I think I've mentioned it before in the past, though in another context: "The Time Ships", by Stephen Baxter. It's an official sequel to H. G. Wells' "The Time Machine".
The book itself is obviously not about education, but that subject is brought about when the protagonist gets into contact with another species (which I'll not name lest I spoil the book). Basically, the approach to education for that species is that children would be taught how to seek information, and then pretty much just told to go educate themselves, seeking out whatever they want.
Of course that would be utopia for humans, because we are mostly hedonistic by nature, but all the same this idea from that book really stuck with me, and made me realize in which way our educational system is a failure: children are usually just told to memorize stuff, a big part of which they will never really use, when they really should be taught how to "think" - how to seek information, stimulate curiosity and solve problems with information they gather themselves.
Fortunately, with initiatives such as these, it seems as though this is slowly changing.
This "plainly" here is difficult to judge. How can we be sure any unbelievable religious text wasn't some sort of "pointed political statement" or "satire"?
You're right, we can't be sure. The difference from this to the FSM cult, however (and what probably influenced the judge's ruling), is historical context - yes, we can't be sure the biblical accounts were derived as a political statement or satire, and what little historical context we have don't seem to point this way, so there's plenty of room for interpretation either by religious people or atheists.
The historical context for the FSM cult, however, is well-known, as is known that it is a political statement and satire.
The thing though is - and maybe the GP was trolling and I'm falling for the bait - as far as the rest of the world that is not North America (and maybe Japan too) was concerned, the Spectrum *was* a world-wide phenomenon. He could have used the argument that the C64 was technically superior, and then there would be few people to argue that. But market-wise, as you correctly pointed out, the Spectrum beat the shit out of the C64, and not only on the UK, but most of the world.
Case in point: I'm from Brazil, we didn't have legit Spectrums here back then, but we had locally-made clones, which amounted to the same thing and ended up getting exported to all of South America, and you know what? The only way anyone on Latin America knew the C64 actually existed was that it was often mentioned on computer magazines, and that was it - I never knew anyone who even heard about of the C64 around here, let alone owned one. We heard a lot about the TRS-80, Apple II (or rather its clones that were produced locally), MSX and so on, and it wasn't uncommon to find users of such systems, but the C64? Nada. I understand the C64 actually managed to chew a bit more of the market on some parts of europe, but the Spectrum was still far more popular. IIRC, on Russia the situation was similar to South America, in that they had Spectrum clones, and the C64 was a computer only the US cared about.
However, I'm disappointed they have yet to port it to Android, and instead chose to focus on porting to a browser platform, which to me counts as a useless gimmick since it's only really playable on browsers on a desktop platforms, for which they have ports for all platforms anyway.
I'd like to bring this discussion down a few notches and ask a more "stupid" question, to which I haven't yet seen a definite answer yet, and it looks like you guys could easily answer that. Sorry if it's a too a stupid question.
What is the performance impact of encrypting an Android device? I'm not talking necessarily about Android L, but any previous version that supports encryption (it was added on ICS, I think?). My concern in fact is with older devices, think 2012 or '13 phones. Are any of such older devices capable of hardware accelerating crypto operations (besides RNG)?
The impression I'm getting here is that people is siding with Linus without really understanding what the discussion was about, and how it really unfolded. It's like people are painting the girl who started it as a diva who gets offended at cursing.
I read the whole thread though (gasp!), and it was a rather civil, insightful, and even funny discussion, at both sides of the argument. Ironically, I think the girl who started the discussion cursed even more than Linus during the discussion. Reading the thread, I've got the impression that the subject wasn't so much about cursing and being disproportionately mad at times, but just "toning it a bit down", and minding to whom Linus would burst out, and I can she has a point. The problem is, Linus tends to burst out at top-level developers (because of their greater responsibility, making mistakes for them are more damaging), and this tends to scare away inexperienced developers, while also impacting morale and reducing top-developers authority. And it's hard not to agree with that - there's only so much abuse a top-developer can take (Alan Cox anyone?).
