I thought I'd chime in as a fellow icon-view aficionado. Of course, it all depends on use case. But for sparking some conversation, here's a few cases where I *definitely* prefer icon view over list view:

- A collection of pictures ("pict001.jpg" and "pict1757.jpg" may as well be the same image to me, until I see it, although being in a folder certainly helps narrow it to at least a topic)
- A collection of videos (although a name is probably useful for full-length movies, home movies and youtube downloads are frequently not named, but rather numbered.)
- My home folder (after much trial of all methods, that's what I've found I prefer.)

And some cases where I definitely prefer list view and/or a shell over icon view:
- A collection of source files (Text icons are all but indistinguishable)
- A collection of MP3s (There's nothing to see, and having lots of names, or better yet, the structure of the folders, visible is useful.)

That's really what I do, on a daily basis, and it just adds a little more weight to your point that really, all 3 views (list, icon, tree) have their use case.

As an additional aside, corporations/companies/groups need to be able to patent because many research topics simply can't be handled by individual. Consider medical patents or new methods of silicon fabrication. These are insanely expensive to research (probably for different reasons), such that no individual could reasonably expect to do so. As a result, only companies have enough money, and in that case, no one individual is responsible for the patent.

I graduated from high school this summer (protracted slightly by my concurrent enrollment at a local community college). At least at my high school, a touch-typing course was required. And having been through it, I can say that nobody should be required to take a course where multiple equal possibilities exist.

For me, I learned to touch-type on Qwerty as a child, but I switched to Dvorak when I was about 12. After 4 years of typing exclusively on dvorak keyboards (and typing more in those 4 years than I probably ever typed any time before that), my fingers had significant muscle memory. Today, I can barely touch a Qwerty keyboard. Dvorak was an alternative, and I chose that alternative -- and it works equally well to Qwerty (so, unlike mathematics, where only correct methods give correct answers, touch-typing would have to be less than universally applicable: everybody would learn Qwerty, most likely, even if everybody doesn't use qwerty)

I had to take a typing test and get 30wpm to graduate. I implored them that I could type at about 60wpm on a dvorak keyboard, and in 5 mouse clicks I could switch the keyboard over (and wouldn't even need a different physical keyboard) and pass the test on the first try, without having to take the course. They refused this of me, and while I practiced for a couple of weeks on Qwerty to pass the test, it took me several months to undo the damage to my muscle memory, so that I could type at 60wpm again without wondering whether the comma was at the top-left or bottom-right of the keyboard. To me, at least, this mandatory typing test was a severe impediment to my studies.

I don't pretend I'm not in a minority -- Dvorak is used by probably less than 5% of typists in America. But what if you grew up in Germany, and used a German keyboard? When you came to America in your mid-teens to attend high school, you would be in almost the same situation (and I'm sure a great majority of Germans who type use a German keyboard, not a Qwerty keyboard).

So, although it's certainly amusing to think of layout wars in the near future, it's probably going to be the reality for some students. A reality that doesn't make any sense anyway -- keyboards are something you learn by doing, and these days there are a growing number of other input devices to deal with (voice recognition is available, cell phones, the frogpad, etc.) -- learning to touch-type may not even be useful to someone who plans to use voice recognition for the rest of their career.

This is just a simple matter of accommodating students with differing needs. A mandatory touch-typing class where the student could choose any keyboard layout they desired (or, better yet, choose any input *device* they desired) would be just fine. A mandatory touch-typing class where the student is locked into a technology that is possibly difficult or alien to them, and one which they possibly will never use again (even if it is in a small minority of cases) is just not acceptable. This isn't math; this is touch-typing. Muscle memory for the students who learned a different keyboard first means that trying to teach them a new keyboard layout will, in fact, harm their productivity, not help it.

Just my $0.02 :)

Well, for one thing, self-driving vehicles aren't 100% applicable. They probably won't deal as well with emergency situations or difficult driving conditions, for example. That would require much more research to make a machine superior to a human. At the very least, this means there will need to be a manual operation mode for unusual circumstances. Personally, if the driver only takes over in emergencies or heavy snow, they're not going to have practiced enough beforehand to handle that anyway -- so, automated drivers are bad drivers, and automated cars probably won't be able to completely replace humans for all situations in any near future.

They also remove a significant pleasure for many people. I don't find a lot of pleasure in driving, but a lot of people do -- why do you think luxury cars and nascar wound up so popular? Simple. Driving can be fun. A machine removes some of the "fun factor" from driving down the road.

Finally, there is cost. A cheap, used car costs less than a fancy self-driving car. It's going to be a long time, even *after* self-driving cars are introduced, before they become the majority of cars. Futuristic wishes aside, lots of people already own gas-consuming manually-steered cars, and most won't just go out there to buy a new one. In the mean time, we need laws that keep the roads safe while we're still stuck with manual cars.

In summary: Sure, automatic cars would be nice. They don't (yet) exist, so the goal of safety means implementing laws. That's what lawmakers and policemen can do. While funding automated cars would be nice, it's not going to change the fact that we need safe driving laws.

