I've been running my home servers (web, mail, dialup, other) on Debian since before Debian got to version 1.0. I've run it on single processors and with SMP. I've run it on x86 and on Alpha.
I've never had stability or performance problems due to the software.
Is it possible that FreeBSD is more stable? I'll grant that it's possible, but... "the extra time required to really get a FreeBSD box tuned will come back in spades through performance and stability metrics"? Really? No, I cannot see how. I do not need more than five nines of uptime.
I do install it in a VM sometimes. Heck, that's still how I run "Lotus Improv" when I need to.
Of course no software written in the last ten years will run on it, but the software that did run on it generally still works today (since the assumption of universal networking hadn't been threaded throughout the whole ecosystem yet).
And yeah, NT4 had problems that NT3.51 lacked. NT4 was the first version that put the video drivers into ring 0. Sure, video performance skyrocketed, but that was kinda the beginning of the end for rock-solid stability in the server flavor of the OS.
(As for modern day-to-day use, yeah, I'm sticking with Linux on my servers and a variety of almost-entirely-non-Microsoft systems on my clients. Only Microsoft client I run regularly is my XBox, which hasn't been too bad... but, reports are that they're about to unleash Metro on it, sigh.)
Assertion: it's not the fundamental content of the message that's turning people off. There are certainly at least elements of non-insanity to that. It's the completely tactless delivery, and that is strongly affected by timing concerns.
I have already seen. If he'd waited, no one would have noticed, and his word would have reached no one that doesn't already read his blog.
...and that would have been better, because next time, more people might have listened to what he had to say, instead of just writing him off as a nutjob.
(If, empirically, every time you open your mouth, the result is for other people's minds to shut, then it's better not to speak. Have someone else speak for you instead, if you must.)
Its effect is going to be open discussion, which will reach more people than doing nothing.
We disagree. IMHO, the main effect is going to be fewer people being willing to give what he says a fair listen, even when they'd otherwise have been persuaded by what he says. Over time, fewer people will have been reached because of this. There are times when the right tactic is to stay silent, and this moment was one of those.
And it is critical that we receive this message -- not you and I, maybe, but as many of the wide-eyed legions of Apple as can be reached. Because what Apple represents is precisely the same thing that Microsoft or Sony represents: a dearth of choice. Stallman might be an egotistical ass, but he is certainly the foremost champion of the rights of the user. Some programmers don't like that, so they don't like the GPL, and they don't like Free Software. They call it a virus and they would prefer to stamp it out rather than have to deal with something so confusing.
Other people can make the same point in a month, and a year, and reach other audiences, but this point needs to be made now and it needs to be made well. Stallman has done both.
Your assertion that Stallman has made the point well -- you're incorrect. That isn't intended as a value-judgment, but as an observation. You can tell by paying attention to the effects.
If your assertion is that the point needed to be made to as many of the wide-eyed legions of Apple as can be reached, then Stallman's declaration was counterproductive at this time. It's effect is going to be the closing of more minds than it opens.
I'm speaking purely from the standpoint of rhetoric or "PR tactics" here. I'm sure he felt what he was saying was true, and I'm sure many people here think so as well, but if the goal is to persuade (and not just to say something he thought was true for its own sake, or to "preach to the converted"), then it was downright counterproductive.
(Which -- and this is very important -- is not the same thing as "wrong".)
Which is not exactly an unusual thing for RMS. He is not a rhetorical genius. (Yes, he's an actual genius, I agree that's true, but not in the realm of rhetoric.)
Nope, there is no such thing as an iPhone that works woth 4G, on the market or announced. The Sprint iPhone 4S is a strictly 3G device, just like the Verizon, AT&T, and unlocked iPhone 4S.
You would do this so you could rent access to the dev environment to people who don't have any access to any kind of Macintosh at all and don't want to obtain one. The business model of "XCode is on OnLive" is what I'm talking about.
XCode for IPhone development is free...as long as you already have a Mac running a sufficiently-recent release of OS X.