I'm not sure I'm really siding with the girl either though, because I can see Linus has some valid reasons for acting like that, I could see *myself* acting the same way in his position, and I'm a generally a laid back guy. That's the kind of discussion that it's not all black and white.
There has been a lot of backlash on their blog about this: Why didn't they just go with XMPP? What their protocol have that XMPP doesn't, or couldn't be extended to support?
Personally - just a guess (also, btw, disclaimer: I'm a subscriber) - I think they're dying. Their client haven't been getting any significant development for the past year, current issues with some protocols have been going unaddressed, and new features like Lync protocol support (which there are working OSS implementations) have been going completely ignored despite many people clamoring for it.
So, they have been silent for a long time, and now this. It's fishy.
I agree with this argument on Windows. On OSX though, 32-bit chrome is a problem incidentally because of Java: on recent updates (past year) the 32-bit Java plugin on OSX was disabled. You can say what you want about Java, its vulnerabilities and shortcomings as a platform, but the fact is that many sites (banks or such) still require it, and that means I have to use Safari for those sites. It's not a big problem, but it's incovenient, and OSX, compared to Windows, has a much higher ratio of software and system components with 64-bit binaries than Windows, so there's very little reason to keep Chrome clinging to a legacy binary format on that platform.
Honest question: I do have an updated PS3 (yeah slashdot, judge me). I'm not interested on pirated games, but I may be interested on homebrew stuff (emulators and stuff like that). That leak will make that possible for me?
Last update was on December 2010 - so it's a fair to assume the first version was submitted even earlier. And that's just one example I could find quickly, of course. It wouldn't surprise me it there are many more other apps (for Android or iOS alike) that does the same thing and was made before.
And yeah, as rolfwind said, just because the idea was implemented only after 10 years after Microsoft entered the smartphone market, doesn't mean it's patentable. The technology needed for this idea wasn't ubiquotous on smartphones until some 4 or 5 years ago anyway, so you should rather start making the math at that point in time.
The nice thing is that it's multi-platform, and it actually looks and works nice on other OSs - it's particularly nice on OS X, for those like me that hate iTunes.
Well, for starters, LMDE is related to Debian Testing, not Squeeze (which is currently the Stable distribution). So you should get more updated packages on LMDE vs. Squeeze, yes, and contrary to Debian, LMDE's packages should be updated more frequently and often to the latest upstream software releases.
Basically, LMDE is Debian Testing with some specific Mint packages (that usually are intended to improve and simplify user experience and add the "Mint" branding over Debian) overlayed on it. You should read my reply above to wrook detailing how on the latest snapshot releases LMDE is not really using the Debian Testing repositories directly, but you can easily make it do so and there shouldn't be any problems by doing that - I haven't tried myself though, so YMMV; I'm quite happy as it is.
I tried using it at the end of last year and I didn't get *any* updates for the 3 or 4 months I was using it. Not even security updates.
That's because since the penultimate snapshot release of LMDE (which was at some time on the last quarter of last year) they've switched the repositories to use their own copy of a snapshot of Debian Testing, and every 5 months or so they release what they call "update packs", which is basically a more recent snapshot of Debian Testing with the packages therein more throughly tested for bugs and such. Before they started doing that, LMDE used the standard Debian Testing repositories.
The intention on having a "snapshot repository" is to try to get the best of both worlds of rolling releases distributions and version-based distributions (i.e. Ubuntu). It's certainly not perfect, but works well enough for me, I don't mind it. Anyway, the maintainers says it's fine to swap out those Mint "snapshot repositories" with the standard Debian Testing repositories on sources.conf - of course, you're more likely to stumble into problems from time to time, and hence should be prepared to work around them or muddle through with them, but it shouldn't be any more problematic than using plain Debian Testing.
