Fedora's catching up fast, but Debian and Gentoo are still in the lead with respect to the number of applications available within their main package repositories. That's why their package management tools appear to work better - it's actually down to all the hard work that's been put in by the package maintainers though; the tools are nothing special (rpm provides equal or better functionality to dpkgs and ebuilds, and apt is available for rpm as well as yum).
The trouble is that the lesser number of packages for Fedora/RH encourages newbie and intermediate users to indiscriminately install packages from random places, with the expected results. If, however, you pick a handful of co-operative package repositories (e.g. dag + rpmforge only, or fedora extras + livna only, or ATrpms only), things work out pretty well. For packages that aren't available, it's best to learn to roll your own, either by porting packages from other versions/distros, or upgrading existing packages, or from scratch.
Dangerous scenarios could arise because the robot does not understand that things that are in its field of view, for instance, are not in the field of view of a person. An example might be a robot handling dangerous materials, during a construction task. Perhaps the person can't see that it's handling hot metal. A person would warn the other person, avoiding danger.
Likewise, to answer the scenario posited in the article, an intelligent robot would realise when it was presenting a trip hazard to dumb humans, and especially when such a trip would have serious consequences, and avoid parking itself there when better alternative parking places were available.
Would that be acting in the shareholders' best interests?
Unbelievably, the choice between "Do Evil" and "Do no Evil" is irrelevant as Google is obliged by law to follow the shareholders interests above everything else.
In the short-term or the long-term? That's not a rhetorical question - is there any stipulation in US regulations that mean that taking the long-term option (regardless of how tenuous the justification may be) is not acceptable?
I'd be interested to know what you count as the 'social costs'. I recently moved back to the UK from the US, and frankly regret it. I earn more than I did in the US, but my standard of living is vastly lower: my house is half the size and cost more than my US one
The things you mentioned are exactly the sort of things I'd class as social costs, as well as things like even more car-centric cities and lifestyles, higher levels of violent crime, and a wider gaper between rich and poor (which, even aside from any ethical considerations, results in pretty dysfunctional behaviour - which I've observed for myself in parts of the UK).
I can't believe how bad the NHS has become in my absence. The health care alone is enough to guarantee we'll move back ASAP.
Wouldn't getting a BUPA or PPP private healthcare plan be cheaper than moving back?
If you're an Indian who can do a job as well as an American can, why work for Indian wages in India when you can work for American wages in America?
Because being paid above-average Indian wages in India will buy you a better standard of living than average or below-average wages in America?
I'm from the UK, and I recognise that although, on exchange rate terms, I could probably get a higher income by working in the US, the extra costs (including social costs) would probably cancel out most, if not all, of the benefit. Of course, the smart thing is probably to work in the US for a short period of time, save as much as possible, then either retire in a cheap part of the world or use your previous highly-paid employment as evidence that you should be as highly-paid in a cheaper part of the world. That all sounds like a bit too much hassle for me, though...
Even better would be to allow authors to disclaim liability if their software is released with full unobfuscated source code. This is not merely self-serving bias either; it would allow anyone considering using it to audit it (or pay someone to do so on their behalf), and so any decision to use it would be at their own risk.
But without sendmail.cf foo, how will we distinguish between the best admins and the mediocre? Sendmail was more useful as a litmus test than as an MTA;)
In that the mediocre admins will bodge some hacks into sendmail.cf to make sendmail appear to perform the job they need it to, whilst the best admins will take the presence of sendmail.cf as an indication that they need to remove sendmail and replace it with something that's actually fitforpurpose?:-P
Here's a dirty little secret about the FDIC, which insures the money you "save" at the bank: the FDIC only insures about 1-2% of the nation's fractional reserves. Hence, no matter how you slice it, If another Depression-style bank run occurs, we are all fucked.
In truth, savings and investment are considered more-or-less the same thing by economists today, because in the end, they *are* the same thing.
Point taken, but there's still a difference between savings and investments; with investments, you can lose your shirt even during the good times if you pick the wrong stocks or the wrong combination of stocks.
Of course, the correct blend of savings/investment/retirement fund will depend on the risk profile of the person whose money it is...
