I already have copyright on all possible combinations of 140 characters. You will soon be able to buy an anthology of Znork's collected works on Amazon. It'll be a few trucks and in really fine print, but for the true connaisseur it's definitely worth it.
A lot. When AM2 boards were new I went through a bunch of bad RAM (memory manufacturers hadn't quite gotten their act together yet) and RAM voltage would significantly change the number of bits that were 'bad'. 1.9 V and there were a few bits bad, 1.85, some more, 1.8 and memtest would light up all over.
So certainly, if any component is subpar, even a slight power fluctuation could trigger a borderline bad bit.
Yep, there isn't a big company available that would be capable of running Redhat; most of its value both to customers and community (and through that to shareholders) is derived from the strategic consistency and reliability. IE, the huge amounts of goodwill available to the people that make up and control the company. It's unlikely that that would survive through a takeover, and both customers and employees could easily move along to a derivative.
I hope and think IBM is smart enough to understand that Redhat is worth more to them as an independent than as an assimilated component; one might very well hire IBM consultants to work with Redhat related stuff, but buying an actual distro from them and relying on IBM support? Hmm. No.
Oracle might not be quite smart enough, but the amount of egg on Ellison's face after first starting his own Redhat ripoff, then buying Redhat, then losing those customers would probably be a bit too much to bear.
Microsoft? Well, Microsoft might buy them to put them down, but that'd be pointless due to the derivatives anyway.
In the end, Redhat is largely protected by the ultimate poison pill; the fact that the company intanglibles are it's main worth and part of them are already available to anyone who wants them for free, and the rest can walk any day. Anyone who thinks they could run a better Redhat is free to do it; if they don't think they can there's little purpose in paying good money for something whose value they'll probably lower.
Well, the good thing with Redhat is that as they've been very consistent with avoiding proprietary components, the employees could easily quit and start a new company off CentOS or something, and the customers would probably come along with them.
Redhat is rare in the computing field in the sense that the customers they have are their customers completely by choice. Many other companies wouldn't survive a day in that situation; their customers would slam the door with the sales guys head still inside.
The incidence of lab screwups with DNA is fairly high; I've seen numbers as high as 10%. There have been examples where lab procedures had the same sampler needle used for multiple samples, ie, first suspect DNA, followed by the crime scene sample. Leading, of course, to lots of crime scene samples testing positive for the suspect DNA.
So I have to agree with a previous poster, something like CSI: Botched Jobs with every episode ending with an innocent being put in jail while the perpetrator goes on to the next crime would be a valuable public service.
Blank media taxes are common elsewhere too. Not to mention that copyright in itself is fundamentally a delegated taxation form only called otherwise to keep control in the various publishing industries and for propaganda purposes.
Not that handling the issue as a taxation form is necessarily a bad solution; the simplest way to solve the whole economic part of the copyright issue would be to just slap a sales-tax of, say, 50% on any revenues made from copying/publishing/performing specific material with the proceeds going directly to the artists and then freeing up copying to whoever wants to copy anything.
Of course, that probably wont happen, as it would hand money to artists and songwriters instead of the media companies. Which makes it the exact opposite of what they want.
Don't let your backup system have access to your main system.
Ah, but this is a large company, and presumably they use 'enterprise' products. Some (most?) of those tend to be secured with rexec equivalent security (nsrexecd and.nsrhosts, coincidence eh?) for jobs executed from the backup server, and any better security is an afterthought. But they have good salesmen, and they're 'enterprise', and they cost a lot of money so they must be safe to use right...
The sad state of affairs is that at many large corporations it's probably as easy as sticking a laptop into the right network port, configuring the same IP address as the backup servers and you've got root access on every machine they back up, either through script facilities or through the simple capability of pushing a restore of a (completely unchecked) home-made passwd or any other file.
Perhaps it's changed in the last few years, but last I checked there were no host keys to authenticate servers and clients, there were at best simple passwords for the server to make the client do whatever it wanted and there were no checksums to verify that only previously backed up files got pushed to the client. I may also have missed some more capable system.
Then again, many large companies separate responsibility for backup and storage systems from system administration, and/or the people deciding what backup system to use have little experience with system security. Leading to exactly this kind of problem, with huge gaping security holes. Monitoring software is another one of those huge holes.
Preventing being hacked from the backup system is, of course, fairly trivial. The first part of which would be, don't use any backup system that creates huge security holes on your servers, or at least don't put the security hole, or 'backup client' on the servers.
