It's worth knowing that some of the literature concerning ergonomic keyboarding suggests that you don't incline your keyboard towards you but keep it as flat as possible on the desk, and have your desk as low (close to your knees) as possible.
Personally, my hands get tired more quicky when working on a keyboard that has its 'feet' raised at the back - I'm the guy who always flattens the feet down on lab computers before beginning to type.
Here's what I don't understand: Presumably, at a large cut of $1/song the record companies are spinning a nice profit. Otherwise, why would they be joining iTunes/Napster/everyone else?
Now, if the vendors can't break even, why doesn't a record company (or, say, the RIAA itself) buy an 'unprofitable' online vendor and continue merrily selling songs - sure, the service itself costs money to run, but 100% of the money goes to the label. Is doing this stuff so expensive that it actually costs them more than $1 to let you download a song?
I remember that Napster belonged to Bertelsmann/BMG before, but apparently not now. Hmm.
/james
I wondered what was going on...
on
Kazaa Offices Raided
·
· Score: 2, Informative
This morning I was standing waiting for my bus outside the building that houses the main Kazaa/Sharman/LEF-interactive office and a couple of guys with cameras and microphones went rushing in.
I followed them a bit of the way into the building but couldn't see anything.
Internationally news-worthy stuff doesn't normally happen near my house:)
I really enjoyed Tanenbaum's Operating Systems: Design and Implementation. The whole book is basically a tour of Minix with lots of OS theory along the way. It's very *NIX oriented - signals and system calls etc, but there's some discussion of other ways of doing things.
The nice thing about Minix is it's very easy to make sweeping changes to the OS, recompile the kernel, and see what happens, and the book helps explain how it all works.
I belive Jon's program did a little more: It got into Quicktime after decryption but before decoding, resulting in an AAC file digitally identical to the original one, but freely copyable.
All the methods involving audio capture or burn-then-rip involve some 'transcoding' loss in audio quality as the file is decompressed, captured, then usually recompressed to AAC or mp3./james
When I started my computer science degree, one of the first-year units was about databases, and they used Access as the basis for teaching and assignments. We had to construct all the queries using that visual tool where you drag lines between the columns of the different tables to represent relations.
This sucked. I was confused and had no idea what I was trying to achieve with these queries, and found I was spending most of my time fighting a very lame GUI.
A year later I had to use SQL for some other work I was doing: I found a tutorial page that showed me the basics, and I was up and running in maybe 15 minutes - everything suddenly made sense, and I had vague memories of concepts from the university databases course that were actually clicking into place now that I saw how the queries I'd been drawing in the GUI were really constructed. It was so much simpler and less confusing to see a single-line SQL query than a big complex diagram with lines snaking between representations of tables.
That's interesting:
I'm using a setup with Cyrus-IMAP and squirrelmail to do webmail, and that is kinda slow for just one person on my 200mhz system...
I use cheap 48 CD zip-up folders - they're easy to browse and small enough to carry around when you need to. I buy them from Paddy's Markets in Sydney for about AU$7, which is about US$4 each. They also have 96 CD holders for not much more, but those just look *too* tacky.
At first, I was looking at Case-Logic folders in music shops, and it was working out at something like 50 cents per CD, which is just silly. I guess the answer is to shop around for low quality ones - all they need to do is store your CDs. (just check the zip actually works).
Re:In the tradition of Gonzo Journalism
on
I, Spammer
·
· Score: 1
I reckon that estimate is orders of magnitude too high.
A 1% response rate would mean that for every email advertising penis enlargements or viagra, 1% of recipients are actually buying something, so if each email reaches say 10,000 people, that's 100 responses per email, and if he sends out 180 emails a day that's 18,000 clients for penis enlargement/viagra per day to a single spammer, which would be thoroughly overwhelming, I would say...
I live in Australia, and at my old number I used to get `junk' calls maybe once or twice a month, normally in the evenings. These were normally `market research' (surveys about products and the like) rather than direct sales pitches, although I got those too.
However, my new phone number used to belong to a business, and now I get more calls - one every day or two - from companies selling inkjet refills and other businessy services. So maybe there are different rules for calls to businesses and private numbers?
I've filled a couple of slots with 'mobile rack' (why do they call them that?) removable IDE caddies.
