Incidentally, wtf is a roaming fee? Is it just a revenue source, or is the cell phone network actually that insanely expensive to operate?
It's actually very simple.
The towers that your cell phone communicates with are owned by the different cell carriers. If you're in your home region, you are connected to your own company's tower.
When you travel, you may have to connect to some other company's tower. When this happens (assuming you can connect at all), this other company will be originating your call. They will bill your company for the call, and that bill gets passed on to you. This (possibly with a service fee tacked on) is the roaming charge.
The charges are less today than in the past because the industry is maturing. The companies today usually have enough spare bandwidth on the towers that they don't have to worry about roamers using up all the capacity. And the different companies end up completing each others' calls so much that the monthly settlements (where each company bills the other for roamers) end up balancing out (mostly).
This is why today you can get a plan like the one Verizon gave me recently, where almost all of the US is a covered area, and roaming charges are 40 cents per minute for those few areas that are not covered. A lot higher than land-line calls, but a lot less than they were in the past.
As more towers get put up, and as more carriers decide to allow each others' roamers through without extra charges, I suspect we'll see roaming fees drop even further and eventually disappear altogether. (ISTR a recent ad for a carrier promising absolutely no roaming charges within the US.)
The OEM sample price is about $1,092 USD...but don't ask me what that means for consumers
It means not many will care to have one for a while. At least not until they are comperable to today's 2.5 inch drives (+/- a 100 USD).
It means absolutely nothing. Engineering samples (especially for a product with an expected high demand) are often very expensive. They're not meant for you to stick in your home PC. They're meant for PC makers like Dell and Compaq to use for testing, so they can ship the drives as soon as they become commercially available. The high price guarantees that most of the samples will ship to their intended target audience - those that can't afford to wait for the actual commercial release.
I think we can be pretty certain that this drive will be competitively priced when it actually starts shipping.
in all seriousness, how long until this is priced and packaged to sell in an OEM enviroment?
According to this article, you usually find drives shipping a few months after engineering samples become available. First to computer makers, and then retail.
Since samples will ship in May, I would expect to see them bundled with new systems around August/September, and available retail before the end of the year.
If I could find a 7200 rpm drive that didn't destroy the battery life in my PowerBook I would be very happy.
Which brings to mind an interesting idea. I wonder if anybody's tried making a hard drive with a variable-speed spindle. Provide a bunch of speeds that your operating system can select from. So you can run at 4200 RPM (or maybe even slower) when you're on batteries and spin up to 7200 when you're plugged into an external power source. Make it configurable through a power-management control panel.
Given that drives already have power modes where they completely turn off at times, this might not be a big stretch for an HD company to design.
MacOS X has always used gcc as the main compiler. And GCC has always supported cross-compiling. I would be far more surprised if you couldn't compile Apple's version of GCC to cross-compile from Windows/Linux/Solaris/whatever.
I suppose it's worth noting that someone's distributing such a cross-compiler, but the most newsworthy aspect is that it took somebody this long to get around to doing it.
FYI, when you buy an ASUS motherboard, its utility cdrom boots FreeDOS.
Well, a stripped-down version with a broken installer. The FreeDOS people warn against using this CD for anything other than flashing a corrupted BIOS. Those who try to install FreeDOS from this CD may end up with a trashed boot sector.
I'd have to give this "award" to Tempest 2000. On the Atari Jaguar, this non-stop game plays a very addictive techno soundtrack underneath the entire game.
Also of note is Rayman. Although the background music isn't that spectacular, the game does have some incredibly surreal music-oriented background scenery.
Another game worth mention is Zoop. This game's background music will get lodged in your head and you'll find yourself humming it for a week if you're not careful. This game is worth tracking down, even today. Although it's written for MS-DOS, it's run flawlessly under every DOS environment I've thrown at it (including several different releases of Windows and OS/2's DOS box.)
You can edit the hosts file on any version of Windows. I've used it in 95.
But the proxy autoconfig method is better. With a hosts file, you must individually block specific hosts. With the autoconfig file, you can block entire domains.
If doubleclick renames its server every week, you'll have to update your hosts file every week. But with the proxy file, you just block the entire doubleclick.net domain and be done with it once and for all.
Apparently other people don't like that sort of stuff too. Look at someone's host file that I found.
Yep. Unfortunately, using a hosts file only solves part of the problem. Sometimes the ads are served from the same machine as content you want - so you need to filter at the URL level (more specific). And sometimes the advertiser keeps changing hostnames, so you need to constantly add hostnames or filter at the domain level (less specific).
