You know, HDCP could actually have a use here if the display data are not otherwise encrypted. I don't know how well it would actually protect against eavesdroppers. At the very least it would remove any repeating patterns which eavesdroppers could use for statistical analysis.
It would be silly if it wasn't encrypted, but there are often some flaws in new standards like this one. (I assume it's easier to eavesdrop on a 60 GHz signal than on actual wires, but there may not be that huge a difference.)
The distinction for BitTorrent is incorrect: you can seed and leech behind an unpenetrable NAT, but you can't connect to other peers which are behind NAT
Technically, it's extremely stupid to rely on URLs to remain secret. On many browers, they are transmitted to a search provider as you type them. I've had Googlebot start hitting my personal web server, and I think it's because I type the URL on my Android phone. (can't be 100 % sure, I posted links to single files on it on IRC and once on slashdot. Would be an interesting experiment.). The URLs are also stored in the history, of course, and the cache files are possibly readable by anone on the system.
It doesn't follow that computing becomes ubiquitous when the energy use goes to zero. What's missing is the *price* of the chips. It's not like it's going to be magically cheaper to make a low-power i7 than to make the current one. In fact, even with the current power usage, you could stick an ARM chip in almost anything and have it do useful calculations. I mean, it's not the *power* that prevents us from having internet-connected light switches, locks, smoke alarms etc. now. You could have "the internet of things" making a comeback, now that we have the address space. The "thing" is either going to be hooked up to Ethernet, in which case power is not an issue (PoE), or it's going to be limited by the power use of a radio.
Don't forget GNOME 3 Fallback Mode! I got a new computer at work which runs Fedora and GNOME 3 in fallback mode and it's surprisingly good. The task switcher is the good simple one which switches between windows, and everything else is almost where it should be. It's what should have become GNOME 3 IMHO.
Well, it's DRM by definition. We've been through this hundreds of times on slashdot why DRM doesn't work. The good thing is that Hollywood seems OK with the current situation. The bad news is that they should be super-happy about tablets and consoles, etc., which limit what software you can use, and that could mean the end of cheap PCs. Closing the analog hole (including "the recording device hole") would require regulation at a massive scale, so that seems unlikely.
The nerve of releasing a digital only connector when the only good reason to use 30-pin over micro-USB was analogue audio out. If it has a better mechanical design than micro-USB I'll give them a pass, though (after having mine half destroyed after 2 years).
It's official. Intel on-board video is all you'll ever need for home and general office use.
Agreed, but don't confuse this with "you should recommend integrated graphics to home users", though. Your examples are perfectly tuned for Intel graphics because many people have them. For a business, that's fine, you only need a fixed set of applications. For a home user, it's likely to be some flash game, Google Earth or some software that works much better with a dedicated graphics card. Good ones are quite cheap now, and if your looking at a "i5", spending some extra on a GPU gives more bang for the bukc than getting a faster CPU. For laptops, it's up to the individual user if they want battery power or performance.
It *is* more convenient. Assuming I didn't need USB for anything other than charging, I'd get a charging pad for $ 5 instead of a practically free USB cable. I'd need one at home, one at work and one for travelling, but for the latter it may be more convenient with plain USB. On a fixed "installation" I'd imagine it's one of those things that seem insignificant at first, but really do make a difference.
My USB port was recently broken, so it will charge, but it won't connect as a mass storage device. I still haven't found an application that allows me to transfer to/from my phone as easily as with USB. I use "LazyDroid", but it's quite insecure to do it over plaintext HTTP, and it isn't possible to download multiple files (e.g. pictures) in one go. USB is also significantly faster than WiFi. Perhaps it has improved over 1-2 generations, but my current Nexus S has incredibly slow WiFi, about 300 kB/s. This is with a very strong signal. And if I'm travelling, I don't want to connect to a wireless network (if there's one around) just to sync some things off my phone. If I had a removable SD card, now that would be something else, I wouldn't mind losing USB at all.
It's 2012, we have the bandwidth and storage to go to lossless. There's really no need to sample at anything above 48 kHz, and 16 bit is enough for human ears, so the required space isn't too large.
We're seeing a damned revolution in the ability of people to create content, and you're bitching about "only good for consuming?"
I'd say it's somewhere in between... PCs are excellent for manipulating text and for precision manipulation of graphics. They are obviously best at computing-heavy tasks.
Tablets may be great for interfaces where you have a limited but large number of options, which on a PC would be hidden in menus, but tablets can use the full screen for interactivity + gestures. They also seem good for timing-sensitive input tasks (e.g. real-time music editing). It seems that some of the innovation is not due to the tablet/touch form factor, but instead due to the delivery and payment options -- many small developers try out new things.
Blizzard should have encrypted the info with a public key to solve these problems.
PKI encryption wouldn't solve the authenticity problem. They would have to pull out the big DRM guns and include some secret (time dependent) string that their servers know, and an authentic WoW client has access to, but which the faker tools couldn't get.
This is a minor privacy leak, when someone creates a program to decode the watermarks. It will also be worthless for authenticating screenshots, because when someone can read the watermarks, it doesn't take much to fake one. Blizzard should have encrypted the info with a public key to solve these problems.