I'm not sure the Amiga 500 would be the most appropriate example; maybe either the 1000 (at one end of the scale) or the 4000 (at the other). I'd say the Amiga has a significant place in the history of home/personal microcomputers, as I regard it as the last in the line before x86 PCs took over.
Later, after much pressure and lost credibility, Intel agreed to replace all the defective chips without requiring the customer to "prove" his case.
AMD has a unique opportunity to do the right thing: offering to replace all the defective chips. If AMD does the right thing, then it will only help AMD in its litigation against Intel and in various attempts to increase marketshare. After all, would you not prefer to buy from a reputable company instead of a dishonest, shifty company?
AMD have probably learnt from Intel's PR disaster. Without Intel's FP bug and the precedent it set, there's a good chance AMD would attempt to handle this problem the same way Intel did theirs. Business is business. An indication of this is that every component in the computer you're using probably has both documented and undocumented errata, and recalls are pretty unusual events.
Back in the day, there were heroes and groups of people who stood up and shouted out what they believed in, rather than grumbling and mutting in cramped cubicles.
Like us? (whilst not even daring to suggest that we - yet - face the same anything like the dangers the French Resistance or Anti-Apartheid campaigners faced during their respective struggles).
The kneejerk reaction against a national ID card is being driven by anarchists who favor the legalization of certain things (mostly drugs), and who think that having a national ID card will make it more difficult for them to evade the warrants they accrue when they skip out on court dates.
I don't think that's the case in the UK. The impression I get is that most of us are aware that sometimes it is necessary to break the law in order to preserve essential liberty. (e.g. the examples I gave, together with the well-made point by another poster about the American Revolution). Law != Morality.
Your solution to government criminalization of things that should not be illegal is to make it more difficult for the government to enforce the law?
You're fixing the wrong problem there, buddy.
ID cards are already required for most things.
The problem with ID cards (and particularly with associated ID databases, as we are promised they will be implemented in the UK) is that they can be mined for whatever behaviour the government of the day feels is indicative of 'criminal' behaviour regardless of a) whether it actually is, or is just prejudice of some kind or other and b) whether the majority of a well-adjusted population actually considers the behaviour criminal in the first place. The awareness that everything one does is being watched leads to a 'chilling effect' on legitimate but politically sensitive behaviour (cf. China).
Why does carrying a card stop you from doing what you want to do? How does it stop you from being 'free'? What does it disable, prevent or otherwise hinder you from doing?
The only answer that comes to my mind is "Crime". And I'm all for a government cutting that down.
The French Resistance were 'criminals' under the laws of the Vichy regime during WWII.
Nelson Mandela was a 'criminal' under the laws of Apartheid South Africa.
How about that Froid (Sigmund) study that increased the sales of this all-ingredients-cake-powder by removing the egg part and required the person to cook it to add a fresh egg.
Close; that'll be his nephew Edward Bernays, father of Public Relations:
Bernays revitalized slumping sales of Betty Crocker cake mix by using focus groups which revealed that the traditional just-add-water recipe was so simple that it made housewives feel guilty and inadequate. By simply adding an egg to the recipe, the marketers of the cake mix believed they could convince women that Betty Crocker had made cake-baking a tangible way to express their femininity to their husbands.
Other niggles: There is no support for the Pentium 2, 3 or 4 architectures. RPM is configured to reject anything with an arch above i686, you have to instruct it to specifically ignore architectures for RPMs. It's simply a case of modifying the arch compatiblity test, but they haven't done that.
eh? Most RPMs in FC and RHEL packages are supplied as.i386.rpms, which are optimized for pentium4, but using only i386 instructions. This is because due to the pentium4 architecture's long pipelines, non-optimal instruction ordering hurts performance a lot. For the architectures in between, it's less critical, and running pentium4-optimized binaries on a Pentium 2 is no usually terribly harmful. For exceptional components (the kernel, glibc, openssl libraries), i686.rpms are supplied as well as the i386.rpms. These are also pentium4-optmized, but use the entire i686 instruction set (including CMOV, which, as an aside is why these won't work on some other non-Intel i686 architectures).
