I write songs...am I not supposed to have a chance to make any kind of money at this?
...here's an idea: if you are a programmer, why not give your work away and make money giving readings of your code?
I write a lot of free and open source software and give it away. As a college prof, I also make money giving readings of my code. Coincidentally, I write songs that are performed publicly, even though I don't make any money at it. I'm afraid you're 0 for 3 with me.
Having never done it professionally myself...can you really do CAD/graphics editing as fast as you can move the mouse? For me, at least, thinking time dominates in most mouse-related tasks: it's not how fast I can push the button, but which button to push. If that weren't true, I think I would want a job with more thinking.
(1) An earlier poster gave the canonical example: new disk too big. I've also seen new RAM too big (supported by HW, but not BIOS). Certainly would be nice to be able to boot an older machine from USB, much less USB 2, with the addition of a PCI USB card. Etc.
(2) There's no way HW manufacturers are leaving the flashable-BIOS world anytime soon. Too much support cost for anything else. They could try the kind of code-signing tricks the Xbox uses, but frankly I don't see them bothering, for the same reason: big pain to manage the keys, and no real benefit to the HW vendor.
Offhand, I can think of three strong arguments for an OpenBIOS.
Decreased obsolence: As others have noted, changes in BIOS requirements can be responded to given the BIOS source. BIOS maintenance is no longer at the whim of the hardware vendor. This could extend the useful life of hardware.
Decreased HW lockout: As others have noted, the move toward "Trusted Computing" could easily take a sinister turn. An open BIOS would make it much harder for hardware vendors to lock out libre software.
Increased OS compatibility: Many seem concerned about getting hardware compatibility right in an open BIOS. The flip side of this is that getting the BIOS to work with an open OS would become much easier. The premiere example is ACPI, where the BIOS often has bugs with corresponding Windows workarounds.
I think the goal of producing an open BIOS that works well on a number of machines is quite a difficult one. The rewards of achieving it, however, seem high.
Hmmm. So the fact that some of us were developing quality open source desktop software and giving it away before there was such a thing as Windows supports the theory of Windows-driven OSS development how, exactly? You see, to believe that OSS desktop developers never innovate, you have to believe that all that we do results from copying commercial developers. My previous post tried to document just a few of the many, many ways in which that is patently untrue, under the highly non-controversial theory of temporal direction in causality (stuff can't be caused by something that happens afterward).
I'm not sure what your point is, in fact. I'm quite confused by your accusation of idiocy. It's possible that this is because I'm being stupid, but if so help me to understand.
You understand that the UNIX graphical desktop predates Windows and Mac, right? That the software being used to render a Linux desktop today is the direct descendant of this venerable software, and contains some of the same bits? That Keith Packard, one of the many high-profile heroes of open source, has continuously developed this open source desktop in innovative ways over the last 20 years? That, for example, he developed the secure login protocol (XDMCP) used by X thin clients on the now-open-source BSD UNIX back when Windows and Mac didn't even do multiprocessing, much less multi-user, not to mention being largely incapable of interoperating with the Internet? That more recently Packard co-developed arguably the most advanced desktop rendering system in the world (Xft/fontconfig/Xrender/Composite/Damage/Cairo)--- way more advanced than Windows and in a lot of ways more than the Mac?
Naw, you're a troll---you'd turn into stone if you ever saw the light.
The NMR article is, overall, fine. I have a physics degree, and feel that I can accurately judge an article at this level. If you don't believe me, get a physicist friend to check it for you.
Just as importantly, there is some potential for "ptr == NULL" to be wrong. This can happen when your libraries somehow have NULL defined to something other than 0. In the ANSI standard, ((void *)0) and (I believe) also ((char *)0) are safe. For non-ANSI compilers and for ANSI compilers where the definition of NULL involves some other kind of pointer the compiler is entitled to silently not do what you thought it was going to do. In particular, ((FILE *)0) could conceivably cause problems.
