Finding and posting bricks is a lot of effort. The simpler thing is to just open it all, and fill each business reply envelope with another company's junk. It doesn't cost them as much, but it only takes a minute to go through a week's worth of junk mail and do this. If everyone did it, junk mail would be a lot more expensive to send...
Most online bill systems require you to log in to the provider's site (via https) and then download the PDF bill. The main reason for people to still get paper bills for things is that they count to a lot of companies as proof of address (which isn't completely silly - at least they prove that you can get mail from a particular address and, if they bother to check with the sending company, that you could at a specific point in the past). If you're planning on getting a partner's visa in the UK, then having one bill in your name and one bill in your spouse's name delivered every month satisfies the government's requirements to prove cohabitation (which is a bit crazy, but no one expects bureaucracy to be sensible).
The USPS has a complete monopoly on delivering mail to people that it is unprofitable to deliver mail to in the USA. Of course, that's not really the kind of monopoly that most companies covet...
Who cares if the code is unmaintainable? Many of the things that they can be doing are simple things like 'put the sum of these boxes into this' or 'do this 100 times' and are going to be run once then thrown away, but still be faster (and less error-prone) than doing the same simple task 100 times by hand.
Seconded. I always think of screen as 'that thing people used before tmux that too many bugs to actually be useable'. What's it got now that tmux lacks, other than an obnoxious license?
FireFox with the self-destructing cookies plugin is the first Android browser I've found with a cookie-management system that seems to be designed for users and not for advertisers. I tried one of the early versions of FireFox for Android and wasn't impressed, but the latest ones are very nice. I've now switched to using it as the default browser for my phone.
A Cortex A5 is no speed demon compared to the quad-core beasts powering your smartphone. But it would only have to execute dalvik bytecode for an app, with an amd64 chip running the Windows host OS.
Why would it have to run the Dalvik bytecode? The entire point of Dalvik bytecode is that it's architecture agnostic. The ARM core would only have to run the binary code accessed via JNI, and given how much faster the AMD core is than an A5, it's likely to be faster to just emulate that. Unfortunately, people tend to only use native code on Android for things that are really performance critical...
Seconded. I have an AMD E-350 in my NAS. It has 4 SATA ports (+1 eSATA), so that gives one optical drive and 3 drive bays, but leaves one of the drive bays empty and unusable. I went with AMD over Intel because Intel crippled (not sure if it still does) its Atom boards to 2 SATA slots, making them unusable for RAID-Z.
The problem for vegetarians (and more especially for vegans) is not getting enough proteins, it's getting all of the required amino acids (for some reason, the term 'a whole protein' is used to mean 'all of the essential amino acids'). For future reference, by the way, you need around 25g of protein per day, but it has to be balanced among 9 amino acids that the human body can't synthesise (the other 11 can be synthesised from those 9). It's not particularly hard to get all of them - in fact, if you're meeting your calorific quota and not starving then you probably are. Unfortunately, a lot of hippy-vegan recipes that seem to be closely associated with vegetarianism have a terrible mix, so you end up with 3-4 times RDA for some amino acids but only a small amount of others. This led to a lot of vegans in the '60s suffering from amino acid deficiencies, which has led to a belief that it's hard for vegetarians to get enough protein.
Who ever said the goal was to teach people to be good programmers? The goal of English is not to teach children to be novelists. Have you ever watched a non-prgrammer use MS Office and cringed at the fact that they're manually repeating things that could be recorded as a macro or written in a couple of lines of VBA? The point of teaching programming is to make people realise that these complex machines that they're using can be instructed to do repetitive tasks for them. This is useful far beyond working as a full-time programmer.
Whenever I read this kind of counter-argument, I'm reminded of the people who, a hundred or so years ago, argued that there was no point in teaching most people to write because most of them weren't going to be scribes.
The (economic) problem with slavery is not that you don't know how to do anything, it's that you create an imbalance between producers and consumers. Someone using slaves can produce things more cheaply than someone using free labour, because his costs are lower. He consumes less, because his slaves are paid less. Who buys his products? Only other slave owners, so you end up with a very small market. Many of the same problems appear in a modern setting due to wage inequality.