As it stands, it may be useful for others than Blizzard, to identify the origin of a screenshot (in a non-adversarial situation)
Regarding my second paragraph, an important part was not obvious: Each session in the session database has a unique ID, and each anonymised user in the middle database has a list of sessions, and each user in the top database points to an anonymised user.
Great idea for some cases. If you need "telemetry" data to understand how people are using your application, assign each session a unique ID and don't store which user did it. It also works for some other statistical data. The argument against is that you may need the correlation between sessions later.
Depending on the application, you could have a hierarchical system of databases where the lowest level contains session information, the next contains persistent user information but not personally identifiable info, and the highest level contains username, password, name, etc. You could have just a few components that have access to the top level, including the login component. The latter could load the information into the session state, so you could display the username on every page, for example. It's just something I thought of, I'm not an expert (I wrote a quiz and a page for a small event 8 years ago;).
24-pin RGB/TTL you can drop down even to as low as 15 pin by ignoring the higher-res bits; you can reduce the clock-rate to run a 320x240 LCD or you can ramp it up to run 2048x2048 @ 30fps in full colour.
So they design a new modular computing interface for phones, tablets and computers, but it can't even support the current benchmark iPad 2 (2048×1536 at 60Hz, refresh rate not confirmed)?
I can't really see a huge benefit to battery-backed caches. Scattered writes can be aggregated in RAM to make them sequential, and this is easier for filesystems with copy-on-write. As long as the drives implement NCQ correctly, the filesystem can arrange the writes such that it remains in a consistent state. The throughput is limited by the drives anyway, so it shouldn't matter if the writes are scheduled in the drive's controller or in software.
For synchronous writes you can't buffer in RAM, but software shouldn't be calling fsync() a lot anyway. NFS most certainly does, though. There are filesystems which allow you to have a separate device for synchronous writes, for example journal devices in XFS and log devices in ZFS.
Slashdot Mirror
You know, HDCP could actually have a use here if the display data are not otherwise encrypted. I don't know how well it would actually protect against eavesdroppers. At the very least it would remove any repeating patterns which eavesdroppers could use for statistical analysis.
It would be silly if it wasn't encrypted, but there are often some flaws in new standards like this one. (I assume it's easier to eavesdrop on a 60 GHz signal than on actual wires, but there may not be that huge a difference.)
The distinction for BitTorrent is incorrect: you can seed and leech behind an unpenetrable NAT, but you can't connect to other peers which are behind NAT
Technically, it's extremely stupid to rely on URLs to remain secret. On many browers, they are transmitted to a search provider as you type them. I've had Googlebot start hitting my personal web server, and I think it's because I type the URL on my Android phone. (can't be 100 % sure, I posted links to single files on it on IRC and once on slashdot. Would be an interesting experiment.). The URLs are also stored in the history, of course, and the cache files are possibly readable by anone on the system.
It doesn't follow that computing becomes ubiquitous when the energy use goes to zero. What's missing is the *price* of the chips. It's not like it's going to be magically cheaper to make a low-power i7 than to make the current one. In fact, even with the current power usage, you could stick an ARM chip in almost anything and have it do useful calculations. I mean, it's not the *power* that prevents us from having internet-connected light switches, locks, smoke alarms etc. now. You could have "the internet of things" making a comeback, now that we have the address space. The "thing" is either going to be hooked up to Ethernet, in which case power is not an issue (PoE), or it's going to be limited by the power use of a radio.
Don't forget GNOME 3 Fallback Mode! I got a new computer at work which runs Fedora and GNOME 3 in fallback mode and it's surprisingly good. The task switcher is the good simple one which switches between windows, and everything else is almost where it should be. It's what should have become GNOME 3 IMHO.
Well, it's DRM by definition. We've been through this hundreds of times on slashdot why DRM doesn't work. The good thing is that Hollywood seems OK with the current situation. The bad news is that they should be super-happy about tablets and consoles, etc., which limit what software you can use, and that could mean the end of cheap PCs. Closing the analog hole (including "the recording device hole") would require regulation at a massive scale, so that seems unlikely.
The nerve of releasing a digital only connector when the only good reason to use 30-pin over micro-USB was analogue audio out. If it has a better mechanical design than micro-USB I'll give them a pass, though (after having mine half destroyed after 2 years).
It's official. Intel on-board video is all you'll ever need for home and general office use.
Agreed, but don't confuse this with "you should recommend integrated graphics to home users", though. Your examples are perfectly tuned for Intel graphics because many people have them. For a business, that's fine, you only need a fixed set of applications. For a home user, it's likely to be some flash game, Google Earth or some software that works much better with a dedicated graphics card. Good ones are quite cheap now, and if your looking at a "i5", spending some extra on a GPU gives more bang for the bukc than getting a faster CPU. For laptops, it's up to the individual user if they want battery power or performance.
Sorry, didn't look at the graph. This looks like a great development indeed, for things like VoIP and video streaming.
It *is* more convenient. Assuming I didn't need USB for anything other than charging, I'd get a charging pad for $ 5 instead of a practically free USB cable. I'd need one at home, one at work and one for travelling, but for the latter it may be more convenient with plain USB. On a fixed "installation" I'd imagine it's one of those things that seem insignificant at first, but really do make a difference.