Mint is nice, and it's the Linux flavor I'm using currently (although I use LMDE, not the standard Mint) after having left Ubuntu when they transitioned to Unity. The best thing about it is that the maintainer(s) actually listen to users regarding development directions, which was what drove them to develop Cinnamon and adopt MATE as an option - as opposed to Ubuntu / Canonical, that just forced down the users' throats their ideas and UI decisions, alienating a large part of their user base in the process.
Having said that, there's still one thing that keeps me from recommending it to new users or users migrating out of Ubuntu: lack of automated upgrade procedure to newer major versions - one thing that Ubuntu has and generally works nice there. On Mint, the official procedure is to backup you files/settings using the backup tool, install the newer version from scratch on top of the existing install, and then restore the backup after. That's just too cumbersome. Yes, it's possible to upgrade without reinstalling by manually editing the sources.list file and upgrading manually with apt-get, but it's considered unsafe and error prone by the maintainers and hence not recommended. I did it anyway on a past install, and sure enough I had hiccups - I still had a working install, but there were a lot of rough edges and inconsistencies on the upgraded install. Because of that I ended up installing LMDE so I didn't have to worry about major version upgrades anymore. It's not a fully smooth ride either, but it's far more manageable, and having previous experience with Debian, I'm totally at home with it. But it's obviously not something I'd recommend to casual / new users either.
This. As PJ points out on Groklaw, although the judge alledgedly does code, he still seems to be having a hard time deciding whether APIs can be copyrighted. I wonder if he "does coding" on Basic...:-P
Well, I disagree. I think the motivations and some of the changes are quite nice, but they are badly misguided. Both Ubuntu and MS are making the mistake of trying to cram UI concepts not suitable to the environment they are going to be used: - Metro UI: Awesome for tablets and touch interfaces in general - I'd go as far as saying it's the best UI concept on these cases, better even than iOS. But that concept just doesn't make any sense and it's counter-productive when you're interfacing with your device using a keyboard and mouse. - Unity: Same thing, doesn't translate well to desktops, but it's worse because this UI doesn't really work well with touch interfaces either; it really was made for and works well only on a dying breed of devices: Netbooks. - Gnome 3: I'll refrain from too much comment on this one because I haven't really used it yet, but from what I've seen, this is the one that's actually the saner of the other two in the sense that they are not trying to jam a square in a peg hole like the other two are doing - they are really trying to redefine traditional UI concepts on the desktop, and maybe other environments too. That's not to say that they're on the ball though: I think they've really gone way too radical there. What I'm really saying is that Gnome is the less bad of bunch.
I don't like that my G+ profile shouts my real name everywhere too. So I was looking around in my profile, and guess what: There's a "Nickname" field in there - but the profile form explicitly says that it won't be shown in the profile. Why the hell are they doing that? Why have this field if it's not going to be used?
The dumb part is that Google could be fixing this problem in a much less disrupting way: Make the "Nickname" field actually useful, make it the default field shown for the public, or have the user setup if he wants the Nickname to be shown or his real name. Hell, if Google is so bent into real names, at least make the Nickname the field to be shown to the public, and the real name only to your friends / circles. What a waste.
Hm, that sounds dangerous. I have a triple-boot setup on my MacPro (OSX, Windows 7 and Linux). How will Lion handle that? Sounds like it could trash one (or both) of my other OSes in the process...
Being able to download the kernel, driver, and Android sources directly from Motorola, the maker of my Droid phone, is so prohibitive.
Good luck getting your recompiled kernel+driver+Android sources past the well-locked-down bootloader on any Motorola Android device newer than the original Droid.
Fair enough. However, this is not Android's (the OS) fault - the bootloader locking mechanism is hardware based, so only Motorola's to blame here. As always in these cases, all you can do is vote with your wallet: get a HTC or some other brand that doesn't lock you out of your property.
Please mod the parent up to the moon. I've been reading all the comments on the story, and it baffled me that no one caught this simple fact - GPS logs can be easily tampered with, or even forged. I don't know if the ruling on this case was made with this technical knowledge, but it's nevertheless a good thing a bad precedent wasn't set. Nevermind GPS accuracy, GPS logs just aren't a reliable source of evidence in the first place.