This is why most of the apps I installed on my Android phone come from the F-Droid store. Everything in there was useful to at least one person, not just written in the hope that someone can be persuaded to buy it. Sadly, this does mean that it contains very few games...
OpenSSL has nothing to do with Linux, other than that a number of vendors that bundle it with their products also bundle Linux. The FreeBSD or NetBSD Foundations would have made as much sense (i.e. none).
I switched from giffgaff when they put up their prices and engaged in misleading advertising ('look, we're cheaper than everyone else if you pick the really expensive plans that you have to dig around on their web sites to even find and ignore the ones that are the same price we were offering before we put prices up!'). I guess the difference is what you count as a plan. I regard their goody bags as an add-on, not a plan. On a pre-pay plan you don't get anything included - that's the point. Given that Giffgaff's cheapest goody bag (which expires after a month) costs what I spend on my phone in 3-4 months, I think it reinforces my point. You get unlimited texts only if you buy them in bulk. The 200 minutes and 250MB that the £7.50 goody bag gives you would cost me £8.50, so if I used that much it would be a good deal (although I'd then be paying 7p/minute more for calls above that, so I'd have to be making close to exactly 200 minutes of calls a month for it to make sense). I spend under £2/month on my phone currently though, so it's a pretty poor deal with that in mind.
If you start in London, you don't have to go that far. Cost of living drops by a factor of 3-4 when you get outside London and, as an added bonus, you aren't living in London. Oh, average life expectancy also increases by two years once you stop breathing London 'air' too.
They did release it DRM free if you bought it from them. If you bought it via another publisher then you got some extra crap and had to go back to them to get the DRM-free version. How about next time giving money directly to the company that sells DRM-free games, instead of to a company whose only contribution was to add some DRM crap and put it in a box?
You're assuming that everyone who wants to get an illegal copy needs to crack the DRM. That's not how it works. One person cracks it then releases it on file-sharing sites / networks and everyone copies it. It may prevent casual copying (e.g. I lend a friend the CD), but these days it's easier to give someone a link to a.torrent file than to lend them a CD anyway. More importantly, if someone doesn't know about things like BitTorrent then when they try to copy their game and find that they can't, they're going to ask their favourite search engine and discover that they can get games that they can copy for free. With something like GOG, you get all of the convenience of illegal downloads (actually more - the downloads are a lot faster and they always work), and I get to support the companies that are releasing the games in a way that I want.
The first or the second? I really enjoyed the first, but about the only improvement in the second was the graphics (and my laptop could only handle the lowest detail at a playable rate anyway). The combat was a lot better in the first one and the characters seemed more interesting.
It's a difficult balance in this kind of game between making it open (so the player feels in control of what's happening) and providing a story (because part of the reason for buying the game like this is to be told a story). The first one seemed to get the balance right, but the sequel felt too scripted to me - I was just running from one plot element to the next and then making the four token decisions. There were lots of side-quests in the first one that impacted the story later on and interactions with characters that told you interesting things.
I think the sequel also got off to a bad start, because it let you import your save game from the first one, but after being given a silver sword by a Goddess and a steel sword by a king and finding some legendary armour exploring a tomb, I discovered that the first person I killed had a better sword than me. More importantly, swords and armour made a significant difference in the second. One thing that always annoys me in fantasy games is when the equipment makes more of a difference in fights than the skill. In The Witcher, the difference between a crappy sword stolen from a low-paid henchman and the amazing sword forged for the kind was about 10-20%. Enough to give you a slight edge, but not enough to make a real difference unless a fight was very close. The difference between Geralt at the start and Geralt after he'd (re)learned a load of fighting skills was significant. In contrast, in The Witcher 2, you can get a really good sword and then be easily able to beat monsters that would kill you easily with a less-good sword, without learning any new skills.
The Witcher 1 and 2 were both DRM free. I grabbed both from GOG a while ago and they're just executables that run on any machine and don't have any need to phone home or similar.