So far, I've found that the proxy-autoconfig file/script is the best way to do this. Every URL gets sent to your JavaScript function, which can selectively choose to let the request go out to the internet, or get shunted to a proxy server (which doesn't actually have to proxy anything.)
BTW, using either technique, it pays to run a web server at the address you're shunting the ad hosts/domains/URLs to. Some web sites completely hang if they can't connect to a web server at the ad's URL. But they don't seem to have any problem if they find a server that returns a "404 page not found" error.
It's called right-click, block images from this server.
But this only works for stuff in IMG tags. It doesn't do a thing for the ads based on plugins (like Flash - which is becoming increasingly common.) The proxy-script method (which I just posted, above) works much better, although you do need a web server somewhere that you can shunt your ad requests to.
I leave a copy of Apache running on a Linux box on my LAN for just this purpose.
Mozilla and Mozilla Firefox have a nifty browser extension called AdBlock.
Even simpler. Most browsers (including Mozilla, Firefox and IE) let you use a JavaScript for automatic proxy configuration. You can also use this to block ads. Have the script look at the domains in the URLs, sending the ad sites to a locally-installed web server (that doesn't proxy anything) and send the rest directly to the internet.
Your ads all go to the web server, which returns "404 Page not found" errors for eveything. All the ads end up as broken-image links, error text or stuff that isn't the ad. This is especially nice when the ads are Flash animations and other things that waste gobs of CPU power.
I use something like this in a "proxy.pac" file:
function FindProxyForURL(url, host) { // Variable declarations< var destip = "";
// Blackhole these specific domains by sending them to a web server // that doesn't proxy anything. // if (dnsDomainIs(host, ".247media.com") || dnsDomainIs(host, ".accendo.com") || ... dnsDomainIs(host, "ads.x10.com") || dnsDomainIs(host, "ads.zdnet.com") || return "PROXY 192.168.1.5"; }
return "DIRECT"; }
(Sorry for the lack of indenting. SlashDot doesn't support the "PRE" tag, nor do they support the "nbsp" character, and the ECODE tag strips leading spaces.)
Feel free to use it yourself. Of course, replace the "PROXY" address with the address of your own server and add more ad-domains. Whenever I see an ad, I view the page's source to find out the ad's domain/host name and add it to the list. I currently have about 90 domains on my list, which blocks nearly every ad.
You can also proxy to an address that has no server, but I've found that doing so causes some web sites to become completely unresponsive.
And if I do get that ad I don't want it to be bigger and placed in front of my search results.
It's getting even worse these days. I temporarily switched browsers a few weeks ago and hadn't configured it for my ad-blocking proxy. The first site I linked to had one of the most obnoxious popups I'd ever seen. The ad window (probably produced with JavaScript) floated over the page I was trying to see and kept on moving so I had to chase it down in order to click the "close" box.
What the F$%^&* do these advertisers think they are doing? Do they seriously think I'm going to buy anything from a company that goes that far over the top to screw up my browsing experience? Given a choice between ads like that and completely cutting off all internet access, I think I would prefer the latter option.
<SARCASM>I'm so glad to hear that Microsoft is on the side of the advertisers here</SARCASM>
Please read a comment before moderating it. A subject line that says "Article Text" doesn't necessarily mean that it actually is the article's text. This particular example is really a work of fiction loosely modeled on the article. Rank it as "funny" if you think it's deserves a positive rating, but it's definitely not "informative".
Here's the actual article's text:
Microsoft Concedes Misstep in Search
By ALLISON LINN, AP Business Writer
REDMOND, Wash. - When Microsoft Corp. entered the Internet browser war in the 1990s, Netscape Navigator was the early leader and Microsoft's Internet Explorer the late-blooming upstart.
Now, Microsoft is gearing up for a similar battle in search technology.
Chief executive Steve Ballmer conceded Thursday that one big misstep by Microsoft over the past few years is that the company did not put resources toward in-house research and development of search technology.
"That's probably the thing I feel worst about over the last few years -- not making our own R&D investment," Ballmer said at a conference for online advertisers held at Microsoft's Redmond campus.
Instead, he said, the company had relied on outside sources for that technology. Joking that the pervasive software giant is often stereotyped as "doing it all," he said: "This is a case where we didn't do it all -- and I wish we had."
But Microsoft is now turning its considerable might toward catching up. It's a move that puts Microsoft head to head with Google, the world's most popular Web surfing vehicle, and Yahoo, the Internet's most popular destination, in what many see as an important, growing and lucrative market.
