither, however, would explain Microsoft singling out a specific browser. And neither has to do with "pushing towards more fragmentation"; done properly then detection provides the best possible experience for all browsers.
Applications like OWA are why the specs need to develop. I'm not sure that I'm in favour of turning browsers into runtimes but I'm also not going to play Canute.
I do hope…
…that HAL is running mission control on this one.
Re: Opera and OWA
And this is Opera's fault because…? Not that it doesn't have issues but website problems in Opera are generally caused by servers sniffing for UAs as opposed to feature detection* and only letting their "friends" into the garden. This is sloppy coding and unfortunately unsurprisingly on the rise again.
* check out the support for feature detection with CSS in Opera. Even if you don't use Opera as your standard browser this feature will make site development easier and Opera offers a head start. To me, that's all good.
Got lots of PDFs? Get a Sony
Sony readers come with a special version of Acrobat which is really good at reflowing PDFs to fit the viewport of the device. Head and shoulders above the competition and yet, depending, on the layout of the PDF still often unsatisfying in comparison with a dedicated e-book.
AFAIK the Kobo is one of the few readers that swings both ways and lets you reader both epub and mobi files at the same time. Would have been good to see that covered in the review,
Re: XBMC not as scalable
I guess you've never heard of DNLA then. Works on my telly, phone, tablet, notebook, etc.
Load of crap
What kind of network really needs to worry about this? If it hooks into the TV infrastructure then it can be a very asymmetric service. Not that is really a problem. Currently trialling Maxdome here and HD films check for a max of a 6 Mb/s connection with an SD card used for buffering so will work with less. The "app" itself has a few problems but connectivity isn't a problem - my provider will happily provide me with up to 100 Mb/s.
Any bandwidth hungry device will come a notice informing the user that they'd better pay to have sufficient bandwidth or they won't be cool; revenue share of course. Worked wonders for mobile. Actually provisioning the bandwidth shouldn't be a problem for any half-way decent provider who has fibre all the way to the DSLASM / splitter.
Barely concealed personal information grab
NFT
Praising British trains?
You can tell that Mr Dabbs doesn't travel very much. My god but British Intercities are poky little things with less leg room than a Ryanair Alicante special! Even in first class getting in and out of a seat is best left to contortionists! Treat yourself to a ride on a proper train (ICE 3rd generation say Frankfurt - Amsterdam). Power sockets below the armrests on every chair, intelligent reclining seats to minimise disruption to those behind you, compartments with optimised data connections for those who need them, and quiet compartment for those who don't and prices that don't go sky high just because you want to travel during rush hour.
Re: That's what building codes are for
It's the same in Europe - sockets no more than 2m apart. But then Britain is famous for observing the law in the breach… difficult to beat us Brits for cheapskating.
Bout time someone mandated high-power usb power point on all sockets.
Moore's law
The question is slightly disingenuous because of the power limit. If that has been constant at 26 amps throughout the history then you need significant changes in architecture for boosts. Might be better to plot FLOPs/Watt over time for a better comparison. If the cap has been in place all the time and the improvements have been gained by the move to GPUs then Intel should start to worry.
Extreme measures
Seems like Apple is going to extreme lengths to be able to boost the number of I-pad Minis shipped…
Clarification required
The article seems to make a distinction between hotspots and access points. I'm not that familiar with WiFi topology but I've always considered the two to be the same though usage seems to imply that hotspots are public access points for a particular network over a large area such as a city and access points generally relating to more or less closed networks such as hotels and conferences.
This sounds like network management optimisation which is unlikely to have much effect with just one device. I'm also not convinced that if everyone is downloading you can increase the yield. This approach sounds like load balancing across access points. Surely, before any effort is made in that direction, you need to make sure that the access points are set up to consider both the effects of the environment, user density and interference from each other? Not really up on much of this so would appreciate an explanation.
Has there been any work done on Bluetooth 3 networks which use Bluetooth as a d-channel to manage clients while data is carried on WiFi?
Re: Why they did it
So, don't use IE would seem to be a better solution.
The same export restrictions also applied to the other browser of the time: Netscape and there was no workaround for it at the time. So, the decision at the time was either use ActiveX and have encryption or don't use ActiveX and have no encryption.