Another way to put this: the C programming language defines the sole invalid value of pointer type to be represented as 0. Thus, operations such as ! on pointer values are required to treat whatever internal bit pattern is chosen by the compiler to be the machine's "null pointer" as if it were a 0. It makes about as much sense to use some bizarre NULL symbol as a representation of 0 as it does to define BEGIN to be "{" and END to be "}". It will distract the reader who understands C, and the reader who doesn't needs fixing anyhow.
What is NULL for, then? Simple: it's the value returned by certain library calls when they want to indicate an unusual condition. Use it in comparisons where the library specification says it is to be used. Note that this was probably also a bad idea on the part of the specifiers, but it's too late now to fix it.
Apparently you've never worked on a commercial fishing boat, or painted houses, or worked in a doctor's office. As someone who did all of these as well as doing system and network administration for 20 years, you can trust me that the computer-using segment of the public is at least as smart as the rest of it.
I did understand what Gorman was claiming: I didn't have the energy at the time for a detailed commentary on his critique of Google's work. My claim is that he fundamentally misunderstands (perhaps deliberately) how Google works, and gives little credit to their demonstrated ability to improve how things work in the future. I would be shocked, first of all, if Google's book search were not able to find "other references to the subject in the book"---related references are provided for most of their other search media. Secondly, I would be rather surprised if Google doesn't find some way to make the scanned texts available as a whole online---perhaps on a pay-the-publisher-to-view model. Once all those books are scanned, it's going to be quite hard, IMHO, to keep them "locked up" from being used more generally: they just represent too valuable a resource.
When Gorman "endorses" libraries in your quoted passage, he is actually endorsing a vision of libraries that I (and many others) believe will kill them. I happen to believe that electronic communication will "supplant and obliterate all previous forms" for scholarly and literary work, just as hand-copied books supplanted the oral tradition, printed books supplanted hand-copied ones, recorded music largely supplanted sheet music, and movies largely supplanted plays. None of the "supplanted" media are dead, but in my opinion any library that concentrates on them at the expense of the newer ones will rapidly be discarded as irrelevant. Yet this is exactly what Gorman seems to be advocating.
What amuses and annoys me about Michael Gorman's comments (and yes, I did read them and understand them) is how arrogant they are. Gorman, as President-Elect of the American Library Association, is not just proud enough to say how much smarter he is than other commentators about managing information. No, he's proud enough to dream of telling Google how to manage their money. He's proud enough to characterize a whole class of people intelligent enough to operate a computer as mouth-breathing idiots.
Best of all, he's very proud of how the Universal Bibliographic Control scheme he endorses will solve the world's information access problems. Now please understand: UBC doesn't actually give anyone access to source materials. In point of fact, it seems to be a scheme for trying to assemble a meta-bibliography---in other words, a list of what printed materials you could read if you could get your hands on them. This is unlike Google, an organization crass enough to actually digitize the text of books, to put you one click away from the primary source of the information it indexes, and to maintain backup copies of that information against the loss of the primary source. It is unlike Project Gutenberg, an organization that has already published a huge number of digitized texts that are now available to anyone with Internet access. It is unlike even the bloggers, who at least make their own work fully available online. Gorman apparently has the more limited goal of indexing materials without providing access to them, while mocking the efforts of these other organizations to provide access.
On the offhand chance that Michael Gorman is reading this, let me make my position as clear as possible. I am a scientific research and (if I do say so myself) a fairly literate writer. I use Google, Wikipedia, Citeseer, Project Gutenberg, and other online information resources on a daily basis, because I've found them to be quite effective for me. I read about five fiction novels a month. The last time I used a library card catalog was about 6 months ago. The reasons for this have nothing to do with the comprehensiveness of my University library's bibliography, and everything to do with the paucity of its actual content.
I support our American public libraries, because I think they're an important bulwark in our fight for free speech. In terms of effectiveness in serving my needs and the needs of my family and friends, they are so bad that I fear for their future. Mr. Gorman, please keep in mind that when public library funding comes up for public discussion, your comments, especially given your position, are extraordinarily unhelpful. So, in the jargon of the "blog people" you so despise, please STFU.
more importantly, we could always recommend the latest bits to newbies without any download hassle.
