A good example of a person who hasn't done well out of piracy is Martin Korth, and his Gameboy Advance emulator and debugger, NO$GBA. It's a fantastic bit of software, written entirely in x86 assembly. It comes in a Windows and DOS version, and the DOS version runs on his 66Mhz processor. He says it's the result of 9 years (full time) of programming and tweaking. The kick in the tail is he charges up to US$1750 for a single commercial license! He harbours a bitter resentment towards people who pirate his software because, despite his exorbitant prices, he doesn't get many sales (he relies on the occasional high-profile one), and he definitely isn't rich. I seriously doubt he made any decent sales as a result of this "free publicity".
As I wrote previously, many development tools are much too expensive to justify buying only for hobby/self-training purposes. The market for commercial development of homebrew games using NO$GBA (how many hobbyists could afford to pay $1750 for it?) is tiny... serious development houses will simply buy the official development platform from Nintendo months before the platform's launch instead of waiting 10 years for a third party's homebrew debugger. People who only want to play emulated ROMs can stick with one of the many free emulators and skip the $1750 debugger.
The market for NO$GBA (people who need GBA debugging) is very small regardless of the price... this product is several years late on its "Best Before" date as a serious development tool.
If you look at engineering and other specialties, each software house has a near-monopoly on some parts of the tool chain... in general, there rarely are more than three well-known providers for any one particular link and each link often costs (tens/hundreds of) thousands CDN$. The market for these tools is somewhat small but they are necessary for engineers to carry on with their day-to-day business of building the next CPU, HD-DVD appliance, etc. These cost way too much to be purchased for personal use and the physical processes involved in putting their output to actual use is often equally prohibitive.
On the software side, most platforms now have free development tools... instead of pirating Turbo/Boarland C like ~15 years ago, new would-be developers can now download Visual Studio Express straight off Microsoft's official site. Offering free downloads of VSE to string up new developers is M$' way of binding new talent to Windows and Visual Studio - an indirect acknowledgment that Linux and FOSS free development tools threaten its monopoly.
The tools that these programmers/engineers were created under copyright. Those tools, since they weren't distributed free, probably wouldn't have been created, or at least not created in the same time-frame, or as completely as they were, if it weren't for copyright. It's the cannibalistic nature of art and the nature of piracy. It competes with (and eventually destroys) the very thing it relies upon. Allowing piracy results in a very short term free-for-all, after which point there's no fresh art to build upon.
If Microsoft managed a total anti-piracy Windows/Office lock-down, most people would refuse to pay $300 for Windows and $500 for Office and Microsoft's monopoly could face serious erosion - people tend to take their home computing habits to work so workplaces would start switching to dekstop Linux (or other) some time later. Both items are priced seriously beyond reason regardless of the piracy situation. I personally view pirating business/development tools for private use as free publicity and training: if you want to sell to businesses, mindhsare is one of your most valuable assets and one of the key reasons behind Microsoft's continued success... and much of this software is much too expensive for people to buy for self-training or personal projects unless you happen to be particularly wealthy.
As for your problem of music discovery, that's no problem for most people. If your tastes are "mainstream" enough, you can rely on the radio, or internet radio. You can use services like last.fm to see what people with similar tastes are listening to. You can walk into an average CD store and listen to a CD off the shelf. iTunes will give you 30 second previews of a huge variety of signed and indie artists, arranged in genres. There are plenty of ways to try before you buy, all of which %100 legal. If you need to rely on piracy you aren't really trying.
My music discovery process: I simply download what I happen to come across when I feel curious. I rarely hunt down any specific genre, song, group or whatever. I do not crave music and I rarely bother hunting down even songs I like in any form... so, queuing up in music stores or browsing through Apple's catalog is not something I could be bothered with. When I download stuff, it is mostly only because I can and felt like it - there is not much of an a-priori intention behind it.
Probably not many. Certainly not nearly enough to offset everyone who decided that modern musicianship is too risky as a career. It's also not just the people, but the labels too. They're going to be less inclined to risk capital on potential artists, especially the risky ones.
Maybe not many for the music industry but I had a broader scope in mind... like computer science and engineering. How many of today's top senior programmers and engineers used pirate copies of expensive software during college/university years? This figure is very likely closer to 100% than 0%.
Careers in music have always been high-risk and mostly non-profitable and that's why so much of it starts as a hobby and ends as a hobby. Most of the small bands are desperate for exposure but broadcasting contracts make it nearly impossible for them to get any airplay unless they agree to be raped by labels and studios. Some have started to wise-up and realized they can get inexpensive no-strings-attached exposure by posting their stuff for free on internet exchanges to generate publicity and gain brand recognition for themselves. Sometimes they manage to make money off their trouble, sometimes they do not. Internet publishing does net them a broader audience and bypasses the labels, studios and broadcast licenses that do everything in their power to block independents and rob them of any revenues.
Between piracy and labels, I think labels cause the most harm and mislead artists into blaming everything on piracy.
I never buy music I never heard before. In a 0-music-piracy/free-download world, I could quite possibly end up not buying any music ever again... that would hurt music sales even worse, no?
A law firm can propose a $100M class-action suit but that will do the class' plaintifs no/little good if the judge awards their class only $1M in remedies. Lawyers cannot know in advance and afford to make any guaratees about the case's conclusion... making promises could render them liable for misrepresentation.
Judges ultimately decide the class-action's outcome and that determines what actual compensation the class can possibly hope to get, class-action lawyers only need to do business as usual: do their best to convince judges to award them as much as they can.