"I think you'll see some good competition in this area," Ballmer said.
Ballmer mentioned the European Union (news - web sites)'s ruling against Microsoft only in passing, when asked about rumors the company may be making some big acquisitions. He said he hadn't heard that; he'd been too busy paying attention to rumors coming out of Europe.
The European Commission (news - web sites) slapped Microsoft with a $613 million fine Wednesday for abusively wielding its near monopoly in desktop operating systems and ordered sanctions that go well beyond the company's antitrust settlement with the United States. The company has vowed to appeal.
About 114.5 million Americans, or 39 percent of the population, now use search engines, according to Nielsen NetRatings. Also, businesses spent an estimated $2 billion last year on search-related advertising and some analysts expect the market to triple during the next three years.
Microsoft -- which will spend nearly $7 billion this year on overall research and development -- hopes to have some of its own search technology development done in the next 12 months, Ballmer said. It will take longer to develop search technology focused on advertising, he told the advertising executives.
The software titan also has said that better search technology will be a big part of the next version of its dominant Windows operating system, which may not be released for a couple years or more.
At the conference, Microsoft also unveiled a study on the effectiveness of online advertising. The company is using the study as the basis for an argument that companies spending about 1 percent of their advertising budget online should consider increasing that to 4 percent or 5 percent because people are spending more time online.
Ballmer said Microsoft spends about 12 percent of its media budget on online advertising, and that he orders his staff to "saturate" that market first and foremost.
"I want to make sure (a user) can't get through... an online experience without hitting a Microsoft ad," he said.
This reminds me of the voice-enabled version of Netscape that IBM bundled with OS/2 version 4.
That system was simpler, since it couldn't rely on special voice-HTML markup tags. It took advantage of the fact that any UI element (menu item, button, etc.) in the system can be activated by speaking its text. So they added a quick Hack to Netscape so that every link's text (or ALT text) visible on a screeen would be present on a "Links" menu - thus turning the links into speakable keywords.
It worked very well for browsing. Much less well when you want to enter new URLs. The dictation mode left a bit to be desired. But that was to be expected from the hardware of the time. Voice recognition on OS/2 required a minimum of a 150MHz Pentium, IIRC. (It would work - with much latency - on my 80MHz 486, however.)
Definitely not the case and/or setup for those overclockers out there:) especially with heat output getting real close to 100W
Of course, they're not selling computers. They're selling keyboards, mice and LCD monitors. None of which are going to consume that kind of power. Their 19" monitor only consumes 50W.
I admin a small non profit wireless isp, and this is what I generally try to do, although our user base is small enough (~110 right now) that I can do this and call them up to tell them they have a virus. But this wont work for all types of viruses, if you block someones smtp access you might cause more trouble than just shutting them down outright.
Depends on what you're blocking.
Cutting off someone's SMTP altogether is probably a bit too draconian.
On the other hand, cutting off SMTP access to anything other than their own ISP's mail servers (like Earthlink and a great many other ISPs do) doesn't seem that bad. This is something that should be done all the time, not just when a customer is discovered to be hosting a virus.
If a virus is sending its spam directly to recipients, this completely blocks it. If it is reading your system settings and starts sending through the ISP's server, this is something the ISP can catch through standard anti-virus filters.
There is really no good reason why individual hosts need to send SMTP directly to recipients' mailboxes. Even if they want to run a local copy of sendmail (or other MTA), they can configure it to forward all non-local mail to the ISP's server.
I think it's far more likely that Novell is the "owner" of Unix.
Novell bought UNIX. (Actually, they bought USL, which was an AT&T spinoff). Then they spun off Caldera to create and distribute a Linux distribution. Then Caldera was cut loose, along with the UNIX rights. Then Caldera bought the smoking remains of SCO and took over the SCO name.
SCO's claim to ownership of AT&T's UNIX legacy is correct.
You know, if you're really worried about your equipment going through the scanner, at a lot of airports, you can request a manual examination of the equipment.
Have you actually tried to request this?
Last year, I tried requesting a hand inspection of a box of floppy disks, because I was afraid of possible X-ray damage (they were about 20 year-old 5.25" disks with programs that it would be hard to replace these days.) The security person said "sorry, we can't do that. Put it through the machine or you can't take it on the plane."
I decided to not start a fight over this, since I don't relish the idea of being arrested 3000 miles from home. I consider myself lucky that the disks didn't get damaged.
