Your casual dismissal of the facts involved in this controversy is what is mind-boggling.
If you're going to disregard my analogy as, "absurd," you're going to have to give specific reasons. Also, if you want to be taken seriously by your audience, you'll need to refrain from describing unsanctioned copying as, "stealing," as they are not at all the same thing.
People who constantly argue that "record companies should adapt their business model to piracy" are missing the point. They shouldn't have to: It's their intellectual property, not yours, and they have every right to dictate the terms of its distribution under existing law in every civilized country, even in Red China.
No. They shouldn't.
The proof of this is precisely because they have no control over independent duplication, nor can they ever reasonably expect to obtain that control.
Consider: Oxygen is a valuable commodity. Indeed, you cannot live without it. Oxygen is exuded from plants every day, including those in your garden. Now, given that there are significant, measurable costs to tending and maintaining your garden, and given that the oxygen it produces has clear market value, shouldn't you be able to charge for it? Aren't those who have received value from the oxygen you produced morally and ethically obliged to stuff money into your wallet? (Or, to take it to more absurd extremes: If Bill Gates bought the entire Amazon rainforest, could he legitimately start billing the world for the oxygen it produces?)
The answer, of course, is a big fat "no," because that's not how reality works. Anyone forming a business model based on this presumption would -- correctly -- be laughed into bankruptcy.
The reality of digital media is that it is easily and cheaply duplicated by anyone, anywhere, any time. It was designed to do this, making it a feature, not a bug, and, despite Micros~1's ambitions with Palladium, it's not going to change any time soon. This reality of digital media has never been a secret -- indeed, it's one of its big selling points -- and for media executives to whine shrilly about it speaks less toward their business acumen and more forcefully toward their stubborn unwillingness to face facts.
When writing a column about graphics driver development for the Be Developer Newsletter (RIP), I needed to come up with a name for an imaginary graphics card. I took a quick glance through the marketing buzzwords of the day, threw them in a blender, punched 'Frappe', and came up with:
Here's what gets me though, why is MS the bad guy here? Obviously there's some demand for MS to fill here.
And obviously there's a demand for heroin makers to fill, but we don't laud them as captains of industry.
The chances are Hollywood is telling MS "we'll start making movies ready for PC when we have the protection we need".
And Bill's only reasonable response is, "Fsck off. You don't need it." (In fact, I'm surprised Bill didn't tell them this; he seems to enjoy telling people to fsck off. Hell, he told the US Government to fsck off and made it stick; Hollywood should be cake in comparison.) Further, Micros~1 doesn't need the support of "Hollywood" to get movies on PCs. Indeed, Micros~1 is in the unique position of being able to fund its own damn studio. Micros~1 can harvest the talent at Sundance and Cannes as well as anyone.
But no, Micros~1 instead chooses to develop Palladium (or whatever it's named this week), which is designed to screw ordinary people by taking away control of their property. Though it may not be ensconced in any law, it is my personal belief that software engineers have a social and ethical responsibility to do right by their users -- the people whose lives are most affected by our work. I am personally revulsed that members of my profession would so easily abandon this responsibility in the name of making a buck.
I have a dual-P3 setup here for my main rig. Though it's starting to show its age -- a mere 1GHz with PC100 RAM -- I still use it for gaming. Unfortunately, while SMP seems to be a consideration among graphics card vendors (especially since Carmack made mention of threading in Q3A), sound card vendors don't appear to be quite as clueful.
I bought a Hercules Game Theater XP. I expected -- reasonably, I thought -- that such a high-end accessory would work solidly on an SMP system. Nope. Despite two major driver revisions since I bought the card, it is still horribly unreliable when all the HW accelerations features are turned on.
If I launch HalfLife with EAX/Sensaura enabled, the game will eventually crash. Leading up to the crash, the echo effects are completely botched. The echo sounds can be heard before the main sound. Sound effects are abbreviated -- the sound will stop before the sample has played out completely (especially true of footsteps). This suggests that buffers are being retired too early, which further suggests that the driver writer isn't locking access to the buffer queues correctly.
If I use the Audio Properties panel to back off HW acceleration one notch, then the card behaves reliably. Of course, I lose 90% of the cool sound effects...