For the second part I'm leaning towards multi-stage backups, particularly if you have separate areas of responsibility; have the server admins being responsible for delivering production data securely to an intermediate storage area (perhaps coalesce the data into a single compressed encrypted, signed file), then let the people responsible for backups dump it to tape/archive it.
It creates a whole other bunch of problems and potential issues of course (and solves another bunch of issues with backup clients that don't understand all file types, file systems, metadata, etc), and you may have to take a serious look at database backups and such, but in the choice between bad and some more bad, at least such a solution won't lube you up for a penetration, it'll be fairly cheap (licenses vs. diskspace), it gives you much flexibility on how to backup which data, and it should keep your data safe. It may cost more man hours on the system side, but it'll probably save them on the storage side with fewer backup clients. The joy the first time you get to do a several-minute restore instead of getting an 'oops-saved-the-data-on-every-tape-in-the-silo' experience waiting a day for less than a hundred gigabytes of small files getting restored will just be another bonus.
a couple of 30 second adverts an hour is a small price to pay.
Don't worry, that wont last. As these services start showing any profitability the music industry will raise the prices and you get either services as expensive as buying the music (spotify arguably already costs as much as emusic for various levels of consumption), or you'll get mostly ads.
See, the fundamental issue is that the music industry is not at all interested in you choosing what music to play. In most places the division of the protection racket money extorted out of resturants, shops, bars, clubs and other venues is decided by radio statistics, and streaming is close to that. With radio, the selection of what to play is largely made by a few companies, through middle-men where legally necessary, and through social and marketing pressure (parties, freebies, etc, the stuff that makes music expensive). IE, the large companies decide who gets that money. Needless to say, it's not the long-tail artists.
If people suddenly start deciding what music to play on their own, that entire revenue stream would risk falling into the broader independent segments and people not beholden to the RIAA corps might get their fair share. Not a desirable outcome from the RIAA corps side.
Bands are going to have to go back to making money off playing live
It's not about the bands. It's never been about the bands. Apart from a few, most have always made any money they get from the live side, the rest gets taken by the label.
So the main interest of the corporations is how they get to continue taking most of the money, and as one part of that is having control over what gets played, that's one thing they care strongly about. And as they control most of the music interest organizations, that's the line that'll continue being driven.
If he's just going to experiment with RAC and virtualization there won't be any serious performance issues on any modern hardware, so he doesn't even really need a dual core.
As his requirements sounds mostly like he needs something cheap that will do the job, I'd recommend he go with a few AM2+ motherboards (ASUS MB, 4 memory slots), some cheap AMD X2 CPU, and throw 8 GB memory in each.
For the shared storage there's either iSCSI (IET, iSCSI enterprise target) or drbd (unbeatable solution for shared-disk semantics on two-node clusters).
It wont be a high-performance solution, but if experimenting with clustering and shared disks I'd rather have several cheap machines than one expensive one, even if I planned to virtualize the nodes.
It's not so much one-man-one-vote that does it, but districted first-past-post winner takes all that results in two party systems.
Proportional representation systems tend to work slightly better for allowing politics to shift according to voters preferences. But considering that such a gain for democracy would come at a price for the players in the current system it's unlikely to happen; it's simply more profitable both for politicians and lobbyists to have only two choices available.
Of course, today it often works out like this: Company A pays Xa to market a trademark. Company A needs something to sell to recoup marketing costs, and know their campaign generate an average product purchase rate of Za. Company A buys generic goods from production company B for Ya. Company A sets price of product to (Xa/Za)+Ya.
Well, that often comes along with it, but in principle it's not necessary unless the original design cost is so low that it is close to the design cost of copying.
Amortized of the series we're talking about design cost is negligible, even if you actually do it inhouse; design isn't a very people-intensive area. Marketing, lobbying and channel control would be far more expensive.
I suspect trademarks are only slightly less damaging to the economy than patents and copyright, but at least their damage isn't infinite as far as to preventing others from distributing similar goods.
There are also reasonable ways to counter the worst of the damage; one could require goods to be designer, producer and product ID marked to be protected by trademarks, ie, if the trademark owner outsources design and production, then customers can buy the exact same unmarked good by buying no-name product ID referenced goods, basically bypassing the scam.
There may be some extra reserves held against further depreciation of assets, but investments in solid assets could certainly be made (and are made in, for example, treasuries) without jeopardizing the reserve requirements. The more likely reason that most banks don't lend much money these days is that nobody who wants loans is qualified and nobody who's qualified wants loans.
