The common accepted price for many products is artificially maintained. Your musicians have a completely different price structure to labels, and yet they sell in the same range, because they can, because that's the accepted price. Same reason a "CD" from iTMS costs "about $10".
Many of the components of that $2.50 you started with are independant of the material on it, and the same whether they're CDs or DVDs or computer software (commercial, freeware, or shareware). The distribution and production is the same, no matter what you put on them. And yet some of these have "common accepted prices" of $5.00, of $10.00, of $20.00,... and the cheapest are "royalty free" movies, on more expensive media.
People need to make profit, and that includes things like making more on the popular CDs so you don't go out of business when you get stuck with a million dollars or so of crappys titles.
Indeed, when the demand's high, you can charge more. We're not talking about charging more for popular CDs, and we're not talking about why they engage in price fixing, or even whether they should or shouldn't. We're just talking about how it happens, and the overly complex distribution mechanism is part of it.
All your model does is move the middleman over to the "middleman" of the warehouses of places like Best Buy and Target.
Their warehousing costs aren't $7.50 per item or they'd go out of business... most of their products cost less than that, and their margins aren't 300% or they'd be undersold. Like Apple, they negotiate with the labels for what they get to buy and sell stuff for, and they get to buy stuff at a price that lets them sell for the price the labels want them to sell it at. If they undercut them too much, they wouldn't get to start that high on the distribution chain.
Even more reason to don't be too much on the RIM side, it itself was sueing handspring just because they used a 'small format' keyboard on their machines.
Fair enough. Handspring was "Palm Version II" and was also started by Jeff Hawkins... and is now part of Palm again. So... Palm sued Royal because they had a handheld with a 68000 and a stylus (and was selling it for less than Palm). Xerox sued Palm because Graffiti (based on Hawkins college thesis) allegedly violated their Unistroke patent (for which Hawkins' thesis was prior art). How far back do you want to take the chain back to find a company that isn't a patent troll?
Like I said, "I don't see people being so much pro-RIM as anti-patent." I still don't see people being particularly "pro-RIM", and I suspect that RIM settled to keep from hurting their ability to abuse patents in the future.
"I always think of how it can cost 99 cents to download a full song from iTunes... but then a ringtone... costs 3 dollars"
That's because p2p networks still keeps prices on downloads down.
$0.99 per song isn't "cheap"... iTMS attraction is that I don't have to buy 10 songs I don't like to get one I do. If I like the music enough to want the whole album, I buy the CD.
It's the album price that limits the iTMS price. They couldn't get away with charging significantly more than CDs on iTMS when they get less for it.
The prices of ringtones are high because you only need to buy a few, maybe even one, and unless you're a total pop culture slut once you find one you like you're unlikely to buy another for six months.
And they don't need P2P ringtones. If you're savvy enough to be using P2P and you have a cellphone, there's bunches of programs out there that will let you take any chunk of a song and turn it into a ringtone for all kinds of phones. People don't care, because the $3 for the ringtone is nothing compared to the $1000+ they're paying for the phone over a 2 year contract.
Speaking of which, is the music or the cellphone industry a bigger rip-off?
The "product" costs $2.50 to get out of the factory to the distributor. That sounds reasonable, I'll buy those prices.
I don't buy two more doublings from there to the stores. If there's 300% profit between the distributor and the public, then someone's going to come in and buy from the distributor and ship directly to their stores, and sell them for $5.00.
If you can't do that, because none of the distributors will sell direct to retail, then guess what... that's price fixing.
You mention Open Office and Firefox. The relationship between these products and "open source" is mixed at best.
Open Office isn't really an Open Source project, it's a commercial product that was open-sourced after it was Gatesed to death.
Mozilla/Firefox is an odd beast. Mosaic out as semi-open-source and benefitted from the same kind of feedback as real open source products. The relationship between Netscape and Mosaic and whether "Netscape Mosaic" shared more than a name with Mosaic aside, Netscape's product was at least a reimplementaion of Mosaic, and Mozilla/Firefox is yet another reimplementation, funded at first as an upgrade for Netscape using the power of the Open Source model... and the result isn't unequivocally good.
Meanwhile, Microsoft started out with the Mosaic code base to produce Internet Explorer.
