I think we need to start consciously distinguishing between fixes and other changes.
If a device doesn't properly do what it was supposed to do when the customer bought it -- there's a security vulnerability, it's not quite in spec and so doesn't work properly with something, that kind of thing -- then like any other purchase, the customer should get what (they thought) they were paying for when they decided to buy, and the manufacturer should fix the defect.
The manufacturer is under no equivalent obligation to offer non-essential changes, like moving the UI around or adding new functionality, but might choose to do so anyway. IMHO in that case they should be free to offer them to others on whatever reasonable terms they want. If a user doesn't want them, they don't have to buy in, just as they don't have to buy the latest version of a device in the first place if they don't like it.
The key point is that if a purchaser chooses not to buy into any extra offers from the manufacturer, that should not be remove the purchaser's basic right to get what they paid for originally or negate the manufacturer's normal obligations to make defects good or compensate in some reasonable way, just like manufacturers of any other product under normal consumer protection and similar laws in many places.
Ultimately, the only way to do that is to give power to the users, and the only way to do that is to offer them a choice or a setting.
There is a middle ground: you separate updates for areas like security and reliability from updates to the UI and functionality. Got an existing system? Sure, you can patch your existing software to keep it as secure as possible, without also getting UI changes or new functionality. Want our latest features? You can update to a new version of the software, but you take it or leave it.
We successfully built software for some number of decades in this way, and in practice it wasn't a prohibitive maintenance burden for the developers to support a previous version or two for essential updates. Unfortunately, this level of customer support and actually making useful products doesn't play nicely with the trendy Agile, X-as-a-service, Lean, insert-other-buzzword-here mentality of a lot of young developers and startups today, and now the established heavyweights are seeing opportunities to save a bit of cash and exert more control over their users by adopting the same techniques.
Unfortunately, there is a trend for updates to be completely automatic and involuntary, both with certain devices and even now with Windows 10 on the desktop. All it takes is some sort of online component it depends on and you have a crank to turn the update wheel, even if the update actually has nothing to do with that online element. Again, it's clear why the developers would prefer only having to support their latest code base, but unfortunately it leaves users with no control over their own devices, including in cases where from their point of view the update makes something worse than it was before.
There are also all kinds of mechanisms that effectively compel updates even if they aren't directly made mandatory and automatic. For example, on iOS devices, you can only get apps from the App Store, and Apple can impose constraints on those apps if they want to be listed. This can drive app developers towards only supporting the latest version of iOS, and again that can be a problem for people who previously had an older version of the app installed on an older version of iOS that worked well on an older device where perhaps the new version does not. These cases are particularly nasty, because all the developers involved can point fingers at each other and say it's someone else causing the problem, yet to the user the reality is the same: their device and software used to work, and now they don't.
A wider issue is the general trend for devices with behaviour that is remotely changed after you buy them thanks to software updates. What is the situation if you bought an e-reader you were happy with and could use comfortably, but then after this kind of update it no longer works for you because, for example, your eyesight isn't good enough to read the new font? It's obvious why hardware and software vendors might want this kind of capability, but how do we protect the buyers who are using the products to make sure they're still getting what they paid for when they decided to buy?
Had it been made plainly obvious what they were pushing, no one would have installed any of it.
Ironically, I suspect that's not true. It seems quite realistic that if they'd pushed telemetry transparently as a recommended update, the average home user would have just said yes with all the others anyway, yet Microsoft would not have lost the confidence of the techie crowd and gained the unwelcome reputation for being deceptive and manipulative that they've managed to cement over the past few months.
Microsoft is certainly doing some very dubious things lately, but it does no-one any good to exaggerate or distort what they're doing. Please stop doing that.
In particular, they have issued telemetry updates for earlier Windows versions, and they have aggressively promoted the update to Windows 10, but they have not forced users of earlier Windows versions to update if they say no.
Assuming that was intended to read "...complaining about not knowing..." the answer seems pretty clear: Microsoft have been serving up updates for all recent versions of Windows with little to no detail of what is actually in them for some time, and lately some of those updates have been outright user-hostile, and consequently a lot of power user or professional sysadmin types simply don't trust them any more.
Just about the one barrier they haven't crossed yet is serving up user-hostile updates under the guise of security updates rather than just recommended ones, which means you can still assume that something marked as a security update is likely to be worth installing with due diligence. Being more explicit about these issues on Win10 goes at least a little way towards maintaining that confidence.