It's vaguely possible that my motherboard may be twitchy (Asus P2B-D with ACPI fixes), but since it's never given trouble in Linux or BeOS, I'm not inclined to think so. So far, Hercules hasn't been very responsive on this issue. (Of course, I haven't pressed them very hard on this, either.)
So, yeah, having a sound card roundup for SMP systems would be a nice thing.
My limited understanding of IPv6 is that you can deploy v6 addresses locally, and advertise them globally via DNS using AAAA records. You can then talk over the larger Internet using a 6-over-4 tunnel.
Assuming this is correct, why doesn't Slashdot simply advertise an AAAA record, then accept connections through a 6-over-4 tunnel (or natively, if their bandwidth provider can speak it)? What are the technical considerations preventing this from working?
My sweetie got Joe-Jobbed a couple of weeks ago. 20K bounces over the course of the day. Thankfully, the payload of the spam was only two lines of text, containing a URL to a (non-existent) pr0n site. So the bounce messages were comparatively short. A cursory look at the headers in the bounces suggested that the attacker -- 'spammer' is too genteel a term for this -- was using a constellation of open relays to spread the stuff.
She came into my office, saying, "Make it stop!" Sadly, there turns out to be little one can do to stop it. The emails were coming from thousands of different legitimate sites, all serving a legitimate bounce to an illegitimate spam. It was very distressing for her. Fortunately, the attack stopped, and things settled down after about 24 hours.
Um, uh... No! Yeah, there's no way to turn off HTML mail in Outlook. Yeah. Outlook has no provisions for safe email reading.
To be completely safe, you should... Uh... delete Outlook entirely. Mmm, yeah, delete it. Outlook gone. Perfectly safe. Yeah, that's it...
Then you can safely install a safe email program, like... Er... Mozilla! Yeah! Or Evolution! Yeah, Evolution. I use it. And so does my wife... Morgan Fairchild...
I'm mostly a user of More QuakeWorld these days. It seems to be what most of the hardcore players are using these days. FuhQuake looks gorgeous, however. I'll be sure to give it a try.
What needs to be done is the Software makers and the Retailers need to sit down and make an effort to make the EULA available BEFORE the sale is made.
...Or, the software vendors could drop the entire license "agreement" charade entirely, and try living in the real world for a change.
Software vendors have been trying to have it both ways for decades. They're trying to enjoy the volume of sales provided by the "low friction" of traditional retail channels, while at the same time reserving privileges for themselves that are only available in one-on-one contract negotiations.
This is absolute bullshit, and I've said so for years. A sale is a sale is a sale. If you allow "licenses" to be applied to a transaction ex post facto, then Caveat Emptor gets raised to ridiculous heights.
I should not have to bring a Ferengi contract lawyer with me every time I go shopping at Fry's If "agreements" of this kind were to disappear tomorrow, software revenues would not decline, and the Republic would not collapse. There is no legitimate reason for them to exist at all.
The honest solution is to ship software with EULA seperate, put a stack of EULA next to the software, and require me to sign off on it, right there in the store, before I fork over my cash. That would be fair.
One of my recurring nightmares is that the credit card companies become, "assent clearinghouses," and issue new cardmember agreements with a clause reading something like, "By using the Card to purchase computer software, the cardmember warrants and represents she/he will comply with all terms of the enclosed Software License Agreement, if any. The cardmember's signature on the receipt at the point of sale shall constitute assent to the terms of the software license."
Because the credit card companies are all colluding anyway, this term will show up on all cardmember agreements at the same time. You won't be able to vote with your feet.
I should go file a patent on this "business practice" so they can't do it for the next 20 years...
A popular quote from most end-user license "agreements" (which are all unethical, anyway). Different tax rules apply for license transactions than sales transactions.
...my system gets significantly better performance. Significantly = Same frame rates even tho i'm playing at 1280x1024 on a 19" monitor and he is playing at 1024x768 on a 17" monitor.
Motherboard - ASUS A7V8X 101.98 [... ]
Motherboard - ASUS P4G8X DELUXE 216.90
Neither of these are dual-CPU machines. You are therefore insufficiently l33t.