The relationship between open source products and commercial ones is complex, but theer's damn few FOSS projects that have produced top notch products that appeal to people other than the software developer crowd that haven't a goodly portion of commercial development involved.
We're talking about patents that could have conceivably covered any possible paging system and turned on things like which servers the messages were stored on before sending them to the customer. Patents so broad should be invalid.
So whether RIM were angels or devils, NTP are no heroes either. This was a battle between two patent abusers. My only complaint is that by caving in RIM made it harder to use this case as precedent for the next time some patent abuser goes after an inventor. Which was probably deliberate, given their track record.
Reminds me of the way Palm boned Royal, and then got boned in turn by Xerox.
Umm if it was their patent, they deserve compensation. Its only fair. ( now the dollar amount, we can debate, but not the concept )
We absolutely can debate the concept.
Even assuming you consider software patents to be legitimate (which is debatable), the holders of many patents are not the people who invented the process or product, they may not even have been involved in it... they're simply the ones who were first to *describe* something that a judge can be convinced includes that process.
Here's how Richard Feynman got the original patent on the nuclear rocket. A "very nice fella" was bugging him about writing down all the things he could think of that nuclear power could be used for:
There are so many ideas about nuclear energy that are so perfectly obvious, that I'd be here all day telling you stuff. Example: nuclear reactor...under water...water goes in...steam goes out the other side...Pshshshsht -- it's a submarine. Or: nuclear reactor...air comes rushing in the front...heated up by nuclear reaction...out the back it goes...Boom! Through the air -- it's an airplane. Or: nuclear reactor...you have hydrogen go through the thing...Zoom! -- it's a rocket....There's a million ideas!" I said, as I went out the door. -- Surely you're Joking, Mr. Feynman
There's millions of patents that are just as obvious.
There's a guy who spent years threatening anyone who made a screen protector for a PDA because he had a patent for attaching a screen protector to a curved surface for a fish radar... and then, after several screen protectors came on the market, he got his patent amended to include flat screens. He didn't invent them, he didn't even file his amendment until after other products were out, but he was able to get people to pay "licensing fees" to use this bogus patent.
And that's all assuming the patent's even valid. If the patent's invalid, they don't "deserve" a thing.
Nah, the explorer downloads an open-source scanning-tunnelling microscope and feeds it into his matter compiler. A few minutes later he's scanning the pit patterns and uploading the resulting images into VR so he can use the CDR-simulator plugin for Mame to read the data.
When HP bought Compaq, they had just come out with a refresh for the Hp Jornada Pocket PC that made it one of the best devices out. While the Rube-Goldberg iPaq had a larger market share by dint of being the first device out using the ARM processor that became the standard with Pocket PC 2002, it was ungainly to actually use... you pretty much had to add a sleeve AND a protective case, making the resulting device too large to conveniantly carry in your pocket, the Jornada 568 with its hard shell and built-in CF slot was only a little bit larger than the "naked" iPaq, but it was complete and self-contained... and the electronics were virtualy identical.
So what did HP do? It killed the Jornada and came out with a stripped down non-expandible baby iPaq for the "Jornada Fans". Me, I took this as an opportunity to go back to PalmOS.
There's plenty of OTHER blame to go around: Microsoft, Palm, Sony, Jaff Hawkins and Handspring's ADD-fueled product-line shuffling. But HP sure owns a fair slice of the blame.
When it comes to software used in education, shouldn't the students be part of the system? When I was in high school and college the computer systems and other specialised equipment that students used, from glassware to oscilloscopes, from sports equipment to metalworking tools, were maintained by students and staff working together.
That's what computers in education should be all about!
Now if you're talking about the administrative systems, the front- and back- office systems that the students don't have anything to do with, then that's just the old question of software in any office environment. The fact that the office is at a school is irrelevant - it comes down to the competance of your staff and your willingness and ability to cross-train people.
One that people have these in their hands, I'd like to see...
1. Duo versus Solo benchmarks, particularly the benchmarks Apple used to tout the performance of the original iMac and Macbook Pro. Multithreaded benchmarks will get WAY better results on a multiprocessor because the saved context switches can give you superlinear speedup.
The slowest part of software development is the developer.
Compiles can be distributed over multiple computers, including servers, when you're using command-line tools (including XCode, which is a wrapper around GCC) so as long as the actual desktop's fast enough to run the editors and test the application local horsepower is less critical.