Otherwise, not free.
This brings up an opportunity that might kinda be interesting.
The latest release of OS X has changed so that you're allowed to run it in a VM, as long as the underlying hardware is Apple. A given box can run three copies of Lion at once (the base OS, and two in VMs).
I'm wondering who will be the first to set up cloud-based access to XCode using something like the VMWare Viewer as the client. Register with the site, get your own Lion VM provisioned, and launch the individual VMs on demand as individual developers connect in.
This wasn't permitted by the licensing in a cost-effective way under 10.6. From what I can see, it is under 10.7. I wonder if anyone will try it.
Previously, if those 11,000 libraries wanted to be able to lend my book, I would have gotten 11,000 sales.
This may seem like a minor nit, but as someone who has depended on it in the past, I have to mention that you're completely ignoring inter-library loans. The number of books paid for depends more on the borrowing load than on the number of institutions involved in the lending -- and I'm talking about paper, not etexts.
I'm pretty sure there will be some kind of compensation model here.
Also note that Amazon isn't the first to do this. They are in no way blazing new territory here. They're partnering with OverDrive, and OverDrive has been lending out etexts for a while now. It's just that in the past, they were EPUB or PDF format with Adobe's DRM, and so were largely unavailable to Kindle users specifically (but Sony and Nook and iOS users could already get 'em).
Even if you weren't missing the point, this wouldn't really be accurate. Even on mobile phones, there are chunks of Android that have never been free. If you want a (relatively) cheap way to see exactly what I mean, get a Nook Color and throw CM7 on there without installing the non-free components (which include Marketplace, YouTube, and the infrastructure for Google accounts). There's a bunch of stuff just missing. (The same stuff is also missing on the Nook Color's stock OS, which again, is running a phone version of Android, not the tablet version.)
Heck, if you insist on limiting the conversation to phones and nothing else, look at the iDroid project. There's a ton of stuff missing that's not because of undocumented hardware (though there's also some stuff missing for exactly that reason, but that doesn't invalidate the point).
Ah, have they said that the system won't let any software get installed unless it's in the Metro app store? I haven't heard a single hint of that yet -- got a pointer?
In any event, that just means it'll require a jailbreak, and that will happen.
Just get the DOSbox core with x86 emulation in there, identify the callouts to libraries that you've got implemented in ARM to improve performance, and you're probably do pretty well.
Heck, a port of DOSbox on its own will be enough to make many GoG games work. (This is how I play "Master of Magic" on my iPad.)
Barring that, well, glue the x86 emulator into WINE and there's probably some potential.
Done right, an extension language that exposes the right interfaces in the right ways could be to GNU software what AppleScript is to Apple software. And that support is what makes Automator work.
To elaborate: the extension language for Macintosh GUI programs isn't built into those programs. They're built keeping it in mind, but it's not built into them. Instead, they essentially expose verbs and nouns (via a "dictionary") that a shared extension language framework then leverages. Some they define themselves (like the way iTunes defines what a "track" is), and some are defined according to standard conventions (like the definition of what a "window" or "menu" is). It's all typed and stuff too.
(In principle, more than one language could take advantage of that stuff, and the frameworks in fact lay the groundwork for that. In practice, nobody actually uses a language other than AppleScript with this stuff, which is kinda too bad. If you're on a mac and want to know what the hell kind of nonsense I'm raving about, read the man page for or output from the "osalang" command-line program.)
The result is incredibly Unix-y in fundamental philosophy, even if it doesn't seem so in detail: you can write scripts that string together small special-purpose programs to do large complex tasks -- and I'm talking about GUI programs, not just command-line utilities.
I make very powerful use of this myself sometimes, just as I've made very powerful use of shell scripting to bind small special-purpose command-line programs together since the 1980s. It's niiiice.
In the United States, this is called a "design patent". It's the mechanism by which such things as the shape of the coca cola bottle or the design of a font are protected.
