As far as external cabling is concerned, I agree with you. But for internal cabling (which is what concerns us here, since we're talking about HDDs), it's not that bad. There are two standards for ribbon-cables (50- and 68-lines), and 3 for connectors (narrow, wide and SCA), all of which are compatible (I should know, I just bought an adapter which carries all three, and wide-to-narrow is even simpler than a gender-changer connector for your serial port).
So, please, let's just compare apples with apples, shall we?
About the protocol variant (SCSI-1/2/3, narrow/wide, async/plain/fast/ultra) there's NEGOTIATION, and it's all in the devices. They'll run slower, but I have yet to see a device which won't talk to a controller in some fashion. Maybe slowly, but they will all nicely cooperate.
It is a fact that here (Italy, EU) in front of banks and other buildings where video surveillance is used, signs stating the fact are appearing more and more often.
Never mind that those cameras are usually plainly visible:)
If you better read my post, I'm not saying that MS's hardware is the best available hardware. Only that's the best products by Microsoft.
Incidentally, I've never had a problem with them. My mouse is responsive enough to be used in business applications, which don't require as much precision as - say - professional graphics. I've never noticed any fault in its pointer handling, and it's much comfortable than any wheel mouse.
The keyboard feels good under the fingers (not as good as ye' ole IBM 'clickety' keyboard, but far better than any other keyboard I've tried). I have no KVM swich though, I use x2vnc.
IMO mouses and keyboards are the BEST products Microsoft EVER did.
I'm writing this on a Microsoft keyboard and I'm clickety-clicking on a Microsoft mouse (both hooked to my main Linux box of course).
They have a great thing: they don't crash.
One I can think of is the ability to join forests (which has some advantage if two enterprises both already running MS Windows 2k merge).
Luna is a step backwards IMO in that it MIGHT be easier for novices, but it's definitely more cumbersome for power users and requires more CPU. Luckily it can be turned off.
Some benchmark demonstrated that IPC (via pipes) is horrible under MS-WinXP as compared to MS-Win2k.
What then? Internet Explorer 6? *cough cough* MS-Smarttags.
The ability to emulate a MS-Win9X registry? Definitely maybe if you use it for games, useless garbage otherwise.
It's ability to force Microsoft's software over the competitors' with Digital cameras? Or the bundling of Media Player (for which Microsoft is risking a $1billion fine from the EU)?
Okay, I'm flamebaiting and I'll stop now, but the question is sincere. What features are you talking about?
If there was a sufficient community response, they MIGHT decide that the OSS users group is wide enough to lift whatever restrictions they have in place.
I sent them a mail too, and I'm absolutely convinced I did the Right Thing(tm).
It's a paradoxycal situation. The point here is that somebody somewhere is treating human lives as monetary value, which doesn't strike me as very ethically appleasing.
Maybe there _is_ a manner to settle this, and it's actually so simple that I'm surprised nobody evern mentioned it.
Some UN-funded agency could simply BUY the patents for some life-saving drugs from the holders (thus ensuring proper compensation for the development) and then license them free-of-charge to anybody willing to produce them (thus ensuring price-lowering via free market competition), or even just putting the patents in the Public Domain.
Sure, such an UN agency would probably be a big drain on UN funding, but I as a citizen of a "developed" country am willing to give some of my tax money for such ethically-compelling matters, and I am sure many others would agree.
Well, some of this _is_ happening.
A friend of mine was involved in a massive MP3-ization of the entire catalgoue of a big shop (to Linux servers, no less - and no, I'm not whoring, it's actually true:). The purpose was to allow customers to swipe the CD label at a listening station and be able to listen samples for all the CD tracks.
Burn in store, I don't think it'll ever happen, given the current (technological) trends: a burned copy can't carry the watermarking and macrovision and whatnot.
The point is that music labels are very set in their ways, and changing those ways would be a big leap into the unknown, with certain disadvantages and only marginal possible benefits (for the labels, of course).