If the wind is too high or the surface too rough it's the cyclists job NOT TO RIDE THERE.
Right. Which is why it's then a problem having cyclists who want to travel at a reasonable speed, but still a speed slower than motor vehicles can achieve, sharing the same main traffic lane. Which brings me back to my main point: substandard facilities like nothing-but-a-paint-job cycle lanes are potentially dangerous, and we need proper, mode-appropriate facilities for all classes of road user. Which in turn comes back to the original point of the discussion: sometimes we're better off without just slapping paint on the road if all it does is create dubious expectations and a false sense of security.
I can never be sure the person in the next lane won't lose control of their vehicle. So I should never pass anyone?
You should only pass when you're sure it's reasonably safe to do so. It's really as simple as that, and that "reasonably safe" means taking into account any significant and predictable risks around other types of vehicle. It's the exact same argument whether we're talking about a two-wheeler at greater risk of being affected by wind or uneven road surface, a horse at risk of being startled, or a large vehicle that is signalling on the approach to a junction and will need to make a wide turn.
I don't know where you are, but this story is about the UK, so it's the UK's Highway Code I'm going by, and everything I've just written should be explained routinely by any driving instructor before their students take the test.
Sorry, but the Highway Code disagrees with you. There are pictures and everything. There is also explicit discussion about the need to allow for more vulnerable road users.
If you go around overtaking other people with insufficient clearance and something bad happens, then again you should expect that to be counted against you in court. Obviously all road users should be able to hold their lane properly under normal conditions, but all road users should also be aware that overtaking is an inherently dangerous manoeuvre and they should not do it unless they are sure they can complete that manoeuvre safely. In the UK, that applies each and every time you pass anyone else, cycle or otherwise, and regardless of which lane(s) you are each following before the overtake.
Just like a car pulling away from a curb can't jump in front of moving traffic and complain about getting rear ended. It's your job to make the lane change safely. You might actually have to stop instead.
Of course. On the other hand, unpredictable things happen on roads, particularly in bad weather. A driver who isn't allowing enough clearance as they pass a cyclist for the cycle to be blown a little off course without getting hit just isn't paying enough attention.
Deliberate? I thought I saw a road hazard.
Then you can try that one on with the court. Perhaps the CPS will throw in a charge of driving without due care and attention as well, since if you had to panic brake in response to a road hazard there's a fair chance you weren't driving carefully and at a suitable speed for the conditions.
It is not illegal to slow down. If you are behind me, it's your job to maintain a safe following distance.
True, but it is illegal to drive dangerously or without reasonable consideration for other road users, among other things, and these would be likely consequences of a sudden "brake check" to harass a following cyclist. In fact, braking without good cause is explicitly included in the CPS guidance for bringing the reasonable consideration charge.
The main facility that can be upgraded is getting 'road bikes' off the road. There is no way to safely ride 100psi+ tire bikes on the street with cars. They basically have to swerve around every bit of glass in the bike lane and are rolling, left and right lurching hazards with unsafely long stopping distances.
Or we could completely prohibit cars from using roads frequented by road bikes, or impose a much lower speed limit where access is still required. In some places, we're approaching the point where there will be more cyclists than cars using a given road, after all, and removing the cars would make it safer for other types of bike as well.
In reality, neither absolutist solution is going to get us anywhere until there are reasonable alternatives for any group that gets displaced.
If you need to get out of the bike lane, it is your job to make sure it is safe to switch lanes, same as a car. You can't just jump into the main traffic lane in front of a car and complain the car was going too fast. You made an unsafe lane change.
That is all true.
However, it is equally true that a driver does not have an automatic right to overtake a cyclist in front of them who wants to use the main traffic lane. A cyclist obviously shouldn't switch lanes right in front of a car, but if a cyclist a little up the road wants to use the main traffic lane, they are perfectly entitled to do so. It is the following driver's responsibility to slow down and maintain a safe distance in that case, just as they would have to if a slower motor vehicle pulled out ahead of them on a multi-lane highway.
