While it's true that the Tungsten T2 was released just a few months ago, there is significant difference between the T2 and the T3. I can summarize the differences as follows:
The T2 has a Texas Instruments OMAP processor (ARM derivative with a built-in DSP), while the T3 has an Intel XScale (a StrongARM chip).
The T2 has a dedicated handwriting area. The T3 has a soft handwriting area that doubles as extra screen real-estate.
The T2 has 32 megs of RAM. The T3 has 64 megs of RAM.
The T3 has somewhat streamlined and updated navigation buttons.
The T3 has the ability to display in landscape mode or portrait, due to the extra pixels revealed by the slide-out navigation controls. The T2 only supports portrait display.
I bought a T2 a couple months back, even though I knew the T3 was going to arrive "soon." (Photos of the T3 had already been leaked, but it was common knowledge that Palm was sitting on the T3 for some reason.) Since I didn't know how soon "soon" meant, I got the T2. I have few regrets. The T2 undoubtedly has better battery life, and I don't have to worry about excess wear on the portion of the display reserved for handwriting, because there are no live pixels in that region.
So why did Palm release the T2 at all? A few reasons. If you look at the difference between the original Tungsten T and the T2, there are very few. The main differences are: The T2 has a better display, the T2 has double the RAM (32 megs versus 16 megs), and the T2 has a newer PalmOS revision. It's this last part that is perhaps the biggest change, since the newer PalmOS comes with Graffiti 2. As you might recall, Palm lost a lawsuit over the original Graffiti handwriting recognition system, and as part of the remedies, they agreed to migrate to the new Graffiti 2 HWR system across their entire product line. I think the T2 was released precisely to comply with the remedies required of them, and it gave Palm an excuse to slip in a few other revisions to the device. The original Tungsten T was, at the time, the only professional PalmOS device from Palm still using the original Graffiti HWR. Even the Zire series had moved to Graffiti 2.
Thanks for the link. I guess Flash is more "open" than I thought. It's still proprietary and therefore driven by a single corporation (and therefore subject to change at that corporation's whims).
A "standard" that requires most people (i.e, IE users) to download a plugin, a "standard" that almost nobody actually uses (unlike other similiar technologies like flash)
Interesting, considering that Flash requires a plugin as well, and the plugin that gets automagically installed with IE and Netscape often isn't the latest version (requiring you to download the latest Flash plugin). Yes, Flash is cool, but it's a closed, proprietary binary format.
SVG is cool because you can freely mix SVG elements with XHTML elements in the same document, and browsers such as Mozilla (with the optional SVG support compiled in) can render such documents appropriately. Because SVG DOM works just like any other XML DOM does, you can manipulate your SVG document in interesting ways in real time using Javascript in a browser environment.
SVG renders on small footprint devices (like PDAs and cell phones, which is what this SVG contest is all about) and requires very little CPU or memory, making it an ideal cross-platform vector graphics solution. This is a real solution to a very real problem, a problem that Flash can't solve (or can't solve well). Hopefully, SVG will replace Flash as more browsers incorporate SVG.
Funny that once upon a time, most PocketPC/Windows CE devices were MIPS based, and a few were SH3/SH4 (Hitachi) based. So much so that MIPS was virtually synonymous with Windows CE device. Now that there are several ARM and StrongARM flavors out there, that seems to be the CPU architecture of choice; even Palm has seen the light, and uses the ARM architecture (although only the ultra-high-end Palm devices get the StrongARM -- the rest get the TI OMAP chip, which has a little less horsepower but has a built-in DSP).
It's also amusing to me that the ARM architecture is what the entire Newton line was built around from day one. It took forever for the other PDA vendors to get around to adopting the ARM processor, perhaps because of the perceived taint of the Newton. (Which is a pity, as I liked the Newton very much.)
I don't know if the major WinCE/PocketPC developers are still cross-compiling their applications for MIPS, SH3/4 and ARM, but up until I gave up on WinCE/PocketPC devices entirely (due to a bad experience with an early iPaq model), I saw that most developers supplied binaries for most or all architectures, and some even supplied binaries specifically tweaked for StrongARM (as opposed to, or in addition to, generic ARM binaries).
Bottom line? I think the emphasis on Linux has more to do with AMD's perception of the industry -- which lately has been pushing embedded Linux as a low cost embedded OS -- and the desire to differentiate their reference platform from everyone else's Microsoft-centric offerings.
I've had my share of bad experiences with recruiters and headhunters. Let me mention a few:
The most obvious situation, and one that happens to quite a few people, is having your resume submitted for a position without the recruter/headhunter informing you first. This has cost me at least one job, because another recruiter submitted me for the same job.
