How many households aren't buying computers anyways? Even with the gradual inclusion of media features into consoles they can't replace computers. So it's more a matter of $800 for a computer with a decent video card vs. $400 for a more basic computer (that will also be less functional for other tasks) + $400 for a console.
(Though personally I still go with a console since it's more convienent, quieter and comes with an input device suited to games.)
Alright isn't exactly the word that springs to mind. Ugly. Overpriced. Ancient. Painful. Openserver 5 was a SVR3 derivative kept alive far too long with absolute minimal support (last I looked they still hadn't fixed bugs that I reported a decade ago). Unixware 2 was okay, but they didn't develop that, they purchased it. Unixware 7 was marred by the introduction of some stupid OSR5isms (ostensibly to ease migration) and incredibly poor quality control. I can't speak for OpenServer 6 since that was released after I ceased having anything to do with SCO.
(For the record - once upon a time I supported SCO Xenix, Unix, OpenDesktop, OpenServer, Unixware, VisionFS, Advanced File and Print Services, Merge, Non-Stop Clusters and Tarantella. Still have the certifications around here somewhere.)
They have OpenServer and Unixware which contain IP beyond the SVR4. They still have a significant installed base of vertical applications (particularly in point of sale and telephony applications). It's emphatically not a growing concern but there's plenty of money to be made in support contracts and migration services for the right company.
Not that I don't appreciate the potential, but as a mail service provider I have to cringe at the prospect of infected machines with fat upstream pipes.
BTX was hamstrung by an incompatability with every case on the market. Case manufacturers don't want to build cases for a marginal form factor. Motherboard manufacturers don't want to build boards for cases which don't exist. Consumers don't want to buy components that are made only by a handful of manufacturers. Stores don't want to stock components consumers are wary of.
DTX, on the other hand, can fit into a standard ATX case. And DTX cases can also fit mini-ITX boards. And there's plenty of markets that would be well-served by a more compact specification than mini-ATX - HTPCs, LAN gaming rigs, high density servers, server appliances, etc.
The slides on the first page of the article clearly state "Target 65W/45W CPUs". If you need a big honking cooling fan you're looking at the wrong form factor.
BTW, a few weeks ago, I read an article about some MoBo manufacturers considering adding 512MB-2GB of flash memory to boot an embedded Linux desktop from the BIOS for disk-less web-browsing and other stuff...
a BIOS with embedded Linux does not seem that far-fetched, we only need 1GB firmware hubs to plug into Intel's chipsets and hope we will not need to flash our 1GB BIOS too often.
People have been replacing the traditional BIOS with Linux for about eight years now, so yeah, I would say it's feasible.:) Would make more sense to just embed a flash drive for the userspace filesystem and just put the bootstrap and kernel in the firmware proper.
If you go to price notebooks from Dell or HP or most other vendors, you have to figure out if you want a Vostro or a Latitude, or a Precision, or whateve, plus you have to pick a model number for each.... and (even after a side-by-side comparison), it can be difficult to work out exactly what's what. For people who value their time, it might actually be worth spending a few hundred bucks more just to be able to say "I want a 17-inch, so that means a MacBook Pro" and then just to quickly select the details of their configuration.
Choice is hard! Paying for only the features that you want is for suckers who don't value their time!;)
The author makes the argument that while 3G networks have more bandwidth than EDGE (they can transfer data at a faster rate), that higher bandwidth comes at a cost of higher latency (the time it takes for the transfer to begin) and more power consumption.
He claims it comes at the cost of higher latency, but fails to present a real argument for it. Benchmarks I've seen peg latency on EDGE four times higher than UTMS, which kind of kills his argument.
It looks like their model is to use the free service as a carrot for their paid users. The storage limit isn't terribly if you're willing to clean up your inbox very regularly, but the bandwidth limit would be a killer for most typical users who are forwarding pictures, video clips, songs and other crap that would be better hosted and linked rather than mailed. I imagine that they have a handful of relatively light users on the free service, a lot of abandoned accounts that eventually get deactivated, and then enough people who like the service enough to upgrade to paid status to make operations profitable.