No shock there -- the carriers are confused as well. They've been referring to stuff that's unambiguously not 4G in marketing materials as if it were 4G. They're using 4G to mean "better than what you probably think of when you hear 3G". It got so bad that 4G had to be formally redefined.
I don't know what the right answer is, except perhaps to discard the short labels and talk about the details. Don't sell me on "4G speeds", sell me on specific speeds (and/or features). For example.
Myself, I'm constantly tempted to disable 3G on my phone so my battery lasts longer. I do not need the higher speeds. If I could drop back to 2G without giving up any features, I would.
(Explanation: on AT&Ts network, apparently some towers do not permit simultaneous voice and data if you connect via 2G. Simultaneous voice and data always work if you connect via 3G. That's the only reason I leave 3G turned on today.)
This should have been seen as inevitable, since it's been made clear that the content owners have been floored by the success of Netflix and are trying to negotiate much better terms for themselves now.
I'm delighted at how little my own costs are going up (less than 8% I think).
I'm also a little frustrated that so many people are reacting so strongly to this -- the content owners would all rather force us into "Hulu+" (with subscription plus advertising), or to "Zune Marketplace" or "iTunes" (per-episode purchase or rentals), or back to cable operators, and all the folks who are dropping their Netflix subscriptions are helping that happen. Ah well.
There's no way you could have widespread common ground among millions of different people, for something as hugely diverse and personal as individual taste in music, without recourse to the lowest common denominator.
, I took for granted that you were being sarcastic, and were attempting to argue exactly the opposite of what you were saying, and were expressing anger at the use of "lowest".
That's what I was reacting to.
(And no, I have not ever listened to much recent popular music. Not sure what I might discover if I did. I haven't yet heard a single song by "Lady GaGa" or "Justin Bieber" or... heck, that's all I can name.)
Larry Fast's solo project, "Synergy", was an early innovation in electronic music. (Not as early as Wendy Carlos, but we're still talking about the 1970s.) For the most part, the albums consist of performances programmed in ahead of time, either via MIDI or via assembly language, and then performed on electronic instruments.
But he had an experimental album years ago, "Computer Experiments Volume 1", in which he programmed the computer to compose the music as well as perform it.
I've got a copy on vinyl. It's remarkable in that instead of the liner notes including lyrics, the liner notes include the Apple ][ assembly language code for the composition algorithms.
(Full disclosure: this one album not very good music. It's mostly interesting. Much of Synergy's other work is totally awesome, though.)
Yes, it's lowest common denominator. There's no way you could have widespread common ground among millions of different people, for something as hugely diverse and personal as individual taste in music, without recourse to the lowest common denominator.
That is "lowest common denominator".
Just because it's "lowest" doesn't necessarily mean it's particularly "low", just that going any "higher" destroys the commonality. If you're talking about widespread common ground among millions of different people, then you're talking about lowest common denominator.
For the moment, not persuaded.
on
Happy Tau Day
·
· Score: 1
I might agree that it makes sense to switch from pi to tau after I agree it makes sense to change from imperial measurement to the metric system.
Note that at this time I do not agree that it makes sense to switch to metric, so we may be in for a bit of a wait...
No, it does not. Much does, but certainly not all.
A lot of streaming buffers like YouTube does. Not all streaming buffers, but most does. But what's buffered is very often not the same as what you'd put into a file for download. (Note that in YouTube, it often is what you'd get in a file download. But YouTube is an exception here.)
For example, streaming and files can behave extremely differently with regard to key frames. And look at Netflix's streaming -- from second to second you can get a different version, depending on your bandwidth fluctuations, which necessarily means your compression 30 minutes in can't refer back to data from 20 minutes ago. Very different from files.
A streaming client often throws away the portion of the data you've already used. Consider rewinding and fast-forwarding in Netflix. Rewinding doesn't give you the same experience you'd get with a downloaded file -- it has to talk to the server and tell it you're doing a "seek" and to start streaming from the new position again.