Being sold a bill of goods about a position, only to find out that the position is completely unlike what you were told by the recruiter (and even by the hiring manager). A recruiter for the now-defunct Blackstone Group got me a position at American Express which I discovered, after taking the job, was not the kind of nightmare I wanted any part of. I discussed the situation with my Blackstone rep, hoping to work out some kind of amicable arrangement which would allow me to gracefully exit; she turned around and stabbed me in the back, and caused me to be escorted from the AmEx property by security. I was never paid for the time I did work on the job, and I've been black-listed by AmEx ever since.
During a job interview at a now-defunct dot-com company in the San Jose area, I was informed that the company was the darling of headhunters because they didn't require the recruiters to profile candidates. When I asked about this practice, I was informed that it was common for companies (especially those with lots of venture capital behind them) to demand that employees fit a certain "profile" -- e.g., Ivy League background, or white protestant, or "looks like an IBM employee," etc. It's illegal to discriminate against applicants on the basis of most of these criteria, so companies that need to do such hiring outsource the recruitment to a headhunter rather than let their own HR department deal with it. The implications should be pretty obvious.
Some recruiters would hardly give me the time of day, but as soon as they jump ship and join a new company, they call me up to let me know that they're with a new firm, and won't I please throw some business their way?
Beware of recruiters and headhunters who are trying to place candidates for IT jobs, but conduct all (or most) of their business on index cards and paper files. There are one or two major firms out there that still do it this way, and they almost invariably lose your information within a few days.
The opposite problem of the recruiter that keeps it all on paper is the recruiter who does nothing but keyword searches on resumes. These people never bother reading your resume; they dump it straight into a database, and even when your name comes up, chances are good the person who called you on the phone still hasn't read your resume, as evidenced by the stupid questions you'll be asked, questions that would have been answered if the recruiter had spent even two minutes looking at your resume.
It seems a bit disingenuous to say you've never found an honest headhunter, when you yourself are not honest.
Yes, but he was honest about his prior dishonesty. Assuming, of course, that you believe taking a bribe to be dishonest.
Incidentally, the word "disingenuous" means insincere or calculating. Look it up. When someone admits to unethical and possibly illegal activity, that doesn't strike me as typical insincere (or calculated) behavior.
Regardless, the tale was meant to provide an example of the sleazy business ethics of headhunters. You can learn a lot from stories told by even the worst scum of the earth.
Crap, you're right. I was confused by the wording of the article I cited.
Damn. Well, the statistics for 2002 are still good numbers, but I agree with you, I'd like to see 2003 sales figures and statistics on iPod market share.
In a recent Newsweek article, Apple let slip that iTunes for Windows has been moved up, apparently because development has progressed faster than expected. They're citing an October release, now, less than a month away:
And Apple, NEWSWEEK learned, quietly informed some music insiders that it's moved up the date for expanding its current Mac-only iTunes for the vast universe of Windows-based PCs to mid-October. Apple couldn't be reached for comment.
Funny, I've owned two iBooks, one of the original Tangerine models (the one that looked kind of like a toilet seat), and later, one of the newer 2001 models (with the white plastics). I only ever had to have my 2001 iBook repaired once, and that was because there was an issue with the plastics on the bottom of the unit. That was repaired for free under warranty, and I haven't had a problem since. (The problem may have been my fault, even; I turned the cam that holds the battery in place in the wrong direction, and it may have cracked the plastics holding the cam in place, but Apple repaired it without complaint.)
I've always felt that the iBooks, while underpowered, were rugged laptops. Considering the abuse I put them through, I think they hold up very well.
On the other hand, I've seen a lot of issues with the new 12" PowerBook, which is based on the iBook design. Go figure.
Yes, I have. Although Apple has very low market share among all personal computer manufacturers, they have very large market share (well over 50%, last I checked) among all vendors of digital audio players.
The iPod pretty much dominates the market for MP3 players. Making the iPod compatible with PCs was a brilliant move, and it'll take a lot more than Dell's share of the personal computer market to shove the Dell DJ down people's throats.
There's been an explosion of iPod sales this year, fueled initially by the availability of an iPod version for Windows, and then later by a version that works with both Macs and Wintel PCs, and supports both FireWire and USB 2.0.
Well, the one of the first big glitches the music industry faced when it first launched this corrupt disks was locking up iMacs. The CD had to be extracted mechanically.
Further underscoring that the record labels really don't care about Mac users. However, consumer backlash did get most of the discs with this particular copy protection scheme yanked from the market. A software and/or firmware update from Apple may help in this case.
Anyway, I'd rather see Mac users able to mount the Red Book audio session directly and be able to rip music for their own personal use (which they're legally entitled to do) than see Mac users unable to even insert the "protected" discs into their computers' CD/DVD drives because of the potential for wreaking havoc.