Possible, yes, but consumers largely don't ask for it. The majority of users, when given the choice, will opt for webmail anyways, followed by POP. Most of the handful who care about IMAP are business users, who providers will want to upsell to a paid (or higher paid) service anyways. Free services, in particular, have even less motivation to provide IMAP since it provides rich capability without the ability to serve ads (unless they inject them as messages which is more likely to generate user complaints). There's little reason to try and compete for users you can't monetize.
Actually, in retrospect, occured to me that AOL provides IMAP and allows free sign-up as well. The responsiveness of their servers leave something to be desired, but it's there at least.
Unlikely to see it. Providing IMAP service is relatively expensive compared to webmail or POP. Persistent connection, encrypted channel, polling for changes, server-side full text search, etc. There are ways to minimize some aspects; e.g. if deliveries and imap are handled by the same process you don't need to poll for changes. Likewise kernel based notification such as inotify is cheaper than polling, but support is platform dependent and not particularly pervasive (of the open source ones, I believe only Dovecot has it currently). The persistent connection is the real killer, though. Not only does that eat into server resources, it also puts an additional burden on the firewalls and load balancers.
Japan averages 340 people per square kilometer. The United States averages 32. Why am I supposed to be surprised that a country with ten times the population density has more pervasive broadband?
You're comparing one time construction costs to four years of operation? The credit card transaction fees alone are estimated at an average of $0.10 per song (once estimated as high as $0.25/song; current figures rely on batch processing, volume discounts and such).
And do you have any idea what rack space costs? Clean power, cooling, bandwidth, opeators, monitoring? Initial build-out is generally the smallest consideration for a large scale site.
I'd say Apple has probably made over $500M so far on music alone, which would put the profit per song cumulatively around 17 cents, and rising every day (the profit per song today could be 20+ cents).
I'll believe an actual analyst. Especially since Apple first mentioned a profit from iTMS in their August 2005 earnings statement.
But nothing else will change a bit. 0% chance that an upgrade may break your configuration file. 0% risks that all the scripts that you manually wrote will suddenly stop functionning because of subtle differences between version 1.8.6.9 and 1.8.6.10 in some obscure software.
And a 100% chance that a change in your timezone will cause your servers to suddenly have the wrong time (assuming default configuration).
No thanks, I'll stick to a platform with a more sane balance between platform stability and not breaking things.
the average mac gets a lot more user life than the average windows box
Yeah, a Core 2 Duo with an Apple logo on it stays faster longer than one with a Dell or HP logo!
Seriously, though, the longevity argument was largely disingenous even before the architecture switch. Apple sells a premium product at a premium price which has the dual effect of encouraging longer retention and bolstering prices in the secondary market. There's nothing keeping someone from using a PC for as many years as a Mac; in fact plenty of people do. On the other hand there's less motivation to stick with the same machine for a decade when every couple of years you can buy a new machine which outperforms the previous generation for only a few hundred dollars.
If it came with 2GB to begin with, you probably started with the "Performance Package", which includes Vista Business and Microsoft Office.
But then again, with the MacBook you can run anything (MacOS, Windows, *nix), so wouldn't the MacBook be a more logical choice?
If you have a need to run MacOS apps, the MacBook. If you don't need to run MacOS apps and you do need to run Windows apps, a PC makes more sense. Especially since you need to add $150 for a Windows license to the Mac, along with $80 for Parallels Desktop or VMWare Fusion (unless you like rebooting).
(Note: I haven't run MacOS or Windows as a desktop in over a decade. However I recognize that both operatings systems can make sense depending on user requirements.)
Slashdot Mirror
How many households aren't buying computers anyways? Even with the gradual inclusion of media features into consoles they can't replace computers. So it's more a matter of $800 for a computer with a decent video card vs. $400 for a more basic computer (that will also be less functional for other tasks) + $400 for a console.