Since $1M is little more than a rounding error to Verizon, it might not be enough to make them fix their definition of "Unlimited" or advertise the limits of their "Unlimitedness".
People should take any claims of something being "Unlimited" with grains of salt: they always have hidden limits which the service providers hope will not be hit often enough to affect service quality and operating costs... but when heavy users start pounding a little too hard, operators are swift to point out the often undisclosed or little known limits of the unlimited.
What is the efficiency of the most efficient fuel-to-mechanical energy conversion currently known to man? State-of-the-art internal combustion is somewhere in the 30-40% range, electric fuel cell may be around 50% with potential for 80% in the not too distant future, what else is there? Even at 90%, there would still be plenty of waste heat to keep hydrogen (or nearly any other fuel) at whatever desired temperature with minimal tank insulation.
Since hydrogen boils at -252C, it would be impossible to pump it out of a rigid tank at -270C (3K) where it is a nearly incompressible liquid. Even at -250C, the H2 vapor pressure would be really low and make pumping difficult. Additionally, I presume trying to burn -250C hydrogen would rob off a fair amount of combustion energy. For efficiency and simplicity, they probably need to regulate the tank pressure by regulating tank temperature: no tank or fuel injection pump needed this way. Waste heat can be reduced and efficiency slightly increased by pre-heating ("interheater") the air/H2 before aspiration/injection - this would cost a modest drop in peak power.
Experimenting on someone else's dime certainly is fun.
I think the point was more like: states can pass all the environment quality improvement laws they can dream of, they will still be adversely affected by neighboring states that have lesser (if any) standards. Since a state cannot force environmental standards on its neighbors, the next best thing they can do is petition the feds for improved national standards to nudge dissident states one step closer to the greener states' standards.
You can ban atmospheric lead all you want in your state, you will never reach your target of 0ppb if anyone within 500km (interstate) uses leaded gasoline or incinerates leaded electronics.
CO2 might not be particularly dangerous but if one state makes efforts to reduce its carbon footprint, I think it would have a legitimate reason to be somewhat cheezed off that its neighbors are increasing theirs. High environmental standards in one state are impossible to achieve if neighbors have ridiculously low standards since they share the same airspace and rivers... sounds interstate enough to me.
Seeding one's favourite Linux distros and other legit stuff can easily account for 150GB/month in upload. On the download side, I imagine someone who bought nearly all Steam-distributed titles and re-downloaded them all over the course of one billing month would get most of the way to 50GB or possibly beyond. Most of the downloading I do is anime fansubs... with a few Linux distros and random other stuff thrown it, I still download less than 40GB/month the vast majority of the time.
Getting to ~50GB/month on 100% legit stuff is not too difficult: internet radio + VoIP + software updates (with a few PCs) can already account for as much as 20GB/month in "baseline" traffic... add movie/game previews, software demos, YouTube and other similar stuff for another easy 20GB/month, we're already in the area of 50GB/month for a somewhat nerdy entertainment freak. By the time downloading reaches 100GB, I start to seriously wonder if people ever find time to use what they downloaded.
If I really wanted to have a glimpse at everything that could possibly go wrong with air-travel, I'd watch Discovery Channel's MAYDAY.
Planes crash or encounter hazards for all sorts of stupid and often wrong reasons... maintenance crew forgetting to return the cabin pressure controller switch to auto, maintenance engineer using matched-by-eye replacement screws, suicidal commander, captain inviting his kids in the cockpit leading to autopilot going from on to partial, etc. One of the episodes is about a captain deciding to take off in near-0-visibility without takeoff authorization and without confirming that the other plane taxiing up the runway cleared it... one crash, two planes, ~550 dead, ~80 survivors. In another episode, a plane crashed on a mountain slope because the landing beacon was 3km ahead of the runway on the mountain top and the pilot presumed the beacon was on the runway instead of checking his six-months-expired charts and following the prescribed descent profile.
Planes are pretty damn safe when properly maintained and operated by competent staff, some pretty unbelievable testing goes into making sure these things are safe to fly. Very few of the worst crashes are caused by factors other than flight and maintenance crew errors.
The problem with releasing new generations of a gaming platform is that it spits your existing installed base. It forces users to decide whether to upgrade (and many won't if the price is significant, as it is with high end graphics/processor upgrades) and it forces developers to decide whether they want to target their games to actually benefit from the advances of the new generation, or to aim for the lowest common denominator.
In past generations, a new generation meant completely new hardware across the board. In the current generation, we have over a dozen variations of PS3 and X360 to confuse the public and developers... add a feature here, remove another there, swap parts all over the place. Some of these changes are potentially disruptive, others are trivial and we have a handful of stupid initial omissions (like no HDMI in some models for the sake of model differentiation and associated price gouging) thrown in for good measure.
On the "using new features" side, Microsoft's X360 guidelines specifically state that the X360's HD-DVD capability shall NOT be used for gaming content since the HD-DVD drive was never part of the X360 base specs. Releasing an X720 with upgraded hardware and HD-DVD would be an interesting option for 2009 and developers would have the choice of releasing for X360 with enhancements for X720 on DVD or doing separate releases.
HD-DVD options Xbox 360 players sold around 150k units Blu-Ray players bundled with the PS3 have sold 7 million units
The HD war is somewhat off-topic as far as console revisions are concerned. In any case, most PS3 owners bought a PS3 to play games and a surprising majority of them did not initially know that the PS3 could play BD movies. I personally do not really care about the HD format war... at least not until titles encoded in 1080p become common fare and the necessary playback devices become affordable since the battle will not be decided before then. For the record, I do own a working Betamax... the one my parents bought back when both VHS and Beta cost over $1000.