One thing to remember is that actually running your software with 64-bit addressing has a performance penalty, and while you can fit more memory in a processor with a wider address bus few applications will actually use it... what it does give you is more cache and the ability to more efficiently run multiple large applications concurrently.
There's only two systems I know of where 64-bit mode gives an actual performance increase on applications that are not actually written to take advantage of large sparse address spaces. That's the Alpha, which doesn't really have a 32-bit mode, and the AMD, where the 32-bit mode is a compatibility mode for Intel's register-crippled x86 architecture.
Power PC didn't have that problem, and I have read that Intel's AMD64 implementation isn't any speed demon... so that means it's worthwhile to look at ways to alleviate the memory crunch.
If you really *are* using that much RAM, setting up an interleaved RAID-0 array just for your swap (since OS X unhappily doesn't seem to support multiple swap files or partitions) might be worthwhile... or even getting a physical RAM disk for it... bringing back the old days where you'd have a high speed (for the time) fixed-head disk for swap. Get the fastest disks you can, and don't worry about how big they are or if you're using them "efficiently".
An Intel Core Solo at 1.5GHz with 2MB of L2 Cache onboard and a 667 Mhz Frontside Bus. (was a PowerPC G4 processor at 1.25GHz with 512K of L2 Cache onboard and a 167 Mhz Frontside Bus.)
An Intel Core Solo at 1.5 GHz with 2MB and 666 MHz FSB (was, PowerPC G4 at 1.25 or 1.33 GHz with 512K of L2 and a 166 Mhz FSB)
-or-
An Intel Core Duo at 1.66 GHz with 2MB and 666 MHzFSB (was, Power PC G4 at 1.42 or 1.5 GHz with 512K of L2 and a 166 MHz FSB)
(I know, round up, but 666 MHz is mildly amusing)
Things You give up:
ATI's Radeon 9200 with 32MB of DDR SDRAM for Intel's GMA950 graphics processor with 64MB of DDR2 SDRAM shared with main memory
If I was working for an antivirus company I'd be happy Apple was starting to go down the Microsoft path, where there's so many security holes in the OS that the whole fundamentally broken approach of "antivirus" software is the only way to stay secure.
Automatic opening of untrusted content is a VERY dangerous operation, and if it is to be implemented at all it MUST be implemented using an interface that's dedicated to the execution of trusted handlers for that content.
Using LaunchServices or Windows Explorer when starting applications to open attachments or downloaded files is like passing CGI variables to programs using system(). The requirements for an API that's used by local applications to open local content are fundamentally different from those for an API used by applications handling untrusted content. Instead of trying to make sure that every local application is crippled and bound about with warnings and alerts, they should be implementing a locked down interface just for browsers, mail software, and so on...
As I predicted, Apple has merely patched the current incarnation of this flaw. It hasn't changed the flawed model of treating files as "safe".
As I noted the last time, there will be future exploits of this type until Apple takes the simple step of establishing a parallel registry of trusted programs, rather than trying (in vain) to echieve a balance between convenience and security using a single list of helpers and URI handlers for both local and untrusted content.
Ironically, the fix in 10.3.9 (do not open the content) is better than the fix in 10.4.5 (warn the user about the content).
I'm not a gamer, but I DO use a lot of OpenGL-based apps, in 3d, and while the existing mini's GPU is limited what it does do it does well. What does the "advanced video" do that I should care about?
I don't want a "performance GPU", but I want one that's going to do OpenGL better than what I already have. They've got a choice of two different Core Image capable GPUs in the Macbook and iBook they could have used. But... no...
Intel must be soaking them on the Core Solo/Duo if they feel they have to risk using Intel GPUs in a Mac to keep the price down.
Well, I guess Apple's decided to terminally cripple the mini. GMA950? As if the limited VRAM in the original model wasn't bad enough, they stick a chip in that only nominally supports 3d in the quote-upgrade-unquote.
I'll keep my old "4 times slower" mini with a functional GPU.
Slashdot Mirror
The common accepted price for CDs is about $10.
... and the cheapest are "royalty free" movies, on more expensive media.
The common accepted price for many products is artificially maintained. Your musicians have a completely different price structure to labels, and yet they sell in the same range, because they can, because that's the accepted price. Same reason a "CD" from iTMS costs "about $10".