Why do I need a USB port anymore?
My USB port was recently broken, so it will charge, but it won't connect as a mass storage device. I still haven't found an application that allows me to transfer to/from my phone as easily as with USB. I use "LazyDroid", but it's quite insecure to do it over plaintext HTTP, and it isn't possible to download multiple files (e.g. pictures) in one go. USB is also significantly faster than WiFi. Perhaps it has improved over 1-2 generations, but my current Nexus S has incredibly slow WiFi, about 300 kB/s. This is with a very strong signal. And if I'm travelling, I don't want to connect to a wireless network (if there's one around) just to sync some things off my phone. If I had a removable SD card, now that would be something else, I wouldn't mind losing USB at all.
It's 2012, we have the bandwidth and storage to go to lossless. There's really no need to sample at anything above 48 kHz, and 16 bit is enough for human ears, so the required space isn't too large.
They're good for consuming, but not much else.
We're seeing a damned revolution in the ability of people to create content, and you're bitching about "only good for consuming?"
I'd say it's somewhere in between ... PCs are excellent for manipulating text and for precision manipulation of graphics. They are obviously best at computing-heavy tasks.
Tablets may be great for interfaces where you have a limited but large number of options, which on a PC would be hidden in menus, but tablets can use the full screen for interactivity + gestures. They also seem good for timing-sensitive input tasks (e.g. real-time music editing). It seems that some of the innovation is not due to the tablet/touch form factor, but instead due to the delivery and payment options -- many small developers try out new things.
Disclaimer: i have barely touched a tablet
Oh, + some hash of the picture data to check that it's the correct picture.
Correcting myself,
Blizzard should have encrypted the info with a public key to solve these problems.
PKI encryption wouldn't solve the authenticity problem. They would have to pull out the big DRM guns and include some secret (time dependent) string that their servers know, and an authentic WoW client has access to, but which the faker tools couldn't get.
Cool discovery.
This is a minor privacy leak, when someone creates a program to decode the watermarks. It will also be worthless for authenticating screenshots, because when someone can read the watermarks, it doesn't take much to fake one. Blizzard should have encrypted the info with a public key to solve these problems.
As it stands, it may be useful for others than Blizzard, to identify the origin of a screenshot (in a non-adversarial situation)
A possible first step would be to make the companies fully responsible when there's a software glitch. That would be a good start anyway. For an example of a bug, see: http://developers.slashdot.org/story/12/08/02/165206/algorithmic-trading-glitch-costs-firm-440-million . They canceled some of the orders -- if they do that for everyone, then OK, but HFTers shouldn't get any preferential treatment.
Avoid local storage inside the device for sensitive information
That does make sense, but it still feels like I've fallen into opposite land.
Avoid writing custom crypto code – prone to vulnerability
Yes! I'll repeat it a couple of times
Avoid writing custom crypto code – prone to vulnerability
Avoid writing custom crypto code – prone to vulnerability
Regarding my second paragraph, an important part was not obvious: Each session in the session database has a unique ID, and each anonymised user in the middle database has a list of sessions, and each user in the top database points to an anonymised user.
Great idea for some cases. If you need "telemetry" data to understand how people are using your application, assign each session a unique ID and don't store which user did it. It also works for some other statistical data. The argument against is that you may need the correlation between sessions later.
Depending on the application, you could have a hierarchical system of databases where the lowest level contains session information, the next contains persistent user information but not personally identifiable info, and the highest level contains username, password, name, etc. You could have just a few components that have access to the top level, including the login component. The latter could load the information into the session state, so you could display the username on every page, for example. It's just something I thought of, I'm not an expert (I wrote a quiz and a page for a small event 8 years ago;).
It shows the ads, and asks you to connect to a WiFi network if you click on them.
Made me chuckle a little.. I suppose they work on the theory that any exposure is good for advertisement
that looks pretty cool!
So we have a new plan: get portable music players like iPods, and plug those into the aux input in the intercom system.
If you haven't already, please check that you get decent sound quality through the intercom.
24-pin RGB/TTL you can drop down even to as low as 15 pin by ignoring the higher-res bits; you can reduce the clock-rate to run a 320x240 LCD or you can ramp it up to run 2048x2048 @ 30fps in full colour.
So they design a new modular computing interface for phones, tablets and computers, but it can't even support the current benchmark iPad 2 (2048×1536 at 60Hz, refresh rate not confirmed)?
I can't really see a huge benefit to battery-backed caches. Scattered writes can be aggregated in RAM to make them sequential, and this is easier for filesystems with copy-on-write. As long as the drives implement NCQ correctly, the filesystem can arrange the writes such that it remains in a consistent state. The throughput is limited by the drives anyway, so it shouldn't matter if the writes are scheduled in the drive's controller or in software.
For synchronous writes you can't buffer in RAM, but software shouldn't be calling fsync() a lot anyway. NFS most certainly does, though. There are filesystems which allow you to have a separate device for synchronous writes, for example journal devices in XFS and log devices in ZFS.