If you were doing that you were being irresponsible. The volume of updates that all major distributions push today means it's not feasable to use Linux on dialup and stay secure at the same time.
The security-critical subset of patches from any major vendor is easily dialup-friendly.
At any rate, who was talking about dialup? I was just talking about convenience. The current Fedora installation process involves downloading a bunch of CD isos totalling about 2.5GB, and then either burning them or loopback-mounting them. BTW, the insns say there's a DVD iso, but it doesn't actually seem to be there. Probably a good thing, since "Note: If you are using HTTP or FTP to download, some download clients cannot handle the DVD image because it is larger than 4 gigabytes." Of course, "Since the Red Hat FTP site can experience extremely high volume, you may have better success using a mirror site. Alternatively, you can join the torrent." Good luck, newbie. Heck, good luck me.
They don't want to do an apt-get upgrade and find a few hours later they've been locked out of their system by a busted PAM upgrade.
I've been running Debian unstable for more than 5 years now, and I've never had anything like that happen. Application breakage, sure, but never an unusable system.
Distros like Ubuntu that actually try hard to manage these things are reportedly even better. On the other hand, I've had to wipe the root partition and try again after big-bang Red Hat upgrades several times in the past. Not very fun.
Seth, you're a good guy. And indeed, all the RH engineers I know are good guys. RH spends an inordinate amount of money on the community, and has been a real focal point for the spread of Linux.
The problem is that several years after the avoidable screwup of claiming "we're walking away from home users and the desktop: go use Windows", I still can't walk into a random computer store, as far as I know, and buy a $20 boxed DVD with any kind of Red Hat Linux on it. This is bad. I can certainly walk into any computer store and buy a copy of Windows. One of the things that made us loyal to Red Hat (and I sysadmined a network of RH boxes for many years) was that we could always get the latest "supported" bits at the corner store without any download hassle---more importantly, we could always recommend the latest bits to newbies without any download hassle.
None of this is the engineers' fault. But it still sends a message to those, like me, who are trying to figure out what Linux distro to use, and that message is "we don't care about you unless you're either rich or have skills we can leverage." The cost of the boxed DVD should be trivial by comparison to the opportunity costs of sending this message, but RH management doesn't seem to see it. That in turn means that we have a hard time trusting RH management when they make us promises about what they're doing and why.
The short version: RH engineers---good, great even. RH management---seems to be out of touch with their customer relationships.
P.S. One thing that is the engineers' fault is this silly big-bang upgrade business. One of the reasons I'm now happily using Debian is that I never need to do the whole "back up the system in the middle of the night, do the upgrade, and pray" dance. Incremental upgrading rules, and for me is a precondition for using a Linux distro. Sure, the incremental upgrade means incremental breakage. But incremental breakage is just a lot easier to manage.
True story: a friend of mine worked at a major European Linux vendor several years ago. At the time, they sold a distro that came on something like 9 CDs and contained everything under the sun, because telecom charges in Europe were extremely high. Customers wanted to be able to get things off the CD rather than paying for network bandwidth.
The distro install process required inserting all 9 CDs.
I whined, and my friend commented that he too had whined, but had gotten an interesting story as a result. Apparently management had tried putting all the important stuff on the first CD, and some customers had complained---seems they felt "ripped off, because they sold me all these extra CDs I didn't need". So they changed the installer to require all the CDs, and the complaints stopped. Problem solved.
Legendary electronics hobbyist Don Lancaster has what I consider to be the must-read page on why patents never help the individual inventor: Patent Avoidance.
Slashdot Mirror
For a different take on the antenna "gun", check out our "Trackmaster 2000" 802.11/ATV rocket tracking antenna for the PSAS launch vehicle.
I believe the tires were weighted internally to achieve this behavior. Counterintuitive, but not cheating.
I write songs...am I not supposed to have a chance to make any kind of money at this?
I write a lot of free and open source software and give it away. As a college prof, I also make money giving readings of my code. Coincidentally, I write songs that are performed publicly, even though I don't make any money at it. I'm afraid you're 0 for 3 with me.