The only way a record label will care if a Linux or OS X user is ripping content (protected or not)is if they make it widely available, and in large quantities. In case you've been asleep for the last year or so, these lawsuits are primarily targeted at users who share thousands or tens of thousands of songs. Yes, some folks run Gnutella supernodes on Linux boxen, and probably are hosting quite a few files. Those people are clearly targets, and will be caught in the dragnet soon enough. But by and large, most of the offenders are running Windows; ergo, statistically, Linux and OS X users are almost insignificant, so a copy protection scheme doesn't need to target them.
Also, the most popular file sharing software, the stuff most people are aware of, is Windows-only. IIRC, Kazaa is Windows-only, and many of the lawsuits are targeted specifically at Kazaa users.
While I don't care about Anthony Hamilton's music, I do care about some other artists whose albums are getting a similar treatment. A Perfect Circle just released Thirteenth Step this week, and the album is mangled/copy protected in some regions. It seems that the United States and Canada are getting the unmolested version of the album (which I confirmed on my G4 Cube by ripping a single track in iTunes for testing purposes), but Europe and other parts of the world are getting the jacked up version.
This isn't right. It creates divisions between fans where there shouldn't be any. Many Europeans are writing on the message boards for A Perfect Circle about their negative experiences with the new album -- apparently, it won't play correctly in certain models of Sony's Discman. I'm having trouble tracking down the specific messages, as these boards get a lot of traffic -- it's possible the negative reports of CDs that won't play on certain players were yanked from the web site. However, it was confirmed on the official fan site (in a news update on September 15th) that this is a very real problem for European users.
Those who think that this problem will go away and can be ignored "because it's only happening to albums/bands/artists that I don't listen to," think again. Eventually, something you care about will be affected.
I'm told that Mike Oldfield re-released Tubular Bells recently, and the initial re-release had some brain damaged copy protection on it. Oldfield didn't understand why legitimate fans would care about this, and actually laughed at one fan who dared to ask why (and told this person that only people who want to pirate music could possibly want a version of the album without copy protection). Nevertheless, consumer pressure forced the label to release an unmolested version of the remastered album.
Based on this information, it's pretty clear to me that the record labels are engaged in a stealth campaign to test which copy protection schemes cause the least uproar among consumers, before they start using these techniques on a wider scale. We can't give an inch. Caving in now means destroying the future market for portable digital music players and possibly destroying the market for esoteric HiFi audio equipment (which often fails to play these mangled discs).
Not to sound like a pro-Mac weenie (although I am), but this is yet another reason why my primary choice of operating systems for day-to-day use at home is Mac OS X. Then again, I suspect that the primary goal of the "copy protection" on this CD was to lock out the majority of music pirates, who run MS Windows. I doubt that the major labels care that Linux and OS X users can rip the audio tracks by mounting the Red Book session directly.
They are under obligation, however, not to market items that are copy protected as 'CDs'. They are laser-readable discs containing music, but 'CD' is a trademark. I beleive it belongs to Pioneer, IIRC.
The owner of the trademark is Philips, not Pioneer. They get to decide who can use the CD logo, and who can call something a "Compact Disc."
This is a legal grey area. Legislation's in U.S. congressional committee right now that requires CD makers to place a prominent 'This CD is copy protected' label on all CDs that contain anti-sharing mechanisms.
Philips has already threatened to sue companies that release such discs and label them as CDs. Philips is rightly worried that such incompatible discs (which often refuse to play correctly on some high-end and some consumer level players) will dilute the Compact Disc trademark, or worse, harm it substantially. Philips' position, which I support, is that a CD must conform to the Red Book standard for audio Compact Discs. Anything else isn't a CD.
Remember when BeOS was for Macs, whose users tend to be artistic? Guess that's not the case now that it's an Intel OS, eh?
Actually, I remember farther back, when Be made its own hardware, the BeBox (which Be, Inc. president Jean Louis Gassee called the "Amiga for the 90's"). It was a dual processor PowerPC machine which was designed to accept PCI and ISA cards originally made for the x86 PC world. This was back when Be's original slogan was "One processor per person is not enough!"
I started working for Metrowerks right around the time when Be was in the beta release cycle for the x86 version of BeOS. My job was supposed to be porting the BeOS debugger to the x86 architecture, a problem compounded by the fact that the BeOS debugger was actually a port of the Macintosh PowerPC debugger. A lot of Mac toolbox calls were being emulated (inefficiently) by BeOS calls. What we really needed to do was a complete rewrite, but nobody wanted to spend the time or money on that.
