You open document A. Your editor / wife also opens document A. You both make edits to the document. You can both see what changes are made, either almost as the other person makes them or shortly thereafter.
WOW That'S approach may not REALLY work so well GREAT! in reality.
Seriously, there is a reason databases have transactions, and that formal documents have revisions and change histories. It's hard to see how a word processor document could reliably be fully edited by two parties at the same time, unless they really were editing guaranteed unique areas of the document. I'm not aware of any word processor that actually supports this concept yet: consider that even if you're editing in different chapters, say, there are still issues of the table of contents expanding and changing your pagination, or adjusting document-wide formatting styles.
Now, if this tool really does support, for example, the concept of having separate text streams running through a document, which can be independently edited, versioned, and so on, then that might be worth something. I can see how that could be combined with a more-DTP-than-WP approach to putting the document together and potentially give good results. But you'd need some pretty powerful formatting and control commands to deal with separate streams that way, well beyond anything any WP today has. Even more specialised tools like web content or LaTeX can't do this particularly well yet, and they reflow the words all the time and therefore need mark-up to specify things like the relative position of pictures.
You make a good point about voting with a wallet. What a shame you dressed it up in so much hyperbole that an average punter would probably classify you as a whacko and ignore you.
How can it be an "inalienable right" if it's being taken away from you?
And in what way is this an assault on our "freedom"? Privacy, sure, but are they locking you up if you don't play enough or something?
IMHO, if you have a good point to make, you can generally make it much more effectively by writing about it in a calm and reasoned fashion.
Do these numbers mean anything? Yes. Due to the sheer volume of specifications out there, it is impossible to physically list every permutation that needs to be validated, but you CAN say what fraction of those permutations have been validated.
In other words, you're choosing an arbitrary selection, and your numbers are subjective. You have not provided an objective measure of compliance.
Moreover, in your next post, you claim that it would be possible to construct all the relevant test cases systematically, and then it would be OK only to examine the rendering of a subset, and to assume if that examination is successful that all other cases also render correctly. Again, this is subjective, and to check all the cases manually and get an objective measure would be impractical because of the scale of the problem.
It seems to me that this is exactly the point that Chris Wilson was making with his original comments, and that Bogtha was highlighting in the post to which you replied.
Er, all three of the things you have listed are tested for in the Acid2 test. Have you actually read it, or did you decide it was about "bragging rights" based on hearsay? It seems specifically geared towards testing exactly what you are concerned with.
I'm not in the habit of criticising things without doing any homework. I read Acid2 the first time a long time ago, and my opinion has not changed significantly since that time. If anything, it's worsened: I'm really quite bored now of scanning past all the "But does it pass Acid2? Cos if not it is da sux0r!" comments in every webby Slashdot discussion to get to informative comments and interesting discussion.
As for the things I'm interested in, yes, Acid2 does test aspects of many of them. I am not saying, and never have said, that the tests in Acid2 aren't worthwhile or that they don't indicate the presence of useful in a browser that passes. I'm simply contending that there are greater priorities.
For example, if memory serves, Acid2 tests whether transparent borders work properly. Now, perhaps I'm just weird, but if I want space around a block I set the margin and/or padding as appropriate, and if I want a visible box round it I set the border. I really don't care whether a browser supports transparent borders, because I've never used one, and I very much doubt whether any web site I've ever visited has either.
The border-radius stuff, on the other hand, is widely used and would remove the need for all kinds of hackery that plagues many popular web sites today, and even my own. It would be rather unfair to criticise Acid2 for not covering this given the status of the W3C's CSS3 work, but nevertheless I'd far rather all concerned worked on this than transparent borders.
Forwards-compatibility with future CSS specifications is a major practical reason why error handling needs to be defined and tested. Would you really want older browsers blowing up whenever you used something in a CSS specification published after they were developed?
The thing is, while that's great in theory, I think that web developers will have to test against all mainstream browsers (including older versions that retain significant user bases) for a very long time, quite possibly forever. Even those browsers that seek to faithfully follow the W3C specifications and be future-proof are bound to have bugs, as many a CSS workaround has demonstrated.
If we were serious about future-proofing, we would have a clear way to identify the browser in use and any broad claims it makes about supporting different versions of different specs, and a mechanism to specify full and degraded content according to these claims. That would be a much easier goal, and without it I think the "future-proofing" ideal is an impractical target anyway. (As an aside, even Microsoft, for all their independence, have done a better job of this than the W3C.)
In any case, I would far rather the browser vendors concentrated on adding support for the more useful features than messing around with technicalities that are unlikely to be relevant to either web developers or users any time soon. After all, if no-one ever gets round to supporting the features I mentioned, I won't need to care about degrading my web site gracefully in older browsers, because I won't be able to use the features anyway!
I don't find the result surprising at all. In the real world, there are more people who will pay up if it's directly in their interests than will pay up simply out of respect/gratitude/charity/whatever, not least because one set is likely to be almost entirely contained within the other.