Having never done it professionally myself...can you really do CAD/graphics editing as fast as you can move the mouse? For me, at least, thinking time dominates in most mouse-related tasks: it's not how fast I can push the button, but which button to push. If that weren't true, I think I would want a job with more thinking.
One way to tell it's time to change jobs: Increasing your mouse speed and accuracy noticeably improves your productivity.
I don't know. I think the real problem was the not giving him the right languages. Made it hard to deal with Kobold...
(1) An earlier poster gave the canonical example: new disk too big. I've also seen new RAM too big (supported by HW, but not BIOS). Certainly would be nice to be able to boot an older machine from USB, much less USB 2, with the addition of a PCI USB card. Etc.
(2) There's no way HW manufacturers are leaving the flashable-BIOS world anytime soon. Too much support cost for anything else. They could try the kind of code-signing tricks the Xbox uses, but frankly I don't see them bothering, for the same reason: big pain to manage the keys, and no real benefit to the HW vendor.
Offhand, I can think of three strong arguments for an OpenBIOS.
- Decreased obsolence: As others have noted, changes in BIOS requirements can be responded to given the BIOS source. BIOS maintenance is no longer at the whim of the hardware vendor. This could extend the useful life of hardware.
- Decreased HW lockout: As others have noted, the move toward "Trusted Computing" could easily take a sinister turn. An open BIOS would make it much harder for hardware vendors to lock out libre software.
- Increased OS compatibility: Many seem concerned about getting hardware compatibility right in an open BIOS. The flip side of this is that getting the BIOS to work with an open OS would become much easier. The premiere example is ACPI, where the BIOS often has bugs with corresponding Windows workarounds.
I think the goal of producing an open BIOS that works well on a number of machines is quite a difficult one. The rewards of achieving it, however, seem high.Hmmm. So the fact that some of us were developing quality open source desktop software and giving it away before there was such a thing as Windows supports the theory of Windows-driven OSS development how, exactly? You see, to believe that OSS desktop developers never innovate, you have to believe that all that we do results from copying commercial developers. My previous post tried to document just a few of the many, many ways in which that is patently untrue, under the highly non-controversial theory of temporal direction in causality (stuff can't be caused by something that happens afterward).
I'm not sure what your point is, in fact. I'm quite confused by your accusation of idiocy. It's possible that this is because I'm being stupid, but if so help me to understand.
You understand that the UNIX graphical desktop predates Windows and Mac, right? That the software being used to render a Linux desktop today is the direct descendant of this venerable software, and contains some of the same bits? That Keith Packard, one of the many high-profile heroes of open source, has continuously developed this open source desktop in innovative ways over the last 20 years? That, for example, he developed the secure login protocol (XDMCP) used by X thin clients on the now-open-source BSD UNIX back when Windows and Mac didn't even do multiprocessing, much less multi-user, not to mention being largely incapable of interoperating with the Internet? That more recently Packard co-developed arguably the most advanced desktop rendering system in the world (Xft/fontconfig/Xrender/Composite/Damage/Cairo)--- way more advanced than Windows and in a lot of ways more than the Mac?
Naw, you're a troll---you'd turn into stone if you ever saw the light.
The NMR article is, overall, fine. I have a physics degree, and feel that I can accurately judge an article at this level. If you don't believe me, get a physicist friend to check it for you.
Excellent troll, though.
Just as importantly, there is some potential for "ptr == NULL" to be wrong. This can happen when your libraries somehow have NULL defined to something other than 0. In the ANSI standard, ((void *)0) and (I believe) also ((char *)0) are safe. For non-ANSI compilers and for ANSI compilers where the definition of NULL involves some other kind of pointer the compiler is entitled to silently not do what you thought it was going to do. In particular, ((FILE *)0) could conceivably cause problems.