It's also worth pointing out that a driver who does overtake a cyclist, regardless of which lane they are cycling in, should be allowing as much space alongside as they would have when passing a car. That will almost certainly require crossing well over the centre line of the road on many urban roads in the UK. If a cyclist veering slightly into the main traffic lane ever actually gets hit by a car overtaking them, even if it was a mistake by the cyclist, it was also self-evidently a mistake by the driver.
Also don't complain when cars 'brake check' you going up hills. That's just payback for you slowing them down.
Doing that deliberately is illegal on several counts, and if you do it in sight of a police car you should rightly expect to get pulled up for it.
Many cyclists seam to think they have the right of way any time the alternative is they lose their inertia.
Some do, certainly, and they are wrong. There is no general right of way on UK roads, for a start.
But equally some drivers seem to think that they have the right to go as fast as they want to regardless of other road users, and those drivers are also wrong.
As I said before, the only credible way to improve this kind of situation is to ensure that the facilities for everyone using the roads are up to scratch and reduce the potential for conflict happening at all.
Part of the trouble is that the kind of on-road cycle lanes we're talking about in the UK aren't normal lanes in various respects, including sometimes legal ones. Even to the extent that they are, they are often created by literally nothing but painting a line down existing roads to mark off an area much smaller than the relevant policies call for. No extra space is created, nor any real physical separation or protection added.
This results in exactly the kind of them-and-us culture I was talking about, where a lot of drivers who don't cycle themselves see a cycle lane and think bikes should stay in it at all times, while anyone who has ever cycled significantly could tell you that this is completely unrealistic because the lanes aren't wide enough for anyone to do so and still make sensible progress even before you consider all the extra hazards that tend to happen towards the side of a road where the cycle lane is.
Consequently a lot of faster and more competent cycles will disregard the lanes and cycle in the main traffic flow when conditions dictate, and a lot of ignorant and selfish drivers will then illegally harass and intimidate the cyclists for riding in the main traffic lane and slowing them down marginally. Many drivers also pass cyclists who are in a cycle lane far too close, and one of the well established benefits of removing road markings for explicit lanes is that drivers do then move out significantly more and pass cyclists at a safer distance.
For me, the only truly credible solutions to today's them-and-us culture involve providing a decent standard of facilities for both groups where conflict is designed out in the first place. Much better designs than what we currently use in the UK are known -- the Dutch typically do these things well, for example -- but they cost significant amounts of money, particularly to implement them retroactively on existing road layouts, and so far the political will in the UK just doesn't seem to be there to spend it. In some places, particularly older cities with historical areas and narrow streets, there simply isn't a good solution as long as so many different types of vehicle are trying to share the same road space.
It's an unfortunate reality of a lot of existing/historical road planning policy that it creates a them-and-us culture one way or another. Cars and cycles. Cars and buses. Buses and cycles. Lorries and everyone. White vans and other white vans.
What a lot of people seem to be missing in this discussion is that over-regulation and excessive road markings and street furniture create a false sense of security and so lead to over-confidence. There's a white line dividing the cycle lane from the main traffic, so of course it's safe for me to fly past at 30mph in my car today when there are 40mph winds gusting as long as I stay my side of the line. Yeah, yeah, I know the cyclist has less than a metre of road width for their lane because the markings don't follow the spec, and I know I'm only leaving half a metre of clearance, and I know that one gust of wind or small fallen branch in their lane could mean they swerve suddenly into mine, but that silly stuff doesn't matter, does it? (Incidentally, this goes both ways, too: a cyclist who races up the cycle lane to the advanced stop line at a junction past dense stationary traffic in today's conditions is just as bad.)
On the evidence so far, the reason that cutting down on the markings and regulations is effective at increasing safety and reducing traffic flows under some conditions is that it forces drivers to pay attention and co-operate instead of assuming. If that means drivers slow right down in places where they didn't before, they probably should have been going slower all along, but weren't because they were trusting the road markings or still under the speed limit or some other rationalization. If it means they can't drive properly and be on the phone at the same time, well, they never could, it's just that now it's blindingly obvious even to them.
We should review the results of these kinds of experiments over the long term of course, just in case the effects turn out to be temporary or they have other unintended consequences. But for now, there is ample credible evidence that this alternative approach may be much better for everyone under some circumstances and it's clearly worth further investigation. The fact that so many people here seem to dismiss it out of hand based on nothing but naive intuition is an excellent demonstration of why these sorts of public policies should be evidence-based.