They're basically a plastic caddy into which you put a hard disk, and a 5.25" size bay with a flip-down cover on the front. You wire your IDE ribbons to the connector on the back of the bay, which stays fixed in the PC, and you can swap the drives in and out to your heart's content (erh.. when the computer is switched off...)
I use them for carrying stuff to friends' houses, and also recently when I had to run windows for a while I just swapped in a new hard disk rather than messing with my partitions.
I always thought it'd be pretty cool to put this on a long wall - it's a big unix/unix clone version family tree.
Though you'd have to print it out yourself, of course.
Well, there was in fact at least one movie proclaiming itself to be compliant with Dogme '95 which I have seen - Festen.
Just keeping the record straight:)
Besides, even if total compliance is near-impossible, rules like this can serve as a useful guideline, or a point to aim for. They can steer you in the right direction.
My experiences:
I run a VIA motherboard (the ABIT KA-7) in my home workstation, exclusively with Linux. I've been running it 24/7 for about 3 months, and I haven't seen it crash (well, except when playing with older Mesa drivers - but that's a separate, non-via issue I believe).
I don't run it particularly hard, it doesn't usually do cpu-intensive things, but I've had it running seti@home for 48 hours straight, ripping and encoding a few CDs in a row, compiling Mozilla etc without any problems.
The longest time I've kept it up for is over 3 weeks, and it was only switched off because I was going away for a few days.
So, I'm not exactly a demanding user, and this isn't a server application, but the stability of the system has yet to prove itself to be anything but perfect in my case.
I really have no idea what I'm talking about here, so just ignore me:)
Isn't dumping 15mb/s, all day every day, going to be prohibitively expensive from your ISP? I understand that that kind of bandwidth is typically billed by volume, so you'll be paying the maximum possible rate.
For the prices I expect you would be paying for the bandwidth, it might be worth checking what kind of alternatives you have: Eg some kind of premium FedEx service instead of standard mail.
Reminds me of the old quote, "never underestimate the bandwidth of a speeding truck full of DAT tapes".
I think this is a very interesting issue - whether a company is primarily those who work for it, or some other beast entirely different from any of its individual parts.
The (thanks Cryptonomicon) mantra 'do my actions increase shareholder value?' is a mercenary idea. So to look at VA from a purely Darwinistic angle, we could say that all of its altruistic actions are performed solely to generate venture capital/revenue. To do otherwise would be irresponsible management, and betrayal of the shareholders. So, in this view, VA's wishes and those of the 'tribe', 'community', whatever, coincide only by chance.
But, of course, this view is hard to stick to once you remind yourself that companies are just groups of people. And all the people that I know of who work for VA are great - I'd entirely trust them all to run the community's stuff.
So where do these two views meet? I don't really have much experience working for corporations, so I'd be interested to get some ideas from people who do!
Errrrr...
What?
I'm not sure I understand;
VA owns Andover! Which of them buys/co-opts/helps out Kuro5hin is kinda immaterial...
Anyway, having said that, VA are 'helping out' a lot of the 'open-source in crowd' with hardware etc; They seem to own:
Linux.com
Slashdot.org
Freshmeat.net
Etc, the whole 'OSDN'...
And they've donated hardware far and wide - jwz's DNA Lounge is running on (donated? not sure) VA hardware, And now Kuro5hin.
Anyway, I guess this is how the corporate world works. We're just lucky at this point that VA is such a nice company, but it still troubles me that so much of our cool stuff relies on a publically owned, money-driven organisation.
Intervideo had a tiny stand at CEBIT in Hannover this year. They had a computer running linux, with the player shown to be working.
At that time it was very flaky, and I guessed it didn't have any CSS decoding built-in - the disc they were demoing was a promo-disc of some sort that I assumed didn't have any encryption.
I tried to play with it a little, but the sales guy shooed me away, saying something like "very new software! Pre-alpha quality!".
Just to add something from my experience of working here in Germany:
I'm a native English speaker, and I often see my German colleagues struggling with things like "SET CONCAT NULL YIELDS NULL OFF" (from sql). It's easy to learn simple keywords for normal programming-language flow, but complex things like that are more difficult - how do you explain the meaning of 'yield' in that context?
