Why do people insist on comparing a celeron to a P4 or athlon?
Generally speaking, whenever anyone's in the market for something, they consider the most famous brands first.. Athlon and P4 in this case.
If they then look at the "low cost" processors as well, the question they'll be asking themselves is "how much a difference is there between the premium version and this", in order to determine if the money saving is worth the somewhat reduced power.
This is where reviews comparing athlons and p4s to the low cost chips comes in handy
I've always thought that the great thing about computers is their mallability - the way you can change the way they act.
Then comes issues like licensing, and the way that proprietary software can only be extended using special macro languages.. These things drop artificial flexibility barriers onto a completely open system - a very sad waste of the potential of such devices..
Because malleability is the best trait of computer programs over specific, fixed systems, it is only at its best if no such artificial barriers are imposed on the system.
The open source model really just seems like a natural method of software development that avoids such wastage...
On the other hand, you could ship discs with multiple versions of the binaries on. This could be the thing that finally makes software distribution on DVDs catch on in a big way.
Myself I'd like to know what kind of manufacturing is available to make this possible.
With silicon semiconductors, a massive crystal of the stuff can be grown by suspending a small crystal "seed" an molten silicon and very slowly pulling the seed upwards while rotating..
Carbon, on the other hand, isn't so obliging - It doesn't melt, it sublimes directly from a solid state into a gaseous one, so this way's out..
Using diamond as a basis for microcircuit manufacture can't seriously take off until we can either find a way to create large crystals, or grow large ones from existing small crystals..
Who will really trust open sourced DRM?
on
Open Source DRM
·
· Score: 1, Flamebait
One of the main reasons corporates seem ready to trust DRM is that the software that can decrypt the DRMd media is relatively immutable..
If one were to release a new single using an open sourced DRM technology, there is simply no reason to believe that the software is "acting in good faith" according to the general idea of DRM.. One could easily adapt such an open-source DRM player so that instead of playing the file, it does something else with the decrypted data. Such as saving it to an un-encrypted version off the ogg file.
Such a file could then easily be put on your favourite peer-to-peer network.. And once one or two people have it, that's it. That one un-encrypted instance can get anywhere and everywhere..
Windows NT4 isn't nearly as big a threat to microsoft going on as win2k could be (properly supported by a third party) in future.
Prior to win2K every version of windows had a major failing.. Win 9x wasn't stable, so that could make people upgrade. WinNT4s driver model was more limited and didn't support adding or removing devices while powered on (iirc).
Windows 2000, on the other hand, has neither of these major failings (granted, linux still has a better reputation in technical regards around these parts). It is stable enough, and it's got a driver model which, as I see it, could certainly be extended to cover future types of hardware.
If the open source movement really wanted to declare war on ms they could start working on projects to extend win2ks longevity, and effectively nullify this "forced upgrade" strategy.
I can see certain problems stemming from this whole challenge-response style address verification. For example, if someone writes a new message to a new person and forgets to add the address to his whitelist, then a situation may arise where the recipient sends a challenge to the sender, and then the sender running a similar scheme recieves the challenge message and decides to challenge its sender..
We have spatulas. But what americans seem to call spatulas we call a "fish slice".
Keeping patches on the install disks
on
Presenting The CDR-ROM
·
· Score: 2, Interesting
I quite like this idea of disks that have rewritable sections of them because they will permit an increased deal of organisational neatness.
If one were to decide to install an old game from a few years ago you would first have to find the install disk, then the various disks on which you've saved patches you've downloaded. Then perhaps if you use any expansion packs with the game you then have to find that disk too. If you're as untidy as I am this could get to be a real nightmare.
By the sound of this, however, you could just store the relevant patches and expansions (and maybe even keep a copy of a file containing your preferences) on the writable portion of the software's install disk and have it all in one place. Could be neat.
Actually I think intellectual property considerations killed it. If you're doing a show regarding devices that will probably hit the market in a couple of years time, as Tomorrows World was, then you're seriously limited as to what you can actually say about it.
They start off by having little clips of the devices they were going to feature on the show, that'd get me thinking "oh, I must see this". Then they talk about the device you were actually interested in as the last item of the show. Not only this but there's very little more detail there than in the clipshow.
I remember watching one, as I rarely do these days, about a year ago. In the clipshow they talked about a device which used microwave radiation to cool stuff. Cool, thought I, I must watch the rest of the show to find out how it works!. It gets to the segment about it and they ask how it works..
"Inverse technology" said the demonstrator.
Frankly, I wasn't massively enlightened. I decided not to watch it again.
I don't know how any such program could have avoided being similarly anticlimactic and not give away far too much information about stuff that's on the drawing board.
The primary market for these things seems to be for people who need to be able to record scribbles in a lecture or meeting. This is probably not a concept that needs huge amounts of computing power. People's buying decision will probably me made not on computing power, but mainly on price.
The main consequence of this is that the prices of the most popular models will go down with competition, not up with increased specs.