Hard Disk 1 - EIDE 200.0GB 7200 RPM 8MB CACHE 255.00
Hard Disk 2 - EIDE 200.0GB 7200 RPM 8MB CACHE 255.00
I don't see a brand name here. If you're buying IBM, you're hosed. If you're buying Western Digital, you're still probably hosed. Even if you're using IDE (a/k/a fake) RAID, you won't like it when one or both of the IBM drives goes toes-up. Get Maxtor or Quantum here.
Better still, ignore the IDE toys completely and go all LVD-SCSI. Faster, more sex appeal, and you can stick in more than a paltry four drives.
Video Card - PNY GeForce FX 399.00
Well, um, okay, I'm among the first to proclaim frames-per-second-uber-alles. But the ATI Radeon 9700 Pro cranks out nearly as many frames, costs half as much, and it won't generate nearly as much waste heat.
Sound Card - Creative Audigy2 Platinum 162.85
Hercules Game Theater XP 7.1 is about as good, and doesn't come with Creative's fsckload of useless but required utilities. Also, Creative couldn't give a toss about SMP users.
Speakers - Creative Inspire 6.1 6600 110.00
I never understood this -- people getting tricked-out audio cards and then cheaping out on the speakers. Real users plug their sound cards into real stereo components with real Dolby/surround decoders and real speakers. With the money you'd save buying ATI instead of bleeding-edge NVidia, you could buy yourself a decent component amp with respectable speakers.
Keyboard/Mouse - Microsoft Wireless Optical KB/M 82.00
Hardcore gamers do not use wireless input peripherals. Wireless == dropouts. Dropouts == lost frags.
If you don't mind washing your mouse balls every so often, Karna still has Razer Boomslang 2000 mice for sale. If you're an optical kinda guy, any of the recent wired USB offerings from Logitech or Micros~1 will serve well.
As for the keyboard, you want something sturdy, since you're going to be beating fairly hard on it (even harder if you're losing:-) ). Go prowl around for offerings from IBM. If you want to be uber-l33t, scrounge up an old IBM PC-XT keyboard:-).
It probably annoys the telcos to no end that a connection can be shared - they are more used to the "telephone" model, where there is one line going into the house and if 2 people want to have separate converations then they need two lines.
Well, the cable company is after me, and I can't understand why.
I picked up a used VAX-11/780 a while back (had the word 'dagobah' scrawled inside the door, never figured out what that was about), and have a couple dozen friends and neighbors hooking up to it via a combinaton of Wyse-50 serial terminals and NDS dedicated X terminals. The terminals are "dumb" and can't do any local processing. All the compute resources are on the VAX, there are no NAT services running, and only one IP address is being consumed. So the connection isn't being shared.
Still, the cableco is giving me static about connection sharing, saying it's tantamount to running NAT. I countered by saying that running NAT is tantamount to running a large multi-user machine. But their lawyers are better dressed than mine, and are threating criminal cable fraud charges. I have no idea how it will turn out. If they decide to go to the mat, it'll be interesting to watch the local constabulary confiscate the VAX for forensic examination.
Schwab
P.S: Anyone know how to compile Quake2 for this thing? It keeps crapping out on the CPU_ARCH #define with the message, "Carmack hits you with a cluestick --more--".
Pardon my curmudgeonliness, but what the hell is wrong with looking at the disk activity light? That's usually a fairly accurate indicator that the machine's in the process of loading and running your program.
It's only just recently that I've tried to understand the vagaries of windowing systems and GUI kits under X. (My previous attempt was by reading the Xlib reference manual. Ugh.) There appears to be a mostly-unstated assumption on which bits of your windowed app are handled by what.
What I've learned so far is that the functional separation seems to based on the "conceptual boundaries" established by the window(s). This appears to have led to the establishment of three major components on X desktops:
Inside The Window: The Widget Toolkit
This is the piece that's responsible for rendering the various buttons, sliders, textboxes, labels, etc. Applications describe in abstract terms what widgets they want and how they want them laid out, and the toolkit is responsible for actually making it happen. An example of a widget toolkit is GTK.
Around The Window: The Window Manager
The Window Manager is responsible for operations on the window proper, allowing the user to depth-arrange, drag, resize, minimize, etc. the windows appearing on the display. To facilitate this, the Window Manager (typically) decorates the borders of the window with control glyphs to accomplish these various tasks. Examples of window managers include WindowMaker and SawMill.