That's great, but as a few moments searching the web could tell you, not everyone has been so fortunate. There have already been several widespread instances of hardware/driver issues, reboot loops, software being uninstalled due to being deemed no longer compatible, and similar problems reported by Windows 10 users.
Shame on Microsoft for making people get off an OS that isn't receiving updates and for pushing for people to get off an OS that will stop receiving them in a handful of years.
That's almost four more years that Microsoft have committed to supporting the OS.
A significant number of computers that haven't even been bought yet could run Windows 7 for their entire working lifetimes and still be within the extended support period.
Also, merely "connecting to the Internet" is highly unlikely to leave a system vulnerable even if it isn't fully patched, and I'll take "outdated and unsupported" over "actively damaged at arbitrary intervals by compulsory updates you can't block".
What do you think happens after the 1 year anniversary of Windows 10 launch?
If adoption rates are still unimpressive, I imagine Nadella gives a mea culpa speech as he decides to spend more time with his family, and the new CEO starts making highly publicised changes in strategic direction as soon as possible to reassure the big corporate customers and to some extent home users that Microsoft is still looking out for them.
What that direction would be is interesting, as Microsoft is one of the few IT giants that probably still has the resources and credibility to shift the entire industry. Apple is another. Both seem to have lost their focus in recent years, but one or two big new ideas could change that.
You make a good point, though it probably requires some degree of actual knowledge and skill, or at least a suitable malware toolkit, to cause damage by playing with electrical levels. Any script kiddie can do 'rm -rf/' and surely it's practically the first thing any mischief-makers will try.
It's still crazily easy to do this by mistake as well, though.
Exactly. The real problem here isn't that root can do stuff. The real problem is that root can do stuff accidentally by sneezing five metres away from the system at lunchtime.
Of course, the other real problem is that anyone is crazy enough to make hardware/firmware where you can delete essential data like this and have no recovery or at least factory reset mechanism, regardless of anything the OS might be doing. People making hardware vulnerable to this should be getting named and shamed as well.
Yes, malware is probably the biggest real danger here.
That said, over the years I've also seen my share of very sheepish-looking engineers whose scripts didn't guard against an empty environment variable...
I didn't write those, so I can't comment on why they have the limitations you're reporting. All I can say is that the similar software I have developed, in some cases also related to networking hardware, has never run into these "must have exactly JRE version X" issues as far as I'm aware, nor can I see any likely reason they ever would (other than now not working with any future versions of some browsers or Java plugins beyond support being cut off, obviously).
I'm not saying the version ties you're complaining about don't happen. I have no reason to doubt what you're telling us. However, I am questioning whether they are due to some inherent problem with Java or just to developers not doing a great job when writing certain specific programs. Without knowing the actual limitation and why it happens in each case you mentioned, it's impossible to say.
Odd, I found about 50% of such things don't work anymore.
A lot of things -- useful things -- provided as Java applets have stopped working lately as the browsers and Oracle itself have increasingly locked down what plugins can do and how they are integrated. There have been ever-increasing numbers of scary warnings about things like who signed what and ever more hoops to jump through just to publish or run an applet. The thing is, those are almost 100% artificial barriers put there by Oracle, Apple, Google, Mozilla, and friends. The underlying Java code that actually made the applet go in each case would probably still work fine today if the artificial barriers were removed again.
I agree about the current state of web app development, but unfortunately there are few organisations with enough influence to significantly affect the course of the industry, and for now their interests seem more aligned with the status quo than radical change. There are some interesting ideas around, web assembly for example, that might open up some more radical options in the future, but then there are always new ideas in the background in web development and all too often they don't achieve the critical mass of interest and support to become established. I guess time will tell.
I'm not saying it doesn't exist, but so far I haven't seen any report of malware that can successfully bypass the click-to-play limitations in both the browser and Java itself. There have been some ingenious attacks on parts of the infrastructure, such as the Pawn Storm issue a few months back, but as far as I'm aware even those required the browser itself to have Java enabled and only compromised the Java plugin's security architecture.
As an aside, if such malware did exist, it would self-evidently indicate a browser bug as well as a Java one. That would not be a good sign for a more secure future when browsers are taking on all the jobs plugins used to get.
Slashdot Mirror
I think we need to start consciously distinguishing between fixes and other changes.