Unless you've done extensive profiling, I doubt you have any solid evidence to indicate that rebuilding other chunks of the OS in ways different to the above will result in significant performance gains. If you have, present your evidence to the distro vendors and the community, and I'm sure they'll listen, willingly or not.:-)
Was it the new parliament bill that allows polititians to change the law without debate?
That'll be the Legislative and Regulatory Reform Bill currently on its second reading. Then there's also the Police and Justice Bill which amongst other nefarious things will probably make posession of nmap, nessus and smbclient illegal and punishable by upto two years in jail and a fine.
1. If you live in the UK (as I do) you can be required to provide ID already, at any time, for practically any reason, by the boys (and girls) in blue. So, no change there.
I agree with you that the card isn't that big a deal. If it was just a printed card (i.e. I can trivially verify the information that it's telling others about me, and conceal elements I don't wish to reveal), then I probably wouldn't be as exercised about the issue as I am.
2. Your records are already a case of "public" record - at least if you've got a driving license or a passport, or visited a hospital in the last few years, or have ever signed on, or if you have a bank account.
This is true, but those records are not easy to cross reference. Once we have the NIR and the associated unique keys it introduces, those with access will conceivably be able to cross reference every individual's entries in each of those databases; healthcare, criminal, background, schooling, political beliefs, assets, purchases... I leave the rest to your imagination.
There is NO way a single device is going to meet many peoples needs without being a bitch to a specific software vendor and doing everything one way.
...and this is yet another reason I built a Linux box running MythTV.
So far, it has a 300G disc, DVD+/-R/RW/RAM, 2xDVB-T cards, 1xAnalogue NICAM PAL TV/FM Radio/Composite/SVideo card, DVD-Rom, RGB SCART output at full PAL resolution and 5.1 sound output (discrete, or via co-ax). It also has an infrared remote control and full-size keyboard/mousepad. Via mplayer, it plays media encoded using most popular codecs.
And, of course, if I decide to add BluRay/HDVD at some point in the future (i.e. when the market has decided which - if either - deserves to survive), I'll be able to do so for probably £30 or less.
Firstly, I've already acknowledged that the specific version of RPM shipped in RH80 (release date of 30 September 2002) was crappy. That was an exception in over ten years of using RH and RPM. Moving on:
It's like this: Back in the late 90s and early 2000s, there was a big campaign to unify Linux, called UnitedLinux, and another one called LSB. Both pushed the sucktastic late 90s/early 2000s RPM as the required package management tool. That was why I mentioned RPM as the kind of crap that gets pushed when decisions are made for corporate or marketing reasons, rather than technical reasons.
RH wasn't involved in UnitedLinux - that was SuSE, Connectiva, SCO and TurboLinux. At least two of those (SuSE and Connectiva) used RPM, and in a heavily modified form in SuSE's case. Also, RH weren't involved in the LSB (including the writing of the spec) as they didn't want to be seen to be pushing their own agenda. Most of the LSB's early work preceded RH80 and its crappy release of RPM. Once the LSB was decided, they went for LSB certification. Presumably, then, RPM was being pushed by SuSE and Mandrake instead, if anyone Why would either of these "push" a package written and maintained by a competitor if it weren't the overall best technical solution they could find?
That's a ridiculous solution. Why should every distribution be responsible for building a binary package for every Free Software program that's out there, and updating them every release? Sounds like a huge scalability problem to me.
Did I say that? Each distro vendor will package the best, most useful packages to its target userbase. It can be seen, however, that the most popular distros (e.g. RH-alikes, Debian, SuSE, Mandriva, Gentoo) also have the largest selection of packages. Most of those do so inside the main tree, Fedora is trying to take a new approach of building a core and letting a close-knit community take care of the rest. It's working out better than I expected only a few years ago.