I'm not saying there's no reason for those aspects of the test for anyone. I am saying that they are basically irrelevant to me as a practising web developer who validates the XHTML and CSS of all my pages using W3C tools anyway. Thus (as a web developer) I care very little about whether any given browser passes Acid2, because the non-compliance aspects will have no significance for the pages I produce.
Now, as a user, I might care more if it turned out that a browser was doing something particularly daft with bad input. But since I don't often find myself running into the sorts of things covered here with any browser today, W3C-friendly or otherwise, I just don't see this as a big problem. When there are still pages that just don't render at all sensibly on popular browsers like Firefox and Opera, and minor layout niggles on numerous sites as a result of things like box model gremlins, I'd far rather the IE team spent their time working on fixing those than some hypothetical problems that I've rarely, if ever, encountered.
The Acid2 stuff is like the browser developer's version of mine's-bigger-than-yours-is. It's about bragging rights, and that's it.
Sure, it's a test of strict compliance with certain aspects of the W3C CSS specs. Speaking as a guy responsible for a web site, though, I care far more about whether IE7 supports everyday, often-useful aspects of W3C specs. Here are some examples that I do care about, all of which have directly affected my work on the site in recent weeks:
Are the various box model gremlins fixed, so I don't have to keep hacking all my dimensions and including extra divs?
Can I use table-* CSS properties reliably, and get rid of the legacy table layouts that are stuffing my semantic mark-up?
Can I use transparency in PNGs, so I don't have to keep recreating bullet point graphics for all my different background shades?
In terms of new features, I'd love for IE to support at least basic SVG, so auto-generated graphics could be available for the majority of my user base. I'd love for someone to drive through the proposed CSS3 border-radius property and friends, so we could drop all the image-based hacks once and for all. Again, these are practical considerations that would directly affect my ability to display visually attractive and informative content for my users.
On the other hand, do you know how many of the Acid2 non-compliance things are relevant to me? None, just like any other web developer who actually writes pages that follow W3C specs.
I was actually quite careful to demonstrate a sense of proportion.
That might have been your intent, but it's certainly not how you came across.
We are talking about a legal mechanism that determines whether human software engineers are free, or not free, to develop software.
No, we're really not. We're talking about a legal mechanism that (in theory, if it were introduced) would temporarily prevent software engineers from developing software to do exactly the same thing as someone else's, on the back of that other party's work. There is no prohibition proposed on developing any other kind of software, nor would the prohibition be permanent. To ask a blunt question in the manner of your own posts, if you can't develop software that's worth something to someone without exploiting others to do your R&D, why do you deserve any reward at all?
Which world do you prefer?
A world in which some people may control what algorithms other people are or are not permitted to utilise in their software (even if they typically independently reinvent them), or a world in which people are free to develop software without any need to obtain anyone else's permission?
That's a loaded question, because if something is so trivial that someone else could readily invent it independently, it shouldn't qualify for a patent anyway.
On the other hand, I spend my working days researching state-of-the-art mathematical algorithms that are then provided to others in the form of software libraries in exchange for money. It can take several smart and well-trained people several weeks to come up with the algorithm behind a single file of finished code. I'm pretty sure that in the kind of niche market I work in, no one company would invest the kind of money required to develop this sort of technology if they knew their competitors would immediately be free to nick the ideas and remove any competitive advantage gained from the investment, yet in the current economic framework, demand for our products is strong. You don't seem to be able to understand why this might matter to the pace of innovation.
Having said all of that, I don't personally support software patents in their usual form, for reasons of practicality that I've outlined elsewhere. In most cases, a combination of copyright and trade secrets laws protect the kind of work we do anyway. But your basic arguments about "slavery" could be applied almost unaltered to these as well.
Freedom has risks. If you have free elections, the "wrong" guys might win. If you have secure communications, "terrorists" might use them to make plans. If you have the right to keep and bear arms, "bad guys" may have guns.
Freedom does have risks, but the danger with issues like this is that for society to function effectively, with freedom must come responsibility, and vice versa. That means if the "wrong guys" win an election, they can be removed from office when they go too far. If "bad guys" walk down the street with a gun and shoot someone, then someone else is likely to shoot them.
While completely secure communication is no doubt a worthy goal in some respects, the danger is that it will be very difficult for society to hold those who abuse the system accountable for their actions if the system also permits anonymous transmission. Contrary to the claim in TFA, there really aren't that many places where anonymity is an effective requirement, and there are certainly plenty of damaging ways to abuse anonymity: think how much nicer the world would be without spam, viruses, phishing/on-line fraud, web sites offering bad advice on critical subjects like finances and medical treatment, and so on.
I have long been in two minds on this subject. On the one hand, I am not generally a supporter of complete anonymity for publicly available material, because for the reasons above I question whether the gains really outweigh the costs in practice. On the other hand, I am a strong believer in the idea that any technology is neutral and what matters is how it's used by those who have access to it. On this basis, we might suppose that even if we were to deny secure, anonymous electronic communication to "bad guys", those people will simply find other, less technologically "clever" ways of communicating privately. The problem isn't the secure communication; it's that bad people want to use it to discuss doing bad things.