To a lesser extent, the same kind of thing can happen within English: I remember, many years ago, seeing my uncle proudly demonstrating a language that was written almost in plain English and thus was very easy to understand, supposedly (it was the scripting language for HyperCard). But imagine this fragment:
"for i from 1 through 8".
That's only 'natural' to an American, to an Englishman it might as well be formatted more like "(i=1;i==8;i++)" as far as ease-of-use is concerned.
Bill Gates is pretty cool... (argh! my karma!)
on
Looking Back At NeXT
·
· Score: 2
I know that he and his company are odious in almost all respects, but seeing his picture in this article made me realise something:
Gates embodies quite a lot of 'geek qualities' - just look at the photo, coffee cup hanging from fingers, dorky glasses, looking decidedly... scruffy (even in a tie).
It might all be part of a carefully sculpted public image, but I get the impression that he doesn't care what impression he gives.
So, I think he's cool (while also being thoroughly dastardly, don't get me wrong...).
Well... Heroine Virtual (xmovie programmers) had a message on their site a few months ago which seemed to state that they'd integrated DeCSS into the player.
They were asking if they should 1) release binary-only players with decss, or 2) remove decss. Someone from the FSF contacted them and said that if they were going to keep the code GPLd they shouldn't release binary only.
So that's why.
Sooo... theoretically you could wrap decss back into the code _fairly_ easily, especially if they left the appropriate hooks in there.
/james.
the 7 minute long story
on
Geek Flavor
·
· Score: 1
Well, some people edited the page, then root came on, the webpage was taken down and now it's refusing connections on port 22:)
Alright, here's something I've never understood about brute-forcing encrypted files (not that I understand the area very much at all):
Presumably the brute-forcing algorithm has made some assumptions about the nature of the encryption it's testing. If, as you say, it's using some kind of dictionary to detect when it has cracked the file:
Why not simply put the plaintext data through some simple cipher that makes it look just like garbage so they can't tell when they have the correct key?
I know there's a reason why this is silly... anyone care to enlighten me?
Slashdot Mirror
It's worth knowing that some of the literature concerning ergonomic keyboarding suggests that you don't incline your keyboard towards you but keep it as flat as possible on the desk, and have your desk as low (close to your knees) as possible.
/james
Personally, my hands get tired more quicky when working on a keyboard that has its 'feet' raised at the back - I'm the guy who always flattens the feet down on lab computers before beginning to type.
Here's what I don't understand: Presumably, at a large cut of $1/song the record companies are spinning a nice profit. Otherwise, why would they be joining iTunes/Napster/everyone else?
/james
Now, if the vendors can't break even, why doesn't a record company (or, say, the RIAA itself) buy an 'unprofitable' online vendor and continue merrily selling songs - sure, the service itself costs money to run, but 100% of the money goes to the label. Is doing this stuff so expensive that it actually costs them more than $1 to let you download a song?
I remember that Napster belonged to Bertelsmann/BMG before, but apparently not now. Hmm.
This morning I was standing waiting for my bus outside the building that houses the main Kazaa/Sharman/LEF-interactive office and a couple of guys with cameras and microphones went rushing in. :)
/james
I followed them a bit of the way into the building but couldn't see anything.
Internationally news-worthy stuff doesn't normally happen near my house
I really enjoyed Tanenbaum's Operating Systems: Design and Implementation. The whole book is basically a tour of Minix with lots of OS theory along the way. It's very *NIX oriented - signals and system calls etc, but there's some discussion of other ways of doing things.
/james
The nice thing about Minix is it's very easy to make sweeping changes to the OS, recompile the kernel, and see what happens, and the book helps explain how it all works.
I belive Jon's program did a little more: It got into Quicktime after decryption but before decoding, resulting in an AAC file digitally identical to the original one, but freely copyable.
/james
All the methods involving audio capture or burn-then-rip involve some 'transcoding' loss in audio quality as the file is decompressed, captured, then usually recompressed to AAC or mp3.
When I started my computer science degree, one of the first-year units was about databases, and they used Access as the basis for teaching and assignments. We had to construct all the queries using that visual tool where you drag lines between the columns of the different tables to represent relations.