Another way to put this: the C programming language defines the sole invalid value of pointer type to be represented as 0. Thus, operations such as ! on pointer values are required to treat whatever internal bit pattern is chosen by the compiler to be the machine's "null pointer" as if it were a 0. It makes about as much sense to use some bizarre NULL symbol as a representation of 0 as it does to define BEGIN to be "{" and END to be "}". It will distract the reader who understands C, and the reader who doesn't needs fixing anyhow.
What is NULL for, then? Simple: it's the value returned by certain library calls when they want to indicate an unusual condition. Use it in comparisons where the library specification says it is to be used. Note that this was probably also a bad idea on the part of the specifiers, but it's too late now to fix it.
Apparently you've never worked on a commercial fishing boat, or painted houses, or worked in a doctor's office. As someone who did all of these as well as doing system and network administration for 20 years, you can trust me that the computer-using segment of the public is at least as smart as the rest of it.
I did understand what Gorman was claiming: I didn't have the energy at the time for a detailed commentary on his critique of Google's work. My claim is that he fundamentally misunderstands (perhaps deliberately) how Google works, and gives little credit to their demonstrated ability to improve how things work in the future. I would be shocked, first of all, if Google's book search were not able to find "other references to the subject in the book"---related references are provided for most of their other search media. Secondly, I would be rather surprised if Google doesn't find some way to make the scanned texts available as a whole online---perhaps on a pay-the-publisher-to-view model. Once all those books are scanned, it's going to be quite hard, IMHO, to keep them "locked up" from being used more generally: they just represent too valuable a resource.
When Gorman "endorses" libraries in your quoted passage, he is actually endorsing a vision of libraries that I (and many others) believe will kill them. I happen to believe that electronic communication will "supplant and obliterate all previous forms" for scholarly and literary work, just as hand-copied books supplanted the oral tradition, printed books supplanted hand-copied ones, recorded music largely supplanted sheet music, and movies largely supplanted plays. None of the "supplanted" media are dead, but in my opinion any library that concentrates on them at the expense of the newer ones will rapidly be discarded as irrelevant. Yet this is exactly what Gorman seems to be advocating.
Oh my. Good catch :-). Thanks for pointing that out.
What amuses and annoys me about Michael Gorman's comments (and yes, I did read them and understand them) is how arrogant they are. Gorman, as President-Elect of the American Library Association, is not just proud enough to say how much smarter he is than other commentators about managing information. No, he's proud enough to dream of telling Google how to manage their money. He's proud enough to characterize a whole class of people intelligent enough to operate a computer as mouth-breathing idiots.
Best of all, he's very proud of how the Universal Bibliographic Control scheme he endorses will solve the world's information access problems. Now please understand: UBC doesn't actually give anyone access to source materials. In point of fact, it seems to be a scheme for trying to assemble a meta-bibliography---in other words, a list of what printed materials you could read if you could get your hands on them. This is unlike Google, an organization crass enough to actually digitize the text of books, to put you one click away from the primary source of the information it indexes, and to maintain backup copies of that information against the loss of the primary source. It is unlike Project Gutenberg, an organization that has already published a huge number of digitized texts that are now available to anyone with Internet access. It is unlike even the bloggers, who at least make their own work fully available online. Gorman apparently has the more limited goal of indexing materials without providing access to them, while mocking the efforts of these other organizations to provide access.
On the offhand chance that Michael Gorman is reading this, let me make my position as clear as possible. I am a scientific research and (if I do say so myself) a fairly literate writer. I use Google, Wikipedia, Citeseer, Project Gutenberg, and other online information resources on a daily basis, because I've found them to be quite effective for me. I read about five fiction novels a month. The last time I used a library card catalog was about 6 months ago. The reasons for this have nothing to do with the comprehensiveness of my University library's bibliography, and everything to do with the paucity of its actual content.
I support our American public libraries, because I think they're an important bulwark in our fight for free speech. In terms of effectiveness in serving my needs and the needs of my family and friends, they are so bad that I fear for their future. Mr. Gorman, please keep in mind that when public library funding comes up for public discussion, your comments, especially given your position, are extraordinarily unhelpful. So, in the jargon of the "blog people" you so despise, please STFU.
