Draw up a league table, but credit all kinds of giving, in relation to their net worth to the community and to their magnitude. I reckon something like 3 points for an application or significant module, 2 for extensive reworking or contributing moderate code, 1 for minor bugfixes and applets, +1 for each significant political or legal hurdle crossed, -1 for every such hurdle added, 2/3 points if it's open source for a closed-source environment, 1/2 points if it is closed source for an open-source environment.
Points have a time-to-live of one year, extended by three months for every major maintenance cycle, two for moderate maintenance, and one if the updates are thrown off the back of a lorry at high speed.
Have three leagues. One league being major corporate entities, one being the smaller companies involved, and one for the genuine collective projects.
My guess is that IBM, SGI etc, would not care in the slightest, but I seriously doubt they'd refuse a plaque commemorating a successful year, either. The smaller companies - now, they might care. Publicity is the lifeblood of the small business, and this is easy free publicity - as well as steering them the right way. Collective projects SHOULDN'T care - if they do, their members have lost sight of the project and are focussed on the kudos. Bad mistake. Projects that do that are doomed, doomed I say!
For the last ten years, you've really been electing a bunch of PET 3032s, Apple Is and ZX-80s. The speech synthesis was by Superior Software and the suits by US Gold. Sometime in the next few months, we are due to be attacked by a large number of mutant camels, the road system already having degenerated into a maze of twisty passages, all alike.
At many key points in history, refugees from the "old system" founded their own, alternative system, from the communities through to the industries needed to maintain them. At one extreme, Cambridge University and the city of Cambridge were founded by students who had escaped being murdered in Oxford by a lynch mob bent on street justice. At the other extreme, the Amish escaped psychotic religious extremism in Europe, fled to the US, and have virtually isolated themselves in their communities - what they need, they provide.
(You have to exclude any commune, settlement or system that depends so much on the system it tries to escape from that the escape is merely an illusion. Interdependence is fine and perfectly respectable, and probably necessary for most such efforts, but it should never be permitted to slip into codependency - which has been the fate of many attempts.)
I'm not proposing that dissatisfied Slashdot geeks get together and build a self-sufficient University-City that can provide its own food, clothing, housing, power, books, etc. The task is not impossible, but the odds of Slashdot having enough sufficiently dissatisfied people with a sufficiently-broad range of skills over a sufficiently small distance to get them together.... It's not going to happen in my lifetime.
Ultimately, the majority of people most desiring of change have no voice and little money. If you expect drastic change to happen, it cannot require what these people don't have. Ergo, if you want change, it must be from the absolute ground up, to the point where the homeless and dispossessed can be valuable, and where there is little or nothing you could need that can't be provided in-house rather than being bought.
Many people have tried to found just such communal enterprises, but most have failed because it's not just not easy, it is unimaginably tough to start and never gets better than phenomenally hard, even after it is not just self-sufficient in all things but capable of "exporting" to the rest of society. Even the organization of such a community system is outside most people's ability, which is why you see most attempting to use charismatic leaders to hold things together, rather than people who know what they're doing. And which is why such communes usually end up in a bad way, or worse.
These problems are very very hard, but they are not unsolvable. If they were, nobody could ever have formed a civilized community in the first place. If you are sick and tired enough to clear the obstacles, solve the problems and keep things together, then nothing is stopping you starting your own civilization. But it really is at that level.
No, it's not "something to worry about", in the sense of looking out for it. There's just no point. Almost all contracts and employment agreements will have such a clause. There's also usually the authority to terminate employment for no reason, which is a great get-out clause for companies that want to violate employment laws. In fact, typical employment contracts give you no rights, no protections and no stability. Your bosses can do what they like and there is nothing you can do about it, even when the law technically says otherwise. There is just no point worrying about conditions you will simply have to accept if you want to continue eating.
Yes, it's something to worry about, in the sense that nobody has any incentive to invent. The employees won't see a dime, if their bright ideas have to be handed over without question. Companies have no incentive, because they should be able to get just as good results for free. Besides, if they ask their new hires to innovate, the new hires will have to give all this neat new stuff to their former boss, not them. R&D has no value, when it is in nobody's interest to carry it out. In their pursuit of instant gratification and the "now" money, the people with business degrees are killing off the people with real knowledge. There is no long-term future for such a mindset. It consumes but never produces. In the end, it will starve itself and all around it to death. Those just graduating damn well should worry that there is a serious danger of there being no long-term future. Not just for a job, but for whole industries.