This sucked. I was confused and had no idea what I was trying to achieve with these queries, and found I was spending most of my time fighting a very lame GUI.
A year later I had to use SQL for some other work I was doing: I found a tutorial page that showed me the basics, and I was up and running in maybe 15 minutes - everything suddenly made sense, and I had vague memories of concepts from the university databases course that were actually clicking into place now that I saw how the queries I'd been drawing in the GUI were really constructed. It was so much simpler and less confusing to see a single-line SQL query than a big complex diagram with lines snaking between representations of tables.
That's interesting:
/james
I'm using a setup with Cyrus-IMAP and squirrelmail to do webmail, and that is kinda slow for just one person on my 200mhz system...
What server software are you running?
I use cheap 48 CD zip-up folders - they're easy to browse and small enough to carry around when you need to. I buy them from Paddy's Markets in Sydney for about AU$7, which is about US$4 each. They also have 96 CD holders for not much more, but those just look *too* tacky.
At first, I was looking at Case-Logic folders in music shops, and it was working out at something like 50 cents per CD, which is just silly. I guess the answer is to shop around for low quality ones - all they need to do is store your CDs. (just check the zip actually works).
I reckon that estimate is orders of magnitude too high.
A 1% response rate would mean that for every email advertising penis enlargements or viagra, 1% of recipients are actually buying something, so if each email reaches say 10,000 people, that's 100 responses per email, and if he sends out 180 emails a day that's 18,000 clients for penis enlargement/viagra per day to a single spammer, which would be thoroughly overwhelming, I would say...
I live in Australia, and at my old number I used to get `junk' calls maybe once or twice a month, normally in the evenings. These were normally `market research' (surveys about products and the like) rather than direct sales pitches, although I got those too.
However, my new phone number used to belong to a business, and now I get more calls - one every day or two - from companies selling inkjet refills and other businessy services.
So maybe there are different rules for calls to businesses and private numbers?
james.
My XP1800+ 'boxed' hits 130F with some regularity, and it's not overclocked, just using the heatsink and thermal tape that came in the box from AMD.
I've filled a couple of slots with 'mobile rack' (why do they call them that?) removable IDE caddies.
/james
They're basically a plastic caddy into which you put a hard disk, and a 5.25" size bay with a flip-down cover on the front. You wire your IDE ribbons to the connector on the back of the bay, which stays fixed in the PC, and you can swap the drives in and out to your heart's content (erh.. when the computer is switched off...)
I use them for carrying stuff to friends' houses, and also recently when I had to run windows for a while I just swapped in a new hard disk rather than messing with my partitions.
I always thought it'd be pretty cool to put this on a long wall - it's a big unix/unix clone version family tree.
/james
Though you'd have to print it out yourself, of course.
Well, there was in fact at least one movie proclaiming itself to be compliant with Dogme '95 which I have seen - Festen.
:)
/James.
Just keeping the record straight
Besides, even if total compliance is near-impossible, rules like this can serve as a useful guideline, or a point to aim for. They can steer you in the right direction.
My experiences:
I run a VIA motherboard (the ABIT KA-7) in my home workstation, exclusively with Linux. I've been running it 24/7 for about 3 months, and I haven't seen it crash (well, except when playing with older Mesa drivers - but that's a separate, non-via issue I believe).
I don't run it particularly hard, it doesn't usually do cpu-intensive things, but I've had it running seti@home for 48 hours straight, ripping and encoding a few CDs in a row, compiling Mozilla etc without any problems.
The longest time I've kept it up for is over 3 weeks, and it was only switched off because I was going away for a few days.
So, I'm not exactly a demanding user, and this isn't a server application, but the stability of the system has yet to prove itself to be anything but perfect in my case.
/james.
I really have no idea what I'm talking about here, so just ignore me :)
Isn't dumping 15mb/s, all day every day, going to be prohibitively expensive from your ISP? I understand that that kind of bandwidth is typically billed by volume, so you'll be paying the maximum possible rate.
For the prices I expect you would be paying for the bandwidth, it might be worth checking what kind of alternatives you have: Eg some kind of premium FedEx service instead of standard mail.
Reminds me of the old quote, "never underestimate the bandwidth of a speeding truck full of DAT tapes".
Chris, thanks for replying to my last message.