Metrowerks stabbed Be in the back and killed all BeOS development, and I was given a choice of leaving Metrowerks or switching to another division. I chose to leave. Be eventually did make the full transition to x86, and PowerPC support became a less compelling concern.
It's interesting to note that BeOS support for Macintosh hardware was begun as a project to work around an asinine restriction placed on Be, Inc., by the folks running Macworld Expo. Apparently, the expo organizers decided that unless BeOS ran on Macintosh hardware, Be wouldn't be allowed to show its product off at Macworld; up to that point, BeOS ran only on the BeBox. An internal team got the OS ported, and soon after, Be's management realized that there was no reason for Be, Inc. to be a hardware vendor. The x86 port pretty much clinched it. I think JLG thought that he'd make Be into a software-only company like NeXT, and indeed, BeOS was one of the OSes being considered by Apple to replace Copland; with Apple's decision to buy NeXT instead, Be's only real avenue of survival was cut off.
The moral of the story is that there's little or no room left in the x86 world for alternative operating systems that you must pay for. Linux survives precisely because it costs nothing, so Microsoft has no leverage against Linux in that arena. I still like BeOS, the same way I liked the Amiga's OS, but I knew the handwriting was on the wall when the efforts to get BeOS pre-bundled on x86 hardware stalled. I noticed that Be was trying to get BeOS ported to MIPS hardware, apparently in a bid to compete against Windows CE (and this seems to be why Palm bought Be's intellectual property), but nothing ever came of that either.
I don't see how a post containing nothing but one biased opinion merits an "Interesting" moderation, especially as it's a one-line statement of opinion about the popularity of a codec.
If nobody cares about Ogg Vorbis, why is Rio including Vorbis support in their new players, such as the Rio Karma?
One could just as easily turn your statement around and claim that nobody cares about any codec other than MP3. Yet Apple and a few other vendors include AAC support, even when the only popular music ripping software to support encoding AAC is iTunes. If it weren't for AAC being the format of choice for the iTunes Music Store, I doubt anyone would "care" about AAC either.
As I understand it, the iPod uses an ARM processor without an FPU which is the main issue with ogg playback
Which is why the Ogg Vorbis folks released an integer-only implementation of the Ogg Vorbis codec (called "Tremor"), to specifically promote the adoption of Vorbis in embedded application (such as portable digital music players). So it would be utterly trivial for Apple to add Ogg Vorbis support and even FLAC support in an iPod software update.
Yep, they play.OGG, natively in iTunes as a matter of fact, and you can use OggDrop to encode.
Not to be contrary, as I'm an avowed Mac fan, but since when has iTunes supported Ogg Vorbis playback without a third-party plug-in? You say iTunes plays Ogg Vorbis natively, but AFAIK, this is only possible with a third party plug-in, which must be installed by the end-user. Apple still doesn't support Vorbis on the iPod, as they're more interested in promoting AAC. I've agitated for Ogg Vorbis support on the iPod and in iTunes since the original iPod was released.
I will say that OggDrop is a great OS X application for encoding Vorbis files. I've used it quite a bit in the past.
In the United States, at least, license agreements which prohibit decompiling and reverse engineering of binaries are not only unenforceable, but probably illegal. Reverse engineering is very well protected in this country. A former employer of mine made a good living at doing exactly this.
As long as your goal is compatibility, and as long as you write a spec based on what you glean from decompiling the code, and make sure the new implementation (open source driver, for instance) is based purely off the spec and not off of any of the decompiled code directly, you're safe. Clean room methodologies are well enough understood and practiced that the only time you really have to worry is if you don't follow the rules.
Oh, and incidentally, paraphrasing and even direct copying from a book is totally legal as long as you are within fair use rights as outlined in copyright law; this means you're not allowed to copy the whole thing, but you are allowed to copy portions for scholarly purposes (e.g., writing a thesis or a journal article) or journalistic purposes (e.g., writing a newspaper or magazine article) and for various other well-defined purposes. Source code is another story, and gets treated a little differently, so analogies between source code and books are not really useful here.
As an MIT graduate (Class of 1992), I'm appalled by this turn of events. But what really bugs me is that Hal Abelson is involved with this ludicrous arrangement between MIT and Microsoft, in an administrative capacity. (This is according to the article.) Hal is co-author of the SICP text book (Structure and Interpretation of Computer Programs), and was one of my professors when I was there. How the hell did they buy him off?
What will this mean for future MIT students? Will SCHEME be replaced by C# as the language of choice for entry level CS classes? The article bemoans that many universities are having their CS departments reduced to little better than vocational schools, where knowledge of proprietary software is prized over theory and general concepts that can be applied anywhere. I think this is a very real threat to future innovation.
