Back in the 80s there was some point in HIGs, and Apple back then was generally felt to lead the way. The reason was that there were, among your users, a very high proportion of new users. So we conflated ease of use with ease of learning, and it was not completely stupid, for much of the market using and learning were the same thing.

Now however HIGs have become part of the problem rather than part of the solution, because they make the implicit assumption that everyone works in the same way, and has the same basic skills. We just do not. And anyone who experiments a bit with end users will find this out in a flash. I have had people who loved spatial browsing because it might be cluttered, but they always knew where they were. Then there are people who love Gnome and the desktop and love to put all their files all over it where they can see them. And then you have the odd case of some totally non-technical person, who you try out with Fluxbox, and you get the reaction that this is great, this is how I always thought Linux was supposed to be, no clutter and very minimalist and above all fast. It turns out that hand edited menus and the explicit startup of the file manager are actually something some non-technical people welcome and find refreshing. Others of course will run a mile. One size does not fit all.

The Gnome ideal, that there is such a thing as the right way to set up a desktop, an application, is the problem. There simply is not, and when you take that approach, the penalty is that you inconvenience and impair working for at least one third of the people using it. Far beter to have a few broad choices, and then let people refine within it, and offer some guidelines. If you are not very computer familiar, start out with this, then see if, a while later, you want to move to this, and here is a very minimalist alternative.

HIGs are a snare and a delusion, very apt that they are sometimes rudely referred to as 'interface fascism'.

At the moment in the UK, subscription to the BBC is compulsory, as a condition of being able to have a TV. And if you watch TV without subscribing, you will be hauled before a magistrate, fined, and maybe imprisoned. People are imprisoned all the time for doing this.

What we need to do is make it voluntary. Everyone should be able to subscribe to the channels of their choice, or not as the case may be. Then, when subscription to the BBC is voluntary, we can just stop arguing about it and let them do what they want. If we don't like it, we would cancel our subscriptions.

This is so simple and obvious, its very difficult to understand why everyone doesn't support it automatically. What possible case can there be for making subscription to one particular broadcaster compulsory, and enforced by criminal law sanctions? Its totally nuts. We don't make subscription to one particular newspaper a condition of being able to read the press. We don't make subscription to one particular web site a condition of being able to have Internet Access. What is the problem here?

Yes, its great news. Now, we have a great way of stopping some of these idiotic ideas in their tracks, which is what patents usually do, or stopping them once they have got going, which is almost as good. But it comes too late for many ideas, which are already in the public domain.

For example, we can no longer patent the idea of killing huge numbers of birds by erecting vast quantities of whirling mobile metal machines on migration routes, on the pretext of generating electricity. The thing I truly wish we could have patented is the idea of allying with industry to develop wild areas while erecting these things, but calling the result conservation. Where was the USPO when we really needed it?

We can no longer patent the idea adjusting the surface station record to show unprecedented warming regardless of what it actually shows. We cannot either patent the famous method of 'hiding the decline' in the proxy record.

We cannot patent the idea of pretending that the Arctic is going to melt and flood the planet sometime very soon, thus raising vast amounts of grant money to do studies to find out how soon.

More serious, we cannot patent the idea of seeding the upper atmosphere with sulfur, and so there is every chance that fools will actually try and produce a new ice age by doing it.

So its a great idea and a most innovative step by the USPO, but they should have done this 20 years ago, and we could have either become enormously rich, or spared the world the greatest mass hysteria since.... Well, certainly since the millenial frenzy around the year 1,000, but maybe one has to go back even earlier.

Still, look on the positive side. If we are a little creative and surreptitiously join some green circles, there is no shortage of truly insane public policy ideas being floated. The best may have gone, but never underestimate the capability of the environmental movement to come up with more. It may be too late for society, but there is probably still time to get very rich if we get busy patenting now.

I work in the climate science department of a well known university in E Anglia, UK, and am proud to be the owner of a 4 x 4 and also an excruciatingly bad programmer. No, sorry, I got that wrong, I have no car, walk to work, and only write in equisitely structured C++.

You can all assuage your guilt from driving those 4 x 4s and writing all that crap code in Python. Ruby or whatever by sending me large sums of money, and I will continue my low guilt lifestyle as long as the cheques keep coming.

You can carry on shopping at malls in your 4 x 4s, and writing your terrible code.

We will all be happy. I will get rich. Everyone wins. We save the planet. What's the problem?

This is a futile effort to compete with London, and indeed with the UK, on this dimension. You may not have realized that one of Britain's great tourist attractions is that it is the most surveilled country, and London the most surveilled city, in the world. People come from all over to experience the magic of proper surveillance, at which we lead the world. There are so many cameras in London that if they were all actively used, everyone in London would be watching them all the time, and there would be no-one to surveil, because we would all be surveilling.