On the gaming side, I have no interest in the X360 and the PS3 games I am interested in won't be out until mid-2008 or later... so I'll be one of those sticking to their PS2 until something like 2009.
Since the 360 came out a year before the PS3/Wii, Microsoft being the first again with its next next-gen console seems like a given. They'd only need to bump the CPU/GPU processing power to provide proper support for 1080p60 games (instead of upscaled 720p), increase RAMs, include the HD-DVD, a user-replaceable 2.5" SATA HDD and HDMI in the base configuration to have something almost directly backwards-compatible that may actually be worth calling the XBox 720 in 2-3 years.
With that said, the way Sony and Microsoft have been introducing new variations of their platforms and discontinuing others at a rate of three or more each year is unprecedented and quite annoying. One new model per year would be ok but replacing most of the lineup over the first year makes it obvious they did not know what they were initially doing.
How many creative careers do you think start with people experimenting with stuff they could not afford or be otherwise unable to get thanks to piracy?
Piracy might hurt creativity in some ways but also helps it in others. Bad material is more likely to get substantial/critical damage from it.
There shouldn't be any need for medical experimentation: take GMO samples, run them through chemical analysis, compare to non-GMO samples, done. If the GMO samples do not present higher concentrations of harmful chemicals than non-GMO ones, the GMO stuff will be every bit as safe as the non-GMO stuff.
Since GMO crops are modified to make them more resilient to diseases, pests, drought, floods, etc., they require fewer environmentally damageable chemicals, less irrigation and less maintenance in general.
The tough part is avoiding cross-pollination since the GMO crops are very likely to displace their originals once out in the wild. Adding a chemically-suppressed kill-switch gene is one possibility but then, this chemical cocktail would indeed need to go through over a decade of medical research to determine its long-term effects on animals... so kill-switch-suppressing chemicals may be interesting but ultimately impractical in principle if they are to be used responsibly.
We're talking investments in the billion$ here... those 'few' would have to be well-established companies that probably already own near-monopoly power on their respective turfs. These players already have the capital necessary to encroach on other players' territory if they wanted to... but most likely won't since they will have low subscriber density in new territories and their competitors may retaliate by encroaching on their territories and thinning the subscriber base there. Having lower subscriber density means less revenue on inflated network maintenance and operating expenses so having competitors for local loops is ultimately bad for both subscribers and network operators... sharing is more cost-effective for everyone and the telcos are too greedy to admit it.
If you drop those middle-weight players from the plan, all that is left is ADSL/phone resellers who lease the telcos' loops. Most of these companies have only few thousand customers with revenues in the low millions. Even if you put 100 of them together, they probably still would have a very hard time financing the deployment of their own local loops or fiber drops and properly maintain them. Also, there would inevitably be power struggle and mergers among them for dominion over at least parts of that network. Forming a non-profit co-op for this task might have a slim chance of working.
Forcing telcos to lease loops at reasonable rates to other service providers is much better, deregulating this was a horrible and unforgivable (but thankfully reversible) mistake. We already have cable, phone/ADSL and FiOS/FTTN/FTTP/FTTH/etc., we do not need multiple sets of local loops and more of everything else: unnecessary infrastructure duplication only increases deployment, maintenance and operating costs.
I agree with this sentiment completely. I think our disagreement was a matter of mere semantics. Not really, I simply have a tendency to be more verbose than necessary and branch off in variously loosely related topics.
Even barring the "obsolete" portions of legacy support in modern x86-64 platforms, they are still built around that original conception of what Intel thought a microprocessor should be 30 years ago. RISC have become more CISC-like, CISC has become more RISC-like to the point that there is no more clear differences between the two.
The x86(_64) ISA is ok feature-wise: it has everything any other general-purpose ISA has. The problem with it is the highly non-uniform instruction format: ~30 years ago, bits were really expensive so Intel designed the x86 ISA to be as compact as they could practically make it... and back then, they had to design the necessary circuits by hand, drawing them transistor by transistors every step of the way to create their lithography masks, makes it hard to blame them for not implementing the most optimal solution conceivable at the time.
If the x86 processor disappeared tomorrow and was replaced with something else engineered to be efficient from the ground up, I think we would all benefit tremendously. It would not need to be completely new... most of the legacy peripheral hardware have modern equivalents and the instruction set is mostly ok. As I said in my other post, a re-engineered x86-64 ISA with all the tack-ons cleanly integrated together and improved ISA orthogonality/regularity would go a long way towards improving the situation. And since that new ISA would be very x86-64-like, legacy emulation should be possible with a moderate performance penalty.
The cost of deploying your own FTTN equipment and stringing your own fibre/wire is pretty prohibitive. Unless you can string up a large number of people in densely populated areas in one go, it will take years to recoup the expenses and turn up a decent return on investment. Only the biggest players can afford to foot the bills for large-scale network (re-)builds.
I said:
EFI simply has fewer kludges and ties to legacy x86 hardware. You said:
Actually, there's quite a bit of legacy crap in your average bios. Obsolete BIOS services/interrupts sound like kludges to me... and I also did call them legacy-BIOS and EFI-BIOS so I obviously implied that the legacy-BIOS had more legacy stuff in it by definition.