This is *not* a hypothetical problem. One project I follow is Battle for Wesnoth, a cool GPL-ed turn based strategy game. They release new versions, in source tarball form, quite frequenty. Typically, the Windows and MacOS X binaries come out within a day or two of release. And the GNU/Linux binaries? The most recent user submitted version might be several versions out of date. The distribution packaged versions are even more out of date that the user submitted ones. For a fast paced project like Wesnoth, being out of date is bad. (I resort to building from source, because I know exactly which libraries I need, but some distros don't package libSDL-dev, ugh)
Pick distros that you're prepared to support and prepare an official package for each, along with any dependencies that don't exist in the standard distro, in the same way as the Windows versions use Installshield or whatever. I can't speak for dpkgs and ebuilds, but once you've done the first RPM spec file, it hardly needs modification from one revision to the next. Packaging doesn't even require much in the way of technical skill to do. If the package has limited requirements, it could even be portable across distros.
Also, I'm not holding FOSS to a "higher" standard. Every other operating system out there besides GNU/Linux *has* a definative baseline you can build applications against, without requiring every application to be integrated into the OS. GNU/Linux is unique in that it has multiple incompatible baselines. If I wanted to release a cool new GPLed application right now I'd have no way of releasing a common binary package that would work for all of my users. There are too many incompatible build environments.
Firstly, it's probably best if you think of FC, Mandriva, SuSE, etc each as separate OSs, even though they share a kernel and many components. Secondly, if you must, build a statically-linked binary. That's often what the commercial vendors do for their 'officially supported' binaries. Alternatively, there are various distro-neutral packaging solutions that might be werth looking into, e.g. AutoPackage.
Slashdot Mirror
Fedora's catching up fast, but Debian and Gentoo are still in the lead with respect to the number of applications available within their main package repositories. That's why their package management tools appear to work better - it's actually down to all the hard work that's been put in by the package maintainers though; the tools are nothing special (rpm provides equal or better functionality to dpkgs and ebuilds, and apt is available for rpm as well as yum).
The trouble is that the lesser number of packages for Fedora/RH encourages newbie and intermediate users to indiscriminately install packages from random places, with the expected results. If, however, you pick a handful of co-operative package repositories (e.g. dag + rpmforge only, or fedora extras + livna only, or ATrpms only), things work out pretty well. For packages that aren't available, it's best to learn to roll your own, either by porting packages from other versions/distros, or upgrading existing packages, or from scratch.
Likewise, to answer the scenario posited in the article, an intelligent robot would realise when it was presenting a trip hazard to dumb humans, and especially when such a trip would have serious consequences, and avoid parking itself there when better alternative parking places were available.
Unbelievably, the choice between "Do Evil" and "Do no Evil" is irrelevant as Google is obliged by law to follow the shareholders interests above everything else.
In the short-term or the long-term? That's not a rhetorical question - is there any stipulation in US regulations that mean that taking the long-term option (regardless of how tenuous the justification may be) is not acceptable?
The things you mentioned are exactly the sort of things I'd class as social costs, as well as things like even more car-centric cities and lifestyles, higher levels of violent crime, and a wider gaper between rich and poor (which, even aside from any ethical considerations, results in pretty dysfunctional behaviour - which I've observed for myself in parts of the UK).
I can't believe how bad the NHS has become in my absence. The health care alone is enough to guarantee we'll move back ASAP.
Wouldn't getting a BUPA or PPP private healthcare plan be cheaper than moving back?
Because being paid above-average Indian wages in India will buy you a better standard of living than average or below-average wages in America?
I'm from the UK, and I recognise that although, on exchange rate terms, I could probably get a higher income by working in the US, the extra costs (including social costs) would probably cancel out most, if not all, of the benefit. Of course, the smart thing is probably to work in the US for a short period of time, save as much as possible, then either retire in a cheap part of the world or use your previous highly-paid employment as evidence that you should be as highly-paid in a cheaper part of the world. That all sounds like a bit too much hassle for me, though...
Even better would be to allow authors to disclaim liability if their software is released with full unobfuscated source code. This is not merely self-serving bias either; it would allow anyone considering using it to audit it (or pay someone to do so on their behalf), and so any decision to use it would be at their own risk.
In that the mediocre admins will bodge some hacks into sendmail.cf to make sendmail appear to perform the job they need it to, whilst the best admins will take the presence of sendmail.cf as an indication that they need to remove sendmail and replace it with something that's actually fit for purpose? :-P
Only to the extent that they regard scientists and engineers as modern-day seers and sorcerers.
