I gotta say I like it - not so much because it's big and HD, but because it's unwieldy and thief-proof.
Don't count on it. A friend works in the big screen business, typically supplying kit for conference centres, office meeting rooms, public buildings, that sort of thing. A couple of years back, they installed a pretty huge screen, something like 15' IIRC. They finished work late one evening, and when they went in the following morning to set up some software to use it, someone had literally lifted the roof off the building and taken the screen! Fortunately for my friend, payment was due for the hardware on completion of installation, and the conference centre took over responsibility for security at that point...
So your argument is that although theoretically Vista has all these weaknesses, the guy who wrote this paper hadn't actually observed them yet at the time, therefore they aren't there?
Wow, that's a pretty far stretch, particularly considering the number of people who do have a release version of Vista now and are seeing exactly the sorts of problems he predicted with performance, interoperability, etc.
My documents will stop being read? Every major word processor in the software world can read Office 97 files just fine. And this isn't about operating systems anyway.
My software will slowly not be supported? IME, most software isn't supported usefully anyway. The only real exception is patches for security things in Windows, which will continue for several more years for WinXP yet.
I'll be criminally liable because my OS got infested with viruses and worms? Well, for one thing, there's a huge market in software to prevent that, which doesn't rely on Microsoft updates; in fact, it pretty much relies on Microsoft not to make good software for its existence. For another thing, no-one serious about security relies on software alone for safety; things like firewalls exist for a reason. And I'd love to know where you got that criminal liability thing from, because I'm looking forward to suing a quarter of the people on the Internet for disrupting my communications with their botnet-joining PCs. I don't suppose you've got a source for your rather headline-grabbing claim?
And as for hardware, most hardware runs much more reliably on an OS from a couple of years ago with mature drivers than it does on the latest and greatest. This has always been the case, and it often takes years for the software to catch up with a new OS. The only exceptions in recent years in Microsoft world have been graphics cards (where no-one smart buys the latest generation without a very good reason anyway, and historically the standard of accompanying software has been pathetic no matter how long you wait) and wireless (where Windows XP was genuinely a lot better than Windows 2000).
The only point you make with any validity at all is the OEM software licensing issue, which does cause a genuine legal problem if Microsoft artificially stop selling old versions of Windows. Of course, as I've argued elsewhere in this discussion, they might say they're going to do that, but in reality it is very, very unlikely that they are actually going to do it any time soon.
In conclusion, your post is pretty good with the sound-bites, but I don't think any of them stands up to closer inspection. Sorry, but I stand by my original claim: no-one is forced to upgrade to Vista.
Microsoft do write a lot of documentation, it's true. Unfortunately, they now write so much of it, and with such a focus on whatever their current trendy thing is (it's currently.Net), that actually finding the information you need is a very, very time-consuming activity, if you can even find all the details at all. I've been watching MSDN degrade from one of the best technical resources in the programming world to one of the worst, and it's been happening steadily for years. These days, if you try to look up a simple C++ detail or Win32 API call in Visual Studio, you are more likely to find details about numerous.Net things you don't care about at all, even with the filters set appropriately. I was trying to look up some details for the automation model for Word the other day, with future-proofing in mind, and it was taking literally minutes for the Microsoft servers to respond (or time out, in most cases). In this context, it's hard to view ISVs as "lazy". They just have better things to do than searching through all the mess Microsoft has made of its documentation, on the off chance that they missed a minor compatibility point that might come back and cause them some difficulty some way off in the future.
Oh, no they won't. Several large parts of the US Federal Government have already implemented blanket bans on the use of Vista. It is highly likely that several very large businesses will do similarly, along with other government departments in the US and elsewhere, for exactly the same reasons. There is no way Microsoft is going to turn down thousands and thousands of future XP sales to customers of that scale just because its new toy isn't selling well.
If you don't believe me, take a look at Microsoft's rhetoric around this time after XP was released, and then take a look at how many major customers were still able to get Win2K years later, and indeed how many still run it. They might say they're going to stop, but they're firing blanks and both they and their major customers know it.
Hint for future credibility on Slashdot: you will be more popular (and look less stupid) if you bother to read to the end of messages before posting sarcastic replies.
The purist in me would love to take the Linux route and force anybody doing weird stuff to fix their software, but in the long run, Microsoft is a business and their customers want compatibility with shitty software.
The thing is, the changes required in Linux world to be compatible with new versions of key libraries are generally minor, well-documented, reasonable, and relatively quick to implement.