If a device doesn't properly do what it was supposed to do when the customer bought it -- there's a security vulnerability, it's not quite in spec and so doesn't work properly with something, that kind of thing -- then like any other purchase, the customer should get what (they thought) they were paying for when they decided to buy, and the manufacturer should fix the defect.
The manufacturer is under no equivalent obligation to offer non-essential changes, like moving the UI around or adding new functionality, but might choose to do so anyway. IMHO in that case they should be free to offer them to others on whatever reasonable terms they want. If a user doesn't want them, they don't have to buy in, just as they don't have to buy the latest version of a device in the first place if they don't like it.
The key point is that if a purchaser chooses not to buy into any extra offers from the manufacturer, that should not be remove the purchaser's basic right to get what they paid for originally or negate the manufacturer's normal obligations to make defects good or compensate in some reasonable way, just like manufacturers of any other product under normal consumer protection and similar laws in many places.
Ultimately, the only way to do that is to give power to the users, and the only way to do that is to offer them a choice or a setting.
There is a middle ground: you separate updates for areas like security and reliability from updates to the UI and functionality. Got an existing system? Sure, you can patch your existing software to keep it as secure as possible, without also getting UI changes or new functionality. Want our latest features? You can update to a new version of the software, but you take it or leave it.
We successfully built software for some number of decades in this way, and in practice it wasn't a prohibitive maintenance burden for the developers to support a previous version or two for essential updates. Unfortunately, this level of customer support and actually making useful products doesn't play nicely with the trendy Agile, X-as-a-service, Lean, insert-other-buzzword-here mentality of a lot of young developers and startups today, and now the established heavyweights are seeing opportunities to save a bit of cash and exert more control over their users by adopting the same techniques.
Unfortunately, there is a trend for updates to be completely automatic and involuntary, both with certain devices and even now with Windows 10 on the desktop. All it takes is some sort of online component it depends on and you have a crank to turn the update wheel, even if the update actually has nothing to do with that online element. Again, it's clear why the developers would prefer only having to support their latest code base, but unfortunately it leaves users with no control over their own devices, including in cases where from their point of view the update makes something worse than it was before.
There are also all kinds of mechanisms that effectively compel updates even if they aren't directly made mandatory and automatic. For example, on iOS devices, you can only get apps from the App Store, and Apple can impose constraints on those apps if they want to be listed. This can drive app developers towards only supporting the latest version of iOS, and again that can be a problem for people who previously had an older version of the app installed on an older version of iOS that worked well on an older device where perhaps the new version does not. These cases are particularly nasty, because all the developers involved can point fingers at each other and say it's someone else causing the problem, yet to the user the reality is the same: their device and software used to work, and now they don't.
A wider issue is the general trend for devices with behaviour that is remotely changed after you buy them thanks to software updates. What is the situation if you bought an e-reader you were happy with and could use comfortably, but then after this kind of update it no longer works for you because, for example, your eyesight isn't good enough to read the new font? It's obvious why hardware and software vendors might want this kind of capability, but how do we protect the buyers who are using the products to make sure they're still getting what they paid for when they decided to buy?
Had it been made plainly obvious what they were pushing, no one would have installed any of it.
Ironically, I suspect that's not true. It seems quite realistic that if they'd pushed telemetry transparently as a recommended update, the average home user would have just said yes with all the others anyway, yet Microsoft would not have lost the confidence of the techie crowd and gained the unwelcome reputation for being deceptive and manipulative that they've managed to cement over the past few months.
Microsoft is certainly doing some very dubious things lately, but it does no-one any good to exaggerate or distort what they're doing. Please stop doing that.
In particular, they have issued telemetry updates for earlier Windows versions, and they have aggressively promoted the update to Windows 10, but they have not forced users of earlier Windows versions to update if they say no.
Assuming that was intended to read "...complaining about not knowing..." the answer seems pretty clear: Microsoft have been serving up updates for all recent versions of Windows with little to no detail of what is actually in them for some time, and lately some of those updates have been outright user-hostile, and consequently a lot of power user or professional sysadmin types simply don't trust them any more.
Just about the one barrier they haven't crossed yet is serving up user-hostile updates under the guise of security updates rather than just recommended ones, which means you can still assume that something marked as a security update is likely to be worth installing with due diligence. Being more explicit about these issues on Win10 goes at least a little way towards maintaining that confidence.