Many of the components of that $2.50 you started with are independant of the material on it, and the same whether they're CDs or DVDs or computer software (commercial, freeware, or shareware). The distribution and production is the same, no matter what you put on them. And yet some of these have "common accepted prices" of $5.00, of $10.00, of $20.00,
People need to make profit, and that includes things like making more on the popular CDs so you don't go out of business when you get stuck with a million dollars or so of crappys titles.
Indeed, when the demand's high, you can charge more. We're not talking about charging more for popular CDs, and we're not talking about why they engage in price fixing, or even whether they should or shouldn't. We're just talking about how it happens, and the overly complex distribution mechanism is part of it.
All your model does is move the middleman over to the "middleman" of the warehouses of places like Best Buy and Target.
Their warehousing costs aren't $7.50 per item or they'd go out of business... most of their products cost less than that, and their margins aren't 300% or they'd be undersold. Like Apple, they negotiate with the labels for what they get to buy and sell stuff for, and they get to buy stuff at a price that lets them sell for the price the labels want them to sell it at. If they undercut them too much, they wouldn't get to start that high on the distribution chain.
they do pay their dues.
Can you name a musician outside Russia that has recieved any dues from AllOfMP3?
It's more to do with the difference in exchange rates/sales prices - allofmp3 prices are *high* compared to the street price of CDs in russia.
Really? You can buy legal CDs of US artists for those kinds of prices in Russia?
Even more reason to don't be too much on the RIM side, it itself was sueing handspring just because they used a 'small format' keyboard on their machines.
Fair enough. Handspring was "Palm Version II" and was also started by Jeff Hawkins... and is now part of Palm again. So... Palm sued Royal because they had a handheld with a 68000 and a stylus (and was selling it for less than Palm). Xerox sued Palm because Graffiti (based on Hawkins college thesis) allegedly violated their Unistroke patent (for which Hawkins' thesis was prior art). How far back do you want to take the chain back to find a company that isn't a patent troll?
Like I said, "I don't see people being so much pro-RIM as anti-patent." I still don't see people being particularly "pro-RIM", and I suspect that RIM settled to keep from hurting their ability to abuse patents in the future.
"I always think of how it can cost 99 cents to download a full song from iTunes... but then a ringtone... costs 3 dollars"
That's because p2p networks still keeps prices on downloads down.
$0.99 per song isn't "cheap"... iTMS attraction is that I don't have to buy 10 songs I don't like to get one I do. If I like the music enough to want the whole album, I buy the CD.
It's the album price that limits the iTMS price. They couldn't get away with charging significantly more than CDs on iTMS when they get less for it.
The prices of ringtones are high because you only need to buy a few, maybe even one, and unless you're a total pop culture slut once you find one you like you're unlikely to buy another for six months.
And they don't need P2P ringtones. If you're savvy enough to be using P2P and you have a cellphone, there's bunches of programs out there that will let you take any chunk of a song and turn it into a ringtone for all kinds of phones. People don't care, because the $3 for the ringtone is nothing compared to the $1000+ they're paying for the phone over a 2 year contract.
Speaking of which, is the music or the cellphone industry a bigger rip-off?
I don't know what the Attorney General is supposed to do about the Russian Mafia. Do any artists actually get any money from them? I don't think so...
Let's cut to the chase, shall we?
The "product" costs $2.50 to get out of the factory to the distributor. That sounds reasonable, I'll buy those prices.
I don't buy two more doublings from there to the stores. If there's 300% profit between the distributor and the public, then someone's going to come in and buy from the distributor and ship directly to their stores, and sell them for $5.00.
If you can't do that, because none of the distributors will sell direct to retail, then guess what... that's price fixing.
You mention Open Office and Firefox. The relationship between these products and "open source" is mixed at best.
Open Office isn't really an Open Source project, it's a commercial product that was open-sourced after it was Gatesed to death.
Mozilla/Firefox is an odd beast. Mosaic out as semi-open-source and benefitted from the same kind of feedback as real open source products. The relationship between Netscape and Mosaic and whether "Netscape Mosaic" shared more than a name with Mosaic aside, Netscape's product was at least a reimplementaion of Mosaic, and Mozilla/Firefox is yet another reimplementation, funded at first as an upgrade for Netscape using the power of the Open Source model... and the result isn't unequivocally good.