I agree it's not the sort of thing one would choose as a main machine, but as a second device for specific tasks, and that £2000 may be a bit much for a machine for this role, but the price they'll be in two years time probably won't be.
As a popfile user, I'm quite impressed with the catch rate possible with bayes theorem spam filters, however I suspect this will decrease in effectiveness over the long term.
Spammers are likely to respond to filters like this by encoding text in ways the filters can't read but humans can (eg having a.gif file of the text, loaded by a HTML statement in the message).
Statistical filters would need to have some kind of built in OCR routine before it could be effective against that trick, and some respectible mailing lists are using images as well, so you can't just filter all mails with images attatched.
In the long term, therefore, I suspect that filters that use a network database of spam will be more successful.
Reading the article I'm reminded of reading of some concept OS or other which claimed to manage both RAM and the hard disc as one large, continuous, memory space.
Thing is, that in systems that model the memory as one large flat space, the difference between what is running and what is not effectively disappears. Programs that are "not running" would simply be waiting for a command to display a window whereas those that were "running" spend most of their time waiting instead for a callback from the user interface manager. Hardly a massive difference.
The RiscOS model of simply displaying an icon that could be dragged to a file system window was indeed a convenient system provided you had the appropriate window open at the time. Particularly nice were tricks like being able to print by dragging this icon to the print queue window.
I don't think it's necessarily the best possible system, but rather just closer to the best than what is often used.
Personally, I think a good direction for interface design would be to attempt to become more project-oriented than desktop-metaphor-oriented. A large percentage of the work most professionals use their computers for is related to a specific project or task at a time, and structuring the operating systems interface to be more similar to that of project management software would be a better way of servicing the majority of people.
Successful as the desktop metaphor has been, it's still mainly a system that's centred around adapting the user to the computers way of doing things rather than adapting the computing environment around the users work plan..
I can think of quite a few situations where saving "scribbles" would be helpful.
The main thing which causes people not to use word processing apps in more situations is primarily the speed hit as soon as any formatting is required. Sure, I might be able to type a lot faster than I write, but as soon as any formatting is required, the "pen-on-paper" scheme of just moving your hand to where you want the text to go is a lot faster and more elegant than the best of WP interfaces.
Then there's other types of annotations - I can't really envisage taking a laptop to lectures for notetaking because as soon as the professor puts up any kind of diagram I'd have my work cut out opening a paint package, using a trackpoint to copy it down and importing it into my "notes" document before he moves to the next slide.
Supporting a new type of data - for an equation or shorthand text would require plugins to be written for the word processor to support that type of information, but you can notate anything as scribbles with much less fuss.
Slashdot Mirror
Why do people insist on comparing a celeron to a P4 or athlon?
Generally speaking, whenever anyone's in the market for something, they consider the most famous brands first.. Athlon and P4 in this case.
If they then look at the "low cost" processors as well, the question they'll be asking themselves is "how much a difference is there between the premium version and this", in order to determine if the money saving is worth the somewhat reduced power.
This is where reviews comparing athlons and p4s to the low cost chips comes in handy
...because they love what computers could be.
I've always thought that the great thing about computers is their mallability - the way you can change the way they act.
Then comes issues like licensing, and the way that proprietary software can only be extended using special macro languages.. These things drop artificial flexibility barriers onto a completely open system - a very sad waste of the potential of such devices..
Because malleability is the best trait of computer programs over specific, fixed systems, it is only at its best if no such artificial barriers are imposed on the system.
The open source model really just seems like a natural method of software development that avoids such wastage...
On the other hand, you could ship discs with multiple versions of the binaries on. This could be the thing that finally makes software distribution on DVDs catch on in a big way.
Myself I'd like to know what kind of manufacturing is available to make this possible.
With silicon semiconductors, a massive crystal of the stuff can be grown by suspending a small crystal "seed" an molten silicon and very slowly pulling the seed upwards while rotating..
Carbon, on the other hand, isn't so obliging - It doesn't melt, it sublimes directly from a solid state into a gaseous one, so this way's out..
Using diamond as a basis for microcircuit manufacture can't seriously take off until we can either find a way to create large crystals, or grow large ones from existing small crystals..
One of the main reasons corporates seem ready to trust DRM is that the software that can decrypt the DRMd media is relatively immutable..
If one were to release a new single using an open sourced DRM technology, there is simply no reason to believe that the software is "acting in good faith" according to the general idea of DRM.. One could easily adapt such an open-source DRM player so that instead of playing the file, it does something else with the decrypted data. Such as saving it to an un-encrypted version off the ogg file.
Such a file could then easily be put on your favourite peer-to-peer network.. And once one or two people have it, that's it. That one un-encrypted instance can get anywhere and everywhere..
Windows NT4 isn't nearly as big a threat to microsoft going on as win2k could be (properly supported by a third party) in future.
Prior to win2K every version of windows had a major failing.. Win 9x wasn't stable, so that could make people upgrade. WinNT4s driver model was more limited and didn't support adding or removing devices while powered on (iirc).
