You bring up an interesting point, so I actually called my attorney and asked him about the points you bring up.
Yes, just getting an IP address is a felony. FCC law says that robbing someone electronically of services or interfering with electronic transmission IS a felony.
Well, actually, my attorney says no it isn't in my case... Because of the following argument:
H*neywell is a corporate entity with known expertise in electronic communication.
H*neywell is on "constructive notice" that they must secure their resources or face the possibility of people "openly and notoriously" using their resources (in this case, wireless network access).
H*neywell remains silent as I and others connect to and use their wireless access point, even though they have the capability to monitor such access, and the ability to lock the electronic "gate" that bars access to this resource. (Locking the gate in this case is equivalent to putting some kind of password protection on the access point.)
H*neywell has, in effect, waived their rights by not voicing objections and putting me and others on notice, and by not securing their resources.
It was [the newspaper's] intention to access the network and they knowingly downloaded files that were sensitive in nature.
Agreed. Intent makes the difference. Confidential information was accessed and stolen, as well.
If you knowingly leave your door unlocked and I willingly open it and walk in, have I committed criminal trespass? According to the law I have... it's called "breaking and entering."
Yes, that's true. I asked my attorney about this, and I learned a few things. First, the "breaking" part of breaking and entering happens when you break the plane of the door frame; the door could be completely wide open, and you're still breaking the law by walking through.
Second, the "breaking and entering" analogy doesn't apply. The laws governing real estate and the laws governing electronic communication are a bit different. My attorney said that a closer real estate analogy to the situation we're discussing would be the following: You own 100 acres of land, and I go and squat on one corner of your property. There are no signs up saying "Do Not Trespass." You see me squatting on one acre of your property but don't do anything for a period of time (months, years). After a time has passed, your silence effectively means that you've waived your rights with respect to the piece of property that I'm squatting on, because I'm "openly and notoriously" utilizing that land. On the other hand, if you take immediate action to notify me, you've asserted your rights, and any further incident where I trespass at that point is a separate crime.
Now, in the case of my dealings with H*neywell, if they put me on notice at any time, and I continued to access their network, then every separate instance where I connected to their network would be a specific felony. But since I was not notified until well after the fact, and because they took no measures to secure the electronic "gate" to their network, H*neywell is clearly at fault in this case.
If I'd taken any data off their internal network, then they'd still be able to nail me for that. (And I would fully expect them to do so!)
In the case of the newspaper accessing the school's network, confidential data was stolen. If the wireless access point was secured in any fashion, then merely breaking that security to gain access would be a crime, yes. But if no measures were taken to secure the access point, then merely obtaining an IP address by connecting to the access point wouldn't be a crime.
Disclaimer: I am not a lawyer, and this is my imperfect understanding of what a lawyer has explained to me. Talk to your lawyer; don't take my word for anything.
How does the fact that you *could* connect to it make it okay to connect to it? Really, why were you surprised you got in trouble?
The same way that a cop walks into a person's house without explicit verbal or written permission, if the cop finds that the person's front door is unlocked and if they have a reason to be at that person's house in the first place.
OK, I may not have had a reason to enter H*neywell's "house," but what they did is tantamount to leaving the barn door open, or leaving their front door unlocked and putting a big neon sign over it that says "this door is unlocked." (My lawyer, incidentally, agrees with me, and not because I paid him to. He helped me with this pro bono.) What I did was stupid, granted, but not technically illegal.
Also keep in mind that, as I stated very clearly, my iBook is configured to automatically connect to any available base station upon waking up, or upon boot. I found out this access point belonged to H*neywell after the (metaphorical)damage was already done. I initially thought that it belonged to the company I am consulting for.
H*neywell might not have been happy, but they have only themselves to blame for running a loose operation at this particular office. I certainly had no way of knowing there was a problem, since I tend to interpret unlocked doors as invitations to entry. If they had put even minimal password protection on their access point, that would have raised a flag saying "Do Not Enter," and I wouldn't have. Simple as that.
When you discover a security venurability ethics dictate that you report it. Otherwise, all that probing is for you cynical interests.
You're right, I should have reported the vulnerability, but I have no personal relationship with H*neywell. My consulting firm does, however. Ethically, I should have reported it anyway, but I wasn't legally obligated to tell them that their barn door was open.
You knew that it was a [H*neywell] network and that access to it was not permitted without permission. In general, you just don't go around taking other peoples stuff and using it without permission.
I knew well after my laptop connected to it that the access point belonged to H*neywell. And I did not know that access to that hot spot was not permitted. How many places run hot spots without security of any kind whatsoever? Of those, how many places do so and demand that you not access their hot spot without permission? Right. Most places that run a WiFi hot spot with no security are coffee shops and neighborhood co-ops.
"Taking people's stuff" implies taking a physical item or information. I did neither. That I used their bandwidth is no different than if I borrowed a phone in their office's reception area. A better analogy to use is that what H*neywell did is tantamount to someone leaving their front door unlocked, and then putting a neon sign above their front door advertising this fact. They might get upset if someone walks into their house from the street, but it's highly questionable whether you could even call it trespassing.
If someone puts a couch on their yard near the road, that doesn't necessarily mean it's free. For all you know they might put it there and sit on it to wait for the bus. If they put a sign on it saying "free" than you can take it.
That's called abandonment. In most municipalities, if you leave something in your yard and it's not secured in any way, especially if it's placed right near a right-of-way (such as a road), it's assumed that the object is either trash or abandoned. Local regulations may dictate whether you or the trash collector has a right to take it, and under what circumstances. In some areas, it isn't free for the taking until it winds up in the trash vehicle. But IANAL, so consult yours before taking any of this as gospel. I already talked to my lawyer, and I'm pretty secure in my position that I violated no laws.
I'm a bit surprised your talking about this. Usually when people strike deals like that, they make you sign a non-dislosure agreemant. Ultimately, if you didn't hack into any of there resources, I don't think they could have had you prosecuted for anything.
I was not forced to sign anything, and I did everything H*neywell asked me to do, so they really have nothing to bitch about. I didn't hack into any of their resources, as they can plainly see by looking at the image of my laptop's hard disk. So no, they had nothing on me to prosecute me with.
The full story is a couple parents up. Basically, they found a blog entry (specifically, my LiveJournal entry where I wrote about being surprised at finding a wide open WiFi access point being run by H*neywell). At least, that's what one person who's "in the know" told me. The official story they fed me during a meeting was that they'd hired a security consulting group, and that they somehow found my "digital fingerprint" as someone who'd accessed their network via WiFi. But that sounded like BS to me, especially when they told me that they could track what I'd done outside their network (i.e., what web sites I went to), but they couldn't track what I did inside H*neywell's network. Yeahhhh, right.