To contrast with what happens when you "stream" a video rental on an AppleTV, or on a pre-Zune XBox 360: you always, always get the full quality of video that you asked for. There's never a down-scaling due to constrained bandwidth. This can mean it takes 20 hours to download a 2-hour movie; the service does not regard this as an problem. In a streamed service, that's an problem of some kind, and on Netflix for example, the solution isn't to wait at "buffering" for 20 hours, it's to transparently make your video quality plummet. It's fundamentally different.
Slashdot Mirror
I've been running my home servers (web, mail, dialup, other) on Debian since before Debian got to version 1.0. I've run it on single processors and with SMP. I've run it on x86 and on Alpha.
I've never had stability or performance problems due to the software.
Is it possible that FreeBSD is more stable? I'll grant that it's possible, but... "the extra time required to really get a FreeBSD box tuned will come back in spades through performance and stability metrics"? Really? No, I cannot see how. I do not need more than five nines of uptime.
I do install it in a VM sometimes. Heck, that's still how I run "Lotus Improv" when I need to.
Of course no software written in the last ten years will run on it, but the software that did run on it generally still works today (since the assumption of universal networking hadn't been threaded throughout the whole ecosystem yet).
And yeah, NT4 had problems that NT3.51 lacked. NT4 was the first version that put the video drivers into ring 0. Sure, video performance skyrocketed, but that was kinda the beginning of the end for rock-solid stability in the server flavor of the OS.
(As for modern day-to-day use, yeah, I'm sticking with Linux on my servers and a variety of almost-entirely-non-Microsoft systems on my clients. Only Microsoft client I run regularly is my XBox, which hasn't been too bad... but, reports are that they're about to unleash Metro on it, sigh.)
I still miss Windows 3.11 (for workgroups) on the desktop, and Windows NT 3.51 on the server. Sigh.
Assertion: it's not the fundamental content of the message that's turning people off. There are certainly at least elements of non-insanity to that. It's the completely tactless delivery, and that is strongly affected by timing concerns.
(If, empirically, every time you open your mouth, the result is for other people's minds to shut, then it's better not to speak. Have someone else speak for you instead, if you must.)
We disagree. IMHO, the main effect is going to be fewer people being willing to give what he says a fair listen, even when they'd otherwise have been persuaded by what he says. Over time, fewer people will have been reached because of this. There are times when the right tactic is to stay silent, and this moment was one of those.
Just wait, you'll see.
Your assertion that Stallman has made the point well -- you're incorrect. That isn't intended as a value-judgment, but as an observation. You can tell by paying attention to the effects.
If your assertion is that the point needed to be made to as many of the wide-eyed legions of Apple as can be reached, then Stallman's declaration was counterproductive at this time. It's effect is going to be the closing of more minds than it opens.
I'm speaking purely from the standpoint of rhetoric or "PR tactics" here. I'm sure he felt what he was saying was true, and I'm sure many people here think so as well, but if the goal is to persuade (and not just to say something he thought was true for its own sake, or to "preach to the converted"), then it was downright counterproductive.
(Which -- and this is very important -- is not the same thing as "wrong".)
Which is not exactly an unusual thing for RMS. He is not a rhetorical genius. (Yes, he's an actual genius, I agree that's true, but not in the realm of rhetoric.)
Nope, there is no such thing as an iPhone that works woth 4G, on the market or announced. The Sprint iPhone 4S is a strictly 3G device, just like the Verizon, AT&T, and unlocked iPhone 4S.
You would do this so you could rent access to the dev environment to people who don't have any access to any kind of Macintosh at all and don't want to obtain one. The business model of "XCode is on OnLive" is what I'm talking about.
This brings up an opportunity that might kinda be interesting.
The latest release of OS X has changed so that you're allowed to run it in a VM, as long as the underlying hardware is Apple. A given box can run three copies of Lion at once (the base OS, and two in VMs).