Also, take into account that given the current price composition schemes, the biggest part of the pie goes to the distributor (only about 30% of the price goes to the label, about 50% to the distributor - those figures are for the books market, but AFAICS they apply to the music market too) and no label wants to piss off a distributor: it's a certain way to trouble.
I think you're missing the point.
Sure, the goods you mention tend to have the same price. But that is not necessarily the work of cartels. It's just that since the manifacturing cost and the perceived value are homogeneous among those goods. Do you care that a CD from a certain artist has been released by company A or company B? Or that the gas in your car comes from company X or Y? Unless you're in some boycott campaign, probably it won't. So since the manifacturing costs are the same, and the distribution costs are the same, the pricess will tend to be the same.
Where the thing sounds fishy is when all companies CHANGE (usually raise) prices together (which is usually the issue with oil companies)those same companies all claim way-above-the-average profit margins (which would be the issue with musci companies).
What I want to say is that probably there are cartels at work, but the homogeneous cost of CDs might very well be caused by other factors.
Software can only do so much.
Suppose that a CPU fails in your precious server: on Intel hardware you have to take the server down, replace the CPU and reboot.
Certain models of Sun hardware have CPU hot-plugging: a faulty CPU gets detected by the OS and is taken offline. A techie comes, replaces the CPU without turning the computer off, and then the new CPU is taken online without breaking a sweat.
Some goes for RAM.
Ever tried changing a broken RAM stick on Intel wihtout powering off? Don't do that, you can't.
The PC platform only has one strong point: it (used to be) is simple and widespread, and those factors mean cheap. Other architectures are not as simple or not as widespread, thus are more expensive. But they are consistent, they are scalable, and they are reliable. All things that the PC architecture is not.
>6) That the patent rights Microsoft is licensing >only apply to the Software, not to any derivatives you make.
> * I don't understand this one, someone care
> to explain? (I think that they mean that if
> I make a derivitave, it is not owned by
> Microsoft? Or, does it mean not protected by Microsoft?)
It's a fair clause actually.
Microsoft MIGHT have licensed patents and included code implementing them in the distribution. Of course they have paid the patent owners to do so. But if you distribute derivative works, you don't own those rights so you have to acquire them yourself.
Let's put this in practice: suppose that there is some code somewhere in WinCE3.0 that generates GIF images. Microsoft has paid Unisys in order to do so. If you redistribute that code, you have to contact Unisys and license the patent yourself, because Microsoft's license does not get transferred to you.
Not that software patents make any sense, mind you.
I used a Motorola cell (I don't remember the details, sorry).
The phone is horrible by itself, and the MS-Windows drivers supplied with it suck big time.
Newer Nokia Communicators should be GPRS-enabled, I think. The Ericsson T39 supports GPRS. And I am sure I am forgetting at least one other terminal.
You say that it will be faster than 56k.
It's not entirely true.
GPRS works by allotting a number of timeslots in the GSM usual time allocation fabric to packet-switched communications in each cell (as in zone covered by one ground [base] station). The maximum number of timeslots that can be allocated in this fashion is (IIRC) 8 downstream and 2 upstream, for a whooping max 115200 bps downstream and 28800 upstream.
What's the catch? The "packet switching" part, of course. If in a cell N phones are active and using all the available bandwidth, then each of those will only get 1/N of 115200 bps.
Why then using GPRS at all then? Well, first it's always-on. Then it has an higher maximum througput than plain circuit-switched (one-slot-up-one-slot-down) phone, and third it's statistics. Circuit-switched lines are losing the battle against packet-switched because while the former give guarranteed-quality services, the latter offer better resource usage, especially in conditions of bursty traffic.
This said, those prices are INSANE. In Italy (where I am writing this from) the major phone providers are offering GPRS service for FREE for a couple of more months to launch it, and the service is already available in most of the country.
Take a look at Pike, it's been covered here in the past.