Yea, I think it's a bit elitist too. I mean, if they don't want a sequel, don't read it!
Case in point, a classic: The Time Machine, from H.G. Wells. A century later, a sequel was authorized and written by Stephen Baxter: The Time Ships. And I like it so much more than the first book, because it expands so much on the idea, concepts and caracter. Granted, there was a lot to expand from given the late 19th century science, and Stephen Baxter is also an excellent SF writer... So the question really is if Mr. Reichert is up to the task, since he's pretty much unkown. But so was Baxter, back when he wrote the The Time Ships. I guess we'll just have to wait and hope that Mr. Reichert does a good job.
Nice, but your Quake has statically linked libraries, which was (and still is) a simple solution for running binaries across wildly different distributions or linux environments without hassle.
Of course, not everyone does that. The common and default way to compile stuff is using dynamic linking (i.e. using the standard libraries you'd already have installed), and if your version of Quake (compiled 10 years ago) was dinamically linked, you'd be in for a big surprise.
And anyway, while it is a good strategy for backwards compatibility and cross-distribution portability, static linking is not fullproof and has its drawbacks - you could still run in to compatibility problems that are kernel-bound (usually, interfaces that have been changed or obsoleted), in which case static linking won't help you, and statically linked binaries are bloated (thus, taking a lot more memory). You really just got lucky with that version of Quake.
Well, I can't vouch for the GP, but my ISP has a very flaky DNS service. For some reason, every 3 out of 10 queries for a given DNS returns a NX - or (in layman's terms), every 3 (at least) out 10 times I try to access a website (that is, one specific website, 10 times), Firefox says the domain doesn't exist. After the first 3 errors the domain is found and cached, and all is well, but this annoyed me to no end.
There were some days when it was bad, and others days the problem never showed up. After trying to diagnose the problem on my end, I finally concluded the problem were my ISP's servers, so I gave up and switched to OpenDNS. Never had the problem again.
I see what you're talking about, and I've noticed this forever, since I've started using Linux, about 10 years ago.
Those "artifacts" you see are a side-effect of the much lauded "modularity" of Linux - the problem being that in the GUI, there's too many layers to go through.
For an application to draw to the screen, it first has to talk to GTK / QT. The widget toolkit will then talk to X, that will talk to the video driver / subsystem, that will then get stuff drawn. Depending on what is being done, there will be the Window manager thrown in the middle of this chain. Now, it all sounds like 3 or 4 layers for an application to go through isn't too much, but when you consider that most of these layers are necessarily bloated because they have to provide a full-blown API to talk to each other, then you start to see where the klunkyness comes from.
Consider too that on other OSes an application has to go through less layers to draw stuff on the GUI - in fact, these "layers" often can't even be well defined, because the GUI design on these OSes are all so tightly integrated (as opposed to abstracted, in a modular approach).
That's an ancient problem on Linux, and personally I think the weakest link on the chain is the widget toolkit (GTK/QT) and X integration. Try using a QT or GTK application compiled for Windows - you'll notice the very same slugishness even on Windows. IMHO, the widget engine should be more integrated with X - ideally, embedded on it, so we could be done with all these "many-widgets-to-choose-from" nonsense that always plagued *nixes. But I'm sure this is not happening any time soon - developers are too stubborn and bent on technicalities to take a user-oriented and pragmatic approach to solving problems.
Wayland looks like it could solve some of the problems, but I've watched such solutions come and go over the years, so I'm not holding my breath. That's a pitty really - I believe Linux can, and eventually to an extent will, find its way into the common user desktop. But it will always be a second or third class citizen in this area, until developers stop holding on to ancient concepts and drag themselves to 21st century.
Slashdot Mirror
I think I've mentioned it before in the past, though in another context: "The Time Ships", by Stephen Baxter. It's an official sequel to H. G. Wells' "The Time Machine".