If the wind is too high or the surface too rough it's the cyclists job NOT TO RIDE THERE.
Right. Which is why it's then a problem having cyclists who want to travel at a reasonable speed, but still a speed slower than motor vehicles can achieve, sharing the same main traffic lane. Which brings me back to my main point: substandard facilities like nothing-but-a-paint-job cycle lanes are potentially dangerous, and we need proper, mode-appropriate facilities for all classes of road user. Which in turn comes back to the original point of the discussion: sometimes we're better off without just slapping paint on the road if all it does is create dubious expectations and a false sense of security.
I can never be sure the person in the next lane won't lose control of their vehicle. So I should never pass anyone?
You should only pass when you're sure it's reasonably safe to do so. It's really as simple as that, and that "reasonably safe" means taking into account any significant and predictable risks around other types of vehicle. It's the exact same argument whether we're talking about a two-wheeler at greater risk of being affected by wind or uneven road surface, a horse at risk of being startled, or a large vehicle that is signalling on the approach to a junction and will need to make a wide turn.
I don't know where you are, but this story is about the UK, so it's the UK's Highway Code I'm going by, and everything I've just written should be explained routinely by any driving instructor before their students take the test.
Sorry, but the Highway Code disagrees with you. There are pictures and everything. There is also explicit discussion about the need to allow for more vulnerable road users.
If you go around overtaking other people with insufficient clearance and something bad happens, then again you should expect that to be counted against you in court. Obviously all road users should be able to hold their lane properly under normal conditions, but all road users should also be aware that overtaking is an inherently dangerous manoeuvre and they should not do it unless they are sure they can complete that manoeuvre safely. In the UK, that applies each and every time you pass anyone else, cycle or otherwise, and regardless of which lane(s) you are each following before the overtake.
Just like a car pulling away from a curb can't jump in front of moving traffic and complain about getting rear ended. It's your job to make the lane change safely. You might actually have to stop instead.
Of course. On the other hand, unpredictable things happen on roads, particularly in bad weather. A driver who isn't allowing enough clearance as they pass a cyclist for the cycle to be blown a little off course without getting hit just isn't paying enough attention.
Deliberate? I thought I saw a road hazard.
Then you can try that one on with the court. Perhaps the CPS will throw in a charge of driving without due care and attention as well, since if you had to panic brake in response to a road hazard there's a fair chance you weren't driving carefully and at a suitable speed for the conditions.
It is not illegal to slow down. If you are behind me, it's your job to maintain a safe following distance.
True, but it is illegal to drive dangerously or without reasonable consideration for other road users, among other things, and these would be likely consequences of a sudden "brake check" to harass a following cyclist. In fact, braking without good cause is explicitly included in the CPS guidance for bringing the reasonable consideration charge.
The main facility that can be upgraded is getting 'road bikes' off the road. There is no way to safely ride 100psi+ tire bikes on the street with cars. They basically have to swerve around every bit of glass in the bike lane and are rolling, left and right lurching hazards with unsafely long stopping distances.
Or we could completely prohibit cars from using roads frequented by road bikes, or impose a much lower speed limit where access is still required. In some places, we're approaching the point where there will be more cyclists than cars using a given road, after all, and removing the cars would make it safer for other types of bike as well.
In reality, neither absolutist solution is going to get us anywhere until there are reasonable alternatives for any group that gets displaced.
If you need to get out of the bike lane, it is your job to make sure it is safe to switch lanes, same as a car. You can't just jump into the main traffic lane in front of a car and complain the car was going too fast. You made an unsafe lane change.
That is all true.
However, it is equally true that a driver does not have an automatic right to overtake a cyclist in front of them who wants to use the main traffic lane. A cyclist obviously shouldn't switch lanes right in front of a car, but if a cyclist a little up the road wants to use the main traffic lane, they are perfectly entitled to do so. It is the following driver's responsibility to slow down and maintain a safe distance in that case, just as they would have to if a slower motor vehicle pulled out ahead of them on a multi-lane highway.
It's also worth pointing out that a driver who does overtake a cyclist, regardless of which lane they are cycling in, should be allowing as much space alongside as they would have when passing a car. That will almost certainly require crossing well over the centre line of the road on many urban roads in the UK. If a cyclist veering slightly into the main traffic lane ever actually gets hit by a car overtaking them, even if it was a mistake by the cyclist, it was also self-evidently a mistake by the driver.