Getting rid of the legacy stuff in the BIOS is one thing... but there is a whole lot more of more complex legacy stuff (logic) embedded in x86 chipsets and CPUs. Dropping hardware support for legacy stuff/kludges (remember "dos=high,umb" and what made it possible?) from the pre-MMX era could considerably simplify chipset and CPU design. The desktop market could use a re-engineered, more orthogonal x86-inspired ISA on a clean platform: it would simplify things across the board (R&D, testing, instruction decoding, compilers, etc.), reduce die sizes, reduce pipeline length, improve frequency scaling, improve IPC, improve performance-per-watt, etc. Intel failed miserably across the line with its first attempt at going clean-slate but the potential benefits are still there.
Had Intel's Itanium plans worked out as initially planned, we would all be running on IA64 desktop hardware by now... kinda sad but seeing how IPF has turned out so far, we are better off this way for now.
What does a BIOS do? It does POST, lets the user customize some low-level system settings and puts the system in a known state before loading the OS's (boot-)loader.
An x86-style legacy BIOS does the same fundamental things as an x86-style EFI BIOS, the only major differences being the BIOS APIs, how the boot process is structured and the fact that EFI is not backwards-compatible on its own. Other than that, a BIOS, by any other name, is still a BIOS. EFI simply has fewer kludges and ties to legacy x86 hardware.
BTW, a few weeks ago, I read an article about some MoBo manufacturers considering adding 512MB-2GB of flash memory to boot an embedded Linux desktop from the BIOS for disk-less web-browsing and other stuff... a BIOS with embedded Linux does not seem that far-fetched, we only need 1GB firmware hubs to plug into Intel's chipsets and hope we will not need to flash our 1GB BIOS too often.
It will clear most stuff and you will see error messages fly by... redirect output to a file for later examination if desired.
I use the good old 'del' whenever I know I will be deleting something like 20k files and do not wish to waste time waiting for windows to prepare for that operation... why the heck does Windows need to scan directories to be deleted before deleting them is beyond me, just delete them and be done with it. Same thing for copying, Windows wastes time scanning the source directory for no apparent reason since it won't tell you you have insufficient disk space to complete the operation until the target drive runs out of disk space... or any other errors for that matter, until it runs into them while carrying out the actual operation.
Linux has quirks, so does Windows. Linux has the excuse of being an relatively immature desktop OS but on the Windows side, it can only be written off as the result of half-ass design decisions.
Hand-writing can also cause RSI... as a kid, I often had to take pauses while writing essays because my hands/wrists started hurting, I would not be too surprised if many people's RSI problems started as early as grade-school. For me, writing became considerably uncomfortable during high-school - that was before I started using computers on a regular basis. Now, my hands' fine motor control and RSI discomfort is so bad that writing legibly is very much like torture. On the other hand, my carpal tunnels have yet to cause me any substantial problems - the RSI pain usually forces me to take extended breaks long before the wrist numbness and "pulsing" from carpal issues I have felt a few times before kick in.
I tried to put my laptop to sleep with Ubuntu x86-64 and my laptop crashed during wake-up. On the other hand, sleep worked fine with both XP 32bits and Vista Ultimate 64bits.... I installed Vista last Tuesday and am pretty glad I got it for only $50 since it looks like I won't be using it much (or at all) in the near future. My laptop is back on XP (too slow and too little RAM to make Vista run sufficiently smoothly for my taste) and I probably won't be booting my dual-boot desktop to Vista very often until SP1 since it appears to become unstable during heavy IO and often/randomly fails to enumerate my external HDDs, quite annoying given that 80% of my storage is near-line.
BTW, installing software is far easier on Linux than Windows. I tried installing VLC under Fedora 7... it took me nearly two days and it did not work. Of course, doing the same thing with Ubuntu takes about five minutes.
What annoys me about Linux is that each distro has its own way of doing things. Some desktop distros just do not get it, others only sort-of get it and some like Ubuntu mostly get it.
For my two primary PCs, I use XP and Vista but for my other PCs, I use Fedora 6/7 (server/development) and Ubuntu. While Linux may excel at various stuff, much of the software I use beyond web browsing, eMail and audio/video playback is either Windows-only or a PITA to setup under Linux.
AGP is basically a souped-up single-drop PCI slot. AGP is fundamentally capable of doing full-speed in any mix of Read-Write as long as the target device is capable of sustaining the operation. The problem is that most GPUs have been designed as output-only devices and their host controller interfaces, be it PCI, AGP or PCIe, have been designed for fast writes to video memory but only implemented a low-cost, small-footprint, low-performance readback path since it is superfluous for normal operation. If you benchmark the readback throughput of early PCIe graphics card - those before the wholesale switch from GPU to GPGPU - probably all things up to Radeon X8xx and GF6 or possibly even X19xx and GF7, you will find that most of them also have relatively slow readback speeds.
Newer GPUs have been designed with GPGPU in mind which has a more symmetric traffic requirement: being able to do very fast computations on-card is pointless when it is impossible to get the results back in a timely manner. Take one of those GPGPUs, slap an AGP bridge onto it, put it in an AGP slot and you will get decent throughput in both directions.
Slow VRAM readback was a chip design decision, not a bus issue and not a bug. While the VRAM path is full-speed between the GPU core and RAM, the path between RAM controller and host controller is asymmetric in nearly all early PCIe GPUs as well of the vast majority of AGP and PCI ones. If you want a GPU with symmetric host-RAM RW capability, all Matrox GPUs, be it on PCI, AGP or PCIe have it IIRC - as should today's X2xxx and GF8.
Windows is primarily a desktop OS and the desktop marketplace is the one that suffers most from Microsoft's monopoly status. In the server space, Windows has a fair bit of competition since it is the new kid on the block in this traditionally *nix market.