I have no idea if the iBook transmitted my real name or not over the air, though I suspect not. The fact that H*neywell was able not only to find me, but did so through my employer (whose name was also mentioned in my blog) lends credence to the theory that they found out totally by accident, by doing a Google search and turning up my blog entry.
A cynical person might say that the real moral of the story is, if you do something that might be considered illicit, don't talk about it. But I had no idea I was doing anything "wrong." As I said earlier, I assumed that the access point was running in a DMZ, because I assumed nobody would be stupid enough to run an unsecured access point behind a corporate firewall.
Hmmm... according to FCC article 15, this newspaper just openly and admittingly committed a felony. Just getting an IP address constitutes committing this felony, [snip]
I'm not familiar with the laws, but which part is the felony exactly? How can "just" getting the IP address constitute a felony? We don't even know whether the newspaper had to crack encryption to get into this network. Maybe the access point was being run wide open, as another poster suggested.
Certainly, if they had to break in, then it's a felony; on the other hand, if the school ran the access point wide open, then there's more of a gray area.
I have a particular interest in this. You see, I recently got in trouble with H*neywell for using their WiFi without permission. I do consulting work for a small company, and there's a H*neywell office just down the hall from where I work. Someone at that office installed a WiFi access point, apparently contrary to company policy. That access point stayed up for many months, then recently came down, and I never thought anything of it. The access point was being run entirely without security of any kind -- no WEP, no password, nothing.
I was only using this to surf the web and download some software updates/patches to my iBook. I didn't go out looking for this access point, but my iBook is configured to find the nearest access point as soon as it wakes up from sleep (or boots up).
Then about a week after the access point went down, I got a call from my consulting firm. It seems that H*neywell had somehow traced my use of their WiFi access point, and wanted to do something about it. I almost lost my job, but ultimately, a deal was struck whereby I surrendered my laptop to have the hard disk imaged; the laptop was returned to me less than 2 days later, fully intact.
The official story I got was that H*neywell hired an outside firm to check their network security, and they identified the WiFi access point as a security hole; the employee who set it up was fired. Then the security firm traced all who had used the access point, and found my "digital fingerprint."
The unofficial story I got from some other folks in-the-know is that I had posted about my discovery in my LiveJournal, and someone did a Google search and found the entry. Apparently, I forgot to make this a non-public entry. So that's how I was really found out. (That entry has been made friends-only now.) I'm still not 100% sure how Google indexed my journal, since I have my prefs set up to prevent indexing, but not all spiders respect that.
I know H*neywell is a defense contractor, so I had assumed, when I discovered the access point, that it must be some sort of public access point for the convenience of vendors, put in a DMZ on their network. Surely, I thought, they wouldn't be dumb enough to put a wide-open WiFi access point behind their firewall! As it turns out, the access point was behind their firewall, and I could have accessed a whole bunch of material I wasn't supposed to. Scary thought.
I think the real reason I got in trouble was that I embarrassed H*neywell. They could have conceivably taken legal action against me personally, but that would have created a weird situation for them, since it would expose them to government scrutiny. And they might lose some favorable government contracts if that happened. Moral of the story: Always check to see what you're connecting to. That hot-spot might not be safe to connect to after all!
After all, writing special cases in the compile for good benchmark results is a _very_ old trick (Sieve OF Erasthones)...
Not to nit-pick too much, but it's the Sieve of Eratosthenes.:-) Or, "One of them Greek guys."
On a more serious note, I remember studying benchmarks in grad school, and how crappy benchmarks are, and how meaningless. Compiler vendors routinely tweak compilers to recognize blocks of code used in benchmarks and generate highly tweaked code to make those benchmark results look better than they should be. Case in point, there was a Fortran benchmark that was used heavily in comparing hardware, and one Fortran compiler was tweaked by the hardware vendor to super-optimize a block of code so it ran far, far faster than usual. The problem was, the optimization, although valid for all possible inputs, was something only a human with advanced knowledge of calculus and transcendental math could have seen/figured out. And this optimization only worked for the benchmark suite, IIRC.
As we've seen lately with video card vendors, this game is still played even today. I, for one, am glad that Apple did their comparisons using the same compiler (and an Open Source one at that) on both platforms. It's a very fair comparison.
Hmmm... USB 2.0 ships on the MDD Macs? First I've heard of it. Are we talking Hi-Speed (480 Mbps) USB 2.0, or merely "Full Speed" (12 Mbps)? It makes a difference.
Regardless, it would be stupid for Apple to not include USB 2.0 (the Hi-Speed variant) on their desktop systems, since they've been berated repeatedly in the press over this "omission." It would be cheap for them to use the NEC chipset for USB 2.0, or some custom variant, and they'd shut the critics up in one fell swoop.
As for the comment about "Optical and analog audio in and out," I have a couple comments. First, I'm not sure I see what's grammatically bad about this in a bulleted list of features. (Not that Apple cares that much for grammar nazis -- "Think Different," anyone?) Second, optical SP/DIF ports have been shipping on mainstream Intel and AMD motherboards for a while now, and also ship standard on all high-end audio cards for the PC.
So the inclusion of this feature on a high-end desktop Mac isn't at all surprising or confusing. It's merely keeping up with the feature set people are growing to expect from their computers. I'm sure all the musicians out there will appreciate the inclusion of digital audio I/O on the new Macs.
Please don't flame me, because I hate SCO as much as most of you do (if not moreso), but they must be doing something right, at least insofar as their shareholders are concerned, because ever since they announced their lawsuit with IBM, their stock price has gone upward. I consult for a company that does option trading and teaching people how to do risk averse trading, and we're working on software that can be used to view stock trends (among other things).
SCO's stock value has continued to climb, so the general perception among investors is that they have something. Whether they actually have a case or not is irrelevant in a sense, because they've created a perception that they have something, and IBM has deep enough pockets that this lawsuit could be very lucrative for SCO. So it's in SCO's best interests to keep shoveling out the FUD and to not reveal any information publicly, because as long as they do this, their stock price will keep climbing.
Now, if investors start to catch wise and realize that SCO is full of crap, maybe this trend will reverse itself. But as of the market close today, their stock was about 1000% higher in value than it was several months back (before they announced their lawsuit with IBM). Before the lawsuit, they were trading at about $1 a share. Now they're up around $11 a share. Periodic fluctuations aside, I don't see the upward trend stopping. Now, I'm not a stock broker, and I'm not giving anyone advice on what to invest in; I'm merely pointing out what appears to be a correlation.
