17 posts • joined 11 Jul 2010
Re: Do it in house
A hybrid approach would seem apposite, where single central platforms are developed by distributed teams.
Take the example of the Linux kernel - initially developed by a single individual, with basic requirements. Very much a single-platform project.
These days, large commercial parties cheerfully contribute to it's development - to support their hardware, to support their business needs. And small independents cheerfully contribute to it's development - in my case, to selfishly serve my own needs and fix driver bugs for my hardware, but *everyone* got to benefit from my changes.
Even relatively long-lived forks for selfish reasons - the Android versions of the kernel, for example - have a tendency to drift back towards the middle.
And if you really can't live with the direction of it any more, you can always fork it as happened with XOrg, OpenOffice / LibreOffice amongst others.
The NHS has certain common, core, needs. PAS should be a total no-brainer.
So ; yes to local dev teams as the best way to work out what local needs are. You pool together enough local needs, and you end up with common needs - which is where the central project maintainers have a role.
Stick a big Git server in the middle and have at it!
UK wanted this too
The Dept for Transport had plans for this published on their website for a while
"Road Pricing" was the name for it.
They ballparked that a GPS logger with cellular modem would cost on the order of £100 a pop for each vehicle. They also noted that their system should be compatible with a Europe-wide system, which is rather creepy.
They're still thinking about it, you can be sure.
Why do you think that Europe sponsored it's own GPS system? With better accuracy for high-density urban areas?
The thing is a total nightmare, both from a practicality and privacy standpoint.
Occam's Razor will now be applied.
- The stated aim of these systems are to "reduce congestion on key roads at peak times".
So you only need a system capable of charging to travel on these key roads, not one that can track every vehicle wherever it goes.
This kind of system already exists - it's called a toll road. We already have toll road systems that don't need a GPS tracker fitted in your car - the London Congestion Charge is such an example. ANPR works just fine.
Even if it didn't, you could make an RFID tag in the number plate a mandatory item to pass your MOT test. RFID tags cost less than a tenner, not more than £100, and only track when you drive on roads with a pickup loop. Pickup loops are cheaper than expanding cellular networks to cope with the log uploads of the 30M vehicles on British roads.
But they wouldn't even achieve the stated aim. Why? Because a traffic jam is already enough incentive not to drive on a certain road at a certain time. Traffic jams form because of inflexibility in working schedules and people living too far from their place of work (for whatever reason).
And why the hell do the government care about traffic jams anyway? Traffic jams cause less damage to roads than equivalent amounts of fast traffic. They don't care about the extra fuel burned because it makes them and their energy company cronies more money. They don't care about the lost productivity because they get their cut of your wages whether you are actually earning them or not.
So ; we have a system proposed to solve a problem, that won't actually solve it, and it's not a problem that they care about. Ergo, the system is designed to do something else, which is provide the ability to track the movements of every vehicle in the country, or to make a lot of money for some consultancy implementing it. Or both.
Re: Is this a problem?
Since it doesn't have to erase a whole cell at a time, I'd say it's longevity is enormously greater than standard flash ; those 100,000 cycles will take a lot longer to accumulate.
The problem with government IT is that they start large. They try and define what it is they want, up front, and spend years doing it. Then they wonder why their Chinese-whispered definition of a giant ball-of-mud process that they've cobbled together from treasury tags and paper mache moistened with the tears of their employees, doesn't produce good software.
Sit the developer with the users. Get them to explain what it is they are actually trying to achieve, rather than trying to define the voodoo ritual they currently use to achieve it. Implement something small. Repeat.
Re: I fix PCs for a living.
Not just the most useless, also the hungriest in terms of system resources.
We used to use Symantec at my employer. It ate about 10-15% CPU time in IO intensive operations. McAfee eats more like 40%. It makes processes that used to take 2 minutes take 8.
But they got it for a good price because it was bundled in with a load of other resource-sucking corporate managementware. *facepalm*
Re: I'll give it a go
It shouldn't affect application portability - there's just some hooks into the standard menu toolkit libraries that allow them to sit at the top of the screen instead of the window, and allows the HUD to access the menu.
