The point is that the default setting is designed to keep entropy low, giving people a false sense of security. Some user know this and can "enable" stronger passwords, but most people don't know this, which is why I pointed it out (and apparently got downvoted for simply doing so). Information is harmful, kids.
A car with a tablet for a dashboard destined to be a classic? No thanks. Real classics are timeless, not those that cash in on stupid (and functionally compromised) fads.
I have a dash cam. It's terrific. It has a dedicated memory card, no WiFi, and does not integrate into the car's electronics. I am not doing anything for "them" if I own and directly control it myself.
If my insurance company offered to give me a proprietary dash cam in exchange for a discount, I'd tell them to screw off. If car manufacturers make them standard equipment and don't let you control recording or retrieve video on your own, I'd be worried.
Apple, on the other hand, stores photos using end-to-end encryption.
One problem I have with this is that the default passcode is limited to a 6-digit number (digits only). You can change the passcode settings to enable alphanumerics, but the default is just digits.
Given this default lack of entropy, and the fact that the secret keys in the hardware are known to Apple, it's trivial for the company to break the encryption on the vast majority of devices if they really want to. I'm not suggesting they're secretly complying with the government, but they probably do mine the "easy" information for statistical reasons just like every other company.
"Drivers have little incentive to be as careful with company trucks as they would with their own."
I agree that statement is bull. Most warehouses use rental trucks, but I worked for a Fortune 500 company that owned its own trucks. Despite owning the vehicles, the company cared less for them than the drivers did, given how difficult it was to schedule regular maintenance.
One box truck in particular, the one I used to load, had no functioning air brakes and would regularly roll forward out of the loading dock while trying to stack pallets.
This is why you use an arbitrary precision math library for calculators. Programmers need to take floating point precision into consideration, users shouldn't have to.
Don't get me started about Apple "magic". The biggest problem I had with my first Mini (PPC) was that the power cable wouldn't stay attached, because there was no retention clip or even enough friction to keep the cord in the socket. It wasn't broken -- just an entirely defective design. I had to literally duct tape it in place. Just one of many, many problems that the machine had that Apple never admitted or fixed (other problems included an out-of-spec DVI-D connector and a leafblower for a cooler).
It's not just the Mini that's dead to me these days. After 15+ years of experience with various Macs, I've sworn off all Apple products completely.
Hopefully there's more than one account of the incident, which allows for cross-referencing. If the public can upload videos, there's a good chance more than one video will be available.
I've seen plenty of dash cam videos on YouTube where more than one camera angle of a crash was available (from building windows, even). The reason is because it was recorded on multiple cameras by completely different people.
At this point, I pretty much just refuse to buy any game that has a 3rd-party EULA. Almost all the time developers have an extended EULA, it's to explain the extra DRM.
While I agree with that sentiment, I have to wonder why this is such a big deal?
Among the other replies, battery life is an issue. Even if a browser doesn't use every CPU core for a single web page, imagine if every tab has a coin miner built in. That will surely max out your CPU... continuously.
I can't wait for GPU compute acceleration to become universal in web browsers. That means web pages will be able to max out your GPU, too.
We are just a few short steps from asm.js becoming a reality, and all the benefits that will flow from there.
The web was designed to be document centric, not application centric. Every UI designer is supposed to know this and why using executable code to load and render pages is a really dumb thing to do. All JS and WebAssembly are doing is recreating all the bad things about Flash and Java all over again (but these technologies are standards compliant, and that makes it okay!)
UX people don't care about these fundamentals, of course. That's why more and more web pages won't backtrack correctly when I hit the Back button, allow me to open links in new windows, and select text correctly, among many other really frustrating issues.
For most of my needs, I found new software that was compatible with the new OS.
I'm not sure if my comment is applicable to this context, but I'm sure we all understand that the compatibility lifecycle of software is getting shorter. Switching to newer products compatible with the latest OS was easy when you could expect 5-10 years of service out of it. Today, major compatibility-breaking changed are rammed down our throats every 6 months. I don't have the time or energy to completely rebuild my workflow that often, to say nothing about new versions of software being available, let alone as good as the old ones!
As stupid as it sounds, I currently use Photoshop 5.5 on a Win7 system for all my web graphics work. Yes, I'm literally using an 18-year-old copy of Photoshop on an 8-year-old OS, running on 7-year-old hardware. It works for my needs, so I'd rather just not upgrade at all if it were to all break on Kaby Lake, Win10, Firefox 57, etc.