Personally, I hope people in the investment community realize what we in the software development community already know -- SCO has nothing.
I used to be a BeOS enthusiast, and I know BeOS had a database integrated with the file system. I also know the performance wasn't quite what everyone hoped for, so Be re-engineered the file system. (Dominic Giampolo had his hands full, no doubt.) I vaguely recall that the newer file system wasn't quite as "slick" and "cool" in terms of features, but it had better performance.
I really wonder about Microsoft's motivation for this. Are they trying to make Windows "the media OS" that BeOS promised to be? I somehow think it's more than that.
The article is rather thin on details, which has been noted elsewhere. One line at the very end of the article stands out:
In the end, Win FS will probably emerge as an optional file system beside FAT and NTFS. It's also possible that Win FS will supersede its predecessors, however. That would most likely produce problems for multi-boot systems, since the only way Windows XP, Longhorn and Linux would all be able to access one and the same volume would be through complex methods - if at all.
What this tells me is that Microsoft is looking for a way to promote lock-in on the desktop, by making multi-boot systems difficult or impossible, and by making cross-OS reading of file systems difficult or impossible. But I think there's even more to the story than that.
Orrin Hatch recently proposed that some kind of measure be placed on computers (he doesn't specify how to accomplish this, just the behavior he'd like to see) to warn the user if they download illegal music the first or second time; the third time that the user downloads an illegal music file, the software that oversees this would essentially trash the user's computer. Now, anyone with two brain cells to rub together can see that this presupposes the ability to distinguish between a legally downloaded MP3 and an illegally downloaded MP3 -- virtually impossible, since MP3 has no built-in DRM. Many bands put MP3 files on their web sites. (Even major acts like the Offspring have made songs available in MP3 format prior to releasing albums.) The only practical way to implement Hatch's proposal is to essentially assume that all MP3 files are illicit, or to assume that all MP3 files downloaded from the network are illicit. This would essentially mandate a closed file format for music, forcing DRM down people's throats and effectively ridding the world of MP3, Ogg Vorbis, and so forth.
How is this relevant? Because if you put a stranglehold on the file system for an OS, you suddenly have the ability to police what goes into that file system, and where it comes from. And maybe, just maybe, this push by Microsoft to add more database-like functionality to their file system is a way to make it easier for a "watchdog" application to perform its cop duties.
The sad part is, if any other developer were to create a file system like this for any other OS, I wouldn't even blink. But because Microsoft comes up with this idea, I automatically start spinning conspiracy theories about their motives, and how they could pervert what could be a great idea and use it against the end users.
Check any site stating the specs of present Power Macs and you will find that the Power Macs being sold for some time now to not be using PC133, but the same DDR SDRAM as many of your stated Intel mobos.
But the front-side bus architecture of all current G4 motherboards limits the effective bandwidth of the CPU talking to the RAM on the motherboard. Therefore, most of the bandwidth of the DDR SDRAM goes to waste. You might as well have SDR RAM (PC133) instead of DDR. Apple and Motorola have attempted to work around this by including a faster L3 cache that talks to the G4 over a separate bus from the front-side bus, and Apple puts a big heaping amount of L3 cache on all of their high-end machines.
For certain types of applications, though, even the L3 cache won't save you. And on Apple's dual-CPU machines, both processors contend for the same bandwidth-choked front-side bus, so each processor gets fed even less under load. Bottom line, even with the use of DDR RAM on current G4 motherboards, the front-side bus bottleneck is killing performance and limiting it to about the level you'd get with SDR RAM.
Actually if you care to do a little research, the NT kernel WAS based on "UNIX" and was posix compliant when it was initially created way back when.
More like the NT kernel is VMS based, not UNIX based, although Microsoft freely snagged a lot of BSD code (mostly networking code). The team that developed NT were all ex-DEC people who had deep VMS roots.
Your suggestion that coding for Windows and UNIX is more similar than different is also pretty laughable. I've done both, and they're not that similar. Microsoft included a POSIX compatibility layer, as did most other OS vendors, because it helps coders port things, and it's useful. Even BeOS had a POSIX layer, and BeOS was most assuredly not based on UNIX, Linux, or any other *nix.
The problem is that without browser support this is like having a CD library in the 70s... Useless. And as long as browsers don't handle PNG's properly it's also chicken & egg problem.
The only missing feature is proper alpha channel support in IE under Windows. (Last time I checked, using IE 6 under Windows XP, many PNG images with alpha channel didn't render properly. Also, for some strange reason, while inline PNG images display mostly correctly, directly trying to load a PNG in IE -- usually the result of a direct URL link to an image -- causes some weirdness, because IE doesn't want to load PNG images stand-alone. Go figure.)
However, PNG images with indexed transparency (a feature included in the PNG spec to provide a "cheap" transparency option that made PNG a direct replacement for GIF89a in most applications) should display fine in IE under Windows.
Please note that all other major Windows browsers display PNG correctly. Furthermore, IE for the Mac correctly renders PNG images, even with alpha transparency.
If all you're doing is displaying a simple rectangular image, or an image with "simple" (indexed) transparency, PNG is perfectly usable on all major browsers including IE for Windows. It's just as functional as GIF in those cases, albeit usually with smaller files. This "browser support" argument is getting old, folks...
BMF doesn't appear to be an open, non-proprietary format, and it doesn't have the W3C's blessing. Maybe when the spec is made widely available and the licensing terms for using the file format have been made clear...
What really killed PNG, imnsho, was that the first graphics programs that implemented it simply did not allow users to create indexed PNG files.
Ummm, where did you get your information from? I'm one of the PNG spec co-authors, although my involvement with the project tapered off years ago, and I wrote one of the first commercial implementations of PNG. You may have heard of a company called MasterSoft that used to produce document and graphic conversion utilities. When we were acquired by Frame, and then Frame was acquired by Adobe, our products got released for a while as "Adobe File Utilities by MasterSoft." Quite a mouthfull, but accurate.
My PNG writing code handled indexed (palette based) and truecolor images equally well, and preserved whatever format/color depth was suggested by the original image. As I understand it, my code made its way into several products later on, although it was probably changed.
One of the utilities that came out early on was a small freeware/open source program designed to take GIF files and convert them to PNG. One of the other spec authors cooked that one up, and it worked very well. It created indexed PNG images by default.