Interesting. When my remotes become unresponsive, I find that it is more often than not due to corrosion at the battery contacts. I remove the batteries and clean their contacts with a pencil eraser. Then reinsert and have no problems for several months, when I have to repeat the process.
Of course, my symptoms aren't exactly the same. Mashing the buttons doesn't make them work any better. But tapping the remote on a table or the back of my hand does.
I will agree that OS X handles things fairly well when you are only doing one thing at a time. But if you have, for instance, 5 finder windows, 8 safari windows, acrobat, mail.app, and 5 terminals you start to have problems, serious ones. This is with 640 MB of RAM on a G4 933 ibook. I routinely have around 3 times this many apps open on a GNOME desktop with 256MB of RAM and an 800Mhz Duron and it simply screams in comparison.
I guess this is a YMMV case.
On my Mac (dual 1GHz G4, 512M RAM), 10.3 has never given me those kinds of problems. At any given time, I am typically running:
AOL Instant Messenger (or occasionally iChat AV)
Mozilla (primarily for mail)
FireFox or Safari - usually with one window, but several tabs
Emacs (for text editing and reading news)
iTunes
One or two terminals
iPhoto, Acrobat Reader, Preview, GraphicConverter, AppleWorks and FileMaker launched and quit as necessary.
One possible difference: I assume you shutdown or suspend your iBook when you're not using it. I leave my G4 tower running 24/7. I know that there are a few cron jobs that OS X runs. Since anacron isn't pre-installed, these jobs won't run if the computer is off at night. Some people have reported performance improvements by allowing these to run (by installing anacron, or by manually running the scripts from time to time.)
The scripts are:
/etc/daily
/etc/weekly
/etc/monthly
Run them from a root shell (or via sudo).
But all that being said, you are right that Linux does perform more efficiently. My dual-200MHz PPro PC (192M RAM) running Linux with a stripped-down fvwm2 window manager usually gets comparable Java performance using Sun's JVM package and Mozilla.
If you are going to kill an iPod just to show how '1337 you are...
According to the article, he was figuring out the take-apart procedure in order to begin offering a re-coloring service for those iPod mini owners that don't like Apple's choice of colors.
Obviously, if you're going to fuse new colors into anodized aluminum, you don't want to do it with any electronic or plastic parts still attached.
Re:Server is already slow - here's the text
on
iPod Mini Autopsy
·
· Score: 1
Has anyone thought that the problem reading it is most likely due to the format?
I got the impression that this was what the author was referring to. I thought he was trying to mount it as-is, to read whatever low-level stuff (like system software) the iPod reads that you can't get access to through the normal FireWire interface.
I am a bit curious about what kind of computer he tried to use to mount it with. His Lexar card reader should just pass the drive as-is to the host computer, which will have to understand the file system. I'm guessing he has a PC running Windows, and therefore can't mount HFS volumes without third party software.
If, however, he has a Mac, then we end up with a much more interesting question.
Slashdot Mirror
It's actually very simple.
The towers that your cell phone communicates with are owned by the different cell carriers. If you're in your home region, you are connected to your own company's tower.
When you travel, you may have to connect to some other company's tower. When this happens (assuming you can connect at all), this other company will be originating your call. They will bill your company for the call, and that bill gets passed on to you. This (possibly with a service fee tacked on) is the roaming charge.
The charges are less today than in the past because the industry is maturing. The companies today usually have enough spare bandwidth on the towers that they don't have to worry about roamers using up all the capacity. And the different companies end up completing each others' calls so much that the monthly settlements (where each company bills the other for roamers) end up balancing out (mostly).
This is why today you can get a plan like the one Verizon gave me recently, where almost all of the US is a covered area, and roaming charges are 40 cents per minute for those few areas that are not covered. A lot higher than land-line calls, but a lot less than they were in the past.
As more towers get put up, and as more carriers decide to allow each others' roamers through without extra charges, I suspect we'll see roaming fees drop even further and eventually disappear altogether. (ISTR a recent ad for a carrier promising absolutely no roaming charges within the US.)
It means not many will care to have one for a while. At least not until they are comperable to today's 2.5 inch drives (+/- a 100 USD).
It means absolutely nothing. Engineering samples (especially for a product with an expected high demand) are often very expensive. They're not meant for you to stick in your home PC. They're meant for PC makers like Dell and Compaq to use for testing, so they can ship the drives as soon as they become commercially available. The high price guarantees that most of the samples will ship to their intended target audience - those that can't afford to wait for the actual commercial release.