Outside The Windows: The Desktop Manager
The space not occupied by visible windows is the Desktop. The Desktop Manager gives functionality to the regions of the screen not occupied by windows. This might include setting the background image, drawing shortcut icons, displaying pop-up menus to launch applications, etc.
Near as I can tell, each of these components exists (mostly) independently of each other -- you can have an app using the GTK toolkit running in the KDE Window Manager on an unmanaged desktop. As such, there appears to be a huge opportunity for similar or duplicate code to accomplish the smae thing.
Each component appears to be independently and variably "theme-able". For example, WindowMaker has relatively little theme flexibility, whereas SawMill apparently has tons. Each manager accomplishes theme-ability in its own way, further contributing to duplicated code.
Further confusing the issue is the use of a single term to refer to all of these components in aggregate. For example, "GNOME" typically refers collectively to the Widget Toolkit, the Window Manager, and the Desktop Manager....Except that GNOME actually seems to be mostly an API specification. It is possible for Window Managers to be GNOME-compliant without actually being part of GNOME. Nautilus, SawMill, and WindowMaker are all GNOME-compliant, but not all of them are officially part of GNOME.
So. Does that sound right, or am I completely off-base?
Well, gee, what to say, except that hindsight is always 20/20. Armchair analysts of today haven't had the benefit of experiencing Trip's Reality Distortion Field(tm), where the idea of a $700.00 console actually seems fairly reasonable.
I have my own ideas as to why the 3DO platform failed. One is that the development system was hosted on NuBus-based Macintosh systems (this was in the 68K era, before Apple jumped entirely over to PowerPC). Despite screams of developers everywhere, no effort was made to port to the PC until very late. Further, once Apple announced they were abandoning NuBus in favor of PCI, no effort was made to convert the development hardware, forcing developers to find increasingly scarce (and slow) older Macs. And, despite the protestations of enthusiasts everywhere, the Mac was just agonizingly slow. (3DO developers should count themselves fortunate, however. Had the original system developers had their way, development would have been hosted on the Amiga. Commodore declared bankruptcy about six months before the 3DO was launched.)
The other big problem was that the development software and tools were, for the most part, utter $(EXPLETIVE) $(EXPLETIVE) $(EXPLETIVE) garbage. 3DODebug was little better than a program loader and dumb command terminal. Being in the system software group, I was fortunate in that I got to use a Philips logic analyzer to debug the thornier problems, rather than suffer with the never-did-work-right symbolic debugger. 3DOAnimator was a very crufty hack on top of EA's Studio32, and it would regularly crash, destroying all work. There were a couple of Photoshop plugins, but their use and enhancement was discouraged, as they were considered "stopgap" measures until 3DOAnimator came up to snuff (it never did). And the Norcroft C compiler sucked rocks. It generated bad code and kicked out stupid and incorrect warnings that couldn't be turned off. That so many titles were developed in this apalling environment is a tribute to the dedication and talent of all the developers we had.
At the end of it all, though, I don't really know why 3DO failed. We had more than enough money, and a charismatic leader who could convince people of the most astonishing things -- a formula for sure-fire success in anybody's book. Except ours.
Get me drunk sometime and I'll tell you all about Jurassic Park Interactive...
And so under the leadership of Pong-creator Trip Hawkins, and the backing of Matsushita, AT&T, Time Warner, Electronic Arts and MCA, 3DO was born.[emphasis mine]
Uh, no. Not by a longshot. Nolan Bushnell is the shepherd behind Pong. Trip Hawkins is the founder of Electronic Arts.
"When it gets down to it -- talking trade balances here -- once we've brain-drained all
our technology into other countries, once things have evened out, they're making cars in Bolivia and microwave ovens in Tadzhikistan and selling them here -- once our edge in natural resources has been made irrelevant by giant Hong Kong ships and dirigibles that can ship North Dakota all the way to New Zealand for a nickel -- once the Invisible Hand has taken all those historical inequities and smeared them out into a broad global layer of what a Pakistani brickmaker would consider to be prosperity -- y'know what? There's only four things we do better than anyone else:
You just have to explicitly mprotect(2) [openbsd.org] the memory where it happen with PROT_EXEC|PROT_WRITE. The fact that on some OSes it can work without doing that is actually a bug in these OSes.