While it's true that the PNG spec doesn't exactly demand that you write an indexed color image when the source data is best represented with indexed color, my early survey of PNG-supporting applications seemed to suggest to me that most PNG writing code out there generated good indexed color PNG images. So I'm not sure where this notion came from that the first programs to implement PNG didn't write indexed color. That doesn't jive with my experience.
I have noticed that some applications will generate truecolor PNG images unless you force your application to use indexed color, or downconvert from 24-bit color to indexed color. That's a function of the application software (usually image editing software) not second-guessing the intent of the user. If you've got your application set to do all editing in a 24-bit RGB color space (and some applications will promote loaded images to 24-bit RGB regardless of the pixel format of the original image), don't be surprised when you go to save as PNG and the resulting file contains 24-bit RGB pixels. Downconvert to an indexed color palette before saving. Some application software supports downconversion to indexed color during the save process.
GPS and Galileo are great, and you can't really substitute something else for them. But it seems to me that you might be able to augment them with ground-based transmitters.
This is already being done. There's such a thing as Differential GPS (DGPS), which allows you to take one or more fixed reference transmitters on the ground and integrate their signal(s) with the signals received from GPS satellites; this "integration" is done in the GPS receiver, and provides a way to differentially correct your position data. As long as your fixed ground-based transmitters have a location known to high accuracy, your GPS data can be corrected to whatever accuracy is required for your commercial application.
Before anyone flames me, I'm using the term "integration" very loosely, since IIRC the signal from the ground-based transmitters is fundamentally different from what the GPS satellites provide. And not all commercial GPS engines support differential correction. I was working for a company that was using GPS for vehicle tracking, and one of the hardware vendors we were in discussion with had no clue that differential correction was even a possibility, even though they were using off-the-shelf GPS chips that had the capability. The hardware vendor's firmware programming skills just weren't up to the task, and none of their engineers really understood the technology as well as they should have.
I think you missed the point... that post was so obviously full of factual holes, it had to be written as a humorous piece! Nobody could be that stupid, especially not someone claiming to be the CTO of a Fortune 500 company. (That claim just made the post all the more delicious.)
Come on, a CTO of a Fortune 500 company posting as Anonymous Coward?
My real problem with the current G4e situation, aside from the 167 SDR FSB, is the fact that it's a shared bus topology, which is just ridiculous.
Thank you for clarifying, Hannibal! I agree, the shared bus topology sucks dead donkey d*ck, and I have suspected for a while that it was simply the result of poor engineering at Apple. After all, they touted their new motherboard chipset in the latest G4 towers and Xserves, which supports DDR RAM and other goodies. Why couldn't they simply have designed their Northbridge-equivalent to have a separate data path to each CPU in a dual processor machine?
The only conclusion I've been able to come to is that this is the result of engineering laziness, coupled with a desire to use the same motherboard chipset on single- and dual-CPU configurations. This way, you force both CPUs on a dual processor system to cohabit on a daughterboard and reduce costs by forcing them to share a single bus out to the rest of the system; for a single-CPU setup, one processor gets the entire bus to itself. I believe (though I'm not 100% certain of this) that the low-end single-CPU G4 tower machine that Apple sells shares the same motherboard design with the middle- and high-end models which have dual processors.
No doubt the device will support a 1/8" mini-stereo jack that will let users listen with headphones or earbuds. This way, you can get a decent audio experience, and 3D positional audio is possible.
Well, that's not strictly true. SACD seems to be winning over DVD-Audio in the high-end audio realm, although SACD is jointly developed by Sony and Philips.
The CD itself was jointly developed by Sony and Philips, and is doing just fine.
The MiniDisc never really took off, but it never died out either. Its major competitor, DCC (Digital Compact Cassette), boasted backwards compatibility with standard analog cassette tapes, but it died a horrible death. MiniDisc is still widely sold and supported, and seems to be popular in portable recorders and as a cheaper alternative to portable players of compressed digital audio (e.g., MP3 players) that rely on expensive solid-state media or hard disks. MiniDisc also is popular in home studio applications for multi-track recorders, although in that application it's used to store raw audio data, not ATRAC-compressed music.
I can't see why Sony wanted flippers - I assume they'd milk the market with more expensive dual laser drives for profit or something. I have Goodfellas on a flipper and to be honest, it kills the movie totally.
I have an early release of Seven on DVD, and it's a flipper as well. (In the early days of DVD, dual-layer discs were problematic in the production stage.) I agree, dual-sided discs kill a feature film, if you're forced to flip the disc over mid-way in the film.
Where dual-sided discs make sense are when special features are put on side 2, or when the original contents are episodic in nature. The Spawn animated series is released entirely on dual sided discs, for instance. The TV miniseries based on The Stand is on a disc that is double-sided, and each side is dual-layer. Such discs are rare, because of production problems inherent to that format, but they do exist.
Personally, I'd rather have one double-sided disc that has everything on it instead of 2- and 4-disc sets, where the extra discs are just bonus material. Of course, with double sided discs you lose the ability to put artwork on one side of the disc, but I never cared about that anyway. (There was supposed to be a provision in the standard to allow holographic artwork on one side of a double-sided disc, which the laser could still read through, but I haven't seen any such discs, so I suspect that provision got dropped.)
...although his role as Neo was certainly one of his best. No, I think Keanu's shining moment was his role as Prince Siddhartha in the movie Little Buddha. I heartily recommend that anyone who cares at all about good film see that movie! It still makes me cry, years later.
I have a copy on laserdisc, and I break it out about once a year.
And thankfully, he avoids sounding like a surfer dude. The last thing you need in a film like Little Buddha is to have a religious icon like The Buddha saying "Whoa!" when he attains enlightenment.
...but I took issue with a couple points made by Mr. Lloyd. I have a lengthy discussion of it over on my journal, but basically, I thought his discussion of consciousness and artificial intelligence was a bit flawed. Also, while I agree with his assertion that the machines of the Matrix would probably rely more heavily on cold fusion than human biothermal/bioelectric energy, I doubt that the machines would throw away the biothermal/bioelectric energy generated by the humans plugged into the Matrix. Indeed, that energy would probably be used to offset the cost of keeping things running.
If Lloyd had spent less time hand-waving about quantum mechanics and consciousness (and reading way, way too much into a couple lines spoken by Agent Smith during Morpheus' interrogation), the article would have been far better. As it is, Lloyd came across as a kind of philosophy snob.
Funny, all the other rumor sites I'm reading say that Apple's resisting putting DRM into their music service, because they want customers to be able to burn music they purchase to CDs, or copy to their iPods, or whatever.