The book itself is obviously not about education, but that subject is brought about when the protagonist gets into contact with another species (which I'll not name lest I spoil the book). Basically, the approach to education for that species is that children would be taught how to seek information, and then pretty much just told to go educate themselves, seeking out whatever they want.
Of course that would be utopia for humans, because we are mostly hedonistic by nature, but all the same this idea from that book really stuck with me, and made me realize in which way our educational system is a failure: children are usually just told to memorize stuff, a big part of which they will never really use, when they really should be taught how to "think" - how to seek information, stimulate curiosity and solve problems with information they gather themselves.
Fortunately, with initiatives such as these, it seems as though this is slowly changing.
This "plainly" here is difficult to judge. How can we be sure any unbelievable religious text wasn't some sort of "pointed political statement" or "satire"?
You're right, we can't be sure. The difference from this to the FSM cult, however (and what probably influenced the judge's ruling), is historical context - yes, we can't be sure the biblical accounts were derived as a political statement or satire, and what little historical context we have don't seem to point this way, so there's plenty of room for interpretation either by religious people or atheists.
The historical context for the FSM cult, however, is well-known, as is known that it is a political statement and satire.
The thing though is - and maybe the GP was trolling and I'm falling for the bait - as far as the rest of the world that is not North America (and maybe Japan too) was concerned, the Spectrum *was* a world-wide phenomenon. He could have used the argument that the C64 was technically superior, and then there would be few people to argue that. But market-wise, as you correctly pointed out, the Spectrum beat the shit out of the C64, and not only on the UK, but most of the world.
Case in point: I'm from Brazil, we didn't have legit Spectrums here back then, but we had locally-made clones, which amounted to the same thing and ended up getting exported to all of South America, and you know what? The only way anyone on Latin America knew the C64 actually existed was that it was often mentioned on computer magazines, and that was it - I never knew anyone who even heard about of the C64 around here, let alone owned one. We heard a lot about the TRS-80, Apple II (or rather its clones that were produced locally), MSX and so on, and it wasn't uncommon to find users of such systems, but the C64? Nada. I understand the C64 actually managed to chew a bit more of the market on some parts of europe, but the Spectrum was still far more popular. IIRC, on Russia the situation was similar to South America, in that they had Spectrum clones, and the C64 was a computer only the US cared about.
+1 for FTL awesomeness.
However, I'm disappointed they have yet to port it to Android, and instead chose to focus on porting to a browser platform, which to me counts as a useless gimmick since it's only really playable on browsers on a desktop platforms, for which they have ports for all platforms anyway.
I'd like to bring this discussion down a few notches and ask a more "stupid" question, to which I haven't yet seen a definite answer yet, and it looks like you guys could easily answer that. Sorry if it's a too a stupid question.
What is the performance impact of encrypting an Android device? I'm not talking necessarily about Android L, but any previous version that supports encryption (it was added on ICS, I think?). My concern in fact is with older devices, think 2012 or '13 phones. Are any of such older devices capable of hardware accelerating crypto operations (besides RNG)?
This.
The impression I'm getting here is that people is siding with Linus without really understanding what the discussion was about, and how it really unfolded. It's like people are painting the girl who started it as a diva who gets offended at cursing.
I read the whole thread though (gasp!), and it was a rather civil, insightful, and even funny discussion, at both sides of the argument. Ironically, I think the girl who started the discussion cursed even more than Linus during the discussion. Reading the thread, I've got the impression that the subject wasn't so much about cursing and being disproportionately mad at times, but just "toning it a bit down", and minding to whom Linus would burst out, and I can she has a point. The problem is, Linus tends to burst out at top-level developers (because of their greater responsibility, making mistakes for them are more damaging), and this tends to scare away inexperienced developers, while also impacting morale and reducing top-developers authority. And it's hard not to agree with that - there's only so much abuse a top-developer can take (Alan Cox anyone?).
I'm not sure I'm really siding with the girl either though, because I can see Linus has some valid reasons for acting like that, I could see *myself* acting the same way in his position, and I'm a generally a laid back guy. That's the kind of discussion that it's not all black and white.