It has a c-like syntax, it has the preprocessor, and it has classes (full OOP), functional programming if you want, a nice runtime library. It doesn't have pointers, but it has references (no arithmetics though), and high-level constructs for strings, sets and associations. It does manage memory quite fine (refcount + garbage collector) and you can make it as pedantic as you want, even not at all (just use the mixed type everywhere).
This said, you won't get rid of the C syntax anytime soon. It has an advantage over most other syntaxes one can come up with for an imperative language: it's concise. And if used with some discipline, it can be very readable which is one of the most important requirements for a programming language. By adhering to a c-like syntax, any other programming language can be instantly readable by a lot of people.
Many organizations developming free software will ask people to assign them copyright for their contributions. The FSF does that on FSF-sponsored projects to better the represent the free software interests, while others (Aladdin for ghostscript for instance) will do that exactly to be able to dual-license.
But the bazaar ownership might be a _desired_ effect: anybody can improve and distribute, but virtually everybody (and nobody) owns the code. This makes proprietarization virtually impossible, but it's a doube-edged sword.
Suppose for instance that a hole is discovered in GPLv2, and a GPLv3 comes out just to address that hole. In such a scenario consensus from every contributor would be required to change the licensing terms from v2 to v3, and it could be impossible to do so because somebody is unreachable.
If software is a service (and many including Microsoft seem to think it's heading that way) then you're not selling software, but SERVICES.
Red Hat for instance sells training and outsourced system management, just as IBM, Compaq, Sun, or many others do. Many companies will pay lots for that.
It is also worth to notice that the ratio of VC-funded "new economy" companies failing is not higher than it alwyas has been in the "old" economy. Bad business plans existed forever, and VCs always account in advance for this.
VC expect VERY high returns from their successful investiments (100% yearly is not unheard of). And they expect 9 out of 10 of their ventures to fail. The one surviving because it's an hot and well-conducted idea will absorb losses from the others and make extra money.
What's the difference? VISIBILITY. We are just _seeing_ the dot-coms failing, while we didn't (and don't) see less visible ventures.
Funnily enough, having a kernel-space HTTP server might actually improve security.
Follow me: a buffer overflow's exploit works by uploading some malicious assembly code to the target system, and overwriting some memory location to cause the execution flow to execute the uploaded code.
Doing so might trash the stack, but it's not a problem since the malicious code might just not care. Suppose for instance we're cracking Apache: if the 'sploit mangles the stack, the process serving the request will die for segfault or similar. The super-process will just fire up a new one, thinking that the infected process just exited because it wanted to.
Now let's suppose that the same happens to a live kernel. If the kernel crashes, there's no recovery. If the attacker wanted to change something on disk, the changes wouldn't even reach it! Flaws in such a server would be much harder to exploit, because the attacker would have to ensure system integrity while doing her own deeds.
Sure, such a thing could cause availability problems (can you say "ping of death"?) to no end, but it wouldn't be a security problem. I don't know you, but I'd take an availability problem over a security flaw any time of the day.
IMO the problem with Disney plagiarizing (or taking inspiration from) many works is that they refuse to acknowledge proper credits to the original authors. Would it cost them much? Adding a line at the end of the credits reciting "partially inspired from Osamu Tezuka's 'Jungle Taitei'"?
Maybe the original author would like to see a few bucks in royalties, but so what? it's not as if they couldn't afford it.
Somebody in other posts mentioned the GPL, and the apparent contradiction in our community which encouages to share IP when it's computer programs, but doesn't approve of derivative works when it's motion pictures.
The point is that every single GPL program I've seen properly recognizes its ancestors, be it direct and indirect, and acknowledges the work of those who created them. Disney's recent movies don't.
Slashdot Mirror
As far as external cabling is concerned, I agree with you. But for internal cabling (which is what concerns us here, since we're talking about HDDs), it's not that bad. There are two standards for ribbon-cables (50- and 68-lines), and 3 for connectors (narrow, wide and SCA), all of which are compatible (I should know, I just bought an adapter which carries all three, and wide-to-narrow is even simpler than a gender-changer connector for your serial port).