So, the competition I had in mind was all non-MS desktop-oriented OSes... and right now, that means mostly GNU/Linux distros like Ubuntu, Gentoo and countless others.
So far, I have played only with Fedora and Ubuntu. Fedora is fine for software development but it thoroughly sucks at entertainment - I never managed to get something as simple as MP3 playback to work and I wasted at least two days on this alone with more days wasted trying to install VLC only to see it crash after loading. Ubuntu does much better as a desktop distro and I was very much pleased with only having to click through a few dialogs to install all the codecs/filters needed to play pretty much any media file I tried. However, both still have way too many quirks to be suitable for an average user without veteran assistance to smooth things out (like install GPU drivers and making them load/work properly) after initial setup.
Things are SLOWLY improving but IP minefield planted by various factions (M$ is only one of them) will ensure that the parking brakes holding FOSS-OS evolution and adoption (to a lesser extent) back will not come off any time soon.
Until Microsoft discloses the relevant patents, it is all FUD to me - just like the SCO case.
Microsoft is a convicted monopolist in multiple countries and it is seeking protection cash from its competitors to diversify its revenue streams. Since MS makes more than enough money to not depend on patent royalties, RedHat&all should seek a permanent injunction barring MS from collecting OS-related patent royalties from its OS competitors until MS' market share drops below something like 70% - since this is unlikely to happen within the next 10+ years, most core OS and GUI technology patents (the GUI is a mandatory component of a core Windows install after all) will have expired long before then.
Slashdot Mirror
As I wrote previously, many development tools are much too expensive to justify buying only for hobby/self-training purposes. The market for commercial development of homebrew games using NO$GBA (how many hobbyists could afford to pay $1750 for it?) is tiny... serious development houses will simply buy the official development platform from Nintendo months before the platform's launch instead of waiting 10 years for a third party's homebrew debugger. People who only want to play emulated ROMs can stick with one of the many free emulators and skip the $1750 debugger.
The market for NO$GBA (people who need GBA debugging) is very small regardless of the price... this product is several years late on its "Best Before" date as a serious development tool.
If you look at engineering and other specialties, each software house has a near-monopoly on some parts of the tool chain... in general, there rarely are more than three well-known providers for any one particular link and each link often costs (tens/hundreds of) thousands CDN$. The market for these tools is somewhat small but they are necessary for engineers to carry on with their day-to-day business of building the next CPU, HD-DVD appliance, etc. These cost way too much to be purchased for personal use and the physical processes involved in putting their output to actual use is often equally prohibitive.
On the software side, most platforms now have free development tools... instead of pirating Turbo/Boarland C like ~15 years ago, new would-be developers can now download Visual Studio Express straight off Microsoft's official site. Offering free downloads of VSE to string up new developers is M$' way of binding new talent to Windows and Visual Studio - an indirect acknowledgment that Linux and FOSS free development tools threaten its monopoly.
If Microsoft managed a total anti-piracy Windows/Office lock-down, most people would refuse to pay $300 for Windows and $500 for Office and Microsoft's monopoly could face serious erosion - people tend to take their home computing habits to work so workplaces would start switching to dekstop Linux (or other) some time later. Both items are priced seriously beyond reason regardless of the piracy situation. I personally view pirating business/development tools for private use as free publicity and training: if you want to sell to businesses, mindhsare is one of your most valuable assets and one of the key reasons behind Microsoft's continued success... and much of this software is much too expensive for people to buy for self-training or personal projects unless you happen to be particularly wealthy.
My music discovery process: I simply download what I happen to come across when I feel curious. I rarely hunt down any specific genre, song, group or whatever. I do not crave music and I rarely bother hunting down even songs I like in any form... so, queuing up in music stores or browsing through Apple's catalog is not something I could be bothered with. When I download stuff, it is mostly only because I can and felt like it - there is not much of an a-priori intention behind it.
Maybe not many for the music industry but I had a broader scope in mind... like computer science and engineering. How many of today's top senior programmers and engineers used pirate copies of expensive software during college/university years? This figure is very likely closer to 100% than 0%.
Careers in music have always been high-risk and mostly non-profitable and that's why so much of it starts as a hobby and ends as a hobby. Most of the small bands are desperate for exposure but broadcasting contracts make it nearly impossible for them to get any airplay unless they agree to be raped by labels and studios. Some have started to wise-up and realized they can get inexpensive no-strings-attached exposure by posting their stuff for free on internet exchanges to generate publicity and gain brand recognition for themselves. Sometimes they manage to make money off their trouble, sometimes they do not. Internet publishing does net them a broader audience and bypasses the labels, studios and broadcast licenses that do everything in their power to block independents and rob them of any revenues.
Between piracy and labels, I think labels cause the most harm and mislead artists into blaming everything on piracy.
I never buy music I never heard before. In a 0-music-piracy/free-download world, I could quite possibly end up not buying any music ever again... that would hurt music sales even worse, no?
A law firm can propose a $100M class-action suit but that will do the class' plaintifs no/little good if the judge awards their class only $1M in remedies. Lawyers cannot know in advance and afford to make any guaratees about the case's conclusion... making promises could render them liable for misrepresentation.
Judges ultimately decide the class-action's outcome and that determines what actual compensation the class can possibly hope to get, class-action lawyers only need to do business as usual: do their best to convince judges to award them as much as they can.
Since $1M is little more than a rounding error to Verizon, it might not be enough to make them fix their definition of "Unlimited" or advertise the limits of their "Unlimitedness".