I'm wondering who will be the first to set up cloud-based access to XCode using something like the VMWare Viewer as the client. Register with the site, get your own Lion VM provisioned, and launch the individual VMs on demand as individual developers connect in.
This wasn't permitted by the licensing in a cost-effective way under 10.6. From what I can see, it is under 10.7. I wonder if anyone will try it.
This may seem like a minor nit, but as someone who has depended on it in the past, I have to mention that you're completely ignoring inter-library loans. The number of books paid for depends more on the borrowing load than on the number of institutions involved in the lending -- and I'm talking about paper, not etexts.
I'm pretty sure there will be some kind of compensation model here.
Also note that Amazon isn't the first to do this. They are in no way blazing new territory here. They're partnering with OverDrive, and OverDrive has been lending out etexts for a while now. It's just that in the past, they were EPUB or PDF format with Adobe's DRM, and so were largely unavailable to Kindle users specifically (but Sony and Nook and iOS users could already get 'em).
http://www.overdrive.com/
Even if you weren't missing the point, this wouldn't really be accurate. Even on mobile phones, there are chunks of Android that have never been free. If you want a (relatively) cheap way to see exactly what I mean, get a Nook Color and throw CM7 on there without installing the non-free components (which include Marketplace, YouTube, and the infrastructure for Google accounts). There's a bunch of stuff just missing. (The same stuff is also missing on the Nook Color's stock OS, which again, is running a phone version of Android, not the tablet version.)
Heck, if you insist on limiting the conversation to phones and nothing else, look at the iDroid project. There's a ton of stuff missing that's not because of undocumented hardware (though there's also some stuff missing for exactly that reason, but that doesn't invalidate the point).
Ah, have they said that the system won't let any software get installed unless it's in the Metro app store? I haven't heard a single hint of that yet -- got a pointer?
In any event, that just means it'll require a jailbreak, and that will happen.
Just get the DOSbox core with x86 emulation in there, identify the callouts to libraries that you've got implemented in ARM to improve performance, and you're probably do pretty well.
Heck, a port of DOSbox on its own will be enough to make many GoG games work. (This is how I play "Master of Magic" on my iPad.)
Barring that, well, glue the x86 emulator into WINE and there's probably some potential.
Hm... seems like an opportunity...
By social engineering applied to the browser vendors.
Done right, an extension language that exposes the right interfaces in the right ways could be to GNU software what AppleScript is to Apple software. And that support is what makes Automator work.
To elaborate: the extension language for Macintosh GUI programs isn't built into those programs. They're built keeping it in mind, but it's not built into them. Instead, they essentially expose verbs and nouns (via a "dictionary") that a shared extension language framework then leverages. Some they define themselves (like the way iTunes defines what a "track" is), and some are defined according to standard conventions (like the definition of what a "window" or "menu" is). It's all typed and stuff too.
(In principle, more than one language could take advantage of that stuff, and the frameworks in fact lay the groundwork for that. In practice, nobody actually uses a language other than AppleScript with this stuff, which is kinda too bad. If you're on a mac and want to know what the hell kind of nonsense I'm raving about, read the man page for or output from the "osalang" command-line program.)
The result is incredibly Unix-y in fundamental philosophy, even if it doesn't seem so in detail: you can write scripts that string together small special-purpose programs to do large complex tasks -- and I'm talking about GUI programs, not just command-line utilities.
I make very powerful use of this myself sometimes, just as I've made very powerful use of shell scripting to bind small special-purpose command-line programs together since the 1980s. It's niiiice.
In the United States, this is called a "design patent". It's the mechanism by which such things as the shape of the coca cola bottle or the design of a font are protected.
http://en.wikipedia.org/wiki/Design_patent
No shock there -- the carriers are confused as well. They've been referring to stuff that's unambiguously not 4G in marketing materials as if it were 4G. They're using 4G to mean "better than what you probably think of when you hear 3G". It got so bad that 4G had to be formally redefined.
I don't know what the right answer is, except perhaps to discard the short labels and talk about the details. Don't sell me on "4G speeds", sell me on specific speeds (and/or features). For example.