But I'd start with something like pchar, which will tell you the effective bandwidth at each hop on the network. That will tell you how severe the blockage is and, more importantly, where. It's not the "best" tool out there, but it's reasonably non-intrusive (unlike most stress-testing tools) and I've not seen any obvious problems with it at gig speeds. It does need patching for Linux, though. I sent a patch to the maintainer who has sworn he'll someday get around to including it. NetBSD has a faster network stack, though, and is more suitable for such tests. Which I hate, as I prefer Linux, but facts don't change themselves to suit a like.
...Valenti is dead and therefore doesn't give a damn about what we think. In his final days, if he ever thought about those who wanted the right to view the movies they bought in the manner of their choosing, he probably smiled at the fact that his view was the enforced view, worldwide, to the very day he died. Or maybe laughed at the fact that those who protest nonetheless fund the very machinery they protest so much about.
In the end, hate is nothing more than a few chemicals swirling around in a brain. It has no meaning, it has no value, and most people who are in any way successful will have plenty of people who hate them. So will many who will never know success at all. Hate is commonplace.
What he had was money, power and influence - three things far more valuable than any mere emotion, and three things that protesters at the MPAA's excesses are unlikely to achieve to anything like the same degree.
I detest what this one man did, I detest what he stood for, but there's no use denying the fact that his was a story of success on a grandiose scale and that many people will try to emulate him for that very reason.
A federation of computer networks that discriminates only by the ability to exchange information with something in that federation.
(It's not about protocols. If you wired up a network using X.25 or DECNet, fired up some obscure machine, loaded up an information server on it, then provided a gateway to regular TCP/IP and a proxy to HTTP, that machine would be as much "on the internet" as any other. Any definition has to allow for non-standard connections, or it's not a complete definition of the Internet that people use.)
Just tried the latest kernel and it hangs on trying to fire up the second ATA instance. Not even a kernel oops, nothing. That's true whether I use the vanilla kernel or Red Hat's RPM. Something is screwed up, and from the sounds of it, there's more than one of us experiencing a failure at the same point, so that would be the obvious suspect.
Far as I know, Slashdot UIDs are only sold at Christie's Auction House, and I think the 4-digit ones go for a couple of million, if they're in good condition and still in their original packaging. I heard rumors that the US DoD, in an effort to gain greater credibility with geeks, were planning an offer of a few billion on a 2-digit UID.
I've seen T1 and T3 lines go down for a week to a week and a half, due to a difficult-to-trace fault. The ISP (MCI) suffered no penalties, even though it was far and away beyond the SLA terms - I forget exactly why, but IIRC there was provision for exceptional situations. This is not to criticize MCI - much - they are one of the better ISPs and deserve credit for that. I've seen many that were far, far worse.
As for speed, you must bear in mind that the TOP speeds for a cable modem or xDSL line are comparable to a T1 line, but an optical T1 line can be uprated to an optical T3 line - or even a T4 line. (T2 and T4 lines are considered "teleco-only" but they do indeed exist.)
There are times a cable modem will run at the same speed as a 56K traditional modem - or slower. The traditional modem will be cheaper, not because of the speed you are running it at, but because of the speed you CAN run it at. The maximum potential is a function of the grade of the hardware used. Better grade hardware can always be run at slower speeds, but will still cost a lot simply because they are higher grade.
Frankly, I'd love it if I could afford a direct O8 connection to my home. But there's no possible way I could afford an O8, even if I only ever ran it at DSL speeds.
Alan Turing was a genius - no two oways about it - but if you look up at the child prodigies that have gone through British Universities (particularly in maths), you can see that there are plenty who are as far beyond his upper limit as he was above the average University postgrad.
The downgrading of maths is bothersome, but so is the fact that children who are too bright are held back and stunted, in the name of "letting them be kids". (How about letting them be themselves?)
There are other issues, too. British Universities are generally pretty good at forcing people to think - you are expected to do far more hands-on research than at a typical US University - however, the examination system sucks. You get to see past papers, and are advised to look at the past two years. I got the past six years, for one course, and found that only two exam papers had been used in all that time. The lecturer just alternated between them, relying on students not to do their homework. Other exam papers were plain drivel, requiring more memorization than comprehension.