All this talk of Apple pushing DRM sounds like FUD to me. Why don't we wait and see what Apple actually offers?
Others have responded to the assertion that one can't distinguish between a 256 kbps MP3 encoded using the LAME psychoacoustic algorithms. So I won't address that here, except to say that on a decent (read: expensive) stereo system, I can distinguish between subtle nuances of source materials. Any material that's been lossy-compressed (MP3, ATRAC on MiniDisc, etc.) is going to sound inferior for certain types of recordings. There's no one perfect psychoacoustic model that compresses all types of music equally well.
No, what I wanted to really respond to was this:
Now on the other hand, lossless compression would be better to download these files, I totally agree with that. MP3 is good for *listening* only. Even a basic filter as a High/Low button or a band equalizer can make diffences audible.
Excuse me? The whole point of MP3 (and other lossy-compressed audio formats) is to reduce storage requirements for the data, and to reduce bandwidth requirements for its transmission over a network or broadcast medium. Your statement runs completely contrary to the spirit of that engineering design goal for MP3 audio. MP3 is obviously inferior to uncompressed (or losslessly compressed) source material for critical listening; where MP3 shines is in streaming applications and applications where storage space is at a premium. Of course you can jack the bitrate up to 256 kbps, but if you're going to do that with MP3, why not use a better codec that's engineered for musical reproduction, instead of using MP3, which was engineered for digital television broadcast and network streaming? ATRAC seems to get some things right that MP3 doesn't, especially at more modest bitrates. I've been hearing good things about AAC as well, although the patent restrictions may hinder its adoption. I mean, seriously, would you rather listen to an uncompressed CD or DVD-A or SACD on your high end home stereo, or an MP3 compressed copy of the original source material? I don't even think there's a contest here! No, the MP3 copies are good for putting ten hours worth of music on a CD-R that you can play on a portable player or in a car's deck. When you're in a car, or flying cross-country on a plane, or stuck in a hotel room somewhere, or visiting family, or when you're camping somewhere -- these are non-critical listening environments, and highly compressed audio is not a problem.
Slashdot Mirror
Well, actually, my attorney says no it isn't in my case... Because of the following argument:
Agreed. Intent makes the difference. Confidential information was accessed and stolen, as well.
Yes, that's true. I asked my attorney about this, and I learned a few things. First, the "breaking" part of breaking and entering happens when you break the plane of the door frame; the door could be completely wide open, and you're still breaking the law by walking through.
Second, the "breaking and entering" analogy doesn't apply. The laws governing real estate and the laws governing electronic communication are a bit different. My attorney said that a closer real estate analogy to the situation we're discussing would be the following: You own 100 acres of land, and I go and squat on one corner of your property. There are no signs up saying "Do Not Trespass." You see me squatting on one acre of your property but don't do anything for a period of time (months, years). After a time has passed, your silence effectively means that you've waived your rights with respect to the piece of property that I'm squatting on, because I'm "openly and notoriously" utilizing that land. On the other hand, if you take immediate action to notify me, you've asserted your rights, and any further incident where I trespass at that point is a separate crime.
Now, in the case of my dealings with H*neywell, if they put me on notice at any time, and I continued to access their network, then every separate instance where I connected to their network would be a specific felony. But since I was not notified until well after the fact, and because they took no measures to secure the electronic "gate" to their network, H*neywell is clearly at fault in this case.
If I'd taken any data off their internal network, then they'd still be able to nail me for that. (And I would fully expect them to do so!)
In the case of the newspaper accessing the school's network, confidential data was stolen. If the wireless access point was secured in any fashion, then merely breaking that security to gain access would be a crime, yes. But if no measures were taken to secure the access point, then merely obtaining an IP address by connecting to the access point wouldn't be a crime.
Disclaimer: I am not a lawyer, and this is my imperfect understanding of what a lawyer has explained to me. Talk to your lawyer; don't take my word for anything.
The same way that a cop walks into a person's house without explicit verbal or written permission, if the cop finds that the person's front door is unlocked and if they have a reason to be at that person's house in the first place.
OK, I may not have had a reason to enter H*neywell's "house," but what they did is tantamount to leaving the barn door open, or leaving their front door unlocked and putting a big neon sign over it that says "this door is unlocked." (My lawyer, incidentally, agrees with me, and not because I paid him to. He helped me with this pro bono.) What I did was stupid, granted, but not technically illegal.
Also keep in mind that, as I stated very clearly, my iBook is configured to automatically connect to any available base station upon waking up, or upon boot. I found out this access point belonged to H*neywell after the (metaphorical)damage was already done. I initially thought that it belonged to the company I am consulting for.
H*neywell might not have been happy, but they have only themselves to blame for running a loose operation at this particular office. I certainly had no way of knowing there was a problem, since I tend to interpret unlocked doors as invitations to entry. If they had put even minimal password protection on their access point, that would have raised a flag saying "Do Not Enter," and I wouldn't have. Simple as that.
You're right, I should have reported the vulnerability, but I have no personal relationship with H*neywell. My consulting firm does, however. Ethically, I should have reported it anyway, but I wasn't legally obligated to tell them that their barn door was open.
I knew well after my laptop connected to it that the access point belonged to H*neywell. And I did not know that access to that hot spot was not permitted. How many places run hot spots without security of any kind whatsoever? Of those, how many places do so and demand that you not access their hot spot without permission? Right. Most places that run a WiFi hot spot with no security are coffee shops and neighborhood co-ops.
"Taking people's stuff" implies taking a physical item or information. I did neither. That I used their bandwidth is no different than if I borrowed a phone in their office's reception area.
A better analogy to use is that what H*neywell did is tantamount to someone leaving their front door unlocked, and then putting a neon sign above their front door advertising this fact. They might get upset if someone walks into their house from the street, but it's highly questionable whether you could even call it trespassing.
That's called abandonment. In most municipalities, if you leave something in your yard and it's not secured in any way, especially if it's placed right near a right-of-way (such as a road), it's assumed that the object is either trash or abandoned. Local regulations may dictate whether you or the trash collector has a right to take it, and under what circumstances. In some areas, it isn't free for the taking until it winds up in the trash vehicle. But IANAL, so consult yours before taking any of this as gospel. I already talked to my lawyer, and I'm pretty secure in my position that I violated no laws.
I was not forced to sign anything, and I did everything H*neywell asked me to do, so they really have nothing to bitch about. I didn't hack into any of their resources, as they can plainly see by looking at the image of my laptop's hard disk. So no, they had nothing on me to prosecute me with.