Alas, the converse isn't true - not every application works properly with the HUD and global menu, as noted in the article (although for LibreOffice, just install the lo-menubar package), but that can improve without forcing people to patch their applications.
Re: @DryBones yes, yes, that's all very well
Canonical used to sell codec packs in their online shop for those people that felt they needed indemnity, but they no longer seem to stock them. I guess a lawyer may have felt that they represented some kind of admission of guilt for distributing codecs freely as well.
The Free set of ubuntu-restricted-extras is now the top item if you search in the software centre for "codec", and a bit further down if you search for "mp3". I couldn't find a for-pay codec pack in the software centre.
LibreOffice HUD support
FYI, there's an extension for LibreOffice that integrates it with the global menu, and thus into the HUD.
sudo apt-get install lo-menubar
It's not marked as "supported by the distribution".. heaven knows why, you'd think the combination of their flagship new feature and the default applications they shortcut in the launcher would be a total no-brainer.
This is an IMPROVEMENT
For the vast majority of customers, this is probably an improvement.
Why? The previous throttling rate, as someone points out above, was 25% of bandwidth. Yes, they have halved the ceilings before you hit this, and for the first time, have imposed throttling on their top-band customers. BUT
i) Currently rolling out a charge-free doubling of bandwidth to all their customers
ii) Have double their "throttled" rates from 25% to 50%
So essentially, all their long-term customers will now get the speed they originally signed up for, only all the time, with a bonus burst of double speed when they are off peak times. And now, when you don't exercise care about when to do a large download, it's about half the problem when you hit the ceiling.
So the news is, their product is now much more attractive to the vast majority of users, leaving a small vocal minority of whiners who seem to think it's their right to monopolize an optical fibre so they can torrent media as fast as possible.
NIF is a weapons research project
NIF is funded largely by the military. It's primary purpose is to be able to examine the physics of fusion reactions without letting off H-bombs. The whole "fusion energy" angle is merely great PR.
To make this into a commercial fusion generator, you have to scale things up somewhat.
Currently NIF fires a few shots per day. To make a commercial reactor, you'd have to fire about 10 shots per second.
So instead of carefully positioning the fuel target in the centre of the laser array with micrometre precision, you'd essentially need a machine gun. A cryogenically cooled machine gun that never misfires and shoots with micrometre precision.
Oh, and the ammunition. The fact that each round is composed of a fuel pellet contained in a beryllium sphere surrounded by a cylinder of gold-plated uranium is not what makes the fuel expensive. What makes the fuel expensive is that it contains tritium, one of the rarest substances on Earth. The USA has only ever made 225kg of this element, and only 65kg of it is left - most of that loss just from natural decay. Even though each pellet only has a few mg of fuel, at 10 shots a second, you'd burn through that tritium reserve in less than a year, for just one reactor.
And how to extract energy? Well, you have to take the heat and generate electricity... so you have to put a large heat exchanger in the reactor. And somehow not obstruct the laser array that shoots the pellets when you do so.
Oh, and the laser array... currently about 1% efficient. So you'd have to either improve that greatly, or generate more than 200 times the energy you put in (accounting for efficiency of steam turbines). And ramping things up to make the laser fire 10 times a second seems difficult given they currently only fire a few shots a day.
Laser-initiated inertial confinement fusion does not seem practical.
I find the BBC totally worth it - just the fact that it's present forces all the commercial broadcasters in this country to raise their game. Without the BBC we would be facing the same miserable morass of commercial programming and an average of 18 minutes of advertisements per hour, as there is in the USA.
Turn it on it's head ; what if the USA had something like the BBC? Content costs the same no matter how many people are watching it ; therefore the more people watch it, the better cost / benefit ratio it has. Imagine what the BBC could do with the license fees of a population of 300 million people, versus their current 60 million.
That said, I think they should focus less on following the trends of the commercial broadcasters and get back to what they do best ; producing world class programming through experimentation.
As for paying for iPlayer ; I would imagine that both the commercial broadcasters and the ISPs would be ecstatic - the commercial broadcasters will cease to have online competition, and the ISPs will see their networks much less utilised. I'd be happy enough to see them permitting paid iPlayer access abroad, but as a license payer, I paid for that content already. I guess I'll just keep using my excellent MythTV PVR (with three Freeview tuners, I only need online players when I miss the first few episodes of a promising series).