I find my views on anonymity softening, however, in light of one inescapable fact: the one guaranteed constant in the debate is that different people will have different views, and not everyone will respect the same authorities to judge what is and is not responsible use of any freedom to transmit anonymously. In the absence of any absolute standard of authority, it is impossible to define what is and is not responsible in a universally applicable way. Thus we come back to freedom of expression and the threat to it posed by denying anonymity.
Perhaps a better way to deal with the problems caused by anonymity is for society to learn not to trust anonymous sources. If no-one ever gave up their details in phishing attacks, then no-one would bother attempting the attacks. Ditto buying things advertised through spam, etc. And if no-one trusted anonymous sources for important advice, then the damaging web sites would not arise. I suspect that there will always be problems with more personal issues like defamation, because it seems to be human nature to assume that there is no smoke without fire, but at least if we collectively grew up enough not to trust unsubstantiated assertions the problem would be diminished.
So I think my views are swinging towards the long-term benefits of absolute anonymity over the short-term benefits of disallowing it. Of course, this may all be a moot point anyway. If the people want anonymous communication, then they're going to get it sooner or later, whether any particular government likes it or not.
Software patents are manacles imposed on software engineers.
Whilst it's nowhere near as severe, there is a similar principle at stake here to slavery. If you don't believe in slavery (removing the freedom from coders everywhere to reinvent wheels and utilise them) then you really shouldn't tolerate it, and that includes tolerating your employer doing it.
The thing is, I don't personally have a problem with restricting the freedom of coders to rip off the results of others' hard work before those others have a chance to gain some reasonable return on their investment. That's what software patents (or any patents) are supposed to do.
What I have a problem with is a patent system that:
allows someone to apply for a large number of speculative patents for trivial ideas that aren't really representative of any genuine hard work at all
tends to grant a significant proportion of these inappropriate patents
allows a business that has attained such patents to use them as a weapon to prevent other businesses developing, rather than as a defensive measure to protect an investment while they're taking advantage of the patented ideas.
None of these things is necessary to achieve the objective that patents are supposed to support, and it is problems like these that lead to abusive companies that just hoard IP and sue others, and more generally to cross-licensing among the big guys to keep the little guys out of the market.
I oppose software patents because, in practice, these things seem to be inevitable under any scheme similar to what we see around the world today. I don't oppose them because I object to the principle of rewarding worthwhile investment.
I can't remember whether the paper specifically discusses the failure of floating point arithmetic to obey the mathematical laws of arithmetic, but even if not, the background it provides is probably enough for you to understand the reasoning yourself.
All of which makes you wonder why wiretap evidence is still inadmissible in UK courts, which (if recent reports are anything to go by) is a large part of why we now have all these dubious restraint-without-charge laws. If the authorities know someone's a bad guy, from legitimate intelligence, why the hell can't that person be hauled up before a court, tried on the basis of that evidence, and sentenced like any other bad guy if convicted? Surely this is a better scheme than the current "we don't need no stinkin' courts" approach, which pretty much puts executive authority in the hands of <shudder> the Home Secretary?
The LRRB stuff is odd; the reports I read a little before the date of your blog post implied that the whole "ministers could do whatever they want with a rubber stamp" section had been chucked out. Perhaps those reports were misinformed.:-(
Re the ID card bill, Cameron's speech is all very nice and waffly as you say, but I believe David Davis actually gave a public, unequivocal statement on the subject using words like "repeal".
As for Civil Contingencies, I wasn't aware that the powers in it were as broad as you describe; I'd only got as far as restricting freedom of movement, freedom of association, etc. In any case, I think we can safely assume that if the government ever did actually try to do the things you describe, it wouldn't be the government for very much longer. There are still around 60 million of us, and so few of them as to be insignificant, and I rather doubt either the police or the armed services would support such a blatant power grab. We're well down the slope, but not that far down, and one of the few remaining differences between the rise of Tony Blair and the rise of various historical dictators is that our society still has a few lines he can't credibly cross without risking being forcibly removed.
But, from the ISP's perspective the problem is that BT charge them if there is no problem.
I appreciate that; indeed, PlusNet's staff tried that one on me many times. Of course, what they actually mean is BT charge them if they can't find a fault, which isn't the same thing as no fault existing.
The thing is, that's not my problem, it's PlusNet's. In this case, the only equipment in use was supplied by PlusNet and connected according to their instructions (something I was happy to confirm when asked). So either their service is broken or the equipment they supplied me is, but either way it's their responsibility. The fact that they have contractual arrangements with BT is a business arrangement on their side, and nothing to do with me. There was no way I could possibly be responsible for the situation unless I had failed to install the equipment correctly, and I had confirmed to them that I had gone down the requested checklist and all was well.
So, while I understand their wish to pass on such charges, I'm afraid incurring them is a hazard of operating in their business, and they're not going to have any luck trying to pass that buck onto me.
Even now, people don't upgrade their PC every two years like in the mid-90s. People now wait for 4 or 5 years, some even more.