So, please, let's just compare apples with apples, shall we?
About the protocol variant (SCSI-1/2/3, narrow/wide, async/plain/fast/ultra) there's NEGOTIATION, and it's all in the devices. They'll run slower, but I have yet to see a device which won't talk to a controller in some fashion. Maybe slowly, but they will all nicely cooperate.
It is a fact that here (Italy, EU) in front of banks and other buildings where video surveillance is used, signs stating the fact are appearing more and more often. :)
Never mind that those cameras are usually plainly visible
Erm...
Wasn't Thawte bought by Verisign?
Sure, it has been challenged subsequently, but still....
I was denied using Mozilla on Linux, so your diagnosis is incorrect.
If you better read my post, I'm not saying that MS's hardware is the best available hardware. Only that's the best products by Microsoft.
Incidentally, I've never had a problem with them. My mouse is responsive enough to be used in business applications, which don't require as much precision as - say - professional graphics. I've never noticed any fault in its pointer handling, and it's much comfortable than any wheel mouse.
The keyboard feels good under the fingers (not as good as ye' ole IBM 'clickety' keyboard, but far better than any other keyboard I've tried). I have no KVM swich though, I use x2vnc.
IMO mouses and keyboards are the BEST products Microsoft EVER did.
I'm writing this on a Microsoft keyboard and I'm clickety-clicking on a Microsoft mouse (both hooked to my main Linux box of course).
They have a great thing: they don't crash.
Care to detail what those features are?
One I can think of is the ability to join forests (which has some advantage if two enterprises both already running MS Windows 2k merge).
Luna is a step backwards IMO in that it MIGHT be easier for novices, but it's definitely more cumbersome for power users and requires more CPU. Luckily it can be turned off.
Some benchmark demonstrated that IPC (via pipes) is horrible under MS-WinXP as compared to MS-Win2k.
What then? Internet Explorer 6? *cough cough* MS-Smarttags.
The ability to emulate a MS-Win9X registry? Definitely maybe if you use it for games, useless garbage otherwise.
It's ability to force Microsoft's software over the competitors' with Digital cameras? Or the bundling of Media Player (for which Microsoft is risking a $1billion fine from the EU)?
Okay, I'm flamebaiting and I'll stop now, but the question is sincere. What features are you talking about?
He who holds the purse is in command.
Let's hope in the EU - which incidentally recently opened up a second investigation against MS for anticompetitive practices.
Pray tell, Why?
If there was a sufficient community response, they MIGHT decide that the OSS users group is wide enough to lift whatever restrictions they have in place.
I sent them a mail too, and I'm absolutely convinced I did the Right Thing(tm).
It's a paradoxycal situation. The point here is that somebody somewhere is treating human lives as monetary value, which doesn't strike me as very ethically appleasing.
Maybe there _is_ a manner to settle this, and it's actually so simple that I'm surprised nobody evern mentioned it.
Some UN-funded agency could simply BUY the patents for some life-saving drugs from the holders (thus ensuring proper compensation for the development) and then license them free-of-charge to anybody willing to produce them (thus ensuring price-lowering via free market competition), or even just putting the patents in the Public Domain.
Sure, such an UN agency would probably be a big drain on UN funding, but I as a citizen of a "developed" country am willing to give some of my tax money for such ethically-compelling matters, and I am sure many others would agree.
Well, some of this _is_ happening. :). The purpose was to allow customers to swipe the CD label at a listening station and be able to listen samples for all the CD tracks.
A friend of mine was involved in a massive MP3-ization of the entire catalgoue of a big shop (to Linux servers, no less - and no, I'm not whoring, it's actually true
Burn in store, I don't think it'll ever happen, given the current (technological) trends: a burned copy can't carry the watermarking and macrovision and whatnot.
The point is that music labels are very set in their ways, and changing those ways would be a big leap into the unknown, with certain disadvantages and only marginal possible benefits (for the labels, of course).