Ah, that's better. Theo's message didn't make that clear, and suggested that the kernel would globally disallow pages where both PROT_WRITE and PROT_EXEC were set, which would suck for several classes of application.
W^X is a short form for "PROT_WRITE XOR PROT_EXEC". The basic idea here is that most buffer overflow "eggs" rely on a particular feature: That there is memory which they can write to, and then jump to.
What if there was no such memory? Does a normal Unix process have memory that is both writeable and executable? Turns out they do: [... ]
But do they need it? [... ]
If you're running a JIT compiler/interpreter or other dynamic code assembly, you sure as hell do.
I can see how you might be able to write dynamically generated code to a page, then turn off PROT_WRITE and turn on PROT_EXEC before jumping to it. However, this is almost certainly two trips into the kernel, each involving tons of permission checking. So performance will likely suffer, which negates the whole point of doing JIT in the first place.
Slashdot Mirror
Your casual dismissal of the facts involved in this controversy is what is mind-boggling.
If you're going to disregard my analogy as, "absurd," you're going to have to give specific reasons. Also, if you want to be taken seriously by your audience, you'll need to refrain from describing unsanctioned copying as, "stealing," as they are not at all the same thing.
Schwab
No. They shouldn't.
The proof of this is precisely because they have no control over independent duplication, nor can they ever reasonably expect to obtain that control.
Consider: Oxygen is a valuable commodity. Indeed, you cannot live without it. Oxygen is exuded from plants every day, including those in your garden. Now, given that there are significant, measurable costs to tending and maintaining your garden, and given that the oxygen it produces has clear market value, shouldn't you be able to charge for it? Aren't those who have received value from the oxygen you produced morally and ethically obliged to stuff money into your wallet? (Or, to take it to more absurd extremes: If Bill Gates bought the entire Amazon rainforest, could he legitimately start billing the world for the oxygen it produces?)
The answer, of course, is a big fat "no," because that's not how reality works. Anyone forming a business model based on this presumption would -- correctly -- be laughed into bankruptcy.
The reality of digital media is that it is easily and cheaply duplicated by anyone, anywhere, any time. It was designed to do this, making it a feature, not a bug, and, despite Micros~1's ambitions with Palladium, it's not going to change any time soon. This reality of digital media has never been a secret -- indeed, it's one of its big selling points -- and for media executives to whine shrilly about it speaks less toward their business acumen and more forcefully toward their stubborn unwillingness to face facts.
Schwab
Or, to translate it into a well-known aphorism:
"Under Communism, man exploits man. Under Capitalism, it's the other way around."
Schwab
When writing a column about graphics driver development for the Be Developer Newsletter (RIP), I needed to come up with a name for an imaginary graphics card. I took a quick glance through the marketing buzzwords of the day, threw them in a blender, punched 'Frappe', and came up with:
Yoyodyne Monstra VelocElite-LX 128-3D-AGP
Schwab
And obviously there's a demand for heroin makers to fill, but we don't laud them as captains of industry.
And Bill's only reasonable response is, "Fsck off. You don't need it." (In fact, I'm surprised Bill didn't tell them this; he seems to enjoy telling people to fsck off. Hell, he told the US Government to fsck off and made it stick; Hollywood should be cake in comparison.) Further, Micros~1 doesn't need the support of "Hollywood" to get movies on PCs. Indeed, Micros~1 is in the unique position of being able to fund its own damn studio. Micros~1 can harvest the talent at Sundance and Cannes as well as anyone.
But no, Micros~1 instead chooses to develop Palladium (or whatever it's named this week), which is designed to screw ordinary people by taking away control of their property. Though it may not be ensconced in any law, it is my personal belief that software engineers have a social and ethical responsibility to do right by their users -- the people whose lives are most affected by our work. I am personally revulsed that members of my profession would so easily abandon this responsibility in the name of making a buck.
Schwab
I have a dual-P3 setup here for my main rig. Though it's starting to show its age -- a mere 1GHz with PC100 RAM -- I still use it for gaming. Unfortunately, while SMP seems to be a consideration among graphics card vendors (especially since Carmack made mention of threading in Q3A), sound card vendors don't appear to be quite as clueful.