This is why, if you watched the recent Queen's speech, you would have been informed of a remarkable innovation in political thought. The UK government has committed to reducing the deficit by half, and it has also committed to reducing the number of people surveilling to no more than half of the population at any one time.

In order to do this, it is proposing to install people counters in all places where the cameras are watchable, and if the number goes over 50%, as it often does nowadays, then the last entrants will have their names taken and incur heavy fines.

To avoid the problem of people giving false names, face recognition software will be installed at all cameras watching surveillance centers.

This is widely reported as a brilliant initiative, which will cement the British surveillance lead, and ensure that for the next few years, tourists will continue to flock to London, confident that they are seeing a world leader in surveillance.

And causing civil servants in other countries to go home in a state of furious rage, muttering to themselves and their bemused wives about the growing surveillance gap, and worse still, the widening meta surveillance chasm.

Remember, this debate is at bottom about a specific clause. It is about the ability of software suppliers to tell you where you buy the hardware you run their software on.

It is absurd to argue that only by giving software suppliers the right to restrict what brand of hardware is used with their software, can the GPL remain enforceable. It makes as much sense as arguing that only by allowing people to give urine can we allow them to give blood.

Yes, its that Pythonesque.

I don't think that any software supplier should be obliged to sell anything in particular. If Apple wants not to sell their software at retail all over the place, they should not have to. It would also be quite wrong for any software company to be obliged to make their software work on particular hardware they didn't care for. No reason at all why they should do that. Finally, if people want to sell bundled combinations of hardware and software, and perhaps only bundled combinations, that is quite unobjectionable also.

But that is not the public policy issue here. The issue here is whether, solely by contractual conditions on sale and or installation, the software maker should be able to tell you what you may and may not install it on and use it with. From this point of view what counts is not that Apple sells hardware and software. What counts is that it chooses to sell the software independently of the hardware, but then wants to restrict what hardware this software is used with.

This is what is so unfortunate about it from a public policy point of view. The only way to make this happen is by giving all software makers who sell unbundled the exact same right to specify what hardware may or may not be used by buyers. This is why you cannot do it without also giving MS the right to stipulate that Windows may be installed as a dual boot system on any but Apple labeled hardware.

Such a world restricts intellectual freedom for the convenience of corporations and is open to enormous anti competitive abuse. We should not want such a world, even if it happens to be convenient, though not essential, for Apple's business model.

The public policy issue here is control. Back off the details of the case, generalize it, and you see there are two possibilities about the relationship between software and the brand of hardware it runs on.

CASE 1 is that the maker of the software has the legal right to specify what brand of hardware the stuff can be used on.

CASE 2 is that he cannot specify the brand of hardware the stuff can be used on.

The public policy issue is not about EULAs in general, or even copyright, or Apple, or about any technicalities of the software itself. It is about what rights you want to have software makers to have in respect of the brand of hardware.

Let us give two very specific examples. If the GPL were to be revised to say that it was permitted to install GPL software on any machine as long as it was not Apple labeled, would you approve of this? If MS were to forbid the installation of Windows in dual boot mode on Apple labeled equipment, but permit dual booting on all other brands, would you approve of them having such a power?

Would you, for instance, approve of proceedings by MS subsequent to making such a modification to the Windows license, if it took Apple to court over bootcamp, for contributory copyright infringement?

Welcome to the world of control. This is what the key issue is. To me, intellectual freedom requires us a society to opt for CASE 2. As for PJ, she may be right or wrong about the future of Psystar, but when it comes to public policy on this issue, she is in denial.

There are two cases which are in contradiction on this point. The first is Vernor vs Autocad, where it was found that when you buy a copy, you do indeed become the owner of that copy in one particular sense which is relevant to installation of software (as we will see). The second case is Blizzard, in which it was found that you do not become the owner, but are only the licensee.

The sense in which you need to be the owner is that of Title 17 Section 117. If you are the owner of a copy in that sense, you have the right to both make any copies or adaptations essential to use with a machine, and you also have the right to authorize others to do so on your behalf.

Blizzard found you are not the owner in the sense needed. The implication of only being the licensee and not the owner of a copy is that you need some permissions from the owner of the copyright in order to make copies. It is established law since MAI that when you load software into memory to run it, you are making a copy in the sense of the copyright legislation in the US. It is not so clear that this is true outside the US, which may be important.