People should take any claims of something being "Unlimited" with grains of salt: they always have hidden limits which the service providers hope will not be hit often enough to affect service quality and operating costs... but when heavy users start pounding a little too hard, operators are swift to point out the often undisclosed or little known limits of the unlimited.
What is the efficiency of the most efficient fuel-to-mechanical energy conversion currently known to man? State-of-the-art internal combustion is somewhere in the 30-40% range, electric fuel cell may be around 50% with potential for 80% in the not too distant future, what else is there? Even at 90%, there would still be plenty of waste heat to keep hydrogen (or nearly any other fuel) at whatever desired temperature with minimal tank insulation.
Since hydrogen boils at -252C, it would be impossible to pump it out of a rigid tank at -270C (3K) where it is a nearly incompressible liquid. Even at -250C, the H2 vapor pressure would be really low and make pumping difficult. Additionally, I presume trying to burn -250C hydrogen would rob off a fair amount of combustion energy. For efficiency and simplicity, they probably need to regulate the tank pressure by regulating tank temperature: no tank or fuel injection pump needed this way. Waste heat can be reduced and efficiency slightly increased by pre-heating ("interheater") the air/H2 before aspiration/injection - this would cost a modest drop in peak power.
Experimenting on someone else's dime certainly is fun.
I think the point was more like: states can pass all the environment quality improvement laws they can dream of, they will still be adversely affected by neighboring states that have lesser (if any) standards. Since a state cannot force environmental standards on its neighbors, the next best thing they can do is petition the feds for improved national standards to nudge dissident states one step closer to the greener states' standards.
You can ban atmospheric lead all you want in your state, you will never reach your target of 0ppb if anyone within 500km (interstate) uses leaded gasoline or incinerates leaded electronics.
CO2 might not be particularly dangerous but if one state makes efforts to reduce its carbon footprint, I think it would have a legitimate reason to be somewhat cheezed off that its neighbors are increasing theirs. High environmental standards in one state are impossible to achieve if neighbors have ridiculously low standards since they share the same airspace and rivers... sounds interstate enough to me.
Seeding one's favourite Linux distros and other legit stuff can easily account for 150GB/month in upload. On the download side, I imagine someone who bought nearly all Steam-distributed titles and re-downloaded them all over the course of one billing month would get most of the way to 50GB or possibly beyond. Most of the downloading I do is anime fansubs... with a few Linux distros and random other stuff thrown it, I still download less than 40GB/month the vast majority of the time.
Getting to ~50GB/month on 100% legit stuff is not too difficult: internet radio + VoIP + software updates (with a few PCs) can already account for as much as 20GB/month in "baseline" traffic... add movie/game previews, software demos, YouTube and other similar stuff for another easy 20GB/month, we're already in the area of 50GB/month for a somewhat nerdy entertainment freak. By the time downloading reaches 100GB, I start to seriously wonder if people ever find time to use what they downloaded.
If I really wanted to have a glimpse at everything that could possibly go wrong with air-travel, I'd watch Discovery Channel's MAYDAY.
Planes crash or encounter hazards for all sorts of stupid and often wrong reasons... maintenance crew forgetting to return the cabin pressure controller switch to auto, maintenance engineer using matched-by-eye replacement screws, suicidal commander, captain inviting his kids in the cockpit leading to autopilot going from on to partial, etc. One of the episodes is about a captain deciding to take off in near-0-visibility without takeoff authorization and without confirming that the other plane taxiing up the runway cleared it... one crash, two planes, ~550 dead, ~80 survivors. In another episode, a plane crashed on a mountain slope because the landing beacon was 3km ahead of the runway on the mountain top and the pilot presumed the beacon was on the runway instead of checking his six-months-expired charts and following the prescribed descent profile.
Planes are pretty damn safe when properly maintained and operated by competent staff, some pretty unbelievable testing goes into making sure these things are safe to fly. Very few of the worst crashes are caused by factors other than flight and maintenance crew errors.
In past generations, a new generation meant completely new hardware across the board. In the current generation, we have over a dozen variations of PS3 and X360 to confuse the public and developers... add a feature here, remove another there, swap parts all over the place. Some of these changes are potentially disruptive, others are trivial and we have a handful of stupid initial omissions (like no HDMI in some models for the sake of model differentiation and associated price gouging) thrown in for good measure.
On the "using new features" side, Microsoft's X360 guidelines specifically state that the X360's HD-DVD capability shall NOT be used for gaming content since the HD-DVD drive was never part of the X360 base specs. Releasing an X720 with upgraded hardware and HD-DVD would be an interesting option for 2009 and developers would have the choice of releasing for X360 with enhancements for X720 on DVD or doing separate releases.
The HD war is somewhat off-topic as far as console revisions are concerned. In any case, most PS3 owners bought a PS3 to play games and a surprising majority of them did not initially know that the PS3 could play BD movies. I personally do not really care about the HD format war... at least not until titles encoded in 1080p become common fare and the necessary playback devices become affordable since the battle will not be decided before then. For the record, I do own a working Betamax... the one my parents bought back when both VHS and Beta cost over $1000.
On the gaming side, I have no interest in the X360 and the PS3 games I am interested in won't be out until mid-2008 or later... so I'll be one of those sticking to their PS2 until something like 2009.
Since the 360 came out a year before the PS3/Wii, Microsoft being the first again with its next next-gen console seems like a given. They'd only need to bump the CPU/GPU processing power to provide proper support for 1080p60 games (instead of upscaled 720p), increase RAMs, include the HD-DVD, a user-replaceable 2.5" SATA HDD and HDMI in the base configuration to have something almost directly backwards-compatible that may actually be worth calling the XBox 720 in 2-3 years.