In Windows world, that isn't the case, because Windows development is a mess. That is mostly Microsoft's fault. There isn't any good way out of this situation for them now, but they could have seen this coming a decade ago, and they chose to ignore it in the interests of making more profits earlier. Now, having made their bed, they are forced to lie in it.
The irony is that I'm not even sure why home users would move.
I've been following Vista developments for years, since back when there were going to be three big pillars underlying it. As far as I can see, from a technical perspective, the only remaining major functional improvement over XP is that Vista supports DirectX 10, and Microsoft are pretty much guaranteed to restrict that artificially to Vista-only.
Of course, going by the history, that won't even start to affect any games except Microsoft's own for at least a couple of years, since most games software isn't using everything DX9 offers yet. Similarly, DX10-supporting hardware won't be even close to mainstream for at least a year or two. Given that PC games now represent only a quarter or so of the market (the consoles are well and truly in charge today) and the majority of home users still aren't going to have Vista for a while, games companies may be hesitant to tread those waters even as they reach the point where the extra goodies in DX10 may be genuinely useful.
Apart from that, what possible reason is there for a home user to upgrade? There's been a lot of negative press for Vista, not just about DRM but also all the hardware and software compatibility problems. The UI is different, which for many users means "bad" by default, even if with time they might come to prefer it. If home users were really serious about security, the world wouldn't be full of botnets. And the list goes on...
I can understand businesses with professional IT people placing some value on improved security or networking features, so if and when the compatibility is sorted out and the trust issues with phoning home and being activated/disabled/whatever remotely are irrevocably fixed, businesses might move. But home users? Not for years, except for the people who just get it with new PCs. (And even the rate of buying those isn't what it used to be.)
Seriously, why would any organisation upgrade to Windows Vista if it wasn't pretty sure all of its key software would work? It's amazing how many people seem to think there's some sort of obligation on people to upgrade. In fact, if you look at recent history, the big corporations are usually the last people to move on major upgrades like XP->Vista, often taking several years to do it. This is why.
There's nothing out there that can match the usability of Exchange/Office.
Ironically, usability is the last thing Outlook is good at. Word and Excel are far more usable than things like OpenOffice, but Outlook's usability is atrocious compared to many other mail clients. What there isn't, yet, is anything that can (a) match the features of the Outlook+Exchange combination, and (b) reliably migrate all the existing data from the Microsoft products to the alternative.
but until stuff syncs with Outlook, it has no change of defeating it
In the corporate space, I suspect sync with Exchange is more important, FWIW.
It's interesting that the other day on Slashdot we were also discussing the new Thunderbird. I was arguing that it wasn't going to take over from Outlook any time soon, because Lightning can't talk to the Exchange calendar features. Several people defended the OSS projects on the basis that Microsoft uses a proprietary interface between Outlook and Thunderbird, which it was difficult to reverse engineer. That cut little ice with me, and there we stalemated. But in the posts around this subthread, several people are saying that various compatibility is there between the Microsoft and Google products: Outlook syncs with Google Calendar, and GMail can respond to Outlook meeting requests (presumably meaning those connected with Outlook/Exchange calendaring, from the context). I would be surprised if Microsoft had knowingly helped Google here, so are the protocols involved too difficult to implement in Thunderbird/Lightning or not?
Google may not be an open source company like Mozilla, but they have historically been much more supportive to open efforts. Open API's are only one example. Think about Google's summer of code, or the open-sourcing of the Google Web Toolkit.
All of which is absolutely, completely and utterly irrelevant to making a commercial decision based on their future behaviour.
Incidentally, those things are pretty obviously in Google's commercial interests, too. Why anyone sees the Summer of Code or the time working on staffers' personal projects as benevolent, I'm not quite sure, given who gets the rights to huge numbers of potentially lucrative ideas at the end of it all.
Bah, it gets easier the more often you do it. Two and three column layouts are quite trivial at this point in the game
That depends on your definition of trivial, of course. Relative to what people used to try, the more recent techniques are certainly a lot more concise in the CSS, I agree. And it's only taken nearly a decade for us to get from what people really did do trivially with tables to a vaguely transparent CSS approach that gets the same result mostly reliably. Hurrah!
So now think about other things that should be easy with good separation of content and presentation, but aren't. Here's a simple example: I want to write my text, duly marked up logically, and I want to tell a browser to flow it into as many 3–4" wide columns as will fit in my user's browser window to give even widths and 0.5" margins and inter-column spacing, or a single column for more narrow devices. Doing this in print would be pretty routine for any DTP package. Doing it in CSS... well, sorry, you can't, because it's not even close to powerful enough to express that kind of concept. The best attempts so far rely on Javascript hacks. And this is in a world where the web is one of the dominant communication media for millions of people, and where home user display hardware varies from a little 15" CRT someone's had for years to a funky new 24" widescreen TFT, not to mention all the mobile devices that have some web capability.