The permissions given, according to Blizzard, are given in the EULA. It follows that if you break any EULA condition, you no longer have permission to copy, and so your next use will be in breach of copyright. This means that not only can you be sued by the owner for breaching a civil contract with him, which you entered into by click through. You will also be subject to prosecution for breach of copyright, which is a criminal offense.

It follows that breaching the EULA in ways that have intrinsically nothing to do with the act of copying can put you in breach of copyright on subsequent use. If, for instance, the EULA says that you will not have on your computer or install in future a copy of mdbtools, whether this is relevant or not to the software you are installing, then if you do, you are in breach of copyright. Even if what you are installing is, for instance, a calculator program, or a photo retouching program, or a calendar, or whatever. If the EULA says you must wear a particular uniform, be under a certain weight, be of a certain racial or ethnic or religious background, if you use it not being or doing any of those things, you are in breach of copyright.

The Church of Scientology could, for instance, put in the EULA that a freely sold copy of some software was only to be used by paid up members of the Church, and anyone who used it would then be in breach of copyright if he either was not in or had left the Church. Political parties could license material to their members and ban non-members from looking at it.

Remember, its the act of loading into memory that is copyright violating now, not the act of duplicating and distributing. This is because if Blizzard is the binding precedent, the protections of Section 117 no longer cover use of software by a buyer of a retail copy.

On the other hand, if Vernor is the binding precedent, then you are the owner of the copy, and you have S117 rights to do any adaptations necessary to use it with a machine. A machine. This means, if its OSX, the machine of your choice. You will then be liable to civil suit by Apple for breaching the civil contract of the EULA, but you won't be in breach of copyright.

All this has no bearing whatever on the GPL. If you are the owner of a copy, you are only exempted in terms of further copying and the making of derivative works by S117, and this only covers use with a machine. A machine. Not machines, not any particular brand of machine. A machine. One machine of your choice, unless given other permissions.

This means that the GPL sections which regulate what you may do in the way of copying, making new works with, distributing, remain valid. It is a common piece of Cupertino FUD to say that if Apple loses the power to restrain installation of OSX to its own brand of computers, the GPL falls. It does not.

All that falls is Apple's power to sell at retail software than can technically be installed on any brand of machine, while restraining it by either copyright or by EULA clause or both to being installed on its own branded machines. That will not affect the GPL. MS OEM contracts, upgrades, or indeed any other company in the industry besides Apple, because no other company engages in such a bizarre practice.

Its excellent value, its a national treasure, everyone loves it.

So why exactly does it have to be legally compulsory to subscribe to it if you want to watch any TV? Why is this the only subscription TV that you are obliged to subscribe to? Why, if we really want to make it compulsory to subscribe to some TV, do we not allow you to pick the provider of your choice? Why is it, that if you want to watch the English cricket team go down in flames yet again, you are obliged to subscribe to two TV broadcast services, only one of which provides cricket coverage?

Try writing to the BBC and telling them you want to cancel your subscription, because you are choosing to use Sky as your premium content provider. Or you don't want any premium content, you are happy to watch only ad funded TV.

Why exactly can you not do that? What would be so terrible if you could?

The BBC desperately needs competition, and that does not mean other channels that you are permitted to subscribe to in addition to it. It needs competition on equal terms, where you can subscribe to the BBC, to Channel X, neither, or both, as the mood strikes you.

I help, but I restrict carefully what I do. Like, if asked to clean infected Windows machines, I refuse. I will reinstall Windows from scratch for them, or I'll install Linux for them and set it up and show them how to use it, but will not engage in trying to clean Windows machines.

And so far, few problems. Several machines with either Debian or Mandriva installed, and only one bad experience, where a motherboard capacitor slowly died over a couple of months leading to intermittent crashes, and I was not quick enough to realize it must be hardware. Well, intermittent problems are the worst ones always. This person bought a new machine, which of course came with Vista.

If putting in Linux, which is often, but not always, the right thing, you have to be very careful to manage expectations and to choose your clients, but mine are OK, and certainly generate no more calls than they would on Windows or MacOS. The best are ones with restricted applications, no gaming, and of mature years, so they can be expected to stay with the same sort of usage once it all works.

I do get calls every so often. Mostly about uses of spreadsheet or word processor, and every once in a while about file format conversions. I have had to run a few files through awk or perl to get the text out. Its not unreasonable, these are things that ordinary users cannot be expected to do for themselves. So far nothing that has been either terribly difficult or time consuming.

But you do have to set the ground rules. And when you put in something unfamiliar, you have to be prepared to support it, and they have to be prepared to learn, and that needs to be made clear up front.

The various cases on this matter make it clear that there are three different things involved. Read Title 17 Section 117.