Also don't complain when cars 'brake check' you going up hills. That's just payback for you slowing them down.
Doing that deliberately is illegal on several counts, and if you do it in sight of a police car you should rightly expect to get pulled up for it.
Many cyclists seam to think they have the right of way any time the alternative is they lose their inertia.
Some do, certainly, and they are wrong. There is no general right of way on UK roads, for a start.
But equally some drivers seem to think that they have the right to go as fast as they want to regardless of other road users, and those drivers are also wrong.
As I said before, the only credible way to improve this kind of situation is to ensure that the facilities for everyone using the roads are up to scratch and reduce the potential for conflict happening at all.
Part of the trouble is that the kind of on-road cycle lanes we're talking about in the UK aren't normal lanes in various respects, including sometimes legal ones. Even to the extent that they are, they are often created by literally nothing but painting a line down existing roads to mark off an area much smaller than the relevant policies call for. No extra space is created, nor any real physical separation or protection added.
This results in exactly the kind of them-and-us culture I was talking about, where a lot of drivers who don't cycle themselves see a cycle lane and think bikes should stay in it at all times, while anyone who has ever cycled significantly could tell you that this is completely unrealistic because the lanes aren't wide enough for anyone to do so and still make sensible progress even before you consider all the extra hazards that tend to happen towards the side of a road where the cycle lane is.
Consequently a lot of faster and more competent cycles will disregard the lanes and cycle in the main traffic flow when conditions dictate, and a lot of ignorant and selfish drivers will then illegally harass and intimidate the cyclists for riding in the main traffic lane and slowing them down marginally. Many drivers also pass cyclists who are in a cycle lane far too close, and one of the well established benefits of removing road markings for explicit lanes is that drivers do then move out significantly more and pass cyclists at a safer distance.
For me, the only truly credible solutions to today's them-and-us culture involve providing a decent standard of facilities for both groups where conflict is designed out in the first place. Much better designs than what we currently use in the UK are known -- the Dutch typically do these things well, for example -- but they cost significant amounts of money, particularly to implement them retroactively on existing road layouts, and so far the political will in the UK just doesn't seem to be there to spend it. In some places, particularly older cities with historical areas and narrow streets, there simply isn't a good solution as long as so many different types of vehicle are trying to share the same road space.
It's an unfortunate reality of a lot of existing/historical road planning policy that it creates a them-and-us culture one way or another. Cars and cycles. Cars and buses. Buses and cycles. Lorries and everyone. White vans and other white vans.
What a lot of people seem to be missing in this discussion is that over-regulation and excessive road markings and street furniture create a false sense of security and so lead to over-confidence. There's a white line dividing the cycle lane from the main traffic, so of course it's safe for me to fly past at 30mph in my car today when there are 40mph winds gusting as long as I stay my side of the line. Yeah, yeah, I know the cyclist has less than a metre of road width for their lane because the markings don't follow the spec, and I know I'm only leaving half a metre of clearance, and I know that one gust of wind or small fallen branch in their lane could mean they swerve suddenly into mine, but that silly stuff doesn't matter, does it? (Incidentally, this goes both ways, too: a cyclist who races up the cycle lane to the advanced stop line at a junction past dense stationary traffic in today's conditions is just as bad.)
On the evidence so far, the reason that cutting down on the markings and regulations is effective at increasing safety and reducing traffic flows under some conditions is that it forces drivers to pay attention and co-operate instead of assuming. If that means drivers slow right down in places where they didn't before, they probably should have been going slower all along, but weren't because they were trusting the road markings or still under the speed limit or some other rationalization. If it means they can't drive properly and be on the phone at the same time, well, they never could, it's just that now it's blindingly obvious even to them.
We should review the results of these kinds of experiments over the long term of course, just in case the effects turn out to be temporary or they have other unintended consequences. But for now, there is ample credible evidence that this alternative approach may be much better for everyone under some circumstances and it's clearly worth further investigation. The fact that so many people here seem to dismiss it out of hand based on nothing but naive intuition is an excellent demonstration of why these sorts of public policies should be evidence-based.
Windows 10 has been very stable for me.
That's great, but as a few moments searching the web could tell you, not everyone has been so fortunate. There have already been several widespread instances of hardware/driver issues, reboot loops, software being uninstalled due to being deemed no longer compatible, and similar problems reported by Windows 10 users.