So, when I can get a two-column layout in CSS by writing "columns: 2" in the style for a DIV, plus any spacing options I want to set, then I'll accept that getting a two-column display is trivial. Until then, it's just less of a hack than it used to be, it still requires non-semantic mark-up, and it's still beyond most users.
A vertical list where each link is on it's own line followed by a description is confusing?? You need some sort of extra help to discern that those are separate links?!?
No you don't, and that's not what Bogtha said.
However, users navigate around web pages primarily via fairly imprecise mouse movements, usually tracked by a relatively discreet mouse cursor graphic. It is much clearer, when displaying a load of adjacent links, to have the entire link change colour or something than just to have the mouse cursor change and rely on the user to identify correctly which part of the "pointer" is the relevant one. One of these things tells you you're over a link. The other tells you which link you're over. Good usability practice strongly favours the latter, and that's all Bogtha is trying to explain to you.
If they were redoing the internet from scratch, what is wrong with it that ought to be fixed? Can we hear some new-internet wishlists?
OK, here are a few of my "ideal world" wishes. Deciding their technical feasibility in real life is left as an exercise to the reader.
Encrypted-by-default versions of major protocols for e-mail, web browsing, etc.
A serious attempt to provide universal, verifiable ID for both individuals and organisations.
This would automatically provide for single sign-in facilities.
It would also automatically provide for looking up the real world identity of an organisation you're trying to contact, and finding their real web address instead of the fraudsters, typosquatters or porn sites who registered all the other similar ones.
There is no reason such a facility need be mandatory, but it would allow users to choose to browse only to authenticated sites, bulletin boards to choose to accept posts only from authenticated users, and e-mails not signed by a verified source to be given a lower score in spam filters, for example.
Of course it will never be possible to guarantee 100% of the identities match up to real life people/organisations, and some robots/trolls/spammers would still get through temporarily. But we could do *much* better than we have today, and fixing 90% of the problems is better than nothing.
A framework for micropayments.
A built-from-scratch protocol that provides for both logical mark-up and serious presentation in a simple, coherent way.
HTML and CSS have evolved, but are stupidly underpowered in some ways.
The separation between content and presentation has become more important with the rise of different kinds of device for browsing, from increased diversity in screen resolutions to entirely non-visual browsers for those with poor/no vision.
HTML e-mail has never been standard, has never been well-supported even by popular e-mail clients like Outlook or Thunderbird, and obviously doesn't degrade gracefully (without supplying a duplicate, non-HTMLised version) in text-only e-mail readers.
Diverse formatting commands (or a complete lack of them) on the many bulletin boards and Usenet hamper effective communications by forcing new users to learn yet another method of mark-up every time, or restricting to plain text. An alternative protocol, where the designers keep in mind that only certain subsets may be desirable depending on context, could go a long way to fixing this.
I've always looked at it this way - do you want to be the guy who turns away a customer?
Only if I'm getting more or better-paying customers by doing so.
But it's always a return-on-investment question in business. Of course it's desirable to get every sale you can, but realistically, you often have to prioritise. Supporting an older browser used by perhaps 2–3% of your customer base is likely to have a lower priority than improving usability that will affect the remaining 97% and generate an average 10% increase in sales to each.
Similarly, you're quite right that you have to factor in good will and market reputation rather than looking at isolated cases, but that cuts both ways as well. If you improve the usability of your web site or your brand recognition, then this is likely to improve your word-of-mouth sales. Again, whether this is worth more than what you lost by giving up those customers with older browsers is a decision you can only make in the specific circumstances of your particular business, preferably with some kind of research into the pros and cons of each investment you might make to help you decide.
Also in the UK, there's a wonderful out-take on one of those police camera shows, where a couple of motorway cops pull over a car that's tailgating someone and going well over the limit. The officer who went to talk to the driver comes back to speak with his colleague, and says the driver of the other car is a police driving instructor! They do fail to mention what action, if any, was ever taken against said instructor...
I think, for once, the UK has a sensible approach in law on this one. Our traffic law exemptions are often written in terms such as "vehicle being used as an ambulance". Clearly it is expected that emergency paramedic types are covered by this, but the same exemption has been applied to other cases similar to those mentioned in this thread.