Also, take into account that given the current price composition schemes, the biggest part of the pie goes to the distributor (only about 30% of the price goes to the label, about 50% to the distributor - those figures are for the books market, but AFAICS they apply to the music market too) and no label wants to piss off a distributor: it's a certain way to trouble.
I think you're missing the point.
Sure, the goods you mention tend to have the same price. But that is not necessarily the work of cartels. It's just that since the manifacturing cost and the perceived value are homogeneous among those goods. Do you care that a CD from a certain artist has been released by company A or company B? Or that the gas in your car comes from company X or Y? Unless you're in some boycott campaign, probably it won't. So since the manifacturing costs are the same, and the distribution costs are the same, the pricess will tend to be the same.
Where the thing sounds fishy is when all companies CHANGE (usually raise) prices together (which is usually the issue with oil companies)those same companies all claim way-above-the-average profit margins (which would be the issue with musci companies).
What I want to say is that probably there are cartels at work, but the homogeneous cost of CDs might very well be caused by other factors.
Software can only do so much.
Suppose that a CPU fails in your precious server: on Intel hardware you have to take the server down, replace the CPU and reboot.
Certain models of Sun hardware have CPU hot-plugging: a faulty CPU gets detected by the OS and is taken offline. A techie comes, replaces the CPU without turning the computer off, and then the new CPU is taken online without breaking a sweat.
Some goes for RAM.
Ever tried changing a broken RAM stick on Intel wihtout powering off? Don't do that, you can't.
The PC platform only has one strong point: it (used to be) is simple and widespread, and those factors mean cheap. Other architectures are not as simple or not as widespread, thus are more expensive. But they are consistent, they are scalable, and they are reliable. All things that the PC architecture is not.
>6) That the patent rights Microsoft is licensing >only apply to the Software, not to any derivatives you make.
> * I don't understand this one, someone care
> to explain? (I think that they mean that if
> I make a derivitave, it is not owned by
> Microsoft? Or, does it mean not protected by Microsoft?)
It's a fair clause actually.
Microsoft MIGHT have licensed patents and included code implementing them in the distribution. Of course they have paid the patent owners to do so. But if you distribute derivative works, you don't own those rights so you have to acquire them yourself.
Let's put this in practice: suppose that there is some code somewhere in WinCE3.0 that generates GIF images. Microsoft has paid Unisys in order to do so. If you redistribute that code, you have to contact Unisys and license the patent yourself, because Microsoft's license does not get transferred to you.
Not that software patents make any sense, mind you.
I used a Motorola cell (I don't remember the details, sorry).
The phone is horrible by itself, and the MS-Windows drivers supplied with it suck big time.
Newer Nokia Communicators should be GPRS-enabled, I think. The Ericsson T39 supports GPRS. And I am sure I am forgetting at least one other terminal.
You say that it will be faster than 56k.
It's not entirely true.
GPRS works by allotting a number of timeslots in the GSM usual time allocation fabric to packet-switched communications in each cell (as in zone covered by one ground [base] station). The maximum number of timeslots that can be allocated in this fashion is (IIRC) 8 downstream and 2 upstream, for a whooping max 115200 bps downstream and 28800 upstream.
What's the catch? The "packet switching" part, of course. If in a cell N phones are active and using all the available bandwidth, then each of those will only get 1/N of 115200 bps.
Why then using GPRS at all then? Well, first it's always-on. Then it has an higher maximum througput than plain circuit-switched (one-slot-up-one-slot-down) phone, and third it's statistics. Circuit-switched lines are losing the battle against packet-switched because while the former give guarranteed-quality services, the latter offer better resource usage, especially in conditions of bursty traffic.
This said, those prices are INSANE. In Italy (where I am writing this from) the major phone providers are offering GPRS service for FREE for a couple of more months to launch it, and the service is already available in most of the country.
Also, didn't Sun agree to add templates to Java just last week?
Take a look at Pike, it's been covered here in the past.