Seriously. What possible reason would WiFi VoIP work any differently at 80MPH than in the rest Earth reference frame?
P.S. Before you say "Doppler Shift", go do the math and examine the chip specs. We have: we hope to shortly demonstrate 802.11b at Mach 2.
If you were doing that you were being irresponsible. The volume of updates that all major distributions push today means it's not feasable to use Linux on dialup and stay secure at the same time.
The security-critical subset of patches from any major vendor is easily dialup-friendly.
At any rate, who was talking about dialup? I was just talking about convenience. The current Fedora installation process involves downloading a bunch of CD isos totalling about 2.5GB, and then either burning them or loopback-mounting them. BTW, the insns say there's a DVD iso, but it doesn't actually seem to be there. Probably a good thing, since "Note: If you are using HTTP or FTP to download, some download clients cannot handle the DVD image because it is larger than 4 gigabytes." Of course, "Since the Red Hat FTP site can experience extremely high volume, you may have better success using a mirror site. Alternatively, you can join the torrent." Good luck, newbie. Heck, good luck me.
They don't want to do an apt-get upgrade and find a few hours later they've been locked out of their system by a busted PAM upgrade.
I've been running Debian unstable for more than 5 years now, and I've never had anything like that happen. Application breakage, sure, but never an unusable system. Distros like Ubuntu that actually try hard to manage these things are reportedly even better. On the other hand, I've had to wipe the root partition and try again after big-bang Red Hat upgrades several times in the past. Not very fun.
Seth, you're a good guy. And indeed, all the RH engineers I know are good guys. RH spends an inordinate amount of money on the community, and has been a real focal point for the spread of Linux.
The problem is that several years after the avoidable screwup of claiming "we're walking away from home users and the desktop: go use Windows", I still can't walk into a random computer store, as far as I know, and buy a $20 boxed DVD with any kind of Red Hat Linux on it. This is bad. I can certainly walk into any computer store and buy a copy of Windows. One of the things that made us loyal to Red Hat (and I sysadmined a network of RH boxes for many years) was that we could always get the latest "supported" bits at the corner store without any download hassle---more importantly, we could always recommend the latest bits to newbies without any download hassle.
None of this is the engineers' fault. But it still sends a message to those, like me, who are trying to figure out what Linux distro to use, and that message is "we don't care about you unless you're either rich or have skills we can leverage." The cost of the boxed DVD should be trivial by comparison to the opportunity costs of sending this message, but RH management doesn't seem to see it. That in turn means that we have a hard time trusting RH management when they make us promises about what they're doing and why.
The short version: RH engineers---good, great even. RH management---seems to be out of touch with their customer relationships.
P.S. One thing that is the engineers' fault is this silly big-bang upgrade business. One of the reasons I'm now happily using Debian is that I never need to do the whole "back up the system in the middle of the night, do the upgrade, and pray" dance. Incremental upgrading rules, and for me is a precondition for using a Linux distro. Sure, the incremental upgrade means incremental breakage. But incremental breakage is just a lot easier to manage.
True story: a friend of mine worked at a major European Linux vendor several years ago. At the time, they sold a distro that came on something like 9 CDs and contained everything under the sun, because telecom charges in Europe were extremely high. Customers wanted to be able to get things off the CD rather than paying for network bandwidth.
The distro install process required inserting all 9 CDs.
I whined, and my friend commented that he too had whined, but had gotten an interesting story as a result. Apparently management had tried putting all the important stuff on the first CD, and some customers had complained---seems they felt "ripped off, because they sold me all these extra CDs I didn't need". So they changed the installer to require all the CDs, and the complaints stopped. Problem solved.
Humans. Go figure.
Legendary electronics hobbyist Don Lancaster has what I consider to be the must-read page on why patents never help the individual inventor: Patent Avoidance.
...to be able to make a CD with hundreds of thousands of songs. My MP3 CDs seem so limited by comparison.
Eric: YHBT. --Bart
"T-Moblile?" A new low, even for Slashdot editors.
Right you are. Thanks for the correction!