I bought a Hercules Game Theater XP. I expected -- reasonably, I thought -- that such a high-end accessory would work solidly on an SMP system. Nope. Despite two major driver revisions since I bought the card, it is still horribly unreliable when all the HW accelerations features are turned on.
If I launch HalfLife with EAX/Sensaura enabled, the game will eventually crash. Leading up to the crash, the echo effects are completely botched. The echo sounds can be heard before the main sound. Sound effects are abbreviated -- the sound will stop before the sample has played out completely (especially true of footsteps). This suggests that buffers are being retired too early, which further suggests that the driver writer isn't locking access to the buffer queues correctly.
If I use the Audio Properties panel to back off HW acceleration one notch, then the card behaves reliably. Of course, I lose 90% of the cool sound effects...
It's vaguely possible that my motherboard may be twitchy (Asus P2B-D with ACPI fixes), but since it's never given trouble in Linux or BeOS, I'm not inclined to think so. So far, Hercules hasn't been very responsive on this issue. (Of course, I haven't pressed them very hard on this, either.)
So, yeah, having a sound card roundup for SMP systems would be a nice thing.
Schwab
My limited understanding of IPv6 is that you can deploy v6 addresses locally, and advertise them globally via DNS using AAAA records. You can then talk over the larger Internet using a 6-over-4 tunnel.
Assuming this is correct, why doesn't Slashdot simply advertise an AAAA record, then accept connections through a 6-over-4 tunnel (or natively, if their bandwidth provider can speak it)? What are the technical considerations preventing this from working?
Schwab
My sweetie got Joe-Jobbed a couple of weeks ago. 20K bounces over the course of the day. Thankfully, the payload of the spam was only two lines of text, containing a URL to a (non-existent) pr0n site. So the bounce messages were comparatively short. A cursory look at the headers in the bounces suggested that the attacker -- 'spammer' is too genteel a term for this -- was using a constellation of open relays to spread the stuff.
She came into my office, saying, "Make it stop!" Sadly, there turns out to be little one can do to stop it. The emails were coming from thousands of different legitimate sites, all serving a legitimate bounce to an illegitimate spam. It was very distressing for her. Fortunately, the attack stopped, and things settled down after about 24 hours.
I wrote up the experience on Kuro5hin. Feel free to have a look.
Schwab
Um, uh... No! Yeah, there's no way to turn off HTML mail in Outlook. Yeah. Outlook has no provisions for safe email reading.
To be completely safe, you should... Uh... delete Outlook entirely. Mmm, yeah, delete it. Outlook gone. Perfectly safe. Yeah, that's it...
Then you can safely install a safe email program, like... Er... Mozilla! Yeah! Or Evolution! Yeah, Evolution. I use it. And so does my wife... Morgan Fairchild...
Schwab
I'm mostly a user of More QuakeWorld these days. It seems to be what most of the hardcore players are using these days. FuhQuake looks gorgeous, however. I'll be sure to give it a try.
Schwab
Main memory in these devices tends to be slow, to conserve power. How slow is this thing when performing CPU-intensive tasks? What's the battery life?
...What I'm driving at is: Can I slap this thing down on a LAN and have a reasonably well performing, instant QuakeWorld server?
Schwab
...Or, the software vendors could drop the entire license "agreement" charade entirely, and try living in the real world for a change.
Software vendors have been trying to have it both ways for decades. They're trying to enjoy the volume of sales provided by the "low friction" of traditional retail channels, while at the same time reserving privileges for themselves that are only available in one-on-one contract negotiations.
This is absolute bullshit, and I've said so for years. A sale is a sale is a sale. If you allow "licenses" to be applied to a transaction ex post facto, then Caveat Emptor gets raised to ridiculous heights.
I should not have to bring a Ferengi contract lawyer with me every time I go shopping at Fry's If "agreements" of this kind were to disappear tomorrow, software revenues would not decline, and the Republic would not collapse. There is no legitimate reason for them to exist at all.
Schwab
One of my recurring nightmares is that the credit card companies become, "assent clearinghouses," and issue new cardmember agreements with a clause reading something like, "By using the Card to purchase computer software, the cardmember warrants and represents she/he will comply with all terms of the enclosed Software License Agreement, if any. The cardmember's signature on the receipt at the point of sale shall constitute assent to the terms of the software license."