I don't think home users ever really upgraded that often, other than the geeks.
What's changed, IME, is that businesses are breaking the three-year upgrade cycle that used to be accepted without question, and instead asking what measurable benefits will come from splashing out another few thousand on faster desktops for Susie Secretary and Ollie Officeworker. That, combined with major software firms having much longer release cycles for big name products and then not offering much real improvement anyway when they finally turn up, means the bean counters are finally starting to notice that a lot of computer hardware and software is vastly overpriced for the benefits it brings, and they're challenging the expense.
I'm guessing ClearType isn't on by default because by its nature, it would make the display worse if improperly configured.
For example, I too normally use it even on my CRT at work, because it doesn't have an artificial lower threshold below which it won't smooth fonts. The standard anti-aliasing cuts out just around the point where most of my fonts are normally configured, making it pretty much useless. With ClearType, most of the fonts I use regularly do look smoother.
OTOH, a couple of the fonts I use regularly do get mangled by ClearType (I'm surprised it's so few...) and when working with those, or with much smaller text in a spreadsheet or similar, I do switch ClearType off to avoid the artifacts.
I guess it should be possible to enable ClearType automatically for suitable screens, as long as the monitor driver installed tells Windows how the pixels are physically arranged so it can perform the anti-aliasing to match.
FWIW, the most offensive of the LRRB provisions (such as reducing the Commons to a rubber stamp) were quietly dropped a few weeks ago, with little more than an acknowledgement on a government web site. Strangely, this happened on a day with a big news story somewhere else.
The Tories have promised to outright repeal the ID card legislation if voted in at the next general election. I never figured I'd vote Conservative in my lifetime, but if they're prepared to commit to that and Labour have committed to the opposite... Fortunately for all of us, it's completely implausible that any viable, operational implementation will exist before the next election.
As for the Civil Contingencies stuff, it's pretty far-reaching, but also liable to be struck down -- on human rights grounds if nothing else -- the first time anyone tries to abuse it. I'm less worried about theoretical government powers that can't practically be abused than I am about the more invidious things like increasing summary "justice" (ASBOs, more on-the-spot fines) and mass surveillance without good cause (which will inevitably create a huge database, with horrendous consequences when it is inevitably abused).
I agree with much of what you say, but being able to use much better screens, such as the large Dell TFT mentioned in the submission, is definitely a plus as far as I'm concerned.
This is partly because of the productivity benefits of getting more on-screen: try editing a book or magazine in a DTP program when you can actually see two real-size pages side-by-side at a useful resolution, and suddenly the idea of zooming in to part of one page on a 19" CRT to make out the details seems quaint and old-fashioned.
Equally important, newer TFTs tend to be a lot easier on the eyes than the older TFTs and mid-range CRTs that fill offices around the world. As someone whose eyes are degrading because of unfortunate genetics anyway, I want to make sure I do as much as possible to help them, and since I sit in front of a computer screen for a large proportion of my waking hours...
I think the problem alluded to in the submission, where large, hi-res screens become effectively unusable under Windows due to poor UI scaling, is a very real one. So, if a new version of Windows will support proper scaling for things like fonts, icons and UI widgets, and thus make bigger and better screens more usable, that is a clear benefit for me.
The first link when Googling "smoking real cost NHS" is this BBC News article. The formal research it cites is actually Dutch, but draws substantially the same conclusions I've heard in debates in the UK.
Look, I'm not saying that a programmer ever codes strictly to a written specification. But that spec provides the basis for the work. Dealing with implementation issues has to come later.
I understand that, and I'm not disagreeing with you.
I just don't see how the current state of play with IE is any different. Most of the time, HTML and basic CSS stuff does work. There are some places where it doesn't, and things like the box model problem are well-known and well-documented in the web development community. Clearly it is possible to design and implement web sites effectively in this environment, because we have a few million of them to prove it.
This isn't to say life wouldn't be better all round if IE did follow the same W3C specifications as other browsers, of course, and I hope IE7 will catch up on some of the more recent CSS developments that have taken place since IE6 came out. But even if and when that happens, developers will still test their pages against mainstream browsers first, and worry about formal specs and such afterwards.
Unfortunately, there's a limited amount we can do, at least quickly. We do not live in a democracy, and recent events have made this abundantly clear.
However, if things carry on at the present rate, the building public backlash against draconian "anti-terror" legislation and the like will be a big enough issue at the next general election that any party wanting to win will have to come out clearly against it. (The Conservatives have already given a clear statement that they will repeal the ID card legislation if elected, in contrast to Labour's bullish rhetoric about forcing the cards through no matter what. The Lib Dems are also against the scheme.)
The more interesting question is whether electoral reforms will become a big issue as well, in light of the obvious failure of government to act as the people wanted over things like Iraq. (Hint: In a democracy, you don't get nearly 5% of your entire electorate making a journey to the capital on the same day, many from right across the country, and then have the Prime Minister ignore them and do his own thing anyway.) Particular issues might include the structure of the House of Lords, which has been changed significantly under New Labour but with little public involvement, and the West Lothian question, and of course the perennial "should we have PR, and if so in what form?" debate.