Predictably, there have been a few silly court cases that hit the news. Doctors or blood transports on their way to emergencies but not technically ambulances and with drivers not trained in emergency response driving are the most common examples. However, these cases are pretty much always thrown out in court, as long as the driver's behaviour was reasonable under the circumstances.
And yes, the exemptions can also be (and have been) used for private citizens driving people to hospital in genuine emergencies. However, speaking as both police-trained driver and first-aider, you want to be very sure this is better for the patient than just calling an ambulance before you start going all hero and racing across town, for a whole host of reasons:
In a normal car, you don't have the same emergency equipment available to deal with the condition worsening that an ambulance would have.
You also don't have the same warning equipment to get other vehicles to move out of your way and give your journey priority.
You may not have access to some routes that could shorten the journey. In my city, for example, ambulances are fitted with devices that allow them to pass through roadblocks around pedestrianised areas, and things like road humps are designed so that an ambulance's wheel base can pass over them without being bounced.
In most cases, you don't have the kind of medical training a paramedic has in emergency first aid.
And you don't have a direct link right to the hospital A&E department to notify them of who's coming in and any really urgent information.
Put that lot together, and you can pretty much see why the authorities aren't keen to promote civilian alternatives to ambulances, even as the law makes reasonable provision for it if it happens.
Investigation with cause is one thing, but harassment of the general public is another entirely.
I noticed some fascinating changes in the behaviour of other drivers towards me when I changed from driving a small, low-end car to a turbo-charged beast a few years ago. Some — such as the reduced frequency of tailgating by people in large vehicles — were welcome. One of the less welcome ones was an obvious increase in the number of times a police car parked up at the side of the road would pull out and follow me for half a mile or so for no apparent reason.
Since you don't know me, I'll simply tell you that my driving is very well regarded by passengers and professional instructors alike, and I have a completely clean record in terms of both the law and accidents. If there was something illegal that those police cars thought they might catch me doing, I don't do it (knowingly, at least), and they didn't catch me. But it was still intimidating (and, in some cases, clearly dangerous) and it does suggest rather strongly that I'm being judged by the kind of car I drive and not the way I drive it.
I generally have quite a high opinion of traffic police here in the UK. Perhaps it's the fact that they deal with the aftermath of real accidents, but for whatever reason, most of them seem to care a lot more about enforcing useful laws than silly ones. But on this particular count, I've been sorely disappointed.
It used to be pretty standard for people to customize their browsers in order to change the text, link, followed-link, background, hightlight, and other colors; why does the page designer necessarily know better than the users themselves what the user wants?
I'm not sure it was ever true that most people customised their browsers in that way. In any case, the better web sites now get people who know what they're doing to work on the graphic design. It is a very good bet that a professional graphic designer/typographer will produce a page that looks and works better for the user than whatever most users would come up with themselves. This is true whether or not the untrained user actually realises it.
So, I've got a client that runs an e-commerce site. At least a couple hundred orders per day. I did a quick dig into today's stats. So far: 4 orders from people running IE5, and one from a Netscape 4 flavor. All appeared to be on dial-up connections. A little over $1600 worth of business in those 5 orders.
A valid point, to be sure, though I do wonder about the opportunity cost of supporting the older browsers. Depending on the weakest area of the site, it's quite likely that a few days of staff time could instead have been spent on something like usability research, with much better returns than the 2.5% you're quoting.
Anti-DRM sentiment has been around for years, and it still hasn't managed to snowball. Maybe Vista will be the straw that broke the camel's back, but somehow I doubt it.
It's unlikely to be Vista that breaks the camel's back. It may, however, be anything from the story about EMI that was all over the national papers last week (check the BBC News poll to see the proportion of "normal people" who think the limitations are unreasonable) to the probable problems with HDCP and hi-def content stinging early adopters of HDTV who didn't get the right connectivity on their multi-thousand-dollar screens. Vista will probably just be collateral damage.
Vista is the first version of Windows where device drivers don't run in the context of the OS kernel. Instead they run in the user context. This inherently leads to better security and stability. It also, unfortunately, means that device drivers have to be rewritten. There is a cost for security. I find it interesting that people here on Slashdot would be complaining about Windows drivers and software needing to be rewritten for the new Vista security model. Normally they're constantly bitching about how insecure Windows is, and then when Microsoft makes a change to improve security in a major way they're bitching about the fact that there was a change. Sometimes I think that the Slashdot crowd doesn't care about security, or about ways that Windows could be made better, or anything like that. They just like to bitch about Microsoft and will jump on any little thing that they think makes them sound reasonable.
What a horrible, and inaccurate, generalisation.