But this isn't really about new hardware, it's about people with existing hardware being tricked into updating to Windows 10 when they don't want to.
Shame on Microsoft for making people get off an OS that isn't receiving updates and for pushing for people to get off an OS that will stop receiving them in a handful of years.
Windows 7 extended support runs until 14 January 2020.
That's almost four more years that Microsoft have committed to supporting the OS.
A significant number of computers that haven't even been bought yet could run Windows 7 for their entire working lifetimes and still be within the extended support period.
Also, merely "connecting to the Internet" is highly unlikely to leave a system vulnerable even if it isn't fully patched, and I'll take "outdated and unsupported" over "actively damaged at arbitrary intervals by compulsory updates you can't block".
What do you think happens after the 1 year anniversary of Windows 10 launch?
If adoption rates are still unimpressive, I imagine Nadella gives a mea culpa speech as he decides to spend more time with his family, and the new CEO starts making highly publicised changes in strategic direction as soon as possible to reassure the big corporate customers and to some extent home users that Microsoft is still looking out for them.
What that direction would be is interesting, as Microsoft is one of the few IT giants that probably still has the resources and credibility to shift the entire industry. Apple is another. Both seem to have lost their focus in recent years, but one or two big new ideas could change that.
You make a good point, though it probably requires some degree of actual knowledge and skill, or at least a suitable malware toolkit, to cause damage by playing with electrical levels. Any script kiddie can do 'rm -rf /' and surely it's practically the first thing any mischief-makers will try.
It's still crazily easy to do this by mistake as well, though.
Exactly. The real problem here isn't that root can do stuff. The real problem is that root can do stuff accidentally by sneezing five metres away from the system at lunchtime.
Of course, the other real problem is that anyone is crazy enough to make hardware/firmware where you can delete essential data like this and have no recovery or at least factory reset mechanism, regardless of anything the OS might be doing. People making hardware vulnerable to this should be getting named and shamed as well.
Yes, malware is probably the biggest real danger here.
That said, over the years I've also seen my share of very sheepish-looking engineers whose scripts didn't guard against an empty environment variable...
I didn't write those, so I can't comment on why they have the limitations you're reporting. All I can say is that the similar software I have developed, in some cases also related to networking hardware, has never run into these "must have exactly JRE version X" issues as far as I'm aware, nor can I see any likely reason they ever would (other than now not working with any future versions of some browsers or Java plugins beyond support being cut off, obviously).
I'm not saying the version ties you're complaining about don't happen. I have no reason to doubt what you're telling us. However, I am questioning whether they are due to some inherent problem with Java or just to developers not doing a great job when writing certain specific programs. Without knowing the actual limitation and why it happens in each case you mentioned, it's impossible to say.
Odd, I found about 50% of such things don't work anymore.
A lot of things -- useful things -- provided as Java applets have stopped working lately as the browsers and Oracle itself have increasingly locked down what plugins can do and how they are integrated. There have been ever-increasing numbers of scary warnings about things like who signed what and ever more hoops to jump through just to publish or run an applet. The thing is, those are almost 100% artificial barriers put there by Oracle, Apple, Google, Mozilla, and friends. The underlying Java code that actually made the applet go in each case would probably still work fine today if the artificial barriers were removed again.
I agree about the current state of web app development, but unfortunately there are few organisations with enough influence to significantly affect the course of the industry, and for now their interests seem more aligned with the status quo than radical change. There are some interesting ideas around, web assembly for example, that might open up some more radical options in the future, but then there are always new ideas in the background in web development and all too often they don't achieve the critical mass of interest and support to become established. I guess time will tell.
[citation needed]
I'm not saying it doesn't exist, but so far I haven't seen any report of malware that can successfully bypass the click-to-play limitations in both the browser and Java itself. There have been some ingenious attacks on parts of the infrastructure, such as the Pawn Storm issue a few months back, but as far as I'm aware even those required the browser itself to have Java enabled and only compromised the Java plugin's security architecture.
As an aside, if such malware did exist, it would self-evidently indicate a browser bug as well as a Java one. That would not be a good sign for a more secure future when browsers are taking on all the jobs plugins used to get.
The UIs I've worked on in that field aren't. :-)