It has a c-like syntax, it has the preprocessor, and it has classes (full OOP), functional programming if you want, a nice runtime library. It doesn't have pointers, but it has references (no arithmetics though), and high-level constructs for strings, sets and associations. It does manage memory quite fine (refcount + garbage collector) and you can make it as pedantic as you want, even not at all (just use the mixed type everywhere).
This said, you won't get rid of the C syntax anytime soon. It has an advantage over most other syntaxes one can come up with for an imperative language: it's concise. And if used with some discipline, it can be very readable which is one of the most important requirements for a programming language. By adhering to a c-like syntax, any other programming language can be instantly readable by a lot of people.
Many organizations developming free software will ask people to assign them copyright for their contributions. The FSF does that on FSF-sponsored projects to better the represent the free software interests, while others (Aladdin for ghostscript for instance) will do that exactly to be able to dual-license.
But the bazaar ownership might be a _desired_ effect: anybody can improve and distribute, but virtually everybody (and nobody) owns the code. This makes proprietarization virtually impossible, but it's a doube-edged sword.
Suppose for instance that a hole is discovered in GPLv2, and a GPLv3 comes out just to address that hole. In such a scenario consensus from every contributor would be required to change the licensing terms from v2 to v3, and it could be impossible to do so because somebody is unreachable.
If software is a service (and many including Microsoft seem to think it's heading that way) then you're not selling software, but SERVICES.
Red Hat for instance sells training and outsourced system management, just as IBM, Compaq, Sun, or many others do. Many companies will pay lots for that.
It is also worth to notice that the ratio of VC-funded "new economy" companies failing is not higher than it alwyas has been in the "old" economy. Bad business plans existed forever, and VCs always account in advance for this.
VC expect VERY high returns from their successful investiments (100% yearly is not unheard of). And they expect 9 out of 10 of their ventures to fail. The one surviving because it's an hot and well-conducted idea will absorb losses from the others and make extra money.
What's the difference? VISIBILITY. We are just _seeing_ the dot-coms failing, while we didn't (and don't) see less visible ventures.
It is, and it was (at least in Italian).
Funnily enough, having a kernel-space HTTP server might actually improve security.
Follow me: a buffer overflow's exploit works by uploading some malicious assembly code to the target system, and overwriting some memory location to cause the execution flow to execute the uploaded code.
Doing so might trash the stack, but it's not a problem since the malicious code might just not care. Suppose for instance we're cracking Apache: if the 'sploit mangles the stack, the process serving the request will die for segfault or similar. The super-process will just fire up a new one, thinking that the infected process just exited because it wanted to.
Now let's suppose that the same happens to a live kernel. If the kernel crashes, there's no recovery. If the attacker wanted to change something on disk, the changes wouldn't even reach it! Flaws in such a server would be much harder to exploit, because the attacker would have to ensure system integrity while doing her own deeds.
Sure, such a thing could cause availability problems (can you say "ping of death"?) to no end, but it wouldn't be a security problem. I don't know you, but I'd take an availability problem over a security flaw any time of the day.
Tux handles dynamic content simply passing it to an user-level server such as Apache, Roxen, Zeus or whatever else you wish to use.
This means that were this a purely static-content benchmark, the results would have been much,much more in favour of Tux.
IMO the problem with Disney plagiarizing (or taking inspiration from) many works is that they refuse to acknowledge proper credits to the original authors. Would it cost them much? Adding a line at the end of the credits reciting "partially inspired from Osamu Tezuka's 'Jungle Taitei'"?
Maybe the original author would like to see a few bucks in royalties, but so what? it's not as if they couldn't afford it.
Somebody in other posts mentioned the GPL, and the apparent contradiction in our community which encouages to share IP when it's computer programs, but doesn't approve of derivative works when it's motion pictures.
The point is that every single GPL program I've seen properly recognizes its ancestors, be it direct and indirect, and acknowledges the work of those who created them. Disney's recent movies don't.