Most of their Mac games use DOSBox or WINE, so it probably wasn't too much effort for them to get Linux support working for most of them. Even before they announced Mac support, I ran quite a few of their games with WINE and DOSBox on OS X (their older games use DOSBox on Windows too), but it's a lot less hassle to get their configs (although they tend to be quite pessimistic about visual quality, and you can improve some of the older adventure games a lot by changing the scaling mode to hq3x in the DOSBox config that they ship).
I'm very happy with GOG - there are typically 5-10 games on my shelf that I haven't got around to playing yet. I got The Witcher 1 and 2 as a bundle and enjoyed them both, although I enjoyed the first one a lot more. They're DRM-free and let you redownload games, often with significant updates (e.g. I bought Dungeon Keeper, and they later added the expansion pack. FTL is now FTL: Advanced Edition).
No, he's talking about mitigation, which is a well-known security practice. It's not about obscurity - you can have two or more open source implementations, but it's then harder for the same bug to be in both or all.
To give a concrete example, take a look at the DNS root zone servers operated by Verisign. They run a 50:50 mix of Linux and FreeBSD and increasingly a mix of BIND and Unbound. They use a userspace network stack on some and the system network stack on others. If someone wants to take out the root zone, they need to find exploits for each of these systems. A bug that lets you remotely crash a FreeBSD box likely won't affect Linux and vice versa. That gives them a little bit more time to find the fix (they also massively overprovision, so if someone does take out all of the Linux systems then the FreeBSD ones can still handle the load, and vice versa). If someone finds a bug in BIND then the Unbound servers will be fine.
If your web site were running a mixture of OpenSSL and something else, then it would be relatively easy to turn off the servers running OpenSSL as soon as the vulnerability is disclosed and only put them back online when they've been audited for compromises. Of course, it depends a bit on what your threat model is. If a single machine being compromised is a game-over problem, then you're better off with a monoculture (at your organisation, at least). If having all (or a large fraction) compromised is a problem, but individual compromises are fine, then it's better to have diversity.
Slashdot Mirror
Finding and posting bricks is a lot of effort. The simpler thing is to just open it all, and fill each business reply envelope with another company's junk. It doesn't cost them as much, but it only takes a minute to go through a week's worth of junk mail and do this. If everyone did it, junk mail would be a lot more expensive to send...
Most online bill systems require you to log in to the provider's site (via https) and then download the PDF bill. The main reason for people to still get paper bills for things is that they count to a lot of companies as proof of address (which isn't completely silly - at least they prove that you can get mail from a particular address and, if they bother to check with the sending company, that you could at a specific point in the past). If you're planning on getting a partner's visa in the UK, then having one bill in your name and one bill in your spouse's name delivered every month satisfies the government's requirements to prove cohabitation (which is a bit crazy, but no one expects bureaucracy to be sensible).
The USPS has a complete monopoly on delivering mail to people that it is unprofitable to deliver mail to in the USA. Of course, that's not really the kind of monopoly that most companies covet...
Who cares if the code is unmaintainable? Many of the things that they can be doing are simple things like 'put the sum of these boxes into this' or 'do this 100 times' and are going to be run once then thrown away, but still be faster (and less error-prone) than doing the same simple task 100 times by hand.
Seconded. I always think of screen as 'that thing people used before tmux that too many bugs to actually be useable'. What's it got now that tmux lacks, other than an obnoxious license?
FireFox with the self-destructing cookies plugin is the first Android browser I've found with a cookie-management system that seems to be designed for users and not for advertisers. I tried one of the early versions of FireFox for Android and wasn't impressed, but the latest ones are very nice. I've now switched to using it as the default browser for my phone.
That's assuming that the app vendor actually does ship x86 binaries. Most only ship ARM, a few ship ARM and MIPS, very few also ship x86.
A Cortex A5 is no speed demon compared to the quad-core beasts powering your smartphone. But it would only have to execute dalvik bytecode for an app, with an amd64 chip running the Windows host OS.
Why would it have to run the Dalvik bytecode? The entire point of Dalvik bytecode is that it's architecture agnostic. The ARM core would only have to run the binary code accessed via JNI, and given how much faster the AMD core is than an A5, it's likely to be faster to just emulate that. Unfortunately, people tend to only use native code on Android for things that are really performance critical...