Microsoft might win more mind-share in the short run, but they'll be screwing the world out of the next generation of advancements in the long run. I, for one, will have grave doubts about sending my offspring to MIT.
Regarding estimates and their trustworthiness, you have a good point. However, I've found that most power dissipation estimates for microprocessors are usually close to the mark when the part finally does sample. Engineers who design systems around these new parts often have to start the design before the part is available, and those designs rely on things like estimated power consumption; the estimates can't be off by more than, say, 10% or there could be serious thermal or electrical issues.
I mean, if estimates turned out to be a factor of 2 off (i.e., the actual power consumption is double the estimate), it would throw a huge monkey wrench into the works, not just for the part vendor but also for downstream consumers of the part. Apple would have had to re-design the G5 Mac and would have delayed introducing it.
That article got its facts wrong, as I pointed out over a month ago in another Slashdot thread. I tried to dig up the old article or the comment that I posted back then, but the dain-bramaged Slashdot search engine won't let me pull up older comments that I've left. (I can see the last 24 comments I wrote, but that's it.)
So, to reiterate YET AGAIN, the EE times article you are citing is IN ERROR. The combined power dissipation of two 2.0 GHz G5 chips is 97 watts. The fact checking for the article in question was very poor, and this has been debunked many places, not just here in Slashdot.
On a personal note, I'm shocked and dismayed that one piece of disinformation can persist for so long, even when so many articles (such as the excellent ArsTechnica series on the PowerPC 970, aka the G5) obviously contradict the one article you chose to cite as a reference. For instance, in the ArsTechnica article, the power dissipation of the 1.8 GHz G5 is given as 42 Watts. It doesn't take a genius to figure out that a 2.0 GHz G5 shouldn't dissipate 97 Watts, over twice what a 1.8 GHz part dissipates. An 11% increase in clock speed does not more than double the power requirements for a semiconductor!
So do us all a favor, and stop spreading disinformation based on one provably flawed piece of news. Of course, I e-mailed a complaint to the author of the article, but like most people, he doesn't have the intellectual fortitude to own up to a mistake and admit he didn't check his facts. I suppose I should have e-mailed the editor instead...
Slashdot Mirror
I bought a T2 a couple months back, even though I knew the T3 was going to arrive "soon." (Photos of the T3 had already been leaked, but it was common knowledge that Palm was sitting on the T3 for some reason.) Since I didn't know how soon "soon" meant, I got the T2. I have few regrets. The T2 undoubtedly has better battery life, and I don't have to worry about excess wear on the portion of the display reserved for handwriting, because there are no live pixels in that region.
So why did Palm release the T2 at all? A few reasons. If you look at the difference between the original Tungsten T and the T2, there are very few. The main differences are: The T2 has a better display, the T2 has double the RAM (32 megs versus 16 megs), and the T2 has a newer PalmOS revision. It's this last part that is perhaps the biggest change, since the newer PalmOS comes with Graffiti 2. As you might recall, Palm lost a lawsuit over the original Graffiti handwriting recognition system, and as part of the remedies, they agreed to migrate to the new Graffiti 2 HWR system across their entire product line. I think the T2 was released precisely to comply with the remedies required of them, and it gave Palm an excuse to slip in a few other revisions to the device. The original Tungsten T was, at the time, the only professional PalmOS device from Palm still using the original Graffiti HWR. Even the Zire series had moved to Graffiti 2.
Thanks for the link. I guess Flash is more "open" than I thought. It's still proprietary and therefore driven by a single corporation (and therefore subject to change at that corporation's whims).
I stand corrected, although please define what "a good reason" is, with regard to asking Macromedia for Flash player source?
Flash is still proprietary, regardless of how open it is or isn't, and I still think SVG is the better technology in the long run.
Interesting, considering that Flash requires a plugin as well, and the plugin that gets automagically installed with IE and Netscape often isn't the latest version (requiring you to download the latest Flash plugin). Yes, Flash is cool, but it's a closed, proprietary binary format.
SVG is cool because you can freely mix SVG elements with XHTML elements in the same document, and browsers such as Mozilla (with the optional SVG support compiled in) can render such documents appropriately. Because SVG DOM works just like any other XML DOM does, you can manipulate your SVG document in interesting ways in real time using Javascript in a browser environment.
SVG renders on small footprint devices (like PDAs and cell phones, which is what this SVG contest is all about) and requires very little CPU or memory, making it an ideal cross-platform vector graphics solution. This is a real solution to a very real problem, a problem that Flash can't solve (or can't solve well). Hopefully, SVG will replace Flash as more browsers incorporate SVG.