(Though personally I still go with a console since it's more convienent, quieter and comes with an input device suited to games.)
Alright isn't exactly the word that springs to mind. Ugly. Overpriced. Ancient. Painful. Openserver 5 was a SVR3 derivative kept alive far too long with absolute minimal support (last I looked they still hadn't fixed bugs that I reported a decade ago). Unixware 2 was okay, but they didn't develop that, they purchased it. Unixware 7 was marred by the introduction of some stupid OSR5isms (ostensibly to ease migration) and incredibly poor quality control. I can't speak for OpenServer 6 since that was released after I ceased having anything to do with SCO.
(For the record - once upon a time I supported SCO Xenix, Unix, OpenDesktop, OpenServer, Unixware, VisionFS, Advanced File and Print Services, Merge, Non-Stop Clusters and Tarantella. Still have the certifications around here somewhere.)
They have OpenServer and Unixware which contain IP beyond the SVR4. They still have a significant installed base of vertical applications (particularly in point of sale and telephony applications). It's emphatically not a growing concern but there's plenty of money to be made in support contracts and migration services for the right company.
Not that I don't appreciate the potential, but as a mail service provider I have to cringe at the prospect of infected machines with fat upstream pipes.
Or a sign that TDMA signalling kind of sucks. :) Makes me glad I'm a Verizon customer.
BTX was hamstrung by an incompatability with every case on the market. Case manufacturers don't want to build cases for a marginal form factor. Motherboard manufacturers don't want to build boards for cases which don't exist. Consumers don't want to buy components that are made only by a handful of manufacturers. Stores don't want to stock components consumers are wary of.
DTX, on the other hand, can fit into a standard ATX case. And DTX cases can also fit mini-ITX boards. And there's plenty of markets that would be well-served by a more compact specification than mini-ATX - HTPCs, LAN gaming rigs, high density servers, server appliances, etc.
The slides on the first page of the article clearly state "Target 65W/45W CPUs". If you need a big honking cooling fan you're looking at the wrong form factor.
Seven year maintenance cycle. Wide ISV support, since it's RHEL compatible. Backported patches and bugfixes.
BTW, a few weeks ago, I read an article about some MoBo manufacturers considering adding 512MB-2GB of flash memory to boot an embedded Linux desktop from the BIOS for disk-less web-browsing and other stuff...
:) Would make more sense to just embed a flash drive for the userspace filesystem and just put the bootstrap and kernel in the firmware proper.
Considering? Done! In stock!
a BIOS with embedded Linux does not seem that far-fetched, we only need 1GB firmware hubs to plug into Intel's chipsets and hope we will not need to flash our 1GB BIOS too often.
People have been replacing the traditional BIOS with Linux for about eight years now, so yeah, I would say it's feasible.
My wife bought two on pre-order. Both came with a nunchuck.
If you're targeting non-gamers using a term which is universally recognized is a good thing.
If you go to price notebooks from Dell or HP or most other vendors, you have to figure out if you want a Vostro or a Latitude, or a Precision, or whateve, plus you have to pick a model number for each.... and (even after a side-by-side comparison), it can be difficult to work out exactly what's what. For people who value their time, it might actually be worth spending a few hundred bucks more just to be able to say "I want a 17-inch, so that means a MacBook Pro" and then just to quickly select the details of their configuration.
;)
Choice is hard! Paying for only the features that you want is for suckers who don't value their time!
The author makes the argument that while 3G networks have more bandwidth than EDGE (they can transfer data at a faster rate), that higher bandwidth comes at a cost of higher latency (the time it takes for the transfer to begin) and more power consumption.
He claims it comes at the cost of higher latency, but fails to present a real argument for it. Benchmarks I've seen peg latency on EDGE four times higher than UTMS, which kind of kills his argument.