Except, of course, that we didn't elect him to this kind of power. At the last general election, Labour didn't even win the popular vote in England, and they have passed several controversial proposals through Parliament only with the help of their Scottish colleagues (whose constituents will never be affected by a lot of those proposals, because the areas concerned are delegated to the Scottish Parliament).
Moreover, even though Labour did achieve the highest number of votes of any single party, they still received the support of only 22% of the electorate, which was around 1/3 of the actual votes cast. The fact that our first-past-the-post electoral system gave Labour an outright majority for three elections is a damning indictment of the undemocratic system we use. However, the power that Tony Blair has had as leader of an absolute majority in the commons has not been a reflection of the level of popular support Labour enjoyed (or rather, didn't enjoy) at the preceding elections, and it is abundantly clear that many Labour policies do not have popular support (c.f. war in Iraq, Israel/Lebanon, and a rapidly increasing number of issues at home).
Slashdot Mirror
WOW That'S approach may not REALLY work so well GREAT! in reality.
Seriously, there is a reason databases have transactions, and that formal documents have revisions and change histories. It's hard to see how a word processor document could reliably be fully edited by two parties at the same time, unless they really were editing guaranteed unique areas of the document. I'm not aware of any word processor that actually supports this concept yet: consider that even if you're editing in different chapters, say, there are still issues of the table of contents expanding and changing your pagination, or adjusting document-wide formatting styles.
Now, if this tool really does support, for example, the concept of having separate text streams running through a document, which can be independently edited, versioned, and so on, then that might be worth something. I can see how that could be combined with a more-DTP-than-WP approach to putting the document together and potentially give good results. But you'd need some pretty powerful formatting and control commands to deal with separate streams that way, well beyond anything any WP today has. Even more specialised tools like web content or LaTeX can't do this particularly well yet, and they reflow the words all the time and therefore need mark-up to specify things like the relative position of pictures.
You make a good point about voting with a wallet. What a shame you dressed it up in so much hyperbole that an average punter would probably classify you as a whacko and ignore you.
How can it be an "inalienable right" if it's being taken away from you?
And in what way is this an assault on our "freedom"? Privacy, sure, but are they locking you up if you don't play enough or something?
IMHO, if you have a good point to make, you can generally make it much more effectively by writing about it in a calm and reasoned fashion.
In other words, you're choosing an arbitrary selection, and your numbers are subjective. You have not provided an objective measure of compliance.
Moreover, in your next post, you claim that it would be possible to construct all the relevant test cases systematically, and then it would be OK only to examine the rendering of a subset, and to assume if that examination is successful that all other cases also render correctly. Again, this is subjective, and to check all the cases manually and get an objective measure would be impractical because of the scale of the problem.
It seems to me that this is exactly the point that Chris Wilson was making with his original comments, and that Bogtha was highlighting in the post to which you replied.
I'm not in the habit of criticising things without doing any homework. I read Acid2 the first time a long time ago, and my opinion has not changed significantly since that time. If anything, it's worsened: I'm really quite bored now of scanning past all the "But does it pass Acid2? Cos if not it is da sux0r!" comments in every webby Slashdot discussion to get to informative comments and interesting discussion.
As for the things I'm interested in, yes, Acid2 does test aspects of many of them. I am not saying, and never have said, that the tests in Acid2 aren't worthwhile or that they don't indicate the presence of useful in a browser that passes. I'm simply contending that there are greater priorities.
For example, if memory serves, Acid2 tests whether transparent borders work properly. Now, perhaps I'm just weird, but if I want space around a block I set the margin and/or padding as appropriate, and if I want a visible box round it I set the border. I really don't care whether a browser supports transparent borders, because I've never used one, and I very much doubt whether any web site I've ever visited has either.
The border-radius stuff, on the other hand, is widely used and would remove the need for all kinds of hackery that plagues many popular web sites today, and even my own. It would be rather unfair to criticise Acid2 for not covering this given the status of the W3C's CSS3 work, but nevertheless I'd far rather all concerned worked on this than transparent borders.
The thing is, while that's great in theory, I think that web developers will have to test against all mainstream browsers (including older versions that retain significant user bases) for a very long time, quite possibly forever. Even those browsers that seek to faithfully follow the W3C specifications and be future-proof are bound to have bugs, as many a CSS workaround has demonstrated.
If we were serious about future-proofing, we would have a clear way to identify the browser in use and any broad claims it makes about supporting different versions of different specs, and a mechanism to specify full and degraded content according to these claims. That would be a much easier goal, and without it I think the "future-proofing" ideal is an impractical target anyway. (As an aside, even Microsoft, for all their independence, have done a better job of this than the W3C.)
In any case, I would far rather the browser vendors concentrated on adding support for the more useful features than messing around with technicalities that are unlikely to be relevant to either web developers or users any time soon. After all, if no-one ever gets round to supporting the features I mentioned, I won't need to care about degrading my web site gracefully in older browsers, because I won't be able to use the features anyway!