What you say about Vista's security model may be entirely true, and driver rewriting may be an inevitable consequence. But that still isn't the user's problem. Microsoft made their bed, over a period of many years during which they profited greatly from selling insecure software to people who didn't know any better. Now they have to lie in it. I'm not bitter, I just have no sympathy for them.
Making Microsoft look good isn't the only advantage to supporting Vista. The other (and far more important) advantage is the ability to be able to sell new hardware/software that is compatible with the most current Windows version.
That isn't a commercial advantage at all. The commercial advantage is in supporting the most popular operating system(s), which right now and for any time in the next year or two at least is going to mean Windows XP. As I said in my previous post, if Vista offers genuine advantages to users, then the user base will migrate over time and the commercial merit in supporting Vista for other vendors will increase.
But it's always about the users. Supporting Microsoft for its own sake does not pay the bills.
I can't recall a time when Microsoft continued to produce the previous version of their operating system for over a year after the new version was released.
Ah, fooey. They tried this stunt with XP, and then backtracked so fast you had to allow for relativistic effects when major hardware brands were confronted with customers who still wanted Win2K instead. They'll do it again, particularly if the current anti-DRM sentiment snowballs and the related lockdown liabilities and performance issues for corporate customers start to get noticed.
But more to the point, Vista has been in development for years. There were betas and technical previews available for almost a year before it was released. If there are companies out there who, by the end of 2007, still do not have compatible software or device drivers then it is simply because of laziness. Either that, or they have no intention of supporting Vista. Either way, the problem is with the third parties, not Microsoft.
You know what? The users don't care. If their current hardware and software worked with XP and don't work with Vista, then as far as they are concerned, that is going to be 100% on Vista for not being backward compatible.
Frankly, I don't have a problem with this. It was always a conceit on Microsoft's part to think that an entire industry was going to pay up lots of time and money to redo work they'd already done, just to support Microsoft's latest exercise in creative marketing.
If Vista offers compelling advantages to users over XP, then people will want to move to it. This will give those developing hardware and software motivation to support it, either to gain new custom or to retain the confidence of past customers who might one day buy from them again.
But it's the users who dictate this, not Microsoft. Many hardware vendors have obsolete products still in active use by their customers but no longer for sale as new. They already invested the money to write XP drivers for that kit and got their sales as a result. If the only advantage to supporting Vista is making Microsoft look good, then any effort to upgrade drivers to support Vista is nothing but a money sink. Why should the whole industry accept that cost, just for Microsoft's benefit?
Slashdot Mirror
Don't count on it. A friend works in the big screen business, typically supplying kit for conference centres, office meeting rooms, public buildings, that sort of thing. A couple of years back, they installed a pretty huge screen, something like 15' IIRC. They finished work late one evening, and when they went in the following morning to set up some software to use it, someone had literally lifted the roof off the building and taken the screen! Fortunately for my friend, payment was due for the hardware on completion of installation, and the conference centre took over responsibility for security at that point...
So your argument is that although theoretically Vista has all these weaknesses, the guy who wrote this paper hadn't actually observed them yet at the time, therefore they aren't there?
Wow, that's a pretty far stretch, particularly considering the number of people who do have a release version of Vista now and are seeing exactly the sorts of problems he predicted with performance, interoperability, etc.
What a load of FUD.
My documents will stop being read? Every major word processor in the software world can read Office 97 files just fine. And this isn't about operating systems anyway.
My software will slowly not be supported? IME, most software isn't supported usefully anyway. The only real exception is patches for security things in Windows, which will continue for several more years for WinXP yet.
I'll be criminally liable because my OS got infested with viruses and worms? Well, for one thing, there's a huge market in software to prevent that, which doesn't rely on Microsoft updates; in fact, it pretty much relies on Microsoft not to make good software for its existence. For another thing, no-one serious about security relies on software alone for safety; things like firewalls exist for a reason. And I'd love to know where you got that criminal liability thing from, because I'm looking forward to suing a quarter of the people on the Internet for disrupting my communications with their botnet-joining PCs. I don't suppose you've got a source for your rather headline-grabbing claim?
And as for hardware, most hardware runs much more reliably on an OS from a couple of years ago with mature drivers than it does on the latest and greatest. This has always been the case, and it often takes years for the software to catch up with a new OS. The only exceptions in recent years in Microsoft world have been graphics cards (where no-one smart buys the latest generation without a very good reason anyway, and historically the standard of accompanying software has been pathetic no matter how long you wait) and wireless (where Windows XP was genuinely a lot better than Windows 2000).