Funny that once upon a time, most PocketPC/Windows CE devices were MIPS based, and a few were SH3/SH4 (Hitachi) based. So much so that MIPS was virtually synonymous with Windows CE device. Now that there are several ARM and StrongARM flavors out there, that seems to be the CPU architecture of choice; even Palm has seen the light, and uses the ARM architecture (although only the ultra-high-end Palm devices get the StrongARM -- the rest get the TI OMAP chip, which has a little less horsepower but has a built-in DSP).
It's also amusing to me that the ARM architecture is what the entire Newton line was built around from day one. It took forever for the other PDA vendors to get around to adopting the ARM processor, perhaps because of the perceived taint of the Newton. (Which is a pity, as I liked the Newton very much.)
I don't know if the major WinCE/PocketPC developers are still cross-compiling their applications for MIPS, SH3/4 and ARM, but up until I gave up on WinCE/PocketPC devices entirely (due to a bad experience with an early iPaq model), I saw that most developers supplied binaries for most or all architectures, and some even supplied binaries specifically tweaked for StrongARM (as opposed to, or in addition to, generic ARM binaries).
Bottom line? I think the emphasis on Linux has more to do with AMD's perception of the industry -- which lately has been pushing embedded Linux as a low cost embedded OS -- and the desire to differentiate their reference platform from everyone else's Microsoft-centric offerings.
Yes, but he was honest about his prior dishonesty. Assuming, of course, that you believe taking a bribe to be dishonest.
Incidentally, the word "disingenuous" means insincere or calculating. Look it up. When someone admits to unethical and possibly illegal activity, that doesn't strike me as typical insincere (or calculated) behavior.
Regardless, the tale was meant to provide an example of the sleazy business ethics of headhunters. You can learn a lot from stories told by even the worst scum of the earth.
Crap, you're right. I was confused by the wording of the article I cited.
Damn. Well, the statistics for 2002 are still good numbers, but I agree with you, I'd like to see 2003 sales figures and statistics on iPod market share.
Funny, I've owned two iBooks, one of the original Tangerine models (the one that looked kind of like a toilet seat), and later, one of the newer 2001 models (with the white plastics). I only ever had to have my 2001 iBook repaired once, and that was because there was an issue with the plastics on the bottom of the unit. That was repaired for free under warranty, and I haven't had a problem since. (The problem may have been my fault, even; I turned the cam that holds the battery in place in the wrong direction, and it may have cracked the plastics holding the cam in place, but Apple repaired it without complaint.)
I've always felt that the iBooks, while underpowered, were rugged laptops. Considering the abuse I put them through, I think they hold up very well.
On the other hand, I've seen a lot of issues with the new 12" PowerBook, which is based on the iBook design. Go figure.
The iPod pretty much dominates the market for MP3 players. Making the iPod compatible with PCs was a brilliant move, and it'll take a lot more than Dell's share of the personal computer market to shove the Dell DJ down people's throats.
Some market share stats for the iPod:
- Belgium: 38%
- USA (currently): 58%
There's been an explosion of iPod sales this year, fueled initially by the availability of an iPod version for Windows, and then later by a version that works with both Macs and Wintel PCs, and supports both FireWire and USB 2.0.And yeah, the Dell DJ really is ugly as sin.
Further underscoring that the record labels really don't care about Mac users. However, consumer backlash did get most of the discs with this particular copy protection scheme yanked from the market. A software and/or firmware update from Apple may help in this case.
Anyway, I'd rather see Mac users able to mount the Red Book audio session directly and be able to rip music for their own personal use (which they're legally entitled to do) than see Mac users unable to even insert the "protected" discs into their computers' CD/DVD drives because of the potential for wreaking havoc.
The only way a record label will care if a Linux or OS X user is ripping content (protected or not)is if they make it widely available, and in large quantities. In case you've been asleep for the last year or so, these lawsuits are primarily targeted at users who share thousands or tens of thousands of songs. Yes, some folks run Gnutella supernodes on Linux boxen, and probably are hosting quite a few files. Those people are clearly targets, and will be caught in the dragnet soon enough. But by and large, most of the offenders are running Windows; ergo, statistically, Linux and OS X users are almost insignificant, so a copy protection scheme doesn't need to target them.
Also, the most popular file sharing software, the stuff most people are aware of, is Windows-only. IIRC, Kazaa is Windows-only, and many of the lawsuits are targeted specifically at Kazaa users.
While I don't care about Anthony Hamilton's music, I do care about some other artists whose albums are getting a similar treatment. A Perfect Circle just released Thirteenth Step this week, and the album is mangled/copy protected in some regions. It seems that the United States and Canada are getting the unmolested version of the album (which I confirmed on my G4 Cube by ripping a single track in iTunes for testing purposes), but Europe and other parts of the world are getting the jacked up version.