There has since been plenty of time to start moving towards using 128-bit https encryption in the browser but changes like this have very high inertia: you have to maintain both systems while you migrate customers. And, once developers have got used to a monoculture they are even more resistant to change which, let's face it, is not required by the job in hand and is going to incur considerable costs. Of course, it is exactly to avoid this kind of lock in that people advocate open standards for.
For patent issues
A square with rounded corners.
Why they did it
At the tail end of the 1990s, the Korean government decided in its wisdom to develop a home-grown 128-bit SSL encryption standard to increase security around e-commerce.
They didn't choose to do this out of vanity. US export restrictions prevented versions of IE with 128-bit encryption from being distributed to Korea, 64-bit was known to be compromised and ActiveX was a reasonable solution to a big problem then.
'Tis true
Corporates are still moving to Windows 7 and while this means that users are moving from IE 8 to IE 9 they are moving to Chrome faster. Stats from a fairly typical corporate site look a bit like
IE 9 18 %
IE 8 16 %
Chrome 30 %
FF 28 %
Safari 8 % (mainly IOS devices)
In comparison, a year ago IE 8 was around 30 % and Chrome around 10 %
It's worth going back to the
first IE 9 preview release which promised a release early, release often schedule which the first IE 10 preview seemed to continue including availability for Windows 7. I seem to remember fairly firm release dates being mentioned but I may be mistaken.
IE 10 does seem to have got a lot of things right but it's also trailing from the word go. HTML 5 coverage is below the previous generation of the other browsers, and nice as the JS optimisations and hardware acceleration might be, users are more likely to notice and appreciate support for SPDY as companies move to adopt it for their servers now that more and more browsers support it.
In any case it doesn' t really matter. For Microsoft IE 10 is too little too late. For them to stop the possibly irreversible slide in the corporate space they need to release IE 10 for Windows 7 as soon as possible. Corporates are overwhelmingly skipping Windows 8 so the next version change won't be before 2014 by when IE could easily have been overtaken by Chrome, Firefox and IOS.
Re: Ribotro
I'm in total agreement with you about Ribbon - I have to use MS Office with Ribbon for a customer and even after more than a year I am not comfortable with it and I have used MS Word on and off for over 20 years - Word 2 which mixed toolbars and menus was the best I've worked with.
I think Metro has some great ideas which if used sparingly would be part of a great UI but basing everything around it? The tiles are great for pairing back and summarising information. This would be great in a notification centre but plastering the screen with them is counterproductive.
Don't know about Silverlight but I think this kind of animation has been around for a while in things like CD-ROMs, remember them?
It's certainly been in use on http://www.20thingsilearned.com/ since long before the patent was applied for so one slap for the submitters and one slap for the USPTO for approval of something for which prior art was a few mouse clicks away.
Apple should be fined % of turnover for this kind of abuse of the patent system.
Re: It's called competition
Indeed, though as Samsung is supplying the presumably higher margin Nexus 10 they are unlikely to worry too much. Anyway Samsung has announced the Galaxy Premier as a slightly cut-down version of the SIII.
@ Mr Cox - as desirable as "sold out" notices are for PR, most companies like to be able to keep on selling. Engineering shortages is against the law. The most likely thing is that Google still doesn't really know much about selling physical products: shipping, warehousing, etc. an may well have been genuinely caught out by the demand which does imply what many have been saying for a while: good Android devices are considered to be as good as I-Phones.
Re: I wonder..
Chinese companies will pick this up in the next few years
1) Fabs are getting more and more expensive to build and they were never cheap in the first place. The next generation of equipment is already being cofinanced by the exising fabs.
2) In a few years the Chinese will be going "who's Apple?"
The last couple of 5 year plans have been about developing technologies for the domestic the market and using the economies of scale to subsequently dominate the market worldwide. Admittedly, this didn't work too well with the homegrown version of UMTS but hardware is something they are more suited to.
Unless we get 3d printing really under control the whole idea of getting stuff built cheaply in China and sold in the US or Europe at a huge profit is going to become untenable.
Re: They should clearly axe the obsolete standard
The sheer amount of wastage that would be caused by abandoning FM
What wastage would that be?
Re: RDS
Never understood why Radio 5 isn't also broadcast on FM
I seem to recall that at the time it was decided that the BBC would not be given any more nationwide FM frequencies so that national commercial radio would have a chance to develop outside.
Re: Then there is Radio 5
I think power of receivers is the only argument against the change.