In truth, savings and investment are considered more-or-less the same thing by economists today, because in the end, they *are* the same thing.
Point taken, but there's still a difference between savings and investments; with investments, you can lose your shirt even during the good times if you pick the wrong stocks or the wrong combination of stocks.
Of course, the correct blend of savings/investment/retirement fund will depend on the risk profile of the person whose money it is...
I'm not sure the Amiga 500 would be the most appropriate example; maybe either the 1000 (at one end of the scale) or the 4000 (at the other). I'd say the Amiga has a significant place in the history of home/personal microcomputers, as I regard it as the last in the line before x86 PCs took over.
AMD has a unique opportunity to do the right thing: offering to replace all the defective chips. If AMD does the right thing, then it will only help AMD in its litigation against Intel and in various attempts to increase marketshare. After all, would you not prefer to buy from a reputable company instead of a dishonest, shifty company?
AMD have probably learnt from Intel's PR disaster. Without Intel's FP bug and the precedent it set, there's a good chance AMD would attempt to handle this problem the same way Intel did theirs. Business is business. An indication of this is that every component in the computer you're using probably has both documented and undocumented errata, and recalls are pretty unusual events.
Like us? (whilst not even daring to suggest that we - yet - face the same anything like the dangers the French Resistance or Anti-Apartheid campaigners faced during their respective struggles).
I don't think that's the case in the UK. The impression I get is that most of us are aware that sometimes it is necessary to break the law in order to preserve essential liberty. (e.g. the examples I gave, together with the well-made point by another poster about the American Revolution). Law != Morality.
You're fixing the wrong problem there, buddy.
ID cards are already required for most things.
The problem with ID cards (and particularly with associated ID databases, as we are promised they will be implemented in the UK) is that they can be mined for whatever behaviour the government of the day feels is indicative of 'criminal' behaviour regardless of a) whether it actually is, or is just prejudice of some kind or other and b) whether the majority of a well-adjusted population actually considers the behaviour criminal in the first place. The awareness that everything one does is being watched leads to a 'chilling effect' on legitimate but politically sensitive behaviour (cf. China).
Don't get me wrong; NZ is probably the freest english-speaking country there is right now. Please, make sure it stays that way.
The only answer that comes to my mind is "Crime". And I'm all for a government cutting that down.
The French Resistance were 'criminals' under the laws of the Vichy regime during WWII.
Nelson Mandela was a 'criminal' under the laws of Apartheid South Africa.
Do I really need to go on?
Close; that'll be his nephew Edward Bernays, father of Public Relations:
- from this article.eh? Most RPMs in FC and RHEL packages are supplied as .i386.rpms, which are optimized for pentium4, but using only i386 instructions. This is because due to the pentium4 architecture's long pipelines, non-optimal instruction ordering hurts performance a lot. For the architectures in between, it's less critical, and running pentium4-optimized binaries on a Pentium 2 is no usually terribly harmful. For exceptional components (the kernel, glibc, openssl libraries), i686.rpms are supplied as well as the i386.rpms. These are also pentium4-optmized, but use the entire i686 instruction set (including CMOV, which, as an aside is why these won't work on some other non-Intel i686 architectures).
Unless you've done extensive profiling, I doubt you have any solid evidence to indicate that rebuilding other chunks of the OS in ways different to the above will result in significant performance gains. If you have, present your evidence to the distro vendors and the community, and I'm sure they'll listen, willingly or not. :-)
Fedora != Fedora Foundation. RTFA. :-)
'getting'? 'GETTING'?! They've been like this since 1998 when they introduced the Regulation of Investigatory Powers Act (RIPA).
Was it the new parliament bill that allows polititians to change the law without debate?
That'll be the Legislative and Regulatory Reform Bill currently on its second reading. Then there's also the Police and Justice Bill which amongst other nefarious things will probably make posession of nmap, nessus and smbclient illegal and punishable by upto two years in jail and a fine.
I agree with you that the card isn't that big a deal. If it was just a printed card (i.e. I can trivially verify the information that it's telling others about me, and conceal elements I don't wish to reveal), then I probably wouldn't be as exercised about the issue as I am.