Meanwhile, Microsoft started out with the Mosaic code base to produce Internet Explorer.
The relationship between open source products and commercial ones is complex, but theer's damn few FOSS projects that have produced top notch products that appeal to people other than the software developer crowd that haven't a goodly portion of commercial development involved.
I'm a bit surprised of the slashdot crowd being so pro-RIM here, maybe it's an allergic reaction to patents and patent-farming companies in general.
I don't see people being so much pro-RIM as anti-patent.
Furthermore we saw some pretty dubious US government-supports-US corporation things happening in the process
Isn't RIM a Canadian company?
We're talking about patents that could have conceivably covered any possible paging system and turned on things like which servers the messages were stored on before sending them to the customer. Patents so broad should be invalid.
So whether RIM were angels or devils, NTP are no heroes either. This was a battle between two patent abusers. My only complaint is that by caving in RIM made it harder to use this case as precedent for the next time some patent abuser goes after an inventor. Which was probably deliberate, given their track record.
Reminds me of the way Palm boned Royal, and then got boned in turn by Xerox.
We absolutely can debate the concept.
Even assuming you consider software patents to be legitimate (which is debatable), the holders of many patents are not the people who invented the process or product, they may not even have been involved in it... they're simply the ones who were first to *describe* something that a judge can be convinced includes that process.
Here's how Richard Feynman got the original patent on the nuclear rocket. A "very nice fella" was bugging him about writing down all the things he could think of that nuclear power could be used for:There's millions of patents that are just as obvious.
There's a guy who spent years threatening anyone who made a screen protector for a PDA because he had a patent for attaching a screen protector to a curved surface for a fish radar... and then, after several screen protectors came on the market, he got his patent amended to include flat screens. He didn't invent them, he didn't even file his amendment until after other products were out, but he was able to get people to pay "licensing fees" to use this bogus patent.
And that's all assuming the patent's even valid. If the patent's invalid, they don't "deserve" a thing.
Forking code is (relatively) easy. So is rewriting an entire project from scratch.
Some projects are easy to rewrite from scratch. Some aren't.
How long would it take to rewrite gcc or KHTML?
Nah, the explorer downloads an open-source scanning-tunnelling microscope and feeds it into his matter compiler. A few minutes later he's scanning the pit patterns and uploading the resulting images into VR so he can use the CDR-simulator plugin for Mame to read the data.
DRM and Open Source.
How about you solve the Palestinian question first? It's easier.
When HP bought Compaq, they had just come out with a refresh for the Hp Jornada Pocket PC that made it one of the best devices out. While the Rube-Goldberg iPaq had a larger market share by dint of being the first device out using the ARM processor that became the standard with Pocket PC 2002, it was ungainly to actually use... you pretty much had to add a sleeve AND a protective case, making the resulting device too large to conveniantly carry in your pocket, the Jornada 568 with its hard shell and built-in CF slot was only a little bit larger than the "naked" iPaq, but it was complete and self-contained... and the electronics were virtualy identical.
So what did HP do? It killed the Jornada and came out with a stripped down non-expandible baby iPaq for the "Jornada Fans". Me, I took this as an opportunity to go back to PalmOS.
There's plenty of OTHER blame to go around: Microsoft, Palm, Sony, Jaff Hawkins and Handspring's ADD-fueled product-line shuffling. But HP sure owns a fair slice of the blame.
When it comes to software used in education, shouldn't the students be part of the system? When I was in high school and college the computer systems and other specialised equipment that students used, from glassware to oscilloscopes, from sports equipment to metalworking tools, were maintained by students and staff working together.
That's what computers in education should be all about!
Now if you're talking about the administrative systems, the front- and back- office systems that the students don't have anything to do with, then that's just the old question of software in any office environment. The fact that the office is at a school is irrelevant - it comes down to the competance of your staff and your willingness and ability to cross-train people.
One that people have these in their hands, I'd like to see...
1. Duo versus Solo benchmarks, particularly the benchmarks Apple used to tout the performance of the original iMac and Macbook Pro. Multithreaded benchmarks will get WAY better results on a multiprocessor because the saved context switches can give you superlinear speedup.
2. OpenGL benchmarks versus the original mini.
The slowest part of software development is the developer.