This isn't right. It creates divisions between fans where there shouldn't be any. Many Europeans are writing on the message boards for A Perfect Circle about their negative experiences with the new album -- apparently, it won't play correctly in certain models of Sony's Discman. I'm having trouble tracking down the specific messages, as these boards get a lot of traffic -- it's possible the negative reports of CDs that won't play on certain players were yanked from the web site. However, it was confirmed on the official fan site (in a news update on September 15th) that this is a very real problem for European users.
Those who think that this problem will go away and can be ignored "because it's only happening to albums/bands/artists that I don't listen to," think again. Eventually, something you care about will be affected.
I'm told that Mike Oldfield re-released Tubular Bells recently, and the initial re-release had some brain damaged copy protection on it. Oldfield didn't understand why legitimate fans would care about this, and actually laughed at one fan who dared to ask why (and told this person that only people who want to pirate music could possibly want a version of the album without copy protection). Nevertheless, consumer pressure forced the label to release an unmolested version of the remastered album.
Based on this information, it's pretty clear to me that the record labels are engaged in a stealth campaign to test which copy protection schemes cause the least uproar among consumers, before they start using these techniques on a wider scale. We can't give an inch. Caving in now means destroying the future market for portable digital music players and possibly destroying the market for esoteric HiFi audio equipment (which often fails to play these mangled discs).
Someone mod this man up!
Not to sound like a pro-Mac weenie (although I am), but this is yet another reason why my primary choice of operating systems for day-to-day use at home is Mac OS X. Then again, I suspect that the primary goal of the "copy protection" on this CD was to lock out the majority of music pirates, who run MS Windows. I doubt that the major labels care that Linux and OS X users can rip the audio tracks by mounting the Red Book session directly.
The owner of the trademark is Philips, not Pioneer. They get to decide who can use the CD logo, and who can call something a "Compact Disc."
Philips has already threatened to sue companies that release such discs and label them as CDs. Philips is rightly worried that such incompatible discs (which often refuse to play correctly on some high-end and some consumer level players) will dilute the Compact Disc trademark, or worse, harm it substantially. Philips' position, which I support, is that a CD must conform to the Red Book standard for audio Compact Discs. Anything else isn't a CD.
Actually, I remember farther back, when Be made its own hardware, the BeBox (which Be, Inc. president Jean Louis Gassee called the "Amiga for the 90's"). It was a dual processor PowerPC machine which was designed to accept PCI and ISA cards originally made for the x86 PC world. This was back when Be's original slogan was "One processor per person is not enough!"
I started working for Metrowerks right around the time when Be was in the beta release cycle for the x86 version of BeOS. My job was supposed to be porting the BeOS debugger to the x86 architecture, a problem compounded by the fact that the BeOS debugger was actually a port of the Macintosh PowerPC debugger. A lot of Mac toolbox calls were being emulated (inefficiently) by BeOS calls. What we really needed to do was a complete rewrite, but nobody wanted to spend the time or money on that.
Metrowerks stabbed Be in the back and killed all BeOS development, and I was given a choice of leaving Metrowerks or switching to another division. I chose to leave. Be eventually did make the full transition to x86, and PowerPC support became a less compelling concern.
It's interesting to note that BeOS support for Macintosh hardware was begun as a project to work around an asinine restriction placed on Be, Inc., by the folks running Macworld Expo. Apparently, the expo organizers decided that unless BeOS ran on Macintosh hardware, Be wouldn't be allowed to show its product off at Macworld; up to that point, BeOS ran only on the BeBox. An internal team got the OS ported, and soon after, Be's management realized that there was no reason for Be, Inc. to be a hardware vendor. The x86 port pretty much clinched it. I think JLG thought that he'd make Be into a software-only company like NeXT, and indeed, BeOS was one of the OSes being considered by Apple to replace Copland; with Apple's decision to buy NeXT instead, Be's only real avenue of survival was cut off.
The moral of the story is that there's little or no room left in the x86 world for alternative operating systems that you must pay for. Linux survives precisely because it costs nothing, so Microsoft has no leverage against Linux in that arena. I still like BeOS, the same way I liked the Amiga's OS, but I knew the handwriting was on the wall when the efforts to get BeOS pre-bundled on x86 hardware stalled. I noticed that Be was trying to get BeOS ported to MIPS hardware, apparently in a bid to compete against Windows CE (and this seems to be why Palm bought Be's intellectual property), but nothing ever came of that either.
One could just as easily turn your statement around and claim that nobody cares about any codec other than MP3. Yet Apple and a few other vendors include AAC support, even when the only popular music ripping software to support encoding AAC is iTunes. If it weren't for AAC being the format of choice for the iTunes Music Store, I doubt anyone would "care" about AAC either.