Even if this is possible now, it's not a long-term solution. It would be nicer to be able to throttle scripting. Everything is moving away from document design and towards application design, and with WebAssembly coming, I'm pretty sure it will soon be impossible to browse the web without scripting support.
My feeling is that built-in plugins defeat the whole point, and is indeed the wrong way of doing it.
Compare to something like the K-Lite codec pack, which is available in a "typical" bundle, as well as a "full" bundle with enhanced legacy support. I'm sick of the current fad that says developers need to strip all legacy support and eliminate as many options as possible to avoid confusion. It's really not hard to make a few easy default choices.
You know, I really find it curious that Mint is the go-to distro for "normal people". If you go to their web site to download it, you're presented with an ISO (actually, multiple ISOs depending on what DE you want), and the directions (which must be downloaded as a PDF, of course) tell you to burn the ISO to a DVD. A lot of PCs these days don't have disc drives anymore. If you want directions on how to turn the ISO into a Live USB, you have to go to the forums, and we all know how well that usually goes in the Linux community. If it's so simple to install and use, why is it necessary to tell people to seek help from the forums? Are they so lazy that they can't maintain directions right on the download page? It is seriously not even possible to RTFM because they don't bother to include the info you need?
The last time I tried Mint from a DVD ISO, it also took about 22 minutes to boot, most of the time leaving me to stare at a blank screen without even a fake progress indicator. At several points I was wondering if my PC froze and whether I should reboot. Running the installer in a VM instead of natively didn't make things any easier.
Right from step 1, Mint is not very convincing that it's user friendly, just like every other Linux distro I've tried in the last 15 years. These are the things that prevent Linux from being taken seriously (and never get fixed) decade after decade.
Slashdot Mirror
Any time someone says the free market can police itself, refer them to situations like this.
Explain to me why CHOICE isn't a Good Thing(tm)?
Strawman.
The point is that the default setting is designed to keep entropy low, giving people a false sense of security. Some user know this and can "enable" stronger passwords, but most people don't know this, which is why I pointed it out (and apparently got downvoted for simply doing so). Information is harmful, kids.
A car with a tablet for a dashboard destined to be a classic? No thanks. Real classics are timeless, not those that cash in on stupid (and functionally compromised) fads.
Hate the programmer, not the programming language.
Applies to all things except Javascript.
Well, okay, a bad programmer can make Javascript worse, but still...
I have a dash cam. It's terrific. It has a dedicated memory card, no WiFi, and does not integrate into the car's electronics. I am not doing anything for "them" if I own and directly control it myself.
If my insurance company offered to give me a proprietary dash cam in exchange for a discount, I'd tell them to screw off. If car manufacturers make them standard equipment and don't let you control recording or retrieve video on your own, I'd be worried.
Apple, on the other hand, stores photos using end-to-end encryption.
One problem I have with this is that the default passcode is limited to a 6-digit number (digits only). You can change the passcode settings to enable alphanumerics, but the default is just digits.
Given this default lack of entropy, and the fact that the secret keys in the hardware are known to Apple, it's trivial for the company to break the encryption on the vast majority of devices if they really want to. I'm not suggesting they're secretly complying with the government, but they probably do mine the "easy" information for statistical reasons just like every other company.
"Drivers have little incentive to be as careful with company trucks as they would with their own."
I agree that statement is bull. Most warehouses use rental trucks, but I worked for a Fortune 500 company that owned its own trucks. Despite owning the vehicles, the company cared less for them than the drivers did, given how difficult it was to schedule regular maintenance.
One box truck in particular, the one I used to load, had no functioning air brakes and would regularly roll forward out of the loading dock while trying to stack pallets.
Fat chance. Remember how all the browsers riled against the status bar, saying people didn't need it?
This is why you use an arbitrary precision math library for calculators. Programmers need to take floating point precision into consideration, users shouldn't have to.
If software is designed like shit and not properly tested before release, all the updates in the world won't fix a damn thing.
Today's definition of "rapid release" is not the same thing as the old UNIX "release early, release often".
And of course the MagicPower supply.
Don't get me started about Apple "magic". The biggest problem I had with my first Mini (PPC) was that the power cable wouldn't stay attached, because there was no retention clip or even enough friction to keep the cord in the socket. It wasn't broken -- just an entirely defective design. I had to literally duct tape it in place. Just one of many, many problems that the machine had that Apple never admitted or fixed (other problems included an out-of-spec DVI-D connector and a leafblower for a cooler).
It's not just the Mini that's dead to me these days. After 15+ years of experience with various Macs, I've sworn off all Apple products completely.