Myself, I'm constantly tempted to disable 3G on my phone so my battery lasts longer. I do not need the higher speeds. If I could drop back to 2G without giving up any features, I would.
(Explanation: on AT&Ts network, apparently some towers do not permit simultaneous voice and data if you connect via 2G. Simultaneous voice and data always work if you connect via 3G. That's the only reason I leave 3G turned on today.)
This should have been seen as inevitable, since it's been made clear that the content owners have been floored by the success of Netflix and are trying to negotiate much better terms for themselves now.
I'm delighted at how little my own costs are going up (less than 8% I think).
I'm also a little frustrated that so many people are reacting so strongly to this -- the content owners would all rather force us into "Hulu+" (with subscription plus advertising), or to "Zune Marketplace" or "iTunes" (per-episode purchase or rentals), or back to cable operators, and all the folks who are dropping their Netflix subscriptions are helping that happen. Ah well.
Hm, I suspect I actually misunderstood you.
See, this is Slashdot. When you said
There's no way you could have widespread common ground among millions of different people, for something as hugely diverse and personal as individual taste in music, without recourse to the lowest common denominator.
, I took for granted that you were being sarcastic, and were attempting to argue exactly the opposite of what you were saying, and were expressing anger at the use of "lowest".
That's what I was reacting to.
(And no, I have not ever listened to much recent popular music. Not sure what I might discover if I did. I haven't yet heard a single song by "Lady GaGa" or "Justin Bieber" or ... heck, that's all I can name.)
Larry Fast, but not in the sense you mean.
Larry Fast's solo project, "Synergy", was an early innovation in electronic music. (Not as early as Wendy Carlos, but we're still talking about the 1970s.) For the most part, the albums consist of performances programmed in ahead of time, either via MIDI or via assembly language, and then performed on electronic instruments.
But he had an experimental album years ago, "Computer Experiments Volume 1", in which he programmed the computer to compose the music as well as perform it.
I've got a copy on vinyl. It's remarkable in that instead of the liner notes including lyrics, the liner notes include the Apple ][ assembly language code for the composition algorithms.
(Full disclosure: this one album not very good music. It's mostly interesting. Much of Synergy's other work is totally awesome, though.)
That is "lowest common denominator".
Just because it's "lowest" doesn't necessarily mean it's particularly "low", just that going any "higher" destroys the commonality. If you're talking about widespread common ground among millions of different people, then you're talking about lowest common denominator.
I might agree that it makes sense to switch from pi to tau after I agree it makes sense to change from imperial measurement to the metric system.
Note that at this time I do not agree that it makes sense to switch to metric, so we may be in for a bit of a wait...
No, it does not. Much does, but certainly not all.
A lot of streaming buffers like YouTube does. Not all streaming buffers, but most does. But what's buffered is very often not the same as what you'd put into a file for download. (Note that in YouTube, it often is what you'd get in a file download. But YouTube is an exception here.)
For example, streaming and files can behave extremely differently with regard to key frames. And look at Netflix's streaming -- from second to second you can get a different version, depending on your bandwidth fluctuations, which necessarily means your compression 30 minutes in can't refer back to data from 20 minutes ago. Very different from files.
A streaming client often throws away the portion of the data you've already used. Consider rewinding and fast-forwarding in Netflix. Rewinding doesn't give you the same experience you'd get with a downloaded file -- it has to talk to the server and tell it you're doing a "seek" and to start streaming from the new position again.
To contrast with what happens when you "stream" a video rental on an AppleTV, or on a pre-Zune XBox 360: you always, always get the full quality of video that you asked for. There's never a down-scaling due to constrained bandwidth. This can mean it takes 20 hours to download a 2-hour movie; the service does not regard this as an problem. In a streamed service, that's an problem of some kind, and on Netflix for example, the solution isn't to wait at "buffering" for 20 hours, it's to transparently make your video quality plummet. It's fundamentally different.