There has been a lot of backlash on their blog about this: Why didn't they just go with XMPP? What their protocol have that XMPP doesn't, or couldn't be extended to support?
Personally - just a guess (also, btw, disclaimer: I'm a subscriber) - I think they're dying. Their client haven't been getting any significant development for the past year, current issues with some protocols have been going unaddressed, and new features like Lync protocol support (which there are working OSS implementations) have been going completely ignored despite many people clamoring for it.
So, they have been silent for a long time, and now this. It's fishy.
I agree with this argument on Windows. On OSX though, 32-bit chrome is a problem incidentally because of Java: on recent updates (past year) the 32-bit Java plugin on OSX was disabled. You can say what you want about Java, its vulnerabilities and shortcomings as a platform, but the fact is that many sites (banks or such) still require it, and that means I have to use Safari for those sites. It's not a big problem, but it's incovenient, and OSX, compared to Windows, has a much higher ratio of software and system components with 64-bit binaries than Windows, so there's very little reason to keep Chrome clinging to a legacy binary format on that platform.
Honest question: I do have an updated PS3 (yeah slashdot, judge me). I'm not interested on pirated games, but I may be interested on homebrew stuff (emulators and stuff like that). That leak will make that possible for me?
Oh, so you want prior art?
https://play.google.com/store/apps/details?id=br.com.eversource.shake2silence&feature=search_result#?t=W251bGwsMSwxLDEsImJyLmNvbS5ldmVyc291cmNlLnNoYWtlMnNpbGVuY2UiXQ..
Last update was on December 2010 - so it's a fair to assume the first version was submitted even earlier. And that's just one example I could find quickly, of course. It wouldn't surprise me it there are many more other apps (for Android or iOS alike) that does the same thing and was made before.
And yeah, as rolfwind said, just because the idea was implemented only after 10 years after Microsoft entered the smartphone market, doesn't mean it's patentable. The technology needed for this idea wasn't ubiquotous on smartphones until some 4 or 5 years ago anyway, so you should rather start making the math at that point in time.
Ask and ye shall receive: http://www.clementine-player.org/
The nice thing is that it's multi-platform, and it actually looks and works nice on other OSs - it's particularly nice on OS X, for those like me that hate iTunes.
Well, for starters, LMDE is related to Debian Testing, not Squeeze (which is currently the Stable distribution). So you should get more updated packages on LMDE vs. Squeeze, yes, and contrary to Debian, LMDE's packages should be updated more frequently and often to the latest upstream software releases.
Basically, LMDE is Debian Testing with some specific Mint packages (that usually are intended to improve and simplify user experience and add the "Mint" branding over Debian) overlayed on it. You should read my reply above to wrook detailing how on the latest snapshot releases LMDE is not really using the Debian Testing repositories directly, but you can easily make it do so and there shouldn't be any problems by doing that - I haven't tried myself though, so YMMV; I'm quite happy as it is.
I tried using it at the end of last year and I didn't get *any* updates for the 3 or 4 months I was using it. Not even security updates.
That's because since the penultimate snapshot release of LMDE (which was at some time on the last quarter of last year) they've switched the repositories to use their own copy of a snapshot of Debian Testing, and every 5 months or so they release what they call "update packs", which is basically a more recent snapshot of Debian Testing with the packages therein more throughly tested for bugs and such. Before they started doing that, LMDE used the standard Debian Testing repositories.
The intention on having a "snapshot repository" is to try to get the best of both worlds of rolling releases distributions and version-based distributions (i.e. Ubuntu). It's certainly not perfect, but works well enough for me, I don't mind it. Anyway, the maintainers says it's fine to swap out those Mint "snapshot repositories" with the standard Debian Testing repositories on sources.conf - of course, you're more likely to stumble into problems from time to time, and hence should be prepared to work around them or muddle through with them, but it shouldn't be any more problematic than using plain Debian Testing.