2. Your records are already a case of "public" record - at least if you've got a driving license or a passport, or visited a hospital in the last few years, or have ever signed on, or if you have a bank account.
This is true, but those records are not easy to cross reference. Once we have the NIR and the associated unique keys it introduces, those with access will conceivably be able to cross reference every individual's entries in each of those databases; healthcare, criminal, background, schooling, political beliefs, assets, purchases... I leave the rest to your imagination.
Don't believe me?
So far, it has a 300G disc, DVD+/-R/RW/RAM, 2xDVB-T cards, 1xAnalogue NICAM PAL TV/FM Radio/Composite/SVideo card, DVD-Rom, RGB SCART output at full PAL resolution and 5.1 sound output (discrete, or via co-ax). It also has an infrared remote control and full-size keyboard/mousepad. Via mplayer, it plays media encoded using most popular codecs.
And, of course, if I decide to add BluRay/HDVD at some point in the future (i.e. when the market has decided which - if either - deserves to survive), I'll be able to do so for probably £30 or less.
It's like this: Back in the late 90s and early 2000s, there was a big campaign to unify Linux, called UnitedLinux, and another one called LSB. Both pushed the sucktastic late 90s/early 2000s RPM as the required package management tool. That was why I mentioned RPM as the kind of crap that gets pushed when decisions are made for corporate or marketing reasons, rather than technical reasons.
RH wasn't involved in UnitedLinux - that was SuSE, Connectiva, SCO and TurboLinux. At least two of those (SuSE and Connectiva) used RPM, and in a heavily modified form in SuSE's case. Also, RH weren't involved in the LSB (including the writing of the spec) as they didn't want to be seen to be pushing their own agenda. Most of the LSB's early work preceded RH80 and its crappy release of RPM. Once the LSB was decided, they went for LSB certification. Presumably, then, RPM was being pushed by SuSE and Mandrake instead, if anyone Why would either of these "push" a package written and maintained by a competitor if it weren't the overall best technical solution they could find?
Did I say that? Each distro vendor will package the best, most useful packages to its target userbase. It can be seen, however, that the most popular distros (e.g. RH-alikes, Debian, SuSE, Mandriva, Gentoo) also have the largest selection of packages. Most of those do so inside the main tree, Fedora is trying to take a new approach of building a core and letting a close-knit community take care of the rest. It's working out better than I expected only a few years ago.
This is *not* a hypothetical problem. One project I follow is Battle for Wesnoth, a cool GPL-ed turn based strategy game. They release new versions, in source tarball form, quite frequenty. Typically, the Windows and MacOS X binaries come out within a day or two of release. And the GNU/Linux binaries? The most recent user submitted version might be several versions out of date. The distribution packaged versions are even more out of date that the user submitted ones. For a fast paced project like Wesnoth, being out of date is bad. (I resort to building from source, because I know exactly which libraries I need, but some distros don't package libSDL-dev, ugh)
Pick distros that you're prepared to support and prepare an official package for each, along with any dependencies that don't exist in the standard distro, in the same way as the Windows versions use Installshield or whatever. I can't speak for dpkgs and ebuilds, but once you've done the first RPM spec file, it hardly needs modification from one revision to the next. Packaging doesn't even require much in the way of technical skill to do. If the package has limited requirements, it could even be portable across distros.
Also, I'm not holding FOSS to a "higher" standard. Every other operating system out there besides GNU/Linux *has* a definative baseline you can build applications against, without requiring every application to be integrated into the OS. GNU/Linux is unique in that it has multiple incompatible baselines. If I wanted to release a cool new GPLed application right now I'd have no way of releasing a common binary package that would work for all of my users. There are too many incompatible build environments.
Firstly, it's probably best if you think of FC, Mandriva, SuSE, etc each as separate OSs, even though they share a kernel and many components. Secondly, if you must, build a statically-linked binary. That's often what the commercial vendors do for their 'officially supported' binaries. Alternatively, there are various distro-neutral packaging solutions that might be werth looking into, e.g. AutoPackage.