You may own copyright in the software. This gives you the right to control what copies are made, with one exception. This, copyright ownership, is what people usually mean when they talk about owning the software. It is the same as a publisher owning copyright to a book. He may print and sell as many copies as he wants.

You may also own a retail copy. This is what Vernor finds, and what Softman found before that. It has been repeatedly argued by software suppliers that you do not own the copy, that you only own a license to use. It has now been found for the second time that you own it, and the criterion used is whether the supplier has any right to repossess. If not, the copy is yours.

We next come to copies made in way of use. If the software is not supplied 'live', ie running off the installation media, it must be installed. Installation constitutes copying. It would be illegal under copyright law without some explicit permission. In fact the sort of copying which also occurs during use when the software is read into memory was found illegal in the well known MAI case, until Title 17 S 1117 was revised as a result of this case.

The revisions provided that copies and modifications made or authorized by the owner which were essential to use with a machine (notice the article, "a" machine) are permitted. But 117 also provides that if you resell the copy you own, you may only sell with it the copies you have made in way of being essential to use, with the consent of the copyright holder.

So, to summarize the situation, when you buy a retail copy of software, you own that copy. You do not become the copyright holder, your right to make copies is limited by Title 17. You may make copies (or modifications) that are essential to use with "a" machine - by implication, the machine of your choice, not of the copyright holder's choice. But your rights over resale of those copies is limited.

Two things are sometimes argued about this.

(1) It is sometimes argued that you may only use a machine which is essential. For instance, you may not install OSX on a Dell, because a Dell is not an essential machine, you could equally well use a Mac. Wrong. The machine does not have to be essential, and the article is indefinite, "a" machine. What has to be essential is the copying.

(2) It is also sometimes argued that because you have no rights of resale of the copies made in way of being essential to use, the copyright holder owns them, and you do not. There is no ground for this view. The test of repossession does not suggest this. The copyright holder has no right of repossession of those copies, and you have a right to them in perpetuity with no further payments. The situation is, you own them but you have restricted rights of resale.

So where does this leave Psystar and OSX? In a very simple situation. If they installed without having transferred the ownership of the retail copy of OSX to the customer, they were in violation of copyright. If they were made when ownership of machine and copy had been transferred, they were permitted by 117 as having been authorized by the owner, and were not then resold, so no permission for transfer was required, as they were never transferred.

This means that there need not have been any violation of copyright, but there was of course a breach of the Apple EULA. Whether the term of that EULA which obliges you to buy your hardware from Apple is enforceable is a quite different matter. But as far as copyright goes, you are the owner of any retail copy of OSX, or MS Office, that you have lawfully acquired. There is nothing in copyright law to stop you installing it wherever you want, as long as you do not make more than one copy. It says "a" machine, remember.

I've put in three kyoceras, one printer, one MFP, one copier with an add-on print card. Mechanically perfect, very cheap to run because the drum lasts forever. More expensive, yes, but worth it. They have all been in without problem for several years, fairly heavy usage, and I expect them to continue indefinitely.

Imperial gallons vs US gallons. They are different.

The question is, over what time period are we seeing rises and falls in coverage? We have no proper data before the satellite age. So all we know is that there has been recent shrinkage. We have however no idea what the standard deviation is of gains and shrinkages over a period of centuries or millenia, so we have no idea whether we are looking at an event close to the mean or one that is several standard deviations away from it.

At this point people usually ridicule one for not being prepared to take action until there is proof, which is usually projected as being some natural disaster like New Orleans.

The argument is mistaken. It is quite reasonable to wait for proof, because 'doing things' in the absence of proof is a risky and expensive business. It could have quite dramatic and unexpected side effects depending on what the situation really is.

It would enormously help us figure this thing out if all the climate scientists would just publish their raw data and algorithms. That way we could at least verify their work so far. The ones that need to publish? Well, just about all of them. They supposedly have evidence that the present warming is a very rare event, but they decline to publish it. They just publish studies based on it, summaries of it, processed forms of it. We need this data, and we need the code that was applied to it.

Without that, its not science, its arm waving. There is probably nothing more important than to establish the climatic history of the last 2,000 years, and if we could establish ice coverage and density in some way, that too. Without the scientists publishing, I do not see how we take this debate any further. It is, to say the least, curious that the main workers in the field, the ones who find the present trend most alarming, are the ones who refuse to reveal the data that would prove them right.

Where, for instance, is Mann's algorithm, the one he refused to supply to the Wegman Committee? Where is the data underlying the HADCRU series? Where is Thompson's ice core data?

If we cannot see it, how do we even know it exists?