Which is why the Ogg Vorbis folks released an integer-only implementation of the Ogg Vorbis codec (called "Tremor"), to specifically promote the adoption of Vorbis in embedded application (such as portable digital music players). So it would be utterly trivial for Apple to add Ogg Vorbis support and even FLAC support in an iPod software update.
Not to be contrary, as I'm an avowed Mac fan, but since when has iTunes supported Ogg Vorbis playback without a third-party plug-in? You say iTunes plays Ogg Vorbis natively, but AFAIK, this is only possible with a third party plug-in, which must be installed by the end-user. Apple still doesn't support Vorbis on the iPod, as they're more interested in promoting AAC. I've agitated for Ogg Vorbis support on the iPod and in iTunes since the original iPod was released.
I will say that OggDrop is a great OS X application for encoding Vorbis files. I've used it quite a bit in the past.
In the United States, at least, license agreements which prohibit decompiling and reverse engineering of binaries are not only unenforceable, but probably illegal. Reverse engineering is very well protected in this country. A former employer of mine made a good living at doing exactly this.
As long as your goal is compatibility, and as long as you write a spec based on what you glean from decompiling the code, and make sure the new implementation (open source driver, for instance) is based purely off the spec and not off of any of the decompiled code directly, you're safe. Clean room methodologies are well enough understood and practiced that the only time you really have to worry is if you don't follow the rules.
Oh, and incidentally, paraphrasing and even direct copying from a book is totally legal as long as you are within fair use rights as outlined in copyright law; this means you're not allowed to copy the whole thing, but you are allowed to copy portions for scholarly purposes (e.g., writing a thesis or a journal article) or journalistic purposes (e.g., writing a newspaper or magazine article) and for various other well-defined purposes. Source code is another story, and gets treated a little differently, so analogies between source code and books are not really useful here.
As an MIT graduate (Class of 1992), I'm appalled by this turn of events. But what really bugs me is that Hal Abelson is involved with this ludicrous arrangement between MIT and Microsoft, in an administrative capacity. (This is according to the article.) Hal is co-author of the SICP text book (Structure and Interpretation of Computer Programs), and was one of my professors when I was there. How the hell did they buy him off?
What will this mean for future MIT students? Will SCHEME be replaced by C# as the language of choice for entry level CS classes? The article bemoans that many universities are having their CS departments reduced to little better than vocational schools, where knowledge of proprietary software is prized over theory and general concepts that can be applied anywhere. I think this is a very real threat to future innovation.
Microsoft might win more mind-share in the short run, but they'll be screwing the world out of the next generation of advancements in the long run. I, for one, will have grave doubts about sending my offspring to MIT.
Wasn't there a huge outcry at Yale, followed by a retraction of this idiotic policy decision?
Regarding estimates and their trustworthiness, you have a good point. However, I've found that most power dissipation estimates for microprocessors are usually close to the mark when the part finally does sample. Engineers who design systems around these new parts often have to start the design before the part is available, and those designs rely on things like estimated power consumption; the estimates can't be off by more than, say, 10% or there could be serious thermal or electrical issues.
I mean, if estimates turned out to be a factor of 2 off (i.e., the actual power consumption is double the estimate), it would throw a huge monkey wrench into the works, not just for the part vendor but also for downstream consumers of the part. Apple would have had to re-design the G5 Mac and would have delayed introducing it.
That article got its facts wrong, as I pointed out over a month ago in another Slashdot thread. I tried to dig up the old article or the comment that I posted back then, but the dain-bramaged Slashdot search engine won't let me pull up older comments that I've left. (I can see the last 24 comments I wrote, but that's it.)
So, to reiterate YET AGAIN, the EE times article you are citing is IN ERROR. The combined power dissipation of two 2.0 GHz G5 chips is 97 watts. The fact checking for the article in question was very poor, and this has been debunked many places, not just here in Slashdot.
On a personal note, I'm shocked and dismayed that one piece of disinformation can persist for so long, even when so many articles (such as the excellent ArsTechnica series on the PowerPC 970, aka the G5) obviously contradict the one article you chose to cite as a reference. For instance, in the ArsTechnica article, the power dissipation of the 1.8 GHz G5 is given as 42 Watts. It doesn't take a genius to figure out that a 2.0 GHz G5 shouldn't dissipate 97 Watts, over twice what a 1.8 GHz part dissipates. An 11% increase in clock speed does not more than double the power requirements for a semiconductor!
So do us all a favor, and stop spreading disinformation based on one provably flawed piece of news. Of course, I e-mailed a complaint to the author of the article, but like most people, he doesn't have the intellectual fortitude to own up to a mistake and admit he didn't check his facts. I suppose I should have e-mailed the editor instead...