More than once in this comment list, people have talked about returning books, used book stores, whatnot, and mentioned that Amazon would need some sort of deletion to allow this to happen. You know what? I don't really care whether Amazon can refund my books or not. I can't always return a book to Barnes and Noble, either (time I've owned it and damage caused to it by me are factors).

But the bottom line here is that the ability to return something is based on the physical world and the non-duplicability of items therein. Although I refuse to give up any of the advantages of digital content over physical content by submitting myself to DRM, I *am* willing to give up certain perks of physical items for digital items -- in particular, refunds and returns are untenable in an information society.

There is precedent. When you buy an album from Amazon's MP3 store, you get a 20-second or so sample from which to decide whether you want the album. However, you download the album, and there are no returns, no refunds, no re-downloads even. And, by the way, that is a perfectly acceptable arrangement, because the songs have no DRM -- it's the perfect digital transaction.

A similar situation for books could apply. No DRM, no returns, no refunds. I see no problem with this setup. It worked for music, why not books?

So, two things come to mind here.

First, what you're saying is that we shouldn't be able to take full advantage of all the added benefits of digital content. See, with a physical book, I can read it and lend it out or sell it. I can't keep a backup copy in case it gets dropped in the tub. I can't search it for keywords (easily). I can't expand it to fit an indefinitely large number of notes per page. Ad infinitum -- digital content is simply more flexible than physical content.

You seem to be suggesting that this should be stifled. The content providers should be provided with a guarantee that customers can't exceed the boundaries previously set by physical books. But this is nonsense: the limits on books are there inherently, because the technology (printing) doesn't have the capability to do the other things. When you *have* that capability, then to restrict it is a restriction, not a benefit.

The second thing that comes to mind is that you seem to think DRM can be transparent to honest customers. I must argue against this. We could look at fringe markets if you want (Linux users rarely get the DRM software, and of course it's never open source. What if someone creates a new OS and it isn't as popular as Linux? They're out of luck, right?). But I think that mainstream markets also exemplify the problem. Say you have a system where I can sell an e-book to my neighbor. You revoke my license and provide him a license. Somehow, whether it's through a website or a piece of software, I must perform some additional action to sell my copy above and beyond handing them a CD with the file on it. In that situation, I believe that no DRM, no matter how unobtrusive it is when reading the book, can remain hidden. Compare this to either a physical book or a non-DRM e-book. In either case, the only task I have to sell it is to transfer my copy in any suitable way to the person I'm selling it to.

A related problem is that of fair use. Parody is provided as a fair use. As is classroom use, and others (I'll let you look up fair use, but the basic idea here is that copyright is not an absolute right, but a privilege bestowed by the state to encourage authors and artists to make more books and art). With a tape or non-DRM'd music, I can remix and parody, use it in a classroom, or anything else. Of course, illegal activities are possible, but that is just a fact and nothing can be done about it except to sue someone who does something illegal with it (the RIAA seems to have figured at least this much out). For books, taking notes on the margins in a digital situation probably requires modifying the file. This is perfectly allowed under copyright (for software, you may have problems with the EULA, but even that is of questionable legal value). I believe, from both past experience and some thought experiments, that any DRM system that has any effect at all cannot permit such uses by either logical, legal, or practical necessity.

The burden of proof is on you now -- show me a DRM system, even hypothetical, that could remain hidden to the end user, open source, and still allow the continuation of the first sale doctrine, remixing and parody, and other fair uses. (Because this is exactly what I get without DRM)

If you have to keep them honest, they're not honest.

DRM *might* stop ignorant, potential gamers from grabbing a free copy. It's not because they're honest, it's because they don't know how to get around the DRM (yet).

On the other hand, if you actually have an honest customer (which wouldn't be so hard to find or keep if the DRM and other crap were dropped), you won't need to keep them honest, because they already are.

DRM restricts honest customers and does not deter anybody else who has any competency at all.

This doesn't make it any better.

Microsoft could easily do that - and they could also easily cover such a nasty activity by claiming some criminal investigation or what have you would benefit. "Save the children!" attracts politicians to do all sorts of things that are stupid and unnecessary. Customers would be pissed, rightfully (except, of course, the customers who are too ignorant to realize what's going on).

But Linux isn't as susceptible to that problem. It's completely invulnerable to it if you feel like auditing the code before compiling/installing. I suppose it's possible (albeit unlikely) that a Debian package or RPM package could have such a trojan (being binary software in any event), but I know of no examples of this offhand. And you can still find the source code if you think there's a problem.

This is just another reason not to use proprietary software -- not an excuse for Amazon.

Having been both home-schooled and public-schooled for various parts of my education (I attended high school and elementary school, but not middle school), I can say that homeschooling is as good as the student. The "socializing" argument is easily reversed: for the outcasts (like my brother, who was teased to the point of tears on a daily basis because of his writing disability), or for those who have better things to do (I wanted to study my computer science. Learning the same elementary algebra 3 years in a row at a public elementary school just doesn't help that task along), homeschooling is a reprieve from the "socializing" that is doing a lot more harm than good.