The only point you make with any validity at all is the OEM software licensing issue, which does cause a genuine legal problem if Microsoft artificially stop selling old versions of Windows. Of course, as I've argued elsewhere in this discussion, they might say they're going to do that, but in reality it is very, very unlikely that they are actually going to do it any time soon.
In conclusion, your post is pretty good with the sound-bites, but I don't think any of them stands up to closer inspection. Sorry, but I stand by my original claim: no-one is forced to upgrade to Vista.
Microsoft do write a lot of documentation, it's true. Unfortunately, they now write so much of it, and with such a focus on whatever their current trendy thing is (it's currently .Net), that actually finding the information you need is a very, very time-consuming activity, if you can even find all the details at all. I've been watching MSDN degrade from one of the best technical resources in the programming world to one of the worst, and it's been happening steadily for years. These days, if you try to look up a simple C++ detail or Win32 API call in Visual Studio, you are more likely to find details about numerous .Net things you don't care about at all, even with the filters set appropriately. I was trying to look up some details for the automation model for Word the other day, with future-proofing in mind, and it was taking literally minutes for the Microsoft servers to respond (or time out, in most cases). In this context, it's hard to view ISVs as "lazy". They just have better things to do than searching through all the mess Microsoft has made of its documentation, on the off chance that they missed a minor compatibility point that might come back and cause them some difficulty some way off in the future.
Oh, no they won't. Several large parts of the US Federal Government have already implemented blanket bans on the use of Vista. It is highly likely that several very large businesses will do similarly, along with other government departments in the US and elsewhere, for exactly the same reasons. There is no way Microsoft is going to turn down thousands and thousands of future XP sales to customers of that scale just because its new toy isn't selling well.
If you don't believe me, take a look at Microsoft's rhetoric around this time after XP was released, and then take a look at how many major customers were still able to get Win2K years later, and indeed how many still run it. They might say they're going to stop, but they're firing blanks and both they and their major customers know it.
Hint for future credibility on Slashdot: you will be more popular (and look less stupid) if you bother to read to the end of messages before posting sarcastic replies.
The thing is, the changes required in Linux world to be compatible with new versions of key libraries are generally minor, well-documented, reasonable, and relatively quick to implement.
In Windows world, that isn't the case, because Windows development is a mess. That is mostly Microsoft's fault. There isn't any good way out of this situation for them now, but they could have seen this coming a decade ago, and they chose to ignore it in the interests of making more profits earlier. Now, having made their bed, they are forced to lie in it.
That simply isn't true.
The irony is that I'm not even sure why home users would move.
I've been following Vista developments for years, since back when there were going to be three big pillars underlying it. As far as I can see, from a technical perspective, the only remaining major functional improvement over XP is that Vista supports DirectX 10, and Microsoft are pretty much guaranteed to restrict that artificially to Vista-only.
Of course, going by the history, that won't even start to affect any games except Microsoft's own for at least a couple of years, since most games software isn't using everything DX9 offers yet. Similarly, DX10-supporting hardware won't be even close to mainstream for at least a year or two. Given that PC games now represent only a quarter or so of the market (the consoles are well and truly in charge today) and the majority of home users still aren't going to have Vista for a while, games companies may be hesitant to tread those waters even as they reach the point where the extra goodies in DX10 may be genuinely useful.
Apart from that, what possible reason is there for a home user to upgrade? There's been a lot of negative press for Vista, not just about DRM but also all the hardware and software compatibility problems. The UI is different, which for many users means "bad" by default, even if with time they might come to prefer it. If home users were really serious about security, the world wouldn't be full of botnets. And the list goes on...
I can understand businesses with professional IT people placing some value on improved security or networking features, so if and when the compatibility is sorted out and the trust issues with phoning home and being activated/disabled/whatever remotely are irrevocably fixed, businesses might move. But home users? Not for years, except for the people who just get it with new PCs. (And even the rate of buying those isn't what it used to be.)
Damn, beaten to it... :-)
Seriously, why would any organisation upgrade to Windows Vista if it wasn't pretty sure all of its key software would work? It's amazing how many people seem to think there's some sort of obligation on people to upgrade. In fact, if you look at recent history, the big corporations are usually the last people to move on major upgrades like XP->Vista, often taking several years to do it. This is why.
Ironically, usability is the last thing Outlook is good at. Word and Excel are far more usable than things like OpenOffice, but Outlook's usability is atrocious compared to many other mail clients. What there isn't, yet, is anything that can (a) match the features of the Outlook+Exchange combination, and (b) reliably migrate all the existing data from the Microsoft products to the alternative.
In the corporate space, I suspect sync with Exchange is more important, FWIW.