The switch from analogue to digital TV has gone smoothly and, if my mum is anything to go by, people are generally happy with the result (time-shifting has become easier). The same would be true for radio. Of course, people would complain but then they always do.
AM does propagate further and require less power to receive than FM or DAB. Partly because of that there is more money in FM because it allows for a better carve up of the audience (radio is often very local) but there are limits on the number of stations that can run nationwide for the few stations that don't have identikit programming.
Anyway, FM propagation has its own problems: reception of the German version of Radio 4 varies from dreadful to shitty in my flat; it's not on LW and MW receivers are almost non-existent do DAB really is the only choice. I never listen to any of the other stations but then I never watch any of the hundreds of TV channels that are available either (more of the same old shit just at different times and in different languages).
As a society we have embraced digital broadcasting. DAB, or at least DAB+, is a good solution for radio. I suspect that, er, reception would improve if say mobile phones started supporting it instead of / as well as FM so that we could listen to it on speakerphone without having to plug a headset in.
The alternative, of course, given the popularity of radio of DVB-T might simply to extend that.
Pity
I think it was better when there was a built-in bullshit brake!
As for the speed not sure that Java is so relevant. A Java foundation in theory provides a scalable environment because so much money has been spent by Sun, IBM and others on the JVM but no one has ever really claimed Java to be fast and memory efficient (in comparison say with C). Moving to a more functional paradigm probably brings better IO performance with lots and lots of asynchronous clients.
Re: Design goals 101...
You, sir, are a cad! Steam trains is definitely off-limits!
Power consumption
It's not necessarily true that GPUs use less power than CPUs but it is generally true that any instructions carried out directly by hardware will use less power because they use fewer cycles. Hardware acceleration does, of course, depend on support in the hardware which for videos means support for the various codecs. Would be interesting to know what's supported just h264 or also webm.
In other news: Opera 12.10 was released.
http://my.opera.com/desktopteam/blog/2012/11/05/12-10-final-released
Support for SPDY and FlexBox are my favourites.
Security matters
to me at least. Not that I generally have a lot to say or anything that would be a problem for eavesdroppers but I do believe that people should have the most secure communications possible on any particular channel. E-mail is the glaring exception, of course.
Reviews of Wickr (http://mywickr.com) or Silent Circle (http://silentcircle.com) on El Reg would be nice.
Re: Sheltered Life
I'm actually a mobile developer, and as such I have 4 iphones, a couple of ipods, ipads 2, …And yes, since getting the mini I find that's the first thing I reach for, and the thing I carry around with me
Two things: this puts you very much in the same camp as Mr Dabbs who openly admits his long-standing preference for Apple products; it is a more than tacit admission that Jobs was wrong to have launched the I-Pad only as a 10" device. The relative and steadily increasing popularity of the 7" devices such as the Kindle Fire really have forced Apple's hand here. Though, as usual, they have responded with excellent hardware. I'd tend to agree that for doing much other than just consuming 8" is preferable to 7" but then Samsung was there first with the Samsung Galaxy 7.7 which has GPS and now LTE.
As with other products Apple is probably going to sell a lot of these devices to existing owners who have so far not admitted that the I-Pad is actually a little too big and too heavy to be really mobile. And it will no doubt encourage a few non-owners to buy the "original" over the cheaper competition. However, as with the phones on a like-for-like basis the I-Pad mini really does have little to offer over the well-made droids who are only likely to become both cheaper and better and offer even more content and services on a par with Apple.
Isn't this what Microsoft wanted?
What does seem increasingly plain, however, is that Microsoft's stance on DNT has backfired.
DNT is such a braindead approach - it depends entirely on advertisers choosing to play along - and Microsoft is effectively exposing this.
The only effective approach to tracking is that which drove the EU's law on cookies: opt-in only after consumers have been told about the matter and effective sanctions on breaches. Note that the law specifically applies to tracking cookies, session cookies, etc. are generally above suspicion. Personally, I think the law has largely achieved that what it set out to do: shine a light on the extensive abuse of privacy by online advertisers and others.
Contempt
If you go to court you are legallybound to accept the decision of the court, within the appeals process of course. So while you can express all kinds of emotions about the judgements ("we are shocked", etc.) you are not allowed to criticise the court's judgement because to do so is to hold the court in contempt. Apple's legal counsel should be shot.