Because the credit card companies are all colluding anyway, this term will show up on all cardmember agreements at the same time. You won't be able to vote with your feet.
I should go file a patent on this "business practice" so they can't do it for the next 20 years...
Schwab
A popular quote from most end-user license "agreements" (which are all unethical, anyway). Different tax rules apply for license transactions than sales transactions.
Schwab
Did you vote?
Not just for the President, but for your Congressman and Senator? State Governor and legislators? Mayor? City council members?
They're all important, you know...
Schwab
Switch monitors and try again.
Schwab
Neither of these are dual-CPU machines. You are therefore insufficiently l33t.
I don't see a brand name here. If you're buying IBM, you're hosed. If you're buying Western Digital, you're still probably hosed. Even if you're using IDE (a/k/a fake) RAID, you won't like it when one or both of the IBM drives goes toes-up. Get Maxtor or Quantum here.
Better still, ignore the IDE toys completely and go all LVD-SCSI. Faster, more sex appeal, and you can stick in more than a paltry four drives.
Well, um, okay, I'm among the first to proclaim frames-per-second-uber-alles. But the ATI Radeon 9700 Pro cranks out nearly as many frames, costs half as much, and it won't generate nearly as much waste heat.
Hercules Game Theater XP 7.1 is about as good, and doesn't come with Creative's fsckload of useless but required utilities. Also, Creative couldn't give a toss about SMP users.
I never understood this -- people getting tricked-out audio cards and then cheaping out on the speakers. Real users plug their sound cards into real stereo components with real Dolby/surround decoders and real speakers. With the money you'd save buying ATI instead of bleeding-edge NVidia, you could buy yourself a decent component amp with respectable speakers.
Hardcore gamers do not use wireless input peripherals. Wireless == dropouts. Dropouts == lost frags.
If you don't mind washing your mouse balls every so often, Karna still has Razer Boomslang 2000 mice for sale. If you're an optical kinda guy, any of the recent wired USB offerings from Logitech or Micros~1 will serve well.
As for the keyboard, you want something sturdy, since you're going to be beating fairly hard on it (even harder if you're losing :-) ). Go prowl around for offerings from IBM. If you want to be uber-l33t, scrounge up an old IBM PC-XT keyboard :-).
Schwab
Well, the cable company is after me, and I can't understand why.
I picked up a used VAX-11/780 a while back (had the word 'dagobah' scrawled inside the door, never figured out what that was about), and have a couple dozen friends and neighbors hooking up to it via a combinaton of Wyse-50 serial terminals and NDS dedicated X terminals. The terminals are "dumb" and can't do any local processing. All the compute resources are on the VAX, there are no NAT services running, and only one IP address is being consumed. So the connection isn't being shared.
Still, the cableco is giving me static about connection sharing, saying it's tantamount to running NAT. I countered by saying that running NAT is tantamount to running a large multi-user machine. But their lawyers are better dressed than mine, and are threating criminal cable fraud charges. I have no idea how it will turn out. If they decide to go to the mat, it'll be interesting to watch the local constabulary confiscate the VAX for forensic examination.
Schwab
P.S: Anyone know how to compile Quake2 for this thing? It keeps crapping out on the CPU_ARCH #define with the message, "Carmack hits you with a cluestick --more--".
P.P.S: :-)
Pardon my curmudgeonliness, but what the hell is wrong with looking at the disk activity light? That's usually a fairly accurate indicator that the machine's in the process of loading and running your program.
Schwab
It's only just recently that I've tried to understand the vagaries of windowing systems and GUI kits under X. (My previous attempt was by reading the Xlib reference manual. Ugh.) There appears to be a mostly-unstated assumption on which bits of your windowed app are handled by what.
What I've learned so far is that the functional separation seems to based on the "conceptual boundaries" established by the window(s). This appears to have led to the establishment of three major components on X desktops:
This is the piece that's responsible for rendering the various buttons, sliders, textboxes, labels, etc. Applications describe in abstract terms what widgets they want and how they want them laid out, and the toolkit is responsible for actually making it happen. An example of a widget toolkit is GTK.