The full story is a couple parents up. Basically, they found a blog entry (specifically, my LiveJournal entry where I wrote about being surprised at finding a wide open WiFi access point being run by H*neywell). At least, that's what one person who's "in the know" told me. The official story they fed me during a meeting was that they'd hired a security consulting group, and that they somehow found my "digital fingerprint" as someone who'd accessed their network via WiFi. But that sounded like BS to me, especially when they told me that they could track what I'd done outside their network (i.e., what web sites I went to), but they couldn't track what I did inside H*neywell's network. Yeahhhh, right.
I have no idea if the iBook transmitted my real name or not over the air, though I suspect not. The fact that H*neywell was able not only to find me, but did so through my employer (whose name was also mentioned in my blog) lends credence to the theory that they found out totally by accident, by doing a Google search and turning up my blog entry.
A cynical person might say that the real moral of the story is, if you do something that might be considered illicit, don't talk about it. But I had no idea I was doing anything "wrong." As I said earlier, I assumed that the access point was running in a DMZ, because I assumed nobody would be stupid enough to run an unsecured access point behind a corporate firewall.
"Was" being the operative word here.
There's no telling who else had access to their network. They just went after me because I was an easy target, and I gave them a black eye.
I'm not familiar with the laws, but which part is the felony exactly? How can "just" getting the IP address constitute a felony? We don't even know whether the newspaper had to crack encryption to get into this network. Maybe the access point was being run wide open, as another poster suggested.
Certainly, if they had to break in, then it's a felony; on the other hand, if the school ran the access point wide open, then there's more of a gray area.
I have a particular interest in this. You see, I recently got in trouble with H*neywell for using their WiFi without permission. I do consulting work for a small company, and there's a H*neywell office just down the hall from where I work. Someone at that office installed a WiFi access point, apparently contrary to company policy. That access point stayed up for many months, then recently came down, and I never thought anything of it. The access point was being run entirely without security of any kind -- no WEP, no password, nothing.
I was only using this to surf the web and download some software updates/patches to my iBook. I didn't go out looking for this access point, but my iBook is configured to find the nearest access point as soon as it wakes up from sleep (or boots up).
Then about a week after the access point went down, I got a call from my consulting firm. It seems that H*neywell had somehow traced my use of their WiFi access point, and wanted to do something about it. I almost lost my job, but ultimately, a deal was struck whereby I surrendered my laptop to have the hard disk imaged; the laptop was returned to me less than 2 days later, fully intact.
The official story I got was that H*neywell hired an outside firm to check their network security, and they identified the WiFi access point as a security hole; the employee who set it up was fired. Then the security firm traced all who had used the access point, and found my "digital fingerprint."
The unofficial story I got from some other folks in-the-know is that I had posted about my discovery in my LiveJournal, and someone did a Google search and found the entry. Apparently, I forgot to make this a non-public entry. So that's how I was really found out. (That entry has been made friends-only now.) I'm still not 100% sure how Google indexed my journal, since I have my prefs set up to prevent indexing, but not all spiders respect that.
I know H*neywell is a defense contractor, so I had assumed, when I discovered the access point, that it must be some sort of public access point for the convenience of vendors, put in a DMZ on their network. Surely, I thought, they wouldn't be dumb enough to put a wide-open WiFi access point behind their firewall! As it turns out, the access point was behind their firewall, and I could have accessed a whole bunch of material I wasn't supposed to. Scary thought.
I think the real reason I got in trouble was that I embarrassed H*neywell. They could have conceivably taken legal action against me personally, but that would have created a weird situation for them, since it would expose them to government scrutiny. And they might lose some favorable government contracts if that happened. Moral of the story: Always check to see what you're connecting to. That hot-spot might not be safe to connect to after all!
Not to nit-pick too much, but it's the Sieve of Eratosthenes.
On a more serious note, I remember studying benchmarks in grad school, and how crappy benchmarks are, and how meaningless. Compiler vendors routinely tweak compilers to recognize blocks of code used in benchmarks and generate highly tweaked code to make those benchmark results look better than they should be. Case in point, there was a Fortran benchmark that was used heavily in comparing hardware, and one Fortran compiler was tweaked by the hardware vendor to super-optimize a block of code so it ran far, far faster than usual. The problem was, the optimization, although valid for all possible inputs, was something only a human with advanced knowledge of calculus and transcendental math could have seen/figured out. And this optimization only worked for the benchmark suite, IIRC.
As we've seen lately with video card vendors, this game is still played even today. I, for one, am glad that Apple did their comparisons using the same compiler (and an Open Source one at that) on both platforms. It's a very fair comparison.
Hmmm... USB 2.0 ships on the MDD Macs? First I've heard of it. Are we talking Hi-Speed (480 Mbps) USB 2.0, or merely "Full Speed" (12 Mbps)? It makes a difference.
Regardless, it would be stupid for Apple to not include USB 2.0 (the Hi-Speed variant) on their desktop systems, since they've been berated repeatedly in the press over this "omission." It would be cheap for them to use the NEC chipset for USB 2.0, or some custom variant, and they'd shut the critics up in one fell swoop.
As for the comment about "Optical and analog audio in and out," I have a couple comments. First, I'm not sure I see what's grammatically bad about this in a bulleted list of features. (Not that Apple cares that much for grammar nazis -- "Think Different," anyone?) Second, optical SP/DIF ports have been shipping on mainstream Intel and AMD motherboards for a while now, and also ship standard on all high-end audio cards for the PC.
So the inclusion of this feature on a high-end desktop Mac isn't at all surprising or confusing. It's merely keeping up with the feature set people are growing to expect from their computers. I'm sure all the musicians out there will appreciate the inclusion of digital audio I/O on the new Macs.
Please don't flame me, because I hate SCO as much as most of you do (if not moreso), but they must be doing something right, at least insofar as their shareholders are concerned, because ever since they announced their lawsuit with IBM, their stock price has gone upward. I consult for a company that does option trading and teaching people how to do risk averse trading, and we're working on software that can be used to view stock trends (among other things).
SCO's stock value has continued to climb, so the general perception among investors is that they have something. Whether they actually have a case or not is irrelevant in a sense, because they've created a perception that they have something, and IBM has deep enough pockets that this lawsuit could be very lucrative for SCO. So it's in SCO's best interests to keep shoveling out the FUD and to not reveal any information publicly, because as long as they do this, their stock price will keep climbing.