I believe that homeschooling versus public schooling versus any other option that might be available needs to be dealt with on a case-by-case basis. Treating children as if they all learn in the same way, at the same pace, or with their age group just doesn't work. Homeschooling isn't for everyone. Neither are public schools. Both can be equally damaging to someone who isn't suited to the environment. And the "lack of socializing" is becoming less and less of an issue as the internet becomes more prevalent (and, there are plenty of places to go other than a school to interact with your peers. But your peers aren't always those who share your age -- as in my brother's case, or TFA's case, where the age group taunts the kid or is so far behind the kid that there's no comparison).

My $0.02. Probably biased :)

I think it's a mere colloquialism; GPL source code is seen as somehow different from "copyrighted material", and probably rightly so -- although copyright law protects it, when we speak of copyrighted material and copyright violation, the GPL is rarely at the center of that debate. Instead, it's almost always someone who uses copyright to *stop* someone from copying (if this isn't the case, then most arguments about copyright infringement don't really make a lot of sense).

A result is that while it's technically true that being copyrighted doesn't automatically make it illegal to copy, it's easy to simply assume that it is within arguments, leaving GPL, et al to be understood by the reader.

I don't see any particular problem with this as long as it *is* understood that the GPL exists (and I would argue that at least here at /., the majority of people already understand this).

I'm not sure if it's what you're thinking of, but the argument for plausible deniability[1] isn't that you didn't transfer the chunks of data. It's that you didn't know what was in those chunks. Could have been teddy bears or instructions for building a nuclear bomb. Since you don't know what was in the packets -- whether they are stored on your hard drive or transferred on a network -- you shouldn't (theoretically) be punished for the transfer. The physical analogy would be a mailman delivering a letter about a murder to a mailbox, and being blissfully unaware that they just gave away the most important evidence of the crime.

The idea that you seem to have presented above is that if you *did* open the letter and therefore came into contact with that illegal data, you aren't liable because of the means by which you came across that illegal data. That argument seemed fairly common when TPB was on trial, and it is an argument that is in error, for the same reason that keeping high explosives in your basement is illegal even if they got there because a stranger dropped them off one piece at a time.

I hope I'm not misrepresenting your argument; I just wanted to make sure that the right argument goes to the right person, regardless of how fallacious or valid that argument may be :)

[1] http://en.wikipedia.org/wiki/Plausible_deniability

Better yet. Install dvorak, and leave all the keys exactly where they are.

I do that, and the faces you get are pretty funny. Even funnier is when you then go to type after they've completely failed to, and you have no problems at all -- "Works for me! Maybe you should take a typing class..."

I don't personally care what OS people choose to run (unless, of course, it's my job to fix it). However, one thing I *do* care about -- and it's worth caring about for a lot more reasons than just being an enthusiast -- is the *ability* to run open source software.

I mean, for example, if the fritz chip became common practice, so you need to have your operating system cryptographically signed or what have you. Or, hardware vendors that refuse adamantly to release specs for their hardware. I don't even care if they write the driver for me! I'm willing to do that work, if they're willing to tell me how their stuff works.

I want copyright to get out of my way and let me get the work done that I want -- snip and remix for a school project, play my DVDs everywhere, not lock source code away from my tinkering eyes.

So, basically, I agree with you, but I don't want to let the technology industry stomp on our enthusiasm :)

It just means that all of the bugs are still there :).

Kidding aside, having the source code -- so that the libraries can have some wiggle-room without killing off every program -- and just recompiling when it doesn't run any more works to get the program working even after binary compatibility breaks. There's a lot of Linux software today that started out in the late '80s and early '90s (emacs, the shell, X11) -- they still run, obviously, because they still form the underpinnings of the system. But if you grabbed a binary copy of bash from 10 years ago, I have my doubts that it would usefully run on a current Linux system. If, however, you grabbed 10-year-old source code for the same program, you could probably compile it and have it run (less so for GUI programs; but that's because the APIs change more over time than, say, the C library). So, windows deals with compatibility by doing binary compatibility. Linux doesn't do that; but because the source code is usually available, binary compatibility is not such a big issue.

To me, this seems like the single biggest cultural difference between Windows and Linux, which means that proprietary software without source code will never have a strong footing in Linux. Whether that is good or bad depends on whether your business model depends on obscurity and nobody having access to the source code, which is an unfortunate way to run a business anyway. As a hobbyist developer myself, I always want to have access to the source code for the programs I use (and improve). So, except for the nvidia driver, everything on my computer is open source.

I've been pondering this statement since the first comment that mentioned it a ways above.

I think there is an easy resolution to that statistical problem. Instead of *adding* error packets, just use the real errors for your message!

It doesn't seem that far-fetched to see that this would be difficult to detect. The actual errors mean that you should expect the retransmitted packet would be different from the corrupted packet. And frequency analysis wouldn't show an increase in errors for any connection, since only real errors were used.

Of course, a very low error rate would slow transmission significantly.