It's interesting that the other day on Slashdot we were also discussing the new Thunderbird. I was arguing that it wasn't going to take over from Outlook any time soon, because Lightning can't talk to the Exchange calendar features. Several people defended the OSS projects on the basis that Microsoft uses a proprietary interface between Outlook and Thunderbird, which it was difficult to reverse engineer. That cut little ice with me, and there we stalemated. But in the posts around this subthread, several people are saying that various compatibility is there between the Microsoft and Google products: Outlook syncs with Google Calendar, and GMail can respond to Outlook meeting requests (presumably meaning those connected with Outlook/Exchange calendaring, from the context). I would be surprised if Microsoft had knowingly helped Google here, so are the protocols involved too difficult to implement in Thunderbird/Lightning or not?
All of which is absolutely, completely and utterly irrelevant to making a commercial decision based on their future behaviour.
Incidentally, those things are pretty obviously in Google's commercial interests, too. Why anyone sees the Summer of Code or the time working on staffers' personal projects as benevolent, I'm not quite sure, given who gets the rights to huge numbers of potentially lucrative ideas at the end of it all.
That depends on your definition of trivial, of course. Relative to what people used to try, the more recent techniques are certainly a lot more concise in the CSS, I agree. And it's only taken nearly a decade for us to get from what people really did do trivially with tables to a vaguely transparent CSS approach that gets the same result mostly reliably. Hurrah!
So now think about other things that should be easy with good separation of content and presentation, but aren't. Here's a simple example: I want to write my text, duly marked up logically, and I want to tell a browser to flow it into as many 3–4" wide columns as will fit in my user's browser window to give even widths and 0.5" margins and inter-column spacing, or a single column for more narrow devices. Doing this in print would be pretty routine for any DTP package. Doing it in CSS... well, sorry, you can't, because it's not even close to powerful enough to express that kind of concept. The best attempts so far rely on Javascript hacks. And this is in a world where the web is one of the dominant communication media for millions of people, and where home user display hardware varies from a little 15" CRT someone's had for years to a funky new 24" widescreen TFT, not to mention all the mobile devices that have some web capability.
So, when I can get a two-column layout in CSS by writing "columns: 2" in the style for a DIV, plus any spacing options I want to set, then I'll accept that getting a two-column display is trivial. Until then, it's just less of a hack than it used to be, it still requires non-semantic mark-up, and it's still beyond most users.
No you don't, and that's not what Bogtha said.
However, users navigate around web pages primarily via fairly imprecise mouse movements, usually tracked by a relatively discreet mouse cursor graphic. It is much clearer, when displaying a load of adjacent links, to have the entire link change colour or something than just to have the mouse cursor change and rely on the user to identify correctly which part of the "pointer" is the relevant one. One of these things tells you you're over a link. The other tells you which link you're over. Good usability practice strongly favours the latter, and that's all Bogtha is trying to explain to you.
OK, here are a few of my "ideal world" wishes. Deciding their technical feasibility in real life is left as an exercise to the reader.
Only if I'm getting more or better-paying customers by doing so.
But it's always a return-on-investment question in business. Of course it's desirable to get every sale you can, but realistically, you often have to prioritise. Supporting an older browser used by perhaps 2–3% of your customer base is likely to have a lower priority than improving usability that will affect the remaining 97% and generate an average 10% increase in sales to each.
Similarly, you're quite right that you have to factor in good will and market reputation rather than looking at isolated cases, but that cuts both ways as well. If you improve the usability of your web site or your brand recognition, then this is likely to improve your word-of-mouth sales. Again, whether this is worth more than what you lost by giving up those customers with older browsers is a decision you can only make in the specific circumstances of your particular business, preferably with some kind of research into the pros and cons of each investment you might make to help you decide.
Also in the UK, there's a wonderful out-take on one of those police camera shows, where a couple of motorway cops pull over a car that's tailgating someone and going well over the limit. The officer who went to talk to the driver comes back to speak with his colleague, and says the driver of the other car is a police driving instructor! They do fail to mention what action, if any, was ever taken against said instructor...
I think, for once, the UK has a sensible approach in law on this one. Our traffic law exemptions are often written in terms such as "vehicle being used as an ambulance". Clearly it is expected that emergency paramedic types are covered by this, but the same exemption has been applied to other cases similar to those mentioned in this thread.
Predictably, there have been a few silly court cases that hit the news. Doctors or blood transports on their way to emergencies but not technically ambulances and with drivers not trained in emergency response driving are the most common examples. However, these cases are pretty much always thrown out in court, as long as the driver's behaviour was reasonable under the circumstances.