I think we can be pretty certain that this drive will be competitively priced when it actually starts shipping.
According to this article, you usually find drives shipping a few months after engineering samples become available. First to computer makers, and then retail.
Since samples will ship in May, I would expect to see them bundled with new systems around August/September, and available retail before the end of the year.
If I could find a 7200 rpm drive that didn't destroy the battery life in my PowerBook I would be very happy.
Which brings to mind an interesting idea. I wonder if anybody's tried making a hard drive with a variable-speed spindle. Provide a bunch of speeds that your operating system can select from. So you can run at 4200 RPM (or maybe even slower) when you're on batteries and spin up to 7200 when you're plugged into an external power source. Make it configurable through a power-management control panel.
Given that drives already have power modes where they completely turn off at times, this might not be a big stretch for an HD company to design.
I suppose it's worth noting that someone's distributing such a cross-compiler, but the most newsworthy aspect is that it took somebody this long to get around to doing it.
Well, a stripped-down version with a broken installer. The FreeDOS people warn against using this CD for anything other than flashing a corrupted BIOS. Those who try to install FreeDOS from this CD may end up with a trashed boot sector.
Also of note is Rayman. Although the background music isn't that spectacular, the game does have some incredibly surreal music-oriented background scenery.
Another game worth mention is Zoop. This game's background music will get lodged in your head and you'll find yourself humming it for a week if you're not careful. This game is worth tracking down, even today. Although it's written for MS-DOS, it's run flawlessly under every DOS environment I've thrown at it (including several different releases of Windows and OS/2's DOS box.)
But the proxy autoconfig method is better. With a hosts file, you must individually block specific hosts. With the autoconfig file, you can block entire domains.
If doubleclick renames its server every week, you'll have to update your hosts file every week. But with the proxy file, you just block the entire doubleclick.net domain and be done with it once and for all.
I posted about it previously in this thread. Here's the comment
Yep. Unfortunately, using a hosts file only solves part of the problem. Sometimes the ads are served from the same machine as content you want - so you need to filter at the URL level (more specific). And sometimes the advertiser keeps changing hostnames, so you need to constantly add hostnames or filter at the domain level (less specific).
So far, I've found that the proxy-autoconfig file/script is the best way to do this. Every URL gets sent to your JavaScript function, which can selectively choose to let the request go out to the internet, or get shunted to a proxy server (which doesn't actually have to proxy anything.)
BTW, using either technique, it pays to run a web server at the address you're shunting the ad hosts/domains/URLs to. Some web sites completely hang if they can't connect to a web server at the ad's URL. But they don't seem to have any problem if they find a server that returns a "404 page not found" error.
But this only works for stuff in IMG tags. It doesn't do a thing for the ads based on plugins (like Flash - which is becoming increasingly common.) The proxy-script method (which I just posted, above) works much better, although you do need a web server somewhere that you can shunt your ad requests to.
I leave a copy of Apache running on a Linux box on my LAN for just this purpose.
Even simpler. Most browsers (including Mozilla, Firefox and IE) let you use a JavaScript for automatic proxy configuration. You can also use this to block ads. Have the script look at the domains in the URLs, sending the ad sites to a locally-installed web server (that doesn't proxy anything) and send the rest directly to the internet.
Your ads all go to the web server, which returns "404 Page not found" errors for eveything. All the ads end up as broken-image links, error text or stuff that isn't the ad. This is especially nice when the ads are Flash animations and other things that waste gobs of CPU power.
I use something like this in a "proxy.pac" file:
(Sorry for the lack of indenting. SlashDot doesn't support the "PRE" tag, nor do they support the "nbsp" character, and the ECODE tag strips leading spaces.)
Feel free to use it yourself. Of course, replace the "PROXY" address with the address of your own server and add more ad-domains. Whenever I see an ad, I view the page's source to find out the ad's domain/host name and add it to the list. I currently have about 90 domains on my list, which blocks nearly every ad.
You can also proxy to an address that has no server, but I've found that doing so causes some web sites to become completely unresponsive.
It's getting even worse these days. I temporarily switched browsers a few weeks ago and hadn't configured it for my ad-blocking proxy. The first site I linked to had one of the most obnoxious popups I'd ever seen. The ad window (probably produced with JavaScript) floated over the page I was trying to see and kept on moving so I had to chase it down in order to click the "close" box.