Windows 2000, on the other hand, has neither of these major failings (granted, linux still has a better reputation in technical regards around these parts). It is stable enough, and it's got a driver model which, as I see it, could certainly be extended to cover future types of hardware.
If the open source movement really wanted to declare war on ms they could start working on projects to extend win2ks longevity, and effectively nullify this "forced upgrade" strategy.
I can see certain problems stemming from this whole challenge-response style address verification. For example, if someone writes a new message to a new person and forgets to add the address to his whitelist, then a situation may arise where the recipient sends a challenge to the sender, and then the sender running a similar scheme recieves the challenge message and decides to challenge its sender..
Infinite loopsville...
We have spatulas. But what americans seem to call spatulas we call a "fish slice".
I quite like this idea of disks that have rewritable sections of them because they will permit an increased deal of organisational neatness.
If one were to decide to install an old game from a few years ago you would first have to find the install disk, then the various disks on which you've saved patches you've downloaded. Then perhaps if you use any expansion packs with the game you then have to find that disk too. If you're as untidy as I am this could get to be a real nightmare.
By the sound of this, however, you could just store the relevant patches and expansions (and maybe even keep a copy of a file containing your preferences) on the writable portion of the software's install disk and have it all in one place. Could be neat.
Actually I think intellectual property considerations killed it. If you're doing a show regarding devices that will probably hit the market in a couple of years time, as Tomorrows World was, then you're seriously limited as to what you can actually say about it.
They start off by having little clips of the devices they were going to feature on the show, that'd get me thinking "oh, I must see this". Then they talk about the device you were actually interested in as the last item of the show. Not only this but there's very little more detail there than in the clipshow.
I remember watching one, as I rarely do these days, about a year ago. In the clipshow they talked about a device which used microwave radiation to cool stuff. Cool, thought I, I must watch the rest of the show to find out how it works!. It gets to the segment about it and they ask how it works..
"Inverse technology" said the demonstrator.
Frankly, I wasn't massively enlightened. I decided not to watch it again.
I don't know how any such program could have avoided being similarly anticlimactic and not give away far too much information about stuff that's on the drawing board.
The thing about prices is that they change.
The primary market for these things seems to be for people who need to be able to record scribbles in a lecture or meeting. This is probably not a concept that needs huge amounts of computing power. People's buying decision will probably me made not on computing power, but mainly on price.
The main consequence of this is that the prices of the most popular models will go down with competition, not up with increased specs.
I agree it's not the sort of thing one would choose as a main machine, but as a second device for specific tasks, and that £2000 may be a bit much for a machine for this role, but the price they'll be in two years time probably won't be.
As a popfile user, I'm quite impressed with the catch rate possible with bayes theorem spam filters, however I suspect this will decrease in effectiveness over the long term.
Spammers are likely to respond to filters like this by encoding text in ways the filters can't read but humans can (eg having a .gif file of the text, loaded by a HTML statement in the message).
Statistical filters would need to have some kind of built in OCR routine before it could be effective against that trick, and some respectible mailing lists are using images as well, so you can't just filter all mails with images attatched.
In the long term, therefore, I suspect that filters that use a network database of spam will be more successful.
Unless, that is, you have truly exceptional aim.
Reading the article I'm reminded of reading of some concept OS or other which claimed to manage both RAM and the hard disc as one large, continuous, memory space.
Thing is, that in systems that model the memory as one large flat space, the difference between what is running and what is not effectively disappears. Programs that are "not running" would simply be waiting for a command to display a window whereas those that were "running" spend most of their time waiting instead for a callback from the user interface manager. Hardly a massive difference.
The RiscOS model of simply displaying an icon that could be dragged to a file system window was indeed a convenient system provided you had the appropriate window open at the time. Particularly nice were tricks like being able to print by dragging this icon to the print queue window.
I don't think it's necessarily the best possible system, but rather just closer to the best than what is often used.
Personally, I think a good direction for interface design would be to attempt to become more project-oriented than desktop-metaphor-oriented. A large percentage of the work most professionals use their computers for is related to a specific project or task at a time, and structuring the operating systems interface to be more similar to that of project management software would be a better way of servicing the majority of people.
Successful as the desktop metaphor has been, it's still mainly a system that's centred around adapting the user to the computers way of doing things rather than adapting the computing environment around the users work plan..
I can think of quite a few situations where saving "scribbles" would be helpful. The main thing which causes people not to use word processing apps in more situations is primarily the speed hit as soon as any formatting is required. Sure, I might be able to type a lot faster than I write, but as soon as any formatting is required, the "pen-on-paper" scheme of just moving your hand to where you want the text to go is a lot faster and more elegant than the best of WP interfaces. Then there's other types of annotations - I can't really envisage taking a laptop to lectures for notetaking because as soon as the professor puts up any kind of diagram I'd have my work cut out opening a paint package, using a trackpoint to copy it down and importing it into my "notes" document before he moves to the next slide. Supporting a new type of data - for an equation or shorthand text would require plugins to be written for the word processor to support that type of information, but you can notate anything as scribbles with much less fuss.