IMHO, examinations should be divided up something like: 40% subject comprehension, 30% ability to extrapolate and deduce within the subject, 20% ability to transfer subject skills, 10% rote memorization. We have information sources aplenty - rote has no purpose in today's society beyond a few corner cases. The ability to intelligently apply a skill - directly or indirectly - and to use the subject as a tool that is crafted around a problem-space, not a designation... These things can't be found on the Internet.
Since the questioner isn't isolated from the system, if the question is a polarizing one, the questioner must themselves be drawn to one pole or the other. (This is opposed to a merely hotly debated subject, where the initial gap in opinions is huge and may remain huge, but no poles exist and an equilibrium that is also a majority opinion will eventually form. Such a debate is the "ideal" in Classical science, although it almost never happens and is virtually extinct from all modern debate.)
If the questioner is polarized, then the posting can only be overtly or covertly antagonistic to those of the opposite pole. If there is a question that is clearly a dig at the other side, then it's pretty overt. Otherwise, there's an excellent chance that the dig is still there but is worded too well to be clearly seen.
I'm getting a little tired of the chances, myself, but if it gives Microsoft another chance of receiving a record fine, it's hard to disagree with it completely. What would be better, though, is for the EU to accelerate the appeals process and get the whole case dealt with ASAP. And then collect, by whatever means necessary. The fact is, although the EU is prosecuting Microsoft's contempt of court, as well as the practices, they will be seen as toothless if they can't actually get Microsoft to deliver.
In fact, the EU should seriously consider whether it would be advisable to seize assets up to the value of the fines so far, on the grounds that Microsoft's continued contempt of court shows a dangerously low probability of payment once appeals are exhausted.
Q. Is it really necessary that every Slashdot summary ends with a very polarizing question?
A. Yes. It's the sure-fire technique, developed over many years, to get people to post responses and not just click on the links. Haven't you noticed that the non-polarizing articles only get 50 or so replies, but the one that incite all-out flamewars can get several thousand?
...if people suddenly became liable for open access ports, zombies running on their machines, etc, the level of spam on the Internet would plummet overnight, bulk identity theft would virtually disappear, and secure protocols would be used in meaningful ways. This would not be a bad thing.
Seriously, though, the Open WAP seems to have been a failed attempt to prevent a search which, according to other posters, turned up a LOT of highly illegal material which had been identified as having been downloaded to that IP address. The only way this could be compared to a school shooter would be if a homocidal gunman decided to advertise that all their firearms were for rent in an effort to claim the police had no cause to believe the gunman was the one who fired them.
"Probable cause" isn't the best system possible. It gets far too many who are innocent, and lets off far too many who are guilty. It's certainly open to abuse, and there's plenty of evidence that it has been abused. I would love to see more effort put into devising a better system, rather than just patching an inherently flawed approach. Patches only get you so far before the bugs introduced by patching exceed the bugs fixed by the patches. However, in this case it seems to have been entirely applicable and correct.
You're right. They should rope up all the staff involved and force them to work as Walmart greeters for the next year. It's about time real fear was struck into the hearts of those responsible for protecting such information.
I've not seen much development there in the past - is it starting to pick up momentum?
EEG seems like a good candidate for improvement - typical EEG equipment in actual use seems to me to be somewhere around the 7-bit to 8-bit mark. Accurate analog-to-digital converters (as in: good enough for multi-billion-dollar nuclear experiments, where a mistake won't kill - at least, not until the customer has run out of torture techniques to play with) run up to 24-bit. The number of supported channels is generally very small - that can't really be improved on very much, but you could still probably double it without too much trouble.
Getting something certified medically - ah, now that's the killer. Even if you used components that were validated for medical use in deep space when next to a Type I supernova, even if you provided a formal design and formal proof of compliance with that design and all medical requirements, you would still be looking at dying of old age before the equipment was certified. Particularly as the established medical community is extremely closed, highly conservative and very rich as a result of their monopoly. You need only look at the stink raised by an Australian doctor's discovery that peptic ulcers could be cured by bismuth and an antibiotic. They damn near lynched him when he presented his paper in the US. Open Source developers are less public and far easier to hide. It's not as if we reach the surface often enough for anyone to know what any of us look like.
Mind you, that was also a SPARC, albeit a SPARC clone. The ESA developed a fully GPLed rad-hardened SPARC replacement, which I believe Atmel then took over. But they were the first.