I don't find the result surprising at all. In the real world, there are more people who will pay up if it's directly in their interests than will pay up simply out of respect/gratitude/charity/whatever, not least because one set is likely to be almost entirely contained within the other.
I'm not saying there's no reason for those aspects of the test for anyone. I am saying that they are basically irrelevant to me as a practising web developer who validates the XHTML and CSS of all my pages using W3C tools anyway. Thus (as a web developer) I care very little about whether any given browser passes Acid2, because the non-compliance aspects will have no significance for the pages I produce.
Now, as a user, I might care more if it turned out that a browser was doing something particularly daft with bad input. But since I don't often find myself running into the sorts of things covered here with any browser today, W3C-friendly or otherwise, I just don't see this as a big problem. When there are still pages that just don't render at all sensibly on popular browsers like Firefox and Opera, and minor layout niggles on numerous sites as a result of things like box model gremlins, I'd far rather the IE team spent their time working on fixing those than some hypothetical problems that I've rarely, if ever, encountered.
The Acid2 stuff is like the browser developer's version of mine's-bigger-than-yours-is. It's about bragging rights, and that's it.
Sure, it's a test of strict compliance with certain aspects of the W3C CSS specs. Speaking as a guy responsible for a web site, though, I care far more about whether IE7 supports everyday, often-useful aspects of W3C specs. Here are some examples that I do care about, all of which have directly affected my work on the site in recent weeks:
In terms of new features, I'd love for IE to support at least basic SVG, so auto-generated graphics could be available for the majority of my user base. I'd love for someone to drive through the proposed CSS3 border-radius property and friends, so we could drop all the image-based hacks once and for all. Again, these are practical considerations that would directly affect my ability to display visually attractive and informative content for my users.
On the other hand, do you know how many of the Acid2 non-compliance things are relevant to me? None, just like any other web developer who actually writes pages that follow W3C specs.
That might have been your intent, but it's certainly not how you came across.
No, we're really not. We're talking about a legal mechanism that (in theory, if it were introduced) would temporarily prevent software engineers from developing software to do exactly the same thing as someone else's, on the back of that other party's work. There is no prohibition proposed on developing any other kind of software, nor would the prohibition be permanent. To ask a blunt question in the manner of your own posts, if you can't develop software that's worth something to someone without exploiting others to do your R&D, why do you deserve any reward at all?
That's a loaded question, because if something is so trivial that someone else could readily invent it independently, it shouldn't qualify for a patent anyway.
On the other hand, I spend my working days researching state-of-the-art mathematical algorithms that are then provided to others in the form of software libraries in exchange for money. It can take several smart and well-trained people several weeks to come up with the algorithm behind a single file of finished code. I'm pretty sure that in the kind of niche market I work in, no one company would invest the kind of money required to develop this sort of technology if they knew their competitors would immediately be free to nick the ideas and remove any competitive advantage gained from the investment, yet in the current economic framework, demand for our products is strong. You don't seem to be able to understand why this might matter to the pace of innovation.
Having said all of that, I don't personally support software patents in their usual form, for reasons of practicality that I've outlined elsewhere. In most cases, a combination of copyright and trade secrets laws protect the kind of work we do anyway. But your basic arguments about "slavery" could be applied almost unaltered to these as well.
Freedom does have risks, but the danger with issues like this is that for society to function effectively, with freedom must come responsibility, and vice versa. That means if the "wrong guys" win an election, they can be removed from office when they go too far. If "bad guys" walk down the street with a gun and shoot someone, then someone else is likely to shoot them.
While completely secure communication is no doubt a worthy goal in some respects, the danger is that it will be very difficult for society to hold those who abuse the system accountable for their actions if the system also permits anonymous transmission. Contrary to the claim in TFA, there really aren't that many places where anonymity is an effective requirement, and there are certainly plenty of damaging ways to abuse anonymity: think how much nicer the world would be without spam, viruses, phishing/on-line fraud, web sites offering bad advice on critical subjects like finances and medical treatment, and so on.
I have long been in two minds on this subject. On the one hand, I am not generally a supporter of complete anonymity for publicly available material, because for the reasons above I question whether the gains really outweigh the costs in practice. On the other hand, I am a strong believer in the idea that any technology is neutral and what matters is how it's used by those who have access to it. On this basis, we might suppose that even if we were to deny secure, anonymous electronic communication to "bad guys", those people will simply find other, less technologically "clever" ways of communicating privately. The problem isn't the secure communication; it's that bad people want to use it to discuss doing bad things.
I find my views on anonymity softening, however, in light of one inescapable fact: the one guaranteed constant in the debate is that different people will have different views, and not everyone will respect the same authorities to judge what is and is not responsible use of any freedom to transmit anonymously. In the absence of any absolute standard of authority, it is impossible to define what is and is not responsible in a universally applicable way. Thus we come back to freedom of expression and the threat to it posed by denying anonymity.