It looks like their model is to use the free service as a carrot for their paid users. The storage limit isn't terribly if you're willing to clean up your inbox very regularly, but the bandwidth limit would be a killer for most typical users who are forwarding pictures, video clips, songs and other crap that would be better hosted and linked rather than mailed. I imagine that they have a handful of relatively light users on the free service, a lot of abandoned accounts that eventually get deactivated, and then enough people who like the service enough to upgrade to paid status to make operations profitable.
Possible, yes, but consumers largely don't ask for it. The majority of users, when given the choice, will opt for webmail anyways, followed by POP. Most of the handful who care about IMAP are business users, who providers will want to upsell to a paid (or higher paid) service anyways. Free services, in particular, have even less motivation to provide IMAP since it provides rich capability without the ability to serve ads (unless they inject them as messages which is more likely to generate user complaints). There's little reason to try and compete for users you can't monetize.
Actually, in retrospect, occured to me that AOL provides IMAP and allows free sign-up as well. The responsiveness of their servers leave something to be desired, but it's there at least.
Unlikely to see it. Providing IMAP service is relatively expensive compared to webmail or POP. Persistent connection, encrypted channel, polling for changes, server-side full text search, etc. There are ways to minimize some aspects; e.g. if deliveries and imap are handled by the same process you don't need to poll for changes. Likewise kernel based notification such as inotify is cheaper than polling, but support is platform dependent and not particularly pervasive (of the open source ones, I believe only Dovecot has it currently). The persistent connection is the real killer, though. Not only does that eat into server resources, it also puts an additional burden on the firewalls and load balancers.
What "360 game"?
s/your/you're/;s/THEIR/THEY'RE/
Japan averages 340 people per square kilometer. The United States averages 32. Why am I supposed to be surprised that a country with ten times the population density has more pervasive broadband?
Sony's gaming division is hemoraging money - $237M last quarter alone.
You're comparing one time construction costs to four years of operation? The credit card transaction fees alone are estimated at an average of $0.10 per song (once estimated as high as $0.25/song; current figures rely on batch processing, volume discounts and such).
And do you have any idea what rack space costs? Clean power, cooling, bandwidth, opeators, monitoring? Initial build-out is generally the smallest consideration for a large scale site.
I'd say Apple has probably made over $500M so far on music alone, which would put the profit per song cumulatively around 17 cents, and rising every day (the profit per song today could be 20+ cents).
I'll believe an actual analyst. Especially since Apple first mentioned a profit from iTMS in their August 2005 earnings statement.
But nothing else will change a bit. 0% chance that an upgrade may break your configuration file. 0% risks that all the scripts that you manually wrote will suddenly stop functionning because of subtle differences between version 1.8.6.9 and 1.8.6.10 in some obscure software.
And a 100% chance that a change in your timezone will cause your servers to suddenly have the wrong time (assuming default configuration).
No thanks, I'll stick to a platform with a more sane balance between platform stability and not breaking things.
the average mac gets a lot more user life than the average windows box
Yeah, a Core 2 Duo with an Apple logo on it stays faster longer than one with a Dell or HP logo!
Seriously, though, the longevity argument was largely disingenous even before the architecture switch. Apple sells a premium product at a premium price which has the dual effect of encouraging longer retention and bolstering prices in the secondary market. There's nothing keeping someone from using a PC for as many years as a Mac; in fact plenty of people do. On the other hand there's less motivation to stick with the same machine for a decade when every couple of years you can buy a new machine which outperforms the previous generation for only a few hundred dollars.
If it came with 2GB to begin with, you probably started with the "Performance Package", which includes Vista Business and Microsoft Office.
But then again, with the MacBook you can run anything (MacOS, Windows, *nix), so wouldn't the MacBook be a more logical choice?
If you have a need to run MacOS apps, the MacBook. If you don't need to run MacOS apps and you do need to run Windows apps, a PC makes more sense. Especially since you need to add $150 for a Windows license to the Mac, along with $80 for Parallels Desktop or VMWare Fusion (unless you like rebooting).
(Note: I haven't run MacOS or Windows as a desktop in over a decade. However I recognize that both operatings systems can make sense depending on user requirements.)