The DoD switched to using public key encryption as an authentication scheme for virtually all their networks, what, gotta be four or five years ago now. I was finishing a contract position with SPAWAR at the time they started the migration. The encryption keys are unique to each individual and are embedded on their ID card. You couldn't so much as enter a building, log into a terminal or access your e-mail without using your personal digital certificate and personal encryption key.
In short, the DoD has all the infrastructure it needs to prevent unauthorized access without having to ban specific IP addresses. They simply invalidate the public key and certificate for whoever leaves. No more problems. I did say they had the infrastructure, but do they have the cognisance? Apparently not. The DoD failed their recent IT security inspection, if I remember the recent Slashdot story correctly -- would someone mind telling me HOW you can screw up a perfectly standard PKI-based authentication system so badly that (a) an entire department can get a failing grade, and (b) a person who has left for some time still has active certificates in the database?
Look, we all know that bureaucrats are total nutters, but do they all have to be so hellbent on being the nuttiest? Why not place Captain Sensible in charge of the DoD - he can't really do much worse, if the current lot aren't even following their own rules and regulations on IT security.
Oh, I dunno. Let's say I was delusional and lacked real world experience (gronda gronda). I'll bet the real world stuff I'm missing out on ain't half as much fun as the fractured reality in my brain he gets to miss out on.
You seem to be correct, and I don't recall me as an observer being happy when AMD bought ATI. Too big a bite at too dangerous a time. Hell, people talk about engineers being poor managers - seems like MBA managers make the worst managers, the engineers seem vastly superior. AMD is probably not dead yet - but as with the Monty Python sketch, that is something Intel can certainly arrange by continuing on the cost-cutting.
AMD has no hope to compete in a fair fight, and Intel are far better when it comes to unfair fights. So change the arena. AMD's only real hope is to keep producing entire new twists. Not stepwise refinements - entire new directions. That's not cheap, but neither is going bankrupt. AMD's only chance lies in keeping Intel wrong-footed. Intel can outpace AMD in a straight line and will squish it flat if that's the only direction that happens.
Slashdot Mirror
Points have a time-to-live of one year, extended by three months for every major maintenance cycle, two for moderate maintenance, and one if the updates are thrown off the back of a lorry at high speed.
Have three leagues. One league being major corporate entities, one being the smaller companies involved, and one for the genuine collective projects.
My guess is that IBM, SGI etc, would not care in the slightest, but I seriously doubt they'd refuse a plaque commemorating a successful year, either. The smaller companies - now, they might care. Publicity is the lifeblood of the small business, and this is easy free publicity - as well as steering them the right way. Collective projects SHOULDN'T care - if they do, their members have lost sight of the project and are focussed on the kudos. Bad mistake. Projects that do that are doomed, doomed I say!
For the last ten years, you've really been electing a bunch of PET 3032s, Apple Is and ZX-80s. The speech synthesis was by Superior Software and the suits by US Gold. Sometime in the next few months, we are due to be attacked by a large number of mutant camels, the road system already having degenerated into a maze of twisty passages, all alike.
(You have to exclude any commune, settlement or system that depends so much on the system it tries to escape from that the escape is merely an illusion. Interdependence is fine and perfectly respectable, and probably necessary for most such efforts, but it should never be permitted to slip into codependency - which has been the fate of many attempts.)
I'm not proposing that dissatisfied Slashdot geeks get together and build a self-sufficient University-City that can provide its own food, clothing, housing, power, books, etc. The task is not impossible, but the odds of Slashdot having enough sufficiently dissatisfied people with a sufficiently-broad range of skills over a sufficiently small distance to get them together.... It's not going to happen in my lifetime.
Ultimately, the majority of people most desiring of change have no voice and little money. If you expect drastic change to happen, it cannot require what these people don't have. Ergo, if you want change, it must be from the absolute ground up, to the point where the homeless and dispossessed can be valuable, and where there is little or nothing you could need that can't be provided in-house rather than being bought.
Many people have tried to found just such communal enterprises, but most have failed because it's not just not easy, it is unimaginably tough to start and never gets better than phenomenally hard, even after it is not just self-sufficient in all things but capable of "exporting" to the rest of society. Even the organization of such a community system is outside most people's ability, which is why you see most attempting to use charismatic leaders to hold things together, rather than people who know what they're doing. And which is why such communes usually end up in a bad way, or worse.
These problems are very very hard, but they are not unsolvable. If they were, nobody could ever have formed a civilized community in the first place. If you are sick and tired enough to clear the obstacles, solve the problems and keep things together, then nothing is stopping you starting your own civilization. But it really is at that level.