Perhaps a better way to deal with the problems caused by anonymity is for society to learn not to trust anonymous sources. If no-one ever gave up their details in phishing attacks, then no-one would bother attempting the attacks. Ditto buying things advertised through spam, etc. And if no-one trusted anonymous sources for important advice, then the damaging web sites would not arise. I suspect that there will always be problems with more personal issues like defamation, because it seems to be human nature to assume that there is no smoke without fire, but at least if we collectively grew up enough not to trust unsubstantiated assertions the problem would be diminished.
So I think my views are swinging towards the long-term benefits of absolute anonymity over the short-term benefits of disallowing it. Of course, this may all be a moot point anyway. If the people want anonymous communication, then they're going to get it sooner or later, whether any particular government likes it or not.
The thing is, I don't personally have a problem with restricting the freedom of coders to rip off the results of others' hard work before those others have a chance to gain some reasonable return on their investment. That's what software patents (or any patents) are supposed to do.
What I have a problem with is a patent system that:
None of these things is necessary to achieve the objective that patents are supposed to support, and it is problems like these that lead to abusive companies that just hoard IP and sue others, and more generally to cross-licensing among the big guys to keep the little guys out of the market.
I oppose software patents because, in practice, these things seem to be inevitable under any scheme similar to what we see around the world today. I don't oppose them because I object to the principle of rewarding worthwhile investment.
But only because the mice haven't yet revealed their true intelligence.
The first paper recommended for learning more about floating point arithmetic is usually Goldberg's famous What Every Computer Scientist Should Know About Floating-Point Arithmetic .
I can't remember whether the paper specifically discusses the failure of floating point arithmetic to obey the mathematical laws of arithmetic, but even if not, the background it provides is probably enough for you to understand the reasoning yourself.
All of which makes you wonder why wiretap evidence is still inadmissible in UK courts, which (if recent reports are anything to go by) is a large part of why we now have all these dubious restraint-without-charge laws. If the authorities know someone's a bad guy, from legitimate intelligence, why the hell can't that person be hauled up before a court, tried on the basis of that evidence, and sentenced like any other bad guy if convicted? Surely this is a better scheme than the current "we don't need no stinkin' courts" approach, which pretty much puts executive authority in the hands of <shudder> the Home Secretary?
Hey, you think you got it bad? All I wanted was a nice game of chess, and I got sent to this lousy movie review site...
The LRRB stuff is odd; the reports I read a little before the date of your blog post implied that the whole "ministers could do whatever they want with a rubber stamp" section had been chucked out. Perhaps those reports were misinformed. :-(
Re the ID card bill, Cameron's speech is all very nice and waffly as you say, but I believe David Davis actually gave a public, unequivocal statement on the subject using words like "repeal".
As for Civil Contingencies, I wasn't aware that the powers in it were as broad as you describe; I'd only got as far as restricting freedom of movement, freedom of association, etc. In any case, I think we can safely assume that if the government ever did actually try to do the things you describe, it wouldn't be the government for very much longer. There are still around 60 million of us, and so few of them as to be insignificant, and I rather doubt either the police or the armed services would support such a blatant power grab. We're well down the slope, but not that far down, and one of the few remaining differences between the rise of Tony Blair and the rise of various historical dictators is that our society still has a few lines he can't credibly cross without risking being forcibly removed.
I appreciate that; indeed, PlusNet's staff tried that one on me many times. Of course, what they actually mean is BT charge them if they can't find a fault, which isn't the same thing as no fault existing.
The thing is, that's not my problem, it's PlusNet's. In this case, the only equipment in use was supplied by PlusNet and connected according to their instructions (something I was happy to confirm when asked). So either their service is broken or the equipment they supplied me is, but either way it's their responsibility. The fact that they have contractual arrangements with BT is a business arrangement on their side, and nothing to do with me. There was no way I could possibly be responsible for the situation unless I had failed to install the equipment correctly, and I had confirmed to them that I had gone down the requested checklist and all was well.
So, while I understand their wish to pass on such charges, I'm afraid incurring them is a hazard of operating in their business, and they're not going to have any luck trying to pass that buck onto me.
I don't think home users ever really upgraded that often, other than the geeks.
What's changed, IME, is that businesses are breaking the three-year upgrade cycle that used to be accepted without question, and instead asking what measurable benefits will come from splashing out another few thousand on faster desktops for Susie Secretary and Ollie Officeworker. That, combined with major software firms having much longer release cycles for big name products and then not offering much real improvement anyway when they finally turn up, means the bean counters are finally starting to notice that a lot of computer hardware and software is vastly overpriced for the benefits it brings, and they're challenging the expense.
I'm guessing ClearType isn't on by default because by its nature, it would make the display worse if improperly configured.
For example, I too normally use it even on my CRT at work, because it doesn't have an artificial lower threshold below which it won't smooth fonts. The standard anti-aliasing cuts out just around the point where most of my fonts are normally configured, making it pretty much useless. With ClearType, most of the fonts I use regularly do look smoother.