Seconded. I have an AMD E-350 in my NAS. It has 4 SATA ports (+1 eSATA), so that gives one optical drive and 3 drive bays, but leaves one of the drive bays empty and unusable. I went with AMD over Intel because Intel crippled (not sure if it still does) its Atom boards to 2 SATA slots, making them unusable for RAID-Z.
The problem for vegetarians (and more especially for vegans) is not getting enough proteins, it's getting all of the required amino acids (for some reason, the term 'a whole protein' is used to mean 'all of the essential amino acids'). For future reference, by the way, you need around 25g of protein per day, but it has to be balanced among 9 amino acids that the human body can't synthesise (the other 11 can be synthesised from those 9). It's not particularly hard to get all of them - in fact, if you're meeting your calorific quota and not starving then you probably are. Unfortunately, a lot of hippy-vegan recipes that seem to be closely associated with vegetarianism have a terrible mix, so you end up with 3-4 times RDA for some amino acids but only a small amount of others. This led to a lot of vegans in the '60s suffering from amino acid deficiencies, which has led to a belief that it's hard for vegetarians to get enough protein.
Who ever said the goal was to teach people to be good programmers? The goal of English is not to teach children to be novelists. Have you ever watched a non-prgrammer use MS Office and cringed at the fact that they're manually repeating things that could be recorded as a macro or written in a couple of lines of VBA? The point of teaching programming is to make people realise that these complex machines that they're using can be instructed to do repetitive tasks for them. This is useful far beyond working as a full-time programmer.
Whenever I read this kind of counter-argument, I'm reminded of the people who, a hundred or so years ago, argued that there was no point in teaching most people to write because most of them weren't going to be scribes.
I'm not a homophone! Some of my best friends sound the same!
The (economic) problem with slavery is not that you don't know how to do anything, it's that you create an imbalance between producers and consumers. Someone using slaves can produce things more cheaply than someone using free labour, because his costs are lower. He consumes less, because his slaves are paid less. Who buys his products? Only other slave owners, so you end up with a very small market. Many of the same problems appear in a modern setting due to wage inequality.
This is why most of the apps I installed on my Android phone come from the F-Droid store. Everything in there was useful to at least one person, not just written in the hope that someone can be persuaded to buy it. Sadly, this does mean that it contains very few games...
It's a shame then that they chose a name that explicitly excludes large portions of the Free and Open Source Software ecosystem.
OpenSSL has nothing to do with Linux, other than that a number of vendors that bundle it with their products also bundle Linux. The FreeBSD or NetBSD Foundations would have made as much sense (i.e. none).
I switched from giffgaff when they put up their prices and engaged in misleading advertising ('look, we're cheaper than everyone else if you pick the really expensive plans that you have to dig around on their web sites to even find and ignore the ones that are the same price we were offering before we put prices up!'). I guess the difference is what you count as a plan. I regard their goody bags as an add-on, not a plan. On a pre-pay plan you don't get anything included - that's the point. Given that Giffgaff's cheapest goody bag (which expires after a month) costs what I spend on my phone in 3-4 months, I think it reinforces my point. You get unlimited texts only if you buy them in bulk. The 200 minutes and 250MB that the £7.50 goody bag gives you would cost me £8.50, so if I used that much it would be a good deal (although I'd then be paying 7p/minute more for calls above that, so I'd have to be making close to exactly 200 minutes of calls a month for it to make sense). I spend under £2/month on my phone currently though, so it's a pretty poor deal with that in mind.
If you start in London, you don't have to go that far. Cost of living drops by a factor of 3-4 when you get outside London and, as an added bonus, you aren't living in London. Oh, average life expectancy also increases by two years once you stop breathing London 'air' too.
They did release it DRM free if you bought it from them. If you bought it via another publisher then you got some extra crap and had to go back to them to get the DRM-free version. How about next time giving money directly to the company that sells DRM-free games, instead of to a company whose only contribution was to add some DRM crap and put it in a box?