Re: Was he born evil, or did the absolute power corrupt him absolutely?
They're all just bleeding into one another ; I don't perceive any major differences between them now.
Watching the Conservatives lambast Labour for continuing with the PFI schemes that the Conservatives themselves introduced, particularly when you know that they too will continue with them, is the soul of irony.
At least with the fire and brimstone socialists you knew what they stood for. All the current lot will kiss your baby and shake your hand while promising eminently reasonable dreams to the public, while either quaking in their Gucci shoes or salivating internally at the thought of their corporate masters.
And the corporates fear and hate the internet (at the same time as they love it), because it gives the commoners the means to organise and communicate. Hence we have the BBC publishing "opinion pieces" that we should tear it all down and start again with something that can be controlled. They propose legislation that mandates tracking all email, phone, and social media communications. We have computer systems (predominantly phones, for now) designed to ignore the commands of their users and do the bidding of their corporate masters.
What to do, what to do....
I'd prefer the Google approach (subtle side-bar ads relevant to the context), to the Microsoft approach....
* Clippy appears *
"It seems you are about to have a w**k. Would you like some help with that? How about :-
* Call Customer Support over VoIP, now with a "Personal satisfaction" script
* View the latest webcam footage from Steve Ballmer's Palace of Love
* Buy some Micro-soft tissues
There's the flipside to this - it was a feature, for a long while. If a file was already in the Dropbox cloud, you could upload it in a fraction of a second, because your client would just say "I have a file that matches the hash for fedora-linux-16-x86.iso", and the server would say "Already have that one, thanks".
Alas, the content sharing community managed to hack this with a client that just told the server you had a file with the hash for "Latest Hollywood Blockbuster.mp4", which meant that suddenly Dropbox became a file-sharing server ; one person would upload it, and everyone else could copy the hash into their Dropbox folder and download the file. Dropbox, not wanting to be treated like MegaUpload, rapidly nixed this "emergent feature".
Open Source *requires* copyright to function. Copyright is what gives the original author of the code the right to state what license it is distributed under.
What are you really selling when you sell software? An ephemeral pattern of bits on disk, which is cheap? Or the expertise and work you put into writing it? The sale of software as a commercial product is one way to solve the problem that the cost of the labour to produce it is often more expensive than a single customer is willing to bear ; Open Source is another approach which addresses this issue.
It does, indeed, seem to be perceived as old-fashioned to believe that you can rest on your laurels and live off the fruits of your past labour ; this is a common argument against copyright extension for the music industry. I don't think anyone would begrudge me the literal fruits of my labour if I built a really good automated tomato production system, but conversely, I don't believe I would be upset if someone copied my design, improved it, and built their own system. Reproducing copies of software requires very little in the way of materials and labour though. I suspect the real issue is the retail model of software development - it just doesn't seem to be sustainable in an era where people are willing to legally share their software openly.
To address the problem that some software is just too expensive for a single buyer to pay for development, perhaps the solution is to use the distributed patronage systems like Kickstarter - which just managed to raise over $800,000 in 24 hours for a well-respected game studio to develop a point and click adventure game.
LibreOffice / NhsOffice
For a fraction of that £80m you could fund a few C++ programmers, some support staff, and a team to assist migration to LibreOffice.
Perhaps that's the motivation behind ditching the Enterprise Wide Agreement - once you collect all those licenses into one bill, it becomes more obvious what they cost, and what that kind of money could buy you.
The NHS needs to realize that software is not like other purchases, especially commodity software like Office. You can build it once, and copy it as many times as you like.
Britain needs a Public Key Infrastructure to enable email to be useful for trusted communications.
If Estonia can do it, why can't we?
- YARR! Pirates walk the plank: DMCA magnets sink in Google results
- Pics Whisper tracks its users. So we tracked down its LA office. This is what happened next
- Review Xperia Z3: Crikey, Sony – ANOTHER flagship phondleslab?
- Ex-US Navy fighter pilot MIT prof: Drones beat humans - I should know
- Human spacecraft dodge COMET CHUNKS pelting off Mars