Hopefully there's more than one account of the incident, which allows for cross-referencing. If the public can upload videos, there's a good chance more than one video will be available.
I've seen plenty of dash cam videos on YouTube where more than one camera angle of a crash was available (from building windows, even). The reason is because it was recorded on multiple cameras by completely different people.
So, the price will go up (again) or it won't fit its niche anymore?
At this point, I pretty much just refuse to buy any game that has a 3rd-party EULA. Almost all the time developers have an extended EULA, it's to explain the extra DRM.
First of all, most respectable websites will never do anything like that.
...at the moment.
We're not talking about something that's illegal. Companies will (eventually) do anything they can get away with.
While I agree with that sentiment, I have to wonder why this is such a big deal?
Among the other replies, battery life is an issue. Even if a browser doesn't use every CPU core for a single web page, imagine if every tab has a coin miner built in. That will surely max out your CPU... continuously.
I can't wait for GPU compute acceleration to become universal in web browsers. That means web pages will be able to max out your GPU, too.
We are just a few short steps from asm.js becoming a reality, and all the benefits that will flow from there.
The web was designed to be document centric, not application centric. Every UI designer is supposed to know this and why using executable code to load and render pages is a really dumb thing to do. All JS and WebAssembly are doing is recreating all the bad things about Flash and Java all over again (but these technologies are standards compliant, and that makes it okay!)
UX people don't care about these fundamentals, of course. That's why more and more web pages won't backtrack correctly when I hit the Back button, allow me to open links in new windows, and select text correctly, among many other really frustrating issues.
For most of my needs, I found new software that was compatible with the new OS.
I'm not sure if my comment is applicable to this context, but I'm sure we all understand that the compatibility lifecycle of software is getting shorter. Switching to newer products compatible with the latest OS was easy when you could expect 5-10 years of service out of it. Today, major compatibility-breaking changed are rammed down our throats every 6 months. I don't have the time or energy to completely rebuild my workflow that often, to say nothing about new versions of software being available, let alone as good as the old ones!
As stupid as it sounds, I currently use Photoshop 5.5 on a Win7 system for all my web graphics work. Yes, I'm literally using an 18-year-old copy of Photoshop on an 8-year-old OS, running on 7-year-old hardware. It works for my needs, so I'd rather just not upgrade at all if it were to all break on Kaby Lake, Win10, Firefox 57, etc.
As with all technology that works just fine and is obsoleted by force -- give it time.
"We worry about technology because we lack confidence in ourselves, and imagination for the future."
I worry about technology because I lack confidence in business ethics and regulation.
If the AI is cheap enough, it doesn't even have to be as good as real people, either. There's always "good enough."
That's why I think jobs will disappear a LOT faster than people think.
Even if this is possible now, it's not a long-term solution. It would be nicer to be able to throttle scripting. Everything is moving away from document design and towards application design, and with WebAssembly coming, I'm pretty sure it will soon be impossible to browse the web without scripting support.
My feeling is that built-in plugins defeat the whole point, and is indeed the wrong way of doing it.
Compare to something like the K-Lite codec pack, which is available in a "typical" bundle, as well as a "full" bundle with enhanced legacy support. I'm sick of the current fad that says developers need to strip all legacy support and eliminate as many options as possible to avoid confusion. It's really not hard to make a few easy default choices.
You know, I really find it curious that Mint is the go-to distro for "normal people". If you go to their web site to download it, you're presented with an ISO (actually, multiple ISOs depending on what DE you want), and the directions (which must be downloaded as a PDF, of course) tell you to burn the ISO to a DVD. A lot of PCs these days don't have disc drives anymore. If you want directions on how to turn the ISO into a Live USB, you have to go to the forums, and we all know how well that usually goes in the Linux community. If it's so simple to install and use, why is it necessary to tell people to seek help from the forums? Are they so lazy that they can't maintain directions right on the download page? It is seriously not even possible to RTFM because they don't bother to include the info you need?
The last time I tried Mint from a DVD ISO, it also took about 22 minutes to boot, most of the time leaving me to stare at a blank screen without even a fake progress indicator. At several points I was wondering if my PC froze and whether I should reboot. Running the installer in a VM instead of natively didn't make things any easier.
Right from step 1, Mint is not very convincing that it's user friendly, just like every other Linux distro I've tried in the last 15 years. These are the things that prevent Linux from being taken seriously (and never get fixed) decade after decade.
An emulator that runs as a plugin should not result in bloat unless it's done incredibly wrong.
I suspect that web browsers have forever tainted our expectations of how plugins work and how they can actually minimize bloat, not cause it.