Now, if investors start to catch wise and realize that SCO is full of crap, maybe this trend will reverse itself. But as of the market close today, their stock was about 1000% higher in value than it was several months back (before they announced their lawsuit with IBM). Before the lawsuit, they were trading at about $1 a share. Now they're up around $11 a share. Periodic fluctuations aside, I don't see the upward trend stopping. Now, I'm not a stock broker, and I'm not giving anyone advice on what to invest in; I'm merely pointing out what appears to be a correlation.
Personally, I hope people in the investment community realize what we in the software development community already know -- SCO has nothing.
I really wonder about Microsoft's motivation for this. Are they trying to make Windows "the media OS" that BeOS promised to be? I somehow think it's more than that.
The article is rather thin on details, which has been noted elsewhere. One line at the very end of the article stands out:
What this tells me is that Microsoft is looking for a way to promote lock-in on the desktop, by making multi-boot systems difficult or impossible, and by making cross-OS reading of file systems difficult or impossible. But I think there's even more to the story than that.
Orrin Hatch recently proposed that some kind of measure be placed on computers (he doesn't specify how to accomplish this, just the behavior he'd like to see) to warn the user if they download illegal music the first or second time; the third time that the user downloads an illegal music file, the software that oversees this would essentially trash the user's computer. Now, anyone with two brain cells to rub together can see that this presupposes the ability to distinguish between a legally downloaded MP3 and an illegally downloaded MP3 -- virtually impossible, since MP3 has no built-in DRM. Many bands put MP3 files on their web sites. (Even major acts like the Offspring have made songs available in MP3 format prior to releasing albums.) The only practical way to implement Hatch's proposal is to essentially assume that all MP3 files are illicit, or to assume that all MP3 files downloaded from the network are illicit. This would essentially mandate a closed file format for music, forcing DRM down people's throats and effectively ridding the world of MP3, Ogg Vorbis, and so forth.
How is this relevant? Because if you put a stranglehold on the file system for an OS, you suddenly have the ability to police what goes into that file system, and where it comes from. And maybe, just maybe, this push by Microsoft to add more database-like functionality to their file system is a way to make it easier for a "watchdog" application to perform its cop duties.
The sad part is, if any other developer were to create a file system like this for any other OS, I wouldn't even blink. But because Microsoft comes up with this idea, I automatically start spinning conspiracy theories about their motives, and how they could pervert what could be a great idea and use it against the end users.
But the front-side bus architecture of all current G4 motherboards limits the effective bandwidth of the CPU talking to the RAM on the motherboard. Therefore, most of the bandwidth of the DDR SDRAM goes to waste. You might as well have SDR RAM (PC133) instead of DDR. Apple and Motorola have attempted to work around this by including a faster L3 cache that talks to the G4 over a separate bus from the front-side bus, and Apple puts a big heaping amount of L3 cache on all of their high-end machines.
For certain types of applications, though, even the L3 cache won't save you. And on Apple's dual-CPU machines, both processors contend for the same bandwidth-choked front-side bus, so each processor gets fed even less under load. Bottom line, even with the use of DDR RAM on current G4 motherboards, the front-side bus bottleneck is killing performance and limiting it to about the level you'd get with SDR RAM.
More like the NT kernel is VMS based, not UNIX based, although Microsoft freely snagged a lot of BSD code (mostly networking code). The team that developed NT were all ex-DEC people who had deep VMS roots.
Your suggestion that coding for Windows and UNIX is more similar than different is also pretty laughable. I've done both, and they're not that similar. Microsoft included a POSIX compatibility layer, as did most other OS vendors, because it helps coders port things, and it's useful. Even BeOS had a POSIX layer, and BeOS was most assuredly not based on UNIX, Linux, or any other *nix.
The only missing feature is proper alpha channel support in IE under Windows. (Last time I checked, using IE 6 under Windows XP, many PNG images with alpha channel didn't render properly. Also, for some strange reason, while inline PNG images display mostly correctly, directly trying to load a PNG in IE -- usually the result of a direct URL link to an image -- causes some weirdness, because IE doesn't want to load PNG images stand-alone. Go figure.)
However, PNG images with indexed transparency (a feature included in the PNG spec to provide a "cheap" transparency option that made PNG a direct replacement for GIF89a in most applications) should display fine in IE under Windows.
Please note that all other major Windows browsers display PNG correctly. Furthermore, IE for the Mac correctly renders PNG images, even with alpha transparency.
If all you're doing is displaying a simple rectangular image, or an image with "simple" (indexed) transparency, PNG is perfectly usable on all major browsers including IE for Windows. It's just as functional as GIF in those cases, albeit usually with smaller files. This "browser support" argument is getting old, folks...
BMF doesn't appear to be an open, non-proprietary format, and it doesn't have the W3C's blessing. Maybe when the spec is made widely available and the licensing terms for using the file format have been made clear...
Ummm, where did you get your information from? I'm one of the PNG spec co-authors, although my involvement with the project tapered off years ago, and I wrote one of the first commercial implementations of PNG. You may have heard of a company called MasterSoft that used to produce document and graphic conversion utilities. When we were acquired by Frame, and then Frame was acquired by Adobe, our products got released for a while as "Adobe File Utilities by MasterSoft." Quite a mouthfull, but accurate.
My PNG writing code handled indexed (palette based) and truecolor images equally well, and preserved whatever format/color depth was suggested by the original image. As I understand it, my code made its way into several products later on, although it was probably changed.
One of the utilities that came out early on was a small freeware/open source program designed to take GIF files and convert them to PNG. One of the other spec authors cooked that one up, and it worked very well. It created indexed PNG images by default.
While it's true that the PNG spec doesn't exactly demand that you write an indexed color image when the source data is best represented with indexed color, my early survey of PNG-supporting applications seemed to suggest to me that most PNG writing code out there generated good indexed color PNG images. So I'm not sure where this notion came from that the first programs to implement PNG didn't write indexed color. That doesn't jive with my experience.
I have noticed that some applications will generate truecolor PNG images unless you force your application to use indexed color, or downconvert from 24-bit color to indexed color. That's a function of the application software (usually image editing software) not second-guessing the intent of the user. If you've got your application set to do all editing in a 24-bit RGB color space (and some applications will promote loaded images to 24-bit RGB regardless of the pixel format of the original image), don't be surprised when you go to save as PNG and the resulting file contains 24-bit RGB pixels. Downconvert to an indexed color palette before saving. Some application software supports downconversion to indexed color during the save process.