Compiles can be distributed over multiple computers, including servers, when you're using command-line tools (including XCode, which is a wrapper around GCC) so as long as the actual desktop's fast enough to run the editors and test the application local horsepower is less critical.
One thing to remember is that actually running your software with 64-bit addressing has a performance penalty, and while you can fit more memory in a processor with a wider address bus few applications will actually use it... what it does give you is more cache and the ability to more efficiently run multiple large applications concurrently.
There's only two systems I know of where 64-bit mode gives an actual performance increase on applications that are not actually written to take advantage of large sparse address spaces. That's the Alpha, which doesn't really have a 32-bit mode, and the AMD, where the 32-bit mode is a compatibility mode for Intel's register-crippled x86 architecture.
Power PC didn't have that problem, and I have read that Intel's AMD64 implementation isn't any speed demon... so that means it's worthwhile to look at ways to alleviate the memory crunch.
If you really *are* using that much RAM, setting up an interleaved RAID-0 array just for your swap (since OS X unhappily doesn't seem to support multiple swap files or partitions) might be worthwhile... or even getting a physical RAM disk for it... bringing back the old days where you'd have a high speed (for the time) fixed-head disk for swap. Get the fastest disks you can, and don't worry about how big they are or if you're using them "efficiently".
Comparing what's shipping:
An Intel Core Solo at 1.5GHz with 2MB of L2 Cache onboard and a 667 Mhz Frontside Bus. (was a PowerPC G4 processor at 1.25GHz with 512K of L2 Cache onboard and a 167 Mhz Frontside Bus.)
An Intel Core Solo at 1.5 GHz with 2MB and 666 MHz FSB (was, PowerPC G4 at 1.25 or 1.33 GHz with 512K of L2 and a 166 Mhz FSB)
-or-
An Intel Core Duo at 1.66 GHz with 2MB and 666 MHzFSB (was, Power PC G4 at 1.42 or 1.5 GHz with 512K of L2 and a 166 MHz FSB)
(I know, round up, but 666 MHz is mildly amusing)
Things You give up:
ATI's Radeon 9200 with 32MB of DDR SDRAM for Intel's GMA950 graphics processor with 64MB of DDR2 SDRAM shared with main memory
ATI Radeon 9200 with 32MB or 64MB.
If I was working for an antivirus company I'd be happy Apple was starting to go down the Microsoft path, where there's so many security holes in the OS that the whole fundamentally broken approach of "antivirus" software is the only way to stay secure.
Automatic opening of untrusted content is a VERY dangerous operation, and if it is to be implemented at all it MUST be implemented using an interface that's dedicated to the execution of trusted handlers for that content.
Using LaunchServices or Windows Explorer when starting applications to open attachments or downloaded files is like passing CGI variables to programs using system(). The requirements for an API that's used by local applications to open local content are fundamentally different from those for an API used by applications handling untrusted content. Instead of trying to make sure that every local application is crippled and bound about with warnings and alerts, they should be implementing a locked down interface just for browsers, mail software, and so on...
As I predicted, Apple has merely patched the current incarnation of this flaw. It hasn't changed the flawed model of treating files as "safe".
As I noted the last time, there will be future exploits of this type until Apple takes the simple step of establishing a parallel registry of trusted programs, rather than trying (in vain) to echieve a balance between convenience and security using a single list of helpers and URI handlers for both local and untrusted content.
Ironically, the fix in 10.3.9 (do not open the content) is better than the fix in 10.4.5 (warn the user about the content).
this chip is actually more powerful than what was in the older Macmini
Intel GPUs are notoriously horrid at 3d.
So... are you sure?
I'm not a gamer, but I DO use a lot of OpenGL-based apps, in 3d, and while the existing mini's GPU is limited what it does do it does well. What does the "advanced video" do that I should care about?
I don't want a "performance GPU", but I want one that's going to do OpenGL better than what I already have. They've got a choice of two different Core Image capable GPUs in the Macbook and iBook they could have used. But... no...
Intel must be soaking them on the Core Solo/Duo if they feel they have to risk using Intel GPUs in a Mac to keep the price down.
Well, I guess Apple's decided to terminally cripple the mini. GMA950? As if the limited VRAM in the original model wasn't bad enough, they stick a chip in that only nominally supports 3d in the quote-upgrade-unquote.
I'll keep my old "4 times slower" mini with a functional GPU.