Mint is nice, and it's the Linux flavor I'm using currently (although I use LMDE, not the standard Mint) after having left Ubuntu when they transitioned to Unity. The best thing about it is that the maintainer(s) actually listen to users regarding development directions, which was what drove them to develop Cinnamon and adopt MATE as an option - as opposed to Ubuntu / Canonical, that just forced down the users' throats their ideas and UI decisions, alienating a large part of their user base in the process.
Having said that, there's still one thing that keeps me from recommending it to new users or users migrating out of Ubuntu: lack of automated upgrade procedure to newer major versions - one thing that Ubuntu has and generally works nice there. On Mint, the official procedure is to backup you files/settings using the backup tool, install the newer version from scratch on top of the existing install, and then restore the backup after. That's just too cumbersome. Yes, it's possible to upgrade without reinstalling by manually editing the sources.list file and upgrading manually with apt-get, but it's considered unsafe and error prone by the maintainers and hence not recommended. I did it anyway on a past install, and sure enough I had hiccups - I still had a working install, but there were a lot of rough edges and inconsistencies on the upgraded install. Because of that I ended up installing LMDE so I didn't have to worry about major version upgrades anymore. It's not a fully smooth ride either, but it's far more manageable, and having previous experience with Debian, I'm totally at home with it. But it's obviously not something I'd recommend to casual / new users either.
This. As PJ points out on Groklaw, although the judge alledgedly does code, he still seems to be having a hard time deciding whether APIs can be copyrighted. I wonder if he "does coding" on Basic... :-P
Well, I disagree. I think the motivations and some of the changes are quite nice, but they are badly misguided. Both Ubuntu and MS are making the mistake of trying to cram UI concepts not suitable to the environment they are going to be used:
- Metro UI: Awesome for tablets and touch interfaces in general - I'd go as far as saying it's the best UI concept on these cases, better even than iOS. But that concept just doesn't make any sense and it's counter-productive when you're interfacing with your device using a keyboard and mouse.
- Unity: Same thing, doesn't translate well to desktops, but it's worse because this UI doesn't really work well with touch interfaces either; it really was made for and works well only on a dying breed of devices: Netbooks.
- Gnome 3: I'll refrain from too much comment on this one because I haven't really used it yet, but from what I've seen, this is the one that's actually the saner of the other two in the sense that they are not trying to jam a square in a peg hole like the other two are doing - they are really trying to redefine traditional UI concepts on the desktop, and maybe other environments too. That's not to say that they're on the ball though: I think they've really gone way too radical there. What I'm really saying is that Gnome is the less bad of bunch.
... every one of those games would be a thematic variation on GTA. Like this. People don't want reality, they want to be entertained.
I don't like that my G+ profile shouts my real name everywhere too. So I was looking around in my profile, and guess what: There's a "Nickname" field in there - but the profile form explicitly says that it won't be shown in the profile. Why the hell are they doing that? Why have this field if it's not going to be used?
The dumb part is that Google could be fixing this problem in a much less disrupting way: Make the "Nickname" field actually useful, make it the default field shown for the public, or have the user setup if he wants the Nickname to be shown or his real name. Hell, if Google is so bent into real names, at least make the Nickname the field to be shown to the public, and the real name only to your friends / circles. What a waste.
Hm, that sounds dangerous. I have a triple-boot setup on my MacPro (OSX, Windows 7 and Linux). How will Lion handle that? Sounds like it could trash one (or both) of my other OSes in the process...
Being able to download the kernel, driver, and Android sources directly from Motorola, the maker of my Droid phone, is so prohibitive.
Good luck getting your recompiled kernel+driver+Android sources past the well-locked-down bootloader on any Motorola Android device newer than the original Droid.
Fair enough. However, this is not Android's (the OS) fault - the bootloader locking mechanism is hardware based, so only Motorola's to blame here. As always in these cases, all you can do is vote with your wallet: get a HTC or some other brand that doesn't lock you out of your property.