The Window Manager is responsible for operations on the window proper, allowing the user to depth-arrange, drag, resize, minimize, etc. the windows appearing on the display. To facilitate this, the Window Manager (typically) decorates the borders of the window with control glyphs to accomplish these various tasks. Examples of window managers include WindowMaker and SawMill.
The space not occupied by visible windows is the Desktop. The Desktop Manager gives functionality to the regions of the screen not occupied by windows. This might include setting the background image, drawing shortcut icons, displaying pop-up menus to launch applications, etc.
Near as I can tell, each of these components exists (mostly) independently of each other -- you can have an app using the GTK toolkit running in the KDE Window Manager on an unmanaged desktop. As such, there appears to be a huge opportunity for similar or duplicate code to accomplish the smae thing.
Each component appears to be independently and variably "theme-able". For example, WindowMaker has relatively little theme flexibility, whereas SawMill apparently has tons. Each manager accomplishes theme-ability in its own way, further contributing to duplicated code.
Further confusing the issue is the use of a single term to refer to all of these components in aggregate. For example, "GNOME" typically refers collectively to the Widget Toolkit, the Window Manager, and the Desktop Manager. ...Except that GNOME actually seems to be mostly an API specification. It is possible for Window Managers to be GNOME-compliant without actually being part of GNOME. Nautilus, SawMill, and WindowMaker are all GNOME-compliant, but not all of them are officially part of GNOME.
So. Does that sound right, or am I completely off-base?
Schwab
Well, gee, what to say, except that hindsight is always 20/20. Armchair analysts of today haven't had the benefit of experiencing Trip's Reality Distortion Field(tm), where the idea of a $700.00 console actually seems fairly reasonable.
I have my own ideas as to why the 3DO platform failed. One is that the development system was hosted on NuBus-based Macintosh systems (this was in the 68K era, before Apple jumped entirely over to PowerPC). Despite screams of developers everywhere, no effort was made to port to the PC until very late. Further, once Apple announced they were abandoning NuBus in favor of PCI, no effort was made to convert the development hardware, forcing developers to find increasingly scarce (and slow) older Macs. And, despite the protestations of enthusiasts everywhere, the Mac was just agonizingly slow. (3DO developers should count themselves fortunate, however. Had the original system developers had their way, development would have been hosted on the Amiga. Commodore declared bankruptcy about six months before the 3DO was launched.)
The other big problem was that the development software and tools were, for the most part, utter $(EXPLETIVE) $(EXPLETIVE) $(EXPLETIVE) garbage. 3DODebug was little better than a program loader and dumb command terminal. Being in the system software group, I was fortunate in that I got to use a Philips logic analyzer to debug the thornier problems, rather than suffer with the never-did-work-right symbolic debugger. 3DOAnimator was a very crufty hack on top of EA's Studio32, and it would regularly crash, destroying all work. There were a couple of Photoshop plugins, but their use and enhancement was discouraged, as they were considered "stopgap" measures until 3DOAnimator came up to snuff (it never did). And the Norcroft C compiler sucked rocks. It generated bad code and kicked out stupid and incorrect warnings that couldn't be turned off. That so many titles were developed in this apalling environment is a tribute to the dedication and talent of all the developers we had.
At the end of it all, though, I don't really know why 3DO failed. We had more than enough money, and a charismatic leader who could convince people of the most astonishing things -- a formula for sure-fire success in anybody's book. Except ours.
Get me drunk sometime and I'll tell you all about Jurassic Park Interactive...
Schwab
Uh, no. Not by a longshot. Nolan Bushnell is the shepherd behind Pong. Trip Hawkins is the founder of Electronic Arts.
Schwab
Schwab
Ah, that's better. Theo's message didn't make that clear, and suggested that the kernel would globally disallow pages where both PROT_WRITE and PROT_EXEC were set, which would suck for several classes of application.
Schwab
Theo de Raadt writes:
If you're running a JIT compiler/interpreter or other dynamic code assembly, you sure as hell do.
I can see how you might be able to write dynamically generated code to a page, then turn off PROT_WRITE and turn on PROT_EXEC before jumping to it. However, this is almost certainly two trips into the kernel, each involving tons of permission checking. So performance will likely suffer, which negates the whole point of doing JIT in the first place.
I like his survey of MMU architectures, though.
Schwab