You're assuming that everyone who wants to get an illegal copy needs to crack the DRM. That's not how it works. One person cracks it then releases it on file-sharing sites / networks and everyone copies it. It may prevent casual copying (e.g. I lend a friend the CD), but these days it's easier to give someone a link to a .torrent file than to lend them a CD anyway. More importantly, if someone doesn't know about things like BitTorrent then when they try to copy their game and find that they can't, they're going to ask their favourite search engine and discover that they can get games that they can copy for free. With something like GOG, you get all of the convenience of illegal downloads (actually more - the downloads are a lot faster and they always work), and I get to support the companies that are releasing the games in a way that I want.
The first or the second? I really enjoyed the first, but about the only improvement in the second was the graphics (and my laptop could only handle the lowest detail at a playable rate anyway). The combat was a lot better in the first one and the characters seemed more interesting.
It's a difficult balance in this kind of game between making it open (so the player feels in control of what's happening) and providing a story (because part of the reason for buying the game like this is to be told a story). The first one seemed to get the balance right, but the sequel felt too scripted to me - I was just running from one plot element to the next and then making the four token decisions. There were lots of side-quests in the first one that impacted the story later on and interactions with characters that told you interesting things.
I think the sequel also got off to a bad start, because it let you import your save game from the first one, but after being given a silver sword by a Goddess and a steel sword by a king and finding some legendary armour exploring a tomb, I discovered that the first person I killed had a better sword than me. More importantly, swords and armour made a significant difference in the second. One thing that always annoys me in fantasy games is when the equipment makes more of a difference in fights than the skill. In The Witcher, the difference between a crappy sword stolen from a low-paid henchman and the amazing sword forged for the kind was about 10-20%. Enough to give you a slight edge, but not enough to make a real difference unless a fight was very close. The difference between Geralt at the start and Geralt after he'd (re)learned a load of fighting skills was significant. In contrast, in The Witcher 2, you can get a really good sword and then be easily able to beat monsters that would kill you easily with a less-good sword, without learning any new skills.
The Witcher 1 and 2 were both DRM free. I grabbed both from GOG a while ago and they're just executables that run on any machine and don't have any need to phone home or similar.
Most of their Mac games use DOSBox or WINE, so it probably wasn't too much effort for them to get Linux support working for most of them. Even before they announced Mac support, I ran quite a few of their games with WINE and DOSBox on OS X (their older games use DOSBox on Windows too), but it's a lot less hassle to get their configs (although they tend to be quite pessimistic about visual quality, and you can improve some of the older adventure games a lot by changing the scaling mode to hq3x in the DOSBox config that they ship).
I'm very happy with GOG - there are typically 5-10 games on my shelf that I haven't got around to playing yet. I got The Witcher 1 and 2 as a bundle and enjoyed them both, although I enjoyed the first one a lot more. They're DRM-free and let you redownload games, often with significant updates (e.g. I bought Dungeon Keeper, and they later added the expansion pack. FTL is now FTL: Advanced Edition).
To give a concrete example, take a look at the DNS root zone servers operated by Verisign. They run a 50:50 mix of Linux and FreeBSD and increasingly a mix of BIND and Unbound. They use a userspace network stack on some and the system network stack on others. If someone wants to take out the root zone, they need to find exploits for each of these systems. A bug that lets you remotely crash a FreeBSD box likely won't affect Linux and vice versa. That gives them a little bit more time to find the fix (they also massively overprovision, so if someone does take out all of the Linux systems then the FreeBSD ones can still handle the load, and vice versa). If someone finds a bug in BIND then the Unbound servers will be fine.
If your web site were running a mixture of OpenSSL and something else, then it would be relatively easy to turn off the servers running OpenSSL as soon as the vulnerability is disclosed and only put them back online when they've been audited for compromises. Of course, it depends a bit on what your threat model is. If a single machine being compromised is a game-over problem, then you're better off with a monoculture (at your organisation, at least). If having all (or a large fraction) compromised is a problem, but individual compromises are fine, then it's better to have diversity.
If that's the take-home that you got from that video, I suggest you watch it again. You clearly missed the point.