With that said, the way Sony and Microsoft have been introducing new variations of their platforms and discontinuing others at a rate of three or more each year is unprecedented and quite annoying. One new model per year would be ok but replacing most of the lineup over the first year makes it obvious they did not know what they were initially doing.
How many creative careers do you think start with people experimenting with stuff they could not afford or be otherwise unable to get thanks to piracy?
Piracy might hurt creativity in some ways but also helps it in others. Bad material is more likely to get substantial/critical damage from it.
There shouldn't be any need for medical experimentation: take GMO samples, run them through chemical analysis, compare to non-GMO samples, done. If the GMO samples do not present higher concentrations of harmful chemicals than non-GMO ones, the GMO stuff will be every bit as safe as the non-GMO stuff.
Since GMO crops are modified to make them more resilient to diseases, pests, drought, floods, etc., they require fewer environmentally damageable chemicals, less irrigation and less maintenance in general.
The tough part is avoiding cross-pollination since the GMO crops are very likely to displace their originals once out in the wild. Adding a chemically-suppressed kill-switch gene is one possibility but then, this chemical cocktail would indeed need to go through over a decade of medical research to determine its long-term effects on animals... so kill-switch-suppressing chemicals may be interesting but ultimately impractical in principle if they are to be used responsibly.
We're talking investments in the billion$ here... those 'few' would have to be well-established companies that probably already own near-monopoly power on their respective turfs. These players already have the capital necessary to encroach on other players' territory if they wanted to... but most likely won't since they will have low subscriber density in new territories and their competitors may retaliate by encroaching on their territories and thinning the subscriber base there. Having lower subscriber density means less revenue on inflated network maintenance and operating expenses so having competitors for local loops is ultimately bad for both subscribers and network operators... sharing is more cost-effective for everyone and the telcos are too greedy to admit it.
If you drop those middle-weight players from the plan, all that is left is ADSL/phone resellers who lease the telcos' loops. Most of these companies have only few thousand customers with revenues in the low millions. Even if you put 100 of them together, they probably still would have a very hard time financing the deployment of their own local loops or fiber drops and properly maintain them. Also, there would inevitably be power struggle and mergers among them for dominion over at least parts of that network. Forming a non-profit co-op for this task might have a slim chance of working.
Forcing telcos to lease loops at reasonable rates to other service providers is much better, deregulating this was a horrible and unforgivable (but thankfully reversible) mistake. We already have cable, phone/ADSL and FiOS/FTTN/FTTP/FTTH/etc., we do not need multiple sets of local loops and more of everything else: unnecessary infrastructure duplication only increases deployment, maintenance and operating costs.
The x86(_64) ISA is ok feature-wise: it has everything any other general-purpose ISA has. The problem with it is the highly non-uniform instruction format: ~30 years ago, bits were really expensive so Intel designed the x86 ISA to be as compact as they could practically make it... and back then, they had to design the necessary circuits by hand, drawing them transistor by transistors every step of the way to create their lithography masks, makes it hard to blame them for not implementing the most optimal solution conceivable at the time. If the x86 processor disappeared tomorrow and was replaced with something else engineered to be efficient from the ground up, I think we would all benefit tremendously. It would not need to be completely new... most of the legacy peripheral hardware have modern equivalents and the instruction set is mostly ok. As I said in my other post, a re-engineered x86-64 ISA with all the tack-ons cleanly integrated together and improved ISA orthogonality/regularity would go a long way towards improving the situation. And since that new ISA would be very x86-64-like, legacy emulation should be possible with a moderate performance penalty.
The cost of deploying your own FTTN equipment and stringing your own fibre/wire is pretty prohibitive. Unless you can string up a large number of people in densely populated areas in one go, it will take years to recoup the expenses and turn up a decent return on investment. Only the biggest players can afford to foot the bills for large-scale network (re-)builds.
Getting rid of the legacy stuff in the BIOS is one thing... but there is a whole lot more of more complex legacy stuff (logic) embedded in x86 chipsets and CPUs. Dropping hardware support for legacy stuff/kludges (remember "dos=high,umb" and what made it possible?) from the pre-MMX era could considerably simplify chipset and CPU design. The desktop market could use a re-engineered, more orthogonal x86-inspired ISA on a clean platform: it would simplify things across the board (R&D, testing, instruction decoding, compilers, etc.), reduce die sizes, reduce pipeline length, improve frequency scaling, improve IPC, improve performance-per-watt, etc. Intel failed miserably across the line with its first attempt at going clean-slate but the potential benefits are still there.
Had Intel's Itanium plans worked out as initially planned, we would all be running on IA64 desktop hardware by now... kinda sad but seeing how IPF has turned out so far, we are better off this way for now.
What does a BIOS do? It does POST, lets the user customize some low-level system settings and puts the system in a known state before loading the OS's (boot-)loader.
An x86-style legacy BIOS does the same fundamental things as an x86-style EFI BIOS, the only major differences being the BIOS APIs, how the boot process is structured and the fact that EFI is not backwards-compatible on its own. Other than that, a BIOS, by any other name, is still a BIOS. EFI simply has fewer kludges and ties to legacy x86 hardware.
BTW, a few weeks ago, I read an article about some MoBo manufacturers considering adding 512MB-2GB of flash memory to boot an embedded Linux desktop from the BIOS for disk-less web-browsing and other stuff... a BIOS with embedded Linux does not seem that far-fetched, we only need 1GB firmware hubs to plug into Intel's chipsets and hope we will not need to flash our 1GB BIOS too often.