And yes, the exemptions can also be (and have been) used for private citizens driving people to hospital in genuine emergencies. However, speaking as both police-trained driver and first-aider, you want to be very sure this is better for the patient than just calling an ambulance before you start going all hero and racing across town, for a whole host of reasons:
Put that lot together, and you can pretty much see why the authorities aren't keen to promote civilian alternatives to ambulances, even as the law makes reasonable provision for it if it happens.
Investigation with cause is one thing, but harassment of the general public is another entirely.
I noticed some fascinating changes in the behaviour of other drivers towards me when I changed from driving a small, low-end car to a turbo-charged beast a few years ago. Some — such as the reduced frequency of tailgating by people in large vehicles — were welcome. One of the less welcome ones was an obvious increase in the number of times a police car parked up at the side of the road would pull out and follow me for half a mile or so for no apparent reason.
Since you don't know me, I'll simply tell you that my driving is very well regarded by passengers and professional instructors alike, and I have a completely clean record in terms of both the law and accidents. If there was something illegal that those police cars thought they might catch me doing, I don't do it (knowingly, at least), and they didn't catch me. But it was still intimidating (and, in some cases, clearly dangerous) and it does suggest rather strongly that I'm being judged by the kind of car I drive and not the way I drive it.
I generally have quite a high opinion of traffic police here in the UK. Perhaps it's the fact that they deal with the aftermath of real accidents, but for whatever reason, most of them seem to care a lot more about enforcing useful laws than silly ones. But on this particular count, I've been sorely disappointed.
On the bright side, at least my car's not red. :-)
Your proposal is acceptable.
I'm not sure it was ever true that most people customised their browsers in that way. In any case, the better web sites now get people who know what they're doing to work on the graphic design. It is a very good bet that a professional graphic designer/typographer will produce a page that looks and works better for the user than whatever most users would come up with themselves. This is true whether or not the untrained user actually realises it.
A valid point, to be sure, though I do wonder about the opportunity cost of supporting the older browsers. Depending on the weakest area of the site, it's quite likely that a few days of staff time could instead have been spent on something like usability research, with much better returns than the 2.5% you're quoting.
It's unlikely to be Vista that breaks the camel's back. It may, however, be anything from the story about EMI that was all over the national papers last week (check the BBC News poll to see the proportion of "normal people" who think the limitations are unreasonable) to the probable problems with HDCP and hi-def content stinging early adopters of HDTV who didn't get the right connectivity on their multi-thousand-dollar screens. Vista will probably just be collateral damage.
What a horrible, and inaccurate, generalisation.
What you say about Vista's security model may be entirely true, and driver rewriting may be an inevitable consequence. But that still isn't the user's problem. Microsoft made their bed, over a period of many years during which they profited greatly from selling insecure software to people who didn't know any better. Now they have to lie in it. I'm not bitter, I just have no sympathy for them.
That isn't a commercial advantage at all. The commercial advantage is in supporting the most popular operating system(s), which right now and for any time in the next year or two at least is going to mean Windows XP. As I said in my previous post, if Vista offers genuine advantages to users, then the user base will migrate over time and the commercial merit in supporting Vista for other vendors will increase.
But it's always about the users. Supporting Microsoft for its own sake does not pay the bills.
Ah, fooey. They tried this stunt with XP, and then backtracked so fast you had to allow for relativistic effects when major hardware brands were confronted with customers who still wanted Win2K instead. They'll do it again, particularly if the current anti-DRM sentiment snowballs and the related lockdown liabilities and performance issues for corporate customers start to get noticed.
You know what? The users don't care. If their current hardware and software worked with XP and don't work with Vista, then as far as they are concerned, that is going to be 100% on Vista for not being backward compatible.
Frankly, I don't have a problem with this. It was always a conceit on Microsoft's part to think that an entire industry was going to pay up lots of time and money to redo work they'd already done, just to support Microsoft's latest exercise in creative marketing.
If Vista offers compelling advantages to users over XP, then people will want to move to it. This will give those developing hardware and software motivation to support it, either to gain new custom or to retain the confidence of past customers who might one day buy from them again.
But it's the users who dictate this, not Microsoft. Many hardware vendors have obsolete products still in active use by their customers but no longer for sale as new. They already invested the money to write XP drivers for that kit and got their sales as a result. If the only advantage to supporting Vista is making Microsoft look good, then any effort to upgrade drivers to support Vista is nothing but a money sink. Why should the whole industry accept that cost, just for Microsoft's benefit?