OTOH, a couple of the fonts I use regularly do get mangled by ClearType (I'm surprised it's so few...) and when working with those, or with much smaller text in a spreadsheet or similar, I do switch ClearType off to avoid the artifacts.
I guess it should be possible to enable ClearType automatically for suitable screens, as long as the monitor driver installed tells Windows how the pixels are physically arranged so it can perform the anti-aliasing to match.
FWIW, the most offensive of the LRRB provisions (such as reducing the Commons to a rubber stamp) were quietly dropped a few weeks ago, with little more than an acknowledgement on a government web site. Strangely, this happened on a day with a big news story somewhere else.
The Tories have promised to outright repeal the ID card legislation if voted in at the next general election. I never figured I'd vote Conservative in my lifetime, but if they're prepared to commit to that and Labour have committed to the opposite... Fortunately for all of us, it's completely implausible that any viable, operational implementation will exist before the next election.
As for the Civil Contingencies stuff, it's pretty far-reaching, but also liable to be struck down -- on human rights grounds if nothing else -- the first time anyone tries to abuse it. I'm less worried about theoretical government powers that can't practically be abused than I am about the more invidious things like increasing summary "justice" (ASBOs, more on-the-spot fines) and mass surveillance without good cause (which will inevitably create a huge database, with horrendous consequences when it is inevitably abused).
I agree with much of what you say, but being able to use much better screens, such as the large Dell TFT mentioned in the submission, is definitely a plus as far as I'm concerned.
This is partly because of the productivity benefits of getting more on-screen: try editing a book or magazine in a DTP program when you can actually see two real-size pages side-by-side at a useful resolution, and suddenly the idea of zooming in to part of one page on a 19" CRT to make out the details seems quaint and old-fashioned.
Equally important, newer TFTs tend to be a lot easier on the eyes than the older TFTs and mid-range CRTs that fill offices around the world. As someone whose eyes are degrading because of unfortunate genetics anyway, I want to make sure I do as much as possible to help them, and since I sit in front of a computer screen for a large proportion of my waking hours...
I think the problem alluded to in the submission, where large, hi-res screens become effectively unusable under Windows due to poor UI scaling, is a very real one. So, if a new version of Windows will support proper scaling for things like fonts, icons and UI widgets, and thus make bigger and better screens more usable, that is a clear benefit for me.
The first link when Googling "smoking real cost NHS" is this BBC News article. The formal research it cites is actually Dutch, but draws substantially the same conclusions I've heard in debates in the UK.
I understand that, and I'm not disagreeing with you.
I just don't see how the current state of play with IE is any different. Most of the time, HTML and basic CSS stuff does work. There are some places where it doesn't, and things like the box model problem are well-known and well-documented in the web development community. Clearly it is possible to design and implement web sites effectively in this environment, because we have a few million of them to prove it.
This isn't to say life wouldn't be better all round if IE did follow the same W3C specifications as other browsers, of course, and I hope IE7 will catch up on some of the more recent CSS developments that have taken place since IE6 came out. But even if and when that happens, developers will still test their pages against mainstream browsers first, and worry about formal specs and such afterwards.
Unfortunately, there's a limited amount we can do, at least quickly. We do not live in a democracy, and recent events have made this abundantly clear.
However, if things carry on at the present rate, the building public backlash against draconian "anti-terror" legislation and the like will be a big enough issue at the next general election that any party wanting to win will have to come out clearly against it. (The Conservatives have already given a clear statement that they will repeal the ID card legislation if elected, in contrast to Labour's bullish rhetoric about forcing the cards through no matter what. The Lib Dems are also against the scheme.)
The more interesting question is whether electoral reforms will become a big issue as well, in light of the obvious failure of government to act as the people wanted over things like Iraq. (Hint: In a democracy, you don't get nearly 5% of your entire electorate making a journey to the capital on the same day, many from right across the country, and then have the Prime Minister ignore them and do his own thing anyway.) Particular issues might include the structure of the House of Lords, which has been changed significantly under New Labour but with little public involvement, and the West Lothian question, and of course the perennial "should we have PR, and if so in what form?" debate.
Everyone start practising your gun kata, for when the emotion-blockers start getting passed out...
Except, of course, that we didn't elect him to this kind of power. At the last general election, Labour didn't even win the popular vote in England, and they have passed several controversial proposals through Parliament only with the help of their Scottish colleagues (whose constituents will never be affected by a lot of those proposals, because the areas concerned are delegated to the Scottish Parliament).
Moreover, even though Labour did achieve the highest number of votes of any single party, they still received the support of only 22% of the electorate, which was around 1/3 of the actual votes cast. The fact that our first-past-the-post electoral system gave Labour an outright majority for three elections is a damning indictment of the undemocratic system we use. However, the power that Tony Blair has had as leader of an absolute majority in the commons has not been a reflection of the level of popular support Labour enjoyed (or rather, didn't enjoy) at the preceding elections, and it is abundantly clear that many Labour policies do not have popular support (c.f. war in Iraq, Israel/Lebanon, and a rapidly increasing number of issues at home).