I think this is a very interesting issue - whether a company is primarily those who work for it, or some other beast entirely different from any of its individual parts.
The (thanks Cryptonomicon) mantra 'do my actions increase shareholder value?' is a mercenary idea. So to look at VA from a purely Darwinistic angle, we could say that all of its altruistic actions are performed solely to generate venture capital/revenue. To do otherwise would be irresponsible management, and betrayal of the shareholders. So, in this view, VA's wishes and those of the 'tribe', 'community', whatever, coincide only by chance.
But, of course, this view is hard to stick to once you remind yourself that companies are just groups of people. And all the people that I know of who work for VA are great - I'd entirely trust them all to run the community's stuff.
So where do these two views meet? I don't really have much experience working for corporations, so I'd be interested to get some ideas from people who do!
What?
I'm not sure I understand;
VA owns Andover! Which of them buys/co-opts/helps out Kuro5hin is kinda immaterial...
Anyway, having said that, VA are 'helping out' a lot of the 'open-source in crowd' with hardware etc; They seem to own:
- Linux.com
- Slashdot.org
- Freshmeat.net
Etc, the whole 'OSDN'...And they've donated hardware far and wide - jwz's DNA Lounge is running on (donated? not sure) VA hardware, And now Kuro5hin.
Anyway, I guess this is how the corporate world works. We're just lucky at this point that VA is such a nice company, but it still troubles me that so much of our cool stuff relies on a publically owned, money-driven organisation.
Intervideo had a tiny stand at CEBIT in Hannover this year. They had a computer running linux, with the player shown to be working.
At that time it was very flaky, and I guessed it didn't have any CSS decoding built-in - the disc they were demoing was a promo-disc of some sort that I assumed didn't have any encryption.
I tried to play with it a little, but the sales guy shooed me away, saying something like "very new software! Pre-alpha quality!".
Just to add something from my experience of working here in Germany:
I'm a native English speaker, and I often see my German colleagues struggling with things like "SET CONCAT NULL YIELDS NULL OFF" (from sql). It's easy to learn simple keywords for normal programming-language flow, but complex things like that are more difficult - how do you explain the meaning of 'yield' in that context?
To a lesser extent, the same kind of thing can happen within English: I remember, many years ago, seeing my uncle proudly demonstrating a language that was written almost in plain English and thus was very easy to understand, supposedly (it was the scripting language for HyperCard). But imagine this fragment:
"for i from 1 through 8".
That's only 'natural' to an American, to an Englishman it might as well be formatted more like "(i=1;i==8;i++)" as far as ease-of-use is concerned.
I know that he and his company are odious in almost all respects, but seeing his picture in this article made me realise something:
Gates embodies quite a lot of 'geek qualities' - just look at the photo, coffee cup hanging from fingers, dorky glasses, looking decidedly... scruffy (even in a tie).
It might all be part of a carefully sculpted public image, but I get the impression that he doesn't care what impression he gives.
So, I think he's cool (while also being thoroughly dastardly, don't get me wrong...).
Well... Heroine Virtual (xmovie programmers) had a message on their site a few months ago which seemed to state that they'd integrated DeCSS into the player.
They were asking if they should 1) release binary-only players with decss, or 2) remove decss.
Someone from the FSF contacted them and said that if they were going to keep the code GPLd they shouldn't release binary only.
So that's why.
Sooo... theoretically you could wrap decss back into the code _fairly_ easily, especially if they left the appropriate hooks in there.
/james.
Well, some people edited the page, then root came on, the webpage was taken down and now it's refusing connections on port 22 :)
/james.
Yeow, that's what 100volts feels like? I managed to do what you've described a couple of years ago. Painful, but didn't feel like 100 volts.
:)
Of course, that was in Australia so it's probably different.
Anyway, yeah, don't do it
Alright, here's something I've never understood about brute-forcing encrypted files (not that I understand the area very much at all):
Presumably the brute-forcing algorithm has made some assumptions about the nature of the encryption it's testing. If, as you say, it's using some kind of dictionary to detect when it has cracked the file:
Why not simply put the plaintext data through some simple cipher that makes it look just like garbage so they can't tell when they have the correct key?
I know there's a reason why this is silly... anyone care to enlighten me?