For those situations...
/s
Run -> "cmd" -> del %dir\*.*
It will clear most stuff and you will see error messages fly by... redirect output to a file for later examination if desired.
I use the good old 'del' whenever I know I will be deleting something like 20k files and do not wish to waste time waiting for windows to prepare for that operation... why the heck does Windows need to scan directories to be deleted before deleting them is beyond me, just delete them and be done with it. Same thing for copying, Windows wastes time scanning the source directory for no apparent reason since it won't tell you you have insufficient disk space to complete the operation until the target drive runs out of disk space... or any other errors for that matter, until it runs into them while carrying out the actual operation.
Linux has quirks, so does Windows. Linux has the excuse of being an relatively immature desktop OS but on the Windows side, it can only be written off as the result of half-ass design decisions.
Hand-writing can also cause RSI... as a kid, I often had to take pauses while writing essays because my hands/wrists started hurting, I would not be too surprised if many people's RSI problems started as early as grade-school. For me, writing became considerably uncomfortable during high-school - that was before I started using computers on a regular basis. Now, my hands' fine motor control and RSI discomfort is so bad that writing legibly is very much like torture. On the other hand, my carpal tunnels have yet to cause me any substantial problems - the RSI pain usually forces me to take extended breaks long before the wrist numbness and "pulsing" from carpal issues I have felt a few times before kick in.
RSI sucks.
I tried to put my laptop to sleep with Ubuntu x86-64 and my laptop crashed during wake-up. On the other hand, sleep worked fine with both XP 32bits and Vista Ultimate 64bits. ... I installed Vista last Tuesday and am pretty glad I got it for only $50 since it looks like I won't be using it much (or at all) in the near future. My laptop is back on XP (too slow and too little RAM to make Vista run sufficiently smoothly for my taste) and I probably won't be booting my dual-boot desktop to Vista very often until SP1 since it appears to become unstable during heavy IO and often/randomly fails to enumerate my external HDDs, quite annoying given that 80% of my storage is near-line.
What annoys me about Linux is that each distro has its own way of doing things. Some desktop distros just do not get it, others only sort-of get it and some like Ubuntu mostly get it.
For my two primary PCs, I use XP and Vista but for my other PCs, I use Fedora 6/7 (server/development) and Ubuntu. While Linux may excel at various stuff, much of the software I use beyond web browsing, eMail and audio/video playback is either Windows-only or a PITA to setup under Linux.
AGP is basically a souped-up single-drop PCI slot. AGP is fundamentally capable of doing full-speed in any mix of Read-Write as long as the target device is capable of sustaining the operation. The problem is that most GPUs have been designed as output-only devices and their host controller interfaces, be it PCI, AGP or PCIe, have been designed for fast writes to video memory but only implemented a low-cost, small-footprint, low-performance readback path since it is superfluous for normal operation. If you benchmark the readback throughput of early PCIe graphics card - those before the wholesale switch from GPU to GPGPU - probably all things up to Radeon X8xx and GF6 or possibly even X19xx and GF7, you will find that most of them also have relatively slow readback speeds.
Newer GPUs have been designed with GPGPU in mind which has a more symmetric traffic requirement: being able to do very fast computations on-card is pointless when it is impossible to get the results back in a timely manner. Take one of those GPGPUs, slap an AGP bridge onto it, put it in an AGP slot and you will get decent throughput in both directions.
Slow VRAM readback was a chip design decision, not a bus issue and not a bug. While the VRAM path is full-speed between the GPU core and RAM, the path between RAM controller and host controller is asymmetric in nearly all early PCIe GPUs as well of the vast majority of AGP and PCI ones. If you want a GPU with symmetric host-RAM RW capability, all Matrox GPUs, be it on PCI, AGP or PCIe have it IIRC - as should today's X2xxx and GF8.
Windows is primarily a desktop OS and the desktop marketplace is the one that suffers most from Microsoft's monopoly status. In the server space, Windows has a fair bit of competition since it is the new kid on the block in this traditionally *nix market.
So, the competition I had in mind was all non-MS desktop-oriented OSes... and right now, that means mostly GNU/Linux distros like Ubuntu, Gentoo and countless others.
So far, I have played only with Fedora and Ubuntu. Fedora is fine for software development but it thoroughly sucks at entertainment - I never managed to get something as simple as MP3 playback to work and I wasted at least two days on this alone with more days wasted trying to install VLC only to see it crash after loading. Ubuntu does much better as a desktop distro and I was very much pleased with only having to click through a few dialogs to install all the codecs/filters needed to play pretty much any media file I tried. However, both still have way too many quirks to be suitable for an average user without veteran assistance to smooth things out (like install GPU drivers and making them load/work properly) after initial setup.
Things are SLOWLY improving but IP minefield planted by various factions (M$ is only one of them) will ensure that the parking brakes holding FOSS-OS evolution and adoption (to a lesser extent) back will not come off any time soon.
Until Microsoft discloses the relevant patents, it is all FUD to me - just like the SCO case.
Microsoft is a convicted monopolist in multiple countries and it is seeking protection cash from its competitors to diversify its revenue streams. Since MS makes more than enough money to not depend on patent royalties, RedHat&all should seek a permanent injunction barring MS from collecting OS-related patent royalties from its OS competitors until MS' market share drops below something like 70% - since this is unlikely to happen within the next 10+ years, most core OS and GUI technology patents (the GUI is a mandatory component of a core Windows install after all) will have expired long before then.