Yes, it's something to worry about, in the sense that nobody has any incentive to invent. The employees won't see a dime, if their bright ideas have to be handed over without question. Companies have no incentive, because they should be able to get just as good results for free. Besides, if they ask their new hires to innovate, the new hires will have to give all this neat new stuff to their former boss, not them. R&D has no value, when it is in nobody's interest to carry it out. In their pursuit of instant gratification and the "now" money, the people with business degrees are killing off the people with real knowledge. There is no long-term future for such a mindset. It consumes but never produces. In the end, it will starve itself and all around it to death. Those just graduating damn well should worry that there is a serious danger of there being no long-term future. Not just for a job, but for whole industries.
But I'd start with something like pchar, which will tell you the effective bandwidth at each hop on the network. That will tell you how severe the blockage is and, more importantly, where. It's not the "best" tool out there, but it's reasonably non-intrusive (unlike most stress-testing tools) and I've not seen any obvious problems with it at gig speeds. It does need patching for Linux, though. I sent a patch to the maintainer who has sworn he'll someday get around to including it. NetBSD has a faster network stack, though, and is more suitable for such tests. Which I hate, as I prefer Linux, but facts don't change themselves to suit a like.
In the end, hate is nothing more than a few chemicals swirling around in a brain. It has no meaning, it has no value, and most people who are in any way successful will have plenty of people who hate them. So will many who will never know success at all. Hate is commonplace.
What he had was money, power and influence - three things far more valuable than any mere emotion, and three things that protesters at the MPAA's excesses are unlikely to achieve to anything like the same degree.
I detest what this one man did, I detest what he stood for, but there's no use denying the fact that his was a story of success on a grandiose scale and that many people will try to emulate him for that very reason.
Why cash? Shouldn't it be American Expressed?
(It's not about protocols. If you wired up a network using X.25 or DECNet, fired up some obscure machine, loaded up an information server on it, then provided a gateway to regular TCP/IP and a proxy to HTTP, that machine would be as much "on the internet" as any other. Any definition has to allow for non-standard connections, or it's not a complete definition of the Internet that people use.)
Just tried the latest kernel and it hangs on trying to fire up the second ATA instance. Not even a kernel oops, nothing. That's true whether I use the vanilla kernel or Red Hat's RPM. Something is screwed up, and from the sounds of it, there's more than one of us experiencing a failure at the same point, so that would be the obvious suspect.
Far as I know, Slashdot UIDs are only sold at Christie's Auction House, and I think the 4-digit ones go for a couple of million, if they're in good condition and still in their original packaging. I heard rumors that the US DoD, in an effort to gain greater credibility with geeks, were planning an offer of a few billion on a 2-digit UID.
As for speed, you must bear in mind that the TOP speeds for a cable modem or xDSL line are comparable to a T1 line, but an optical T1 line can be uprated to an optical T3 line - or even a T4 line. (T2 and T4 lines are considered "teleco-only" but they do indeed exist.)
There are times a cable modem will run at the same speed as a 56K traditional modem - or slower. The traditional modem will be cheaper, not because of the speed you are running it at, but because of the speed you CAN run it at. The maximum potential is a function of the grade of the hardware used. Better grade hardware can always be run at slower speeds, but will still cost a lot simply because they are higher grade.
Frankly, I'd love it if I could afford a direct O8 connection to my home. But there's no possible way I could afford an O8, even if I only ever ran it at DSL speeds.
Trans: Yes, that's what I meant.
The downgrading of maths is bothersome, but so is the fact that children who are too bright are held back and stunted, in the name of "letting them be kids". (How about letting them be themselves?)
There are other issues, too. British Universities are generally pretty good at forcing people to think - you are expected to do far more hands-on research than at a typical US University - however, the examination system sucks. You get to see past papers, and are advised to look at the past two years. I got the past six years, for one course, and found that only two exam papers had been used in all that time. The lecturer just alternated between them, relying on students not to do their homework. Other exam papers were plain drivel, requiring more memorization than comprehension.
IMHO, examinations should be divided up something like: 40% subject comprehension, 30% ability to extrapolate and deduce within the subject, 20% ability to transfer subject skills, 10% rote memorization. We have information sources aplenty - rote has no purpose in today's society beyond a few corner cases. The ability to intelligently apply a skill - directly or indirectly - and to use the subject as a tool that is crafted around a problem-space, not a designation... These things can't be found on the Internet.