This is already being done. There's such a thing as Differential GPS (DGPS), which allows you to take one or more fixed reference transmitters on the ground and integrate their signal(s) with the signals received from GPS satellites; this "integration" is done in the GPS receiver, and provides a way to differentially correct your position data. As long as your fixed ground-based transmitters have a location known to high accuracy, your GPS data can be corrected to whatever accuracy is required for your commercial application.
Before anyone flames me, I'm using the term "integration" very loosely, since IIRC the signal from the ground-based transmitters is fundamentally different from what the GPS satellites provide. And not all commercial GPS engines support differential correction. I was working for a company that was using GPS for vehicle tracking, and one of the hardware vendors we were in discussion with had no clue that differential correction was even a possibility, even though they were using off-the-shelf GPS chips that had the capability. The hardware vendor's firmware programming skills just weren't up to the task, and none of their engineers really understood the technology as well as they should have.
I think you missed the point... that post was so obviously full of factual holes, it had to be written as a humorous piece! Nobody could be that stupid, especially not someone claiming to be the CTO of a Fortune 500 company. (That claim just made the post all the more delicious.)
Come on, a CTO of a Fortune 500 company posting as Anonymous Coward?
Thank you for clarifying, Hannibal! I agree, the shared bus topology sucks dead donkey d*ck, and I have suspected for a while that it was simply the result of poor engineering at Apple. After all, they touted their new motherboard chipset in the latest G4 towers and Xserves, which supports DDR RAM and other goodies. Why couldn't they simply have designed their Northbridge-equivalent to have a separate data path to each CPU in a dual processor machine?
The only conclusion I've been able to come to is that this is the result of engineering laziness, coupled with a desire to use the same motherboard chipset on single- and dual-CPU configurations. This way, you force both CPUs on a dual processor system to cohabit on a daughterboard and reduce costs by forcing them to share a single bus out to the rest of the system; for a single-CPU setup, one processor gets the entire bus to itself. I believe (though I'm not 100% certain of this) that the low-end single-CPU G4 tower machine that Apple sells shares the same motherboard design with the middle- and high-end models which have dual processors.
No doubt the device will support a 1/8" mini-stereo jack that will let users listen with headphones or earbuds. This way, you can get a decent audio experience, and 3D positional audio is possible.
Well, that's not strictly true. SACD seems to be winning over DVD-Audio in the high-end audio realm, although SACD is jointly developed by Sony and Philips.
The CD itself was jointly developed by Sony and Philips, and is doing just fine.
The MiniDisc never really took off, but it never died out either. Its major competitor, DCC (Digital Compact Cassette), boasted backwards compatibility with standard analog cassette tapes, but it died a horrible death. MiniDisc is still widely sold and supported, and seems to be popular in portable recorders and as a cheaper alternative to portable players of compressed digital audio (e.g., MP3 players) that rely on expensive solid-state media or hard disks. MiniDisc also is popular in home studio applications for multi-track recorders, although in that application it's used to store raw audio data, not ATRAC-compressed music.
I have an early release of Seven on DVD, and it's a flipper as well. (In the early days of DVD, dual-layer discs were problematic in the production stage.) I agree, dual-sided discs kill a feature film, if you're forced to flip the disc over mid-way in the film.
Where dual-sided discs make sense are when special features are put on side 2, or when the original contents are episodic in nature. The Spawn animated series is released entirely on dual sided discs, for instance. The TV miniseries based on The Stand is on a disc that is double-sided, and each side is dual-layer. Such discs are rare, because of production problems inherent to that format, but they do exist.
Personally, I'd rather have one double-sided disc that has everything on it instead of 2- and 4-disc sets, where the extra discs are just bonus material. Of course, with double sided discs you lose the ability to put artwork on one side of the disc, but I never cared about that anyway. (There was supposed to be a provision in the standard to allow holographic artwork on one side of a double-sided disc, which the laser could still read through, but I haven't seen any such discs, so I suspect that provision got dropped.)
...although his role as Neo was certainly one of his best. No, I think Keanu's shining moment was his role as Prince Siddhartha in the movie Little Buddha. I heartily recommend that anyone who cares at all about good film see that movie! It still makes me cry, years later.
I have a copy on laserdisc, and I break it out about once a year.
And thankfully, he avoids sounding like a surfer dude. The last thing you need in a film like Little Buddha is to have a religious icon like The Buddha saying "Whoa!" when he attains enlightenment.
...but I took issue with a couple points made by Mr. Lloyd. I have a lengthy discussion of it over on my journal, but basically, I thought his discussion of consciousness and artificial intelligence was a bit flawed. Also, while I agree with his assertion that the machines of the Matrix would probably rely more heavily on cold fusion than human biothermal/bioelectric energy, I doubt that the machines would throw away the biothermal/bioelectric energy generated by the humans plugged into the Matrix. Indeed, that energy would probably be used to offset the cost of keeping things running.
If Lloyd had spent less time hand-waving about quantum mechanics and consciousness (and reading way, way too much into a couple lines spoken by Agent Smith during Morpheus' interrogation), the article would have been far better. As it is, Lloyd came across as a kind of philosophy snob.
Funny, all the other rumor sites I'm reading say that Apple's resisting putting DRM into their music service, because they want customers to be able to burn music they purchase to CDs, or copy to their iPods, or whatever.
All this talk of Apple pushing DRM sounds like FUD to me. Why don't we wait and see what Apple actually offers?
No, what I wanted to really respond to was this:
Excuse me? The whole point of MP3 (and other lossy-compressed audio formats) is to reduce storage requirements for the data, and to reduce bandwidth requirements for its transmission over a network or broadcast medium. Your statement runs completely contrary to the spirit of that engineering design goal for MP3 audio. MP3 is obviously inferior to uncompressed (or losslessly compressed) source material for critical listening; where MP3 shines is in streaming applications and applications where storage space is at a premium. Of course you can jack the bitrate up to 256 kbps, but if you're going to do that with MP3, why not use a better codec that's engineered for musical reproduction, instead of using MP3, which was engineered for digital television broadcast and network streaming? ATRAC seems to get some things right that MP3 doesn't, especially at more modest bitrates. I've been hearing good things about AAC as well, although the patent restrictions may hinder its adoption.
I mean, seriously, would you rather listen to an uncompressed CD or DVD-A or SACD on your high end home stereo, or an MP3 compressed copy of the original source material? I don't even think there's a contest here! No, the MP3 copies are good for putting ten hours worth of music on a CD-R that you can play on a portable player or in a car's deck. When you're in a car, or flying cross-country on a plane, or stuck in a hotel room somewhere, or visiting family, or when you're camping somewhere -- these are non-critical listening environments, and highly compressed audio is not a problem.