Please mod the parent up to the moon. I've been reading all the comments on the story, and it baffled me that no one caught this simple fact - GPS logs can be easily tampered with, or even forged. I don't know if the ruling on this case was made with this technical knowledge, but it's nevertheless a good thing a bad precedent wasn't set. Nevermind GPS accuracy, GPS logs just aren't a reliable source of evidence in the first place.
Yea, I think it's a bit elitist too. I mean, if they don't want a sequel, don't read it!
Case in point, a classic: The Time Machine, from H.G. Wells. A century later, a sequel was authorized and written by Stephen Baxter: The Time Ships. And I like it so much more than the first book, because it expands so much on the idea, concepts and caracter. Granted, there was a lot to expand from given the late 19th century science, and Stephen Baxter is also an excellent SF writer... So the question really is if Mr. Reichert is up to the task, since he's pretty much unkown. But so was Baxter, back when he wrote the The Time Ships. I guess we'll just have to wait and hope that Mr. Reichert does a good job.
Nice, but your Quake has statically linked libraries, which was (and still is) a simple solution for running binaries across wildly different distributions or linux environments without hassle.
Of course, not everyone does that. The common and default way to compile stuff is using dynamic linking (i.e. using the standard libraries you'd already have installed), and if your version of Quake (compiled 10 years ago) was dinamically linked, you'd be in for a big surprise.
And anyway, while it is a good strategy for backwards compatibility and cross-distribution portability, static linking is not fullproof and has its drawbacks - you could still run in to compatibility problems that are kernel-bound (usually, interfaces that have been changed or obsoleted), in which case static linking won't help you, and statically linked binaries are bloated (thus, taking a lot more memory). You really just got lucky with that version of Quake.
Well, I can't vouch for the GP, but my ISP has a very flaky DNS service. For some reason, every 3 out of 10 queries for a given DNS returns a NX - or (in layman's terms), every 3 (at least) out 10 times I try to access a website (that is, one specific website, 10 times), Firefox says the domain doesn't exist. After the first 3 errors the domain is found and cached, and all is well, but this annoyed me to no end.
There were some days when it was bad, and others days the problem never showed up. After trying to diagnose the problem on my end, I finally concluded the problem were my ISP's servers, so I gave up and switched to OpenDNS. Never had the problem again.
I see what you're talking about, and I've noticed this forever, since I've started using Linux, about 10 years ago.
Those "artifacts" you see are a side-effect of the much lauded "modularity" of Linux - the problem being that in the GUI, there's too many layers to go through.
For an application to draw to the screen, it first has to talk to GTK / QT. The widget toolkit will then talk to X, that will talk to the video driver / subsystem, that will then get stuff drawn. Depending on what is being done, there will be the Window manager thrown in the middle of this chain. Now, it all sounds like 3 or 4 layers for an application to go through isn't too much, but when you consider that most of these layers are necessarily bloated because they have to provide a full-blown API to talk to each other, then you start to see where the klunkyness comes from.
Consider too that on other OSes an application has to go through less layers to draw stuff on the GUI - in fact, these "layers" often can't even be well defined, because the GUI design on these OSes are all so tightly integrated (as opposed to abstracted, in a modular approach).
That's an ancient problem on Linux, and personally I think the weakest link on the chain is the widget toolkit (GTK/QT) and X integration. Try using a QT or GTK application compiled for Windows - you'll notice the very same slugishness even on Windows. IMHO, the widget engine should be more integrated with X - ideally, embedded on it, so we could be done with all these "many-widgets-to-choose-from" nonsense that always plagued *nixes. But I'm sure this is not happening any time soon - developers are too stubborn and bent on technicalities to take a user-oriented and pragmatic approach to solving problems.
Wayland looks like it could solve some of the problems, but I've watched such solutions come and go over the years, so I'm not holding my breath. That's a pitty really - I believe Linux can, and eventually to an extent will, find its way into the common user desktop. But it will always be a second or third class citizen in this area, until developers stop holding on to ancient concepts and drag themselves to 21st century.