If the questioner is polarized, then the posting can only be overtly or covertly antagonistic to those of the opposite pole. If there is a question that is clearly a dig at the other side, then it's pretty overt. Otherwise, there's an excellent chance that the dig is still there but is worded too well to be clearly seen.
In fact, the EU should seriously consider whether it would be advisable to seize assets up to the value of the fines so far, on the grounds that Microsoft's continued contempt of court shows a dangerously low probability of payment once appeals are exhausted.
Q. Is it really necessary that every Slashdot summary ends with a very polarizing question? A. Yes. It's the sure-fire technique, developed over many years, to get people to post responses and not just click on the links. Haven't you noticed that the non-polarizing articles only get 50 or so replies, but the one that incite all-out flamewars can get several thousand?
Seriously, though, the Open WAP seems to have been a failed attempt to prevent a search which, according to other posters, turned up a LOT of highly illegal material which had been identified as having been downloaded to that IP address. The only way this could be compared to a school shooter would be if a homocidal gunman decided to advertise that all their firearms were for rent in an effort to claim the police had no cause to believe the gunman was the one who fired them.
"Probable cause" isn't the best system possible. It gets far too many who are innocent, and lets off far too many who are guilty. It's certainly open to abuse, and there's plenty of evidence that it has been abused. I would love to see more effort put into devising a better system, rather than just patching an inherently flawed approach. Patches only get you so far before the bugs introduced by patching exceed the bugs fixed by the patches. However, in this case it seems to have been entirely applicable and correct.
You're right. They should rope up all the staff involved and force them to work as Walmart greeters for the next year. It's about time real fear was struck into the hearts of those responsible for protecting such information.
EEG seems like a good candidate for improvement - typical EEG equipment in actual use seems to me to be somewhere around the 7-bit to 8-bit mark. Accurate analog-to-digital converters (as in: good enough for multi-billion-dollar nuclear experiments, where a mistake won't kill - at least, not until the customer has run out of torture techniques to play with) run up to 24-bit. The number of supported channels is generally very small - that can't really be improved on very much, but you could still probably double it without too much trouble.
Getting something certified medically - ah, now that's the killer. Even if you used components that were validated for medical use in deep space when next to a Type I supernova, even if you provided a formal design and formal proof of compliance with that design and all medical requirements, you would still be looking at dying of old age before the equipment was certified. Particularly as the established medical community is extremely closed, highly conservative and very rich as a result of their monopoly. You need only look at the stink raised by an Australian doctor's discovery that peptic ulcers could be cured by bismuth and an antibiotic. They damn near lynched him when he presented his paper in the US. Open Source developers are less public and far easier to hide. It's not as if we reach the surface often enough for anyone to know what any of us look like.
Mind you, that was also a SPARC, albeit a SPARC clone. The ESA developed a fully GPLed rad-hardened SPARC replacement, which I believe Atmel then took over. But they were the first.
In short, the DoD has all the infrastructure it needs to prevent unauthorized access without having to ban specific IP addresses. They simply invalidate the public key and certificate for whoever leaves. No more problems. I did say they had the infrastructure, but do they have the cognisance? Apparently not. The DoD failed their recent IT security inspection, if I remember the recent Slashdot story correctly -- would someone mind telling me HOW you can screw up a perfectly standard PKI-based authentication system so badly that (a) an entire department can get a failing grade, and (b) a person who has left for some time still has active certificates in the database?
Look, we all know that bureaucrats are total nutters, but do they all have to be so hellbent on being the nuttiest? Why not place Captain Sensible in charge of the DoD - he can't really do much worse, if the current lot aren't even following their own rules and regulations on IT security.
Oh, I dunno. Let's say I was delusional and lacked real world experience (gronda gronda). I'll bet the real world stuff I'm missing out on ain't half as much fun as the fractured reality in my brain he gets to miss out on.
That's scarily accurate. Do we work for the same company?
AMD has no hope to compete in a fair fight, and Intel are far better when it comes to unfair fights. So change the arena. AMD's only real hope is to keep producing entire new twists. Not stepwise refinements - entire new directions. That's not cheap, but neither is going bankrupt. AMD's only chance lies in keeping Intel wrong-footed. Intel can outpace AMD in a straight line and will squish it flat if that's the only direction that happens.
Tux and I will drive over with large quantities of booze and herrings. You sneak round the back with the schedulers.