What the F$%^&* do these advertisers think they are doing? Do they seriously think I'm going to buy anything from a company that goes that far over the top to screw up my browsing experience? Given a choice between ads like that and completely cutting off all internet access, I think I would prefer the latter option.
<SARCASM>I'm so glad to hear that Microsoft is on the side of the advertisers here</SARCASM>
Here's the actual article's text:
That system was simpler, since it couldn't rely on special voice-HTML markup tags. It took advantage of the fact that any UI element (menu item, button, etc.) in the system can be activated by speaking its text. So they added a quick Hack to Netscape so that every link's text (or ALT text) visible on a screeen would be present on a "Links" menu - thus turning the links into speakable keywords.
It worked very well for browsing. Much less well when you want to enter new URLs. The dictation mode left a bit to be desired. But that was to be expected from the hardware of the time. Voice recognition on OS/2 required a minimum of a 150MHz Pentium, IIRC. (It would work - with much latency - on my 80MHz 486, however.)
Assuming you actually have to speak. Tie it in with something like this and that particular problem goes away.
Of course, they're not selling computers. They're selling keyboards, mice and LCD monitors. None of which are going to consume that kind of power. Their 19" monitor only consumes 50W.
Depends on what you're blocking.
Cutting off someone's SMTP altogether is probably a bit too draconian.
On the other hand, cutting off SMTP access to anything other than their own ISP's mail servers (like Earthlink and a great many other ISPs do) doesn't seem that bad. This is something that should be done all the time, not just when a customer is discovered to be hosting a virus.
If a virus is sending its spam directly to recipients, this completely blocks it. If it is reading your system settings and starts sending through the ISP's server, this is something the ISP can catch through standard anti-virus filters.
There is really no good reason why individual hosts need to send SMTP directly to recipients' mailboxes. Even if they want to run a local copy of sendmail (or other MTA), they can configure it to forward all non-local mail to the ISP's server.
Novell bought UNIX. (Actually, they bought USL, which was an AT&T spinoff). Then they spun off Caldera to create and distribute a Linux distribution. Then Caldera was cut loose, along with the UNIX rights. Then Caldera bought the smoking remains of SCO and took over the SCO name.
SCO's claim to ownership of AT&T's UNIX legacy is correct.
Have you actually tried to request this?
Last year, I tried requesting a hand inspection of a box of floppy disks, because I was afraid of possible X-ray damage (they were about 20 year-old 5.25" disks with programs that it would be hard to replace these days.) The security person said "sorry, we can't do that. Put it through the machine or you can't take it on the plane."
I decided to not start a fight over this, since I don't relish the idea of being arrested 3000 miles from home. I consider myself lucky that the disks didn't get damaged.
Of course, my symptoms aren't exactly the same. Mashing the buttons doesn't make them work any better. But tapping the remote on a table or the back of my hand does.
I guess this is a YMMV case.
On my Mac (dual 1GHz G4, 512M RAM), 10.3 has never given me those kinds of problems. At any given time, I am typically running:
One possible difference: I assume you shutdown or suspend your iBook when you're not using it. I leave my G4 tower running 24/7. I know that there are a few cron jobs that OS X runs. Since anacron isn't pre-installed, these jobs won't run if the computer is off at night. Some people have reported performance improvements by allowing these to run (by installing anacron, or by manually running the scripts from time to time.)
The scripts are:
/etc/daily /etc/weekly /etc/monthly
Run them from a root shell (or via sudo).But all that being said, you are right that Linux does perform more efficiently. My dual-200MHz PPro PC (192M RAM) running Linux with a stripped-down fvwm2 window manager usually gets comparable Java performance using Sun's JVM package and Mozilla.
According to the article, he was figuring out the take-apart procedure in order to begin offering a re-coloring service for those iPod mini owners that don't like Apple's choice of colors.
Obviously, if you're going to fuse new colors into anodized aluminum, you don't want to do it with any electronic or plastic parts still attached.
I got the impression that this was what the author was referring to. I thought he was trying to mount it as-is, to read whatever low-level stuff (like system software) the iPod reads that you can't get access to through the normal FireWire interface.
I am a bit curious about what kind of computer he tried to use to mount it with. His Lexar card reader should just pass the drive as-is to the host computer, which will have to understand the file system. I'm guessing he has a PC running Windows, and therefore can't mount HFS volumes without third party software.
If, however, he has a Mac, then we end up with a much more interesting question.
If you do, point them to this page. The author there recommends buying a Creative Nomad MuVo. It has the same 4G drive and sells for $200.