What 'innocents'? The investment banks who didn't research what they did? The fund managers who wanted higher bonuses and stuck their clients money in more risky papers? The speculators who bought high-yield papers without caring why they were higher yield? The supposedly qualified traders who ignored counterparty risks? The average layman who had money in these companies when the half the internet was shouting about these companies being _dead_ and who apparently learned nothing from the dotcom crash?
It certainly wasn't hard to see exactly what was going to happen, and there were warnings everywhere. It's not like it came as a big surprise for anyone who actually cared at all what happened to their money. Of course, for those who were managing someone elses money there may not have been that much reason to care.
So why should the actual innocents (those who didn't participate in the greed carousel, and who learned their lesson the last time) get the crushing instead? They didn't play the game for the upside, why should they get the downside?
This isn't a choice between innocent getting crushed or not. With the bailout the innocent and the perhaps slightly less guilty (or merely unqualified) get to pay and the guilty get away scot free. Without a bailout the slightly less guilty still get to pay, while the thoroughly guilty parties get burned and may actually change somewhat (or at least not remain in control so they can go on and sabotage more companies).
Wipe these companies out. The fratboy and yes-men culture of these institutions make them a liability for the whole financial system of the world, and we simply cannot afford to let them take us all for any more rides.
What's stopping you? Mine does. Or, to be fair, my server cabinet does, but that's nothing but PC components.
multi-gigabit switch built in
Fair enough. Technically you could stick it in a PC if you wanted to, but 10GBE tech is unfortunately still a tad expensive. Altho that applies wherever you put it; it's not a good price/performance proposition for most cases besides interconnects.
Well, one could argue that Cell is IBM and seriously multi-threaded, but true, they don't have it in a general purpose server line.
As for memory, sure those expensive servers have DDR2 memory and slower at that than what I have in my desktop, but it's very expensive DDR2 memory. And, um, the slot is different.
IBM is able to pick up a good company with solid products
Are there any specific products that Sun sells that IBM doesn't have equivalents of? Sun has some good products, but I'm not sure IBM is after any specific products rather than just buying customers in certain segments and getting rid of some competition as a bonus at a fairly good price.
a good long-term strategy
Sun has a long-term strategy? Not one that's the long-term strategy of the month, but something, eh, more long-term? Having worked with Sun stuff for more than a decade, that's one of the more irritating habits the company has; sudden changes in strategy, often accompanied with a total re-branding of large parts of their product series.
Depends on what you mean with standard PC equipment.
Very low power can be accomplished with, for example, VIA or Atom. You get fast recovery by stripping a linux dist down to the initrd plus whatever you actually need, or possibly even a pre-hibernated image if you're stateless. Stability is best accomplished by using standardized parts tested by millions before you (frankly, I'd say standard PC equipment tends to be significantly more stable than proprietary hardware, but that's just my experience). Less moving parts is solved by moving to SSD disk and using passive cooling and appropriate components.
The equipment that falls into the standard PC range is amazingly wide these days and certainly includes embedded-equivalent systems.
If you go 'right over the head of' the server, network and storage people there's nobody who can tell the CIO that he's paying a 5000% markup on a PC with Linux. It's a great sales strategy.
It doesn't sound much different this time; 'by cramming computer power into the very box that contains storage capacity and the networking'. Uh-huh. Also known as 'a PC'.
I seem to recall a time when 'enterprise' used to mean somewhat of an improvement over what I have at home. These days, not so much.
That's a matter of debate; power in the US congress tends to get wielded by people supported by less than 20% of eligible voters. That's on par with what former one party system rulers tend to get when countries move to democracy.
We should be blaming ourselves for electing politicians
With first past post systems there isn't enough choice to put much blame on the voters. With only two or three realistic choices any candidates with a chance can easily be bought, and any unbuyable candidate smeared enough to not make it.
Perhaps the blame should be for not applying enough tar and feathers to get the system fundamentally changed...
I think those four may be the only people with standing to sue.
Correct me if I'm wrong, but wouldn't the rest of the kernel devs also have standing (for the distribution of the rest of the kernel, not the vfat module specifically)? If the vfat module itself becomes 'encumbered' in TomToms opinion, wouldn't they technically be obliged to distribute it separately from the kernel, ie, the old proprietary driver issue; they can distribute one or the other but not both together?
It's certainly one of those more tricky areas, and as far as I know there hasn't been any specific enforcement, but IIRC I've heard arguments towards that end around the general issue of proprietary drivers distributed together with the kernel.
This is a fairly general rule; any policy that artificially inflates prices beyond competitive market price creates a profit opportunity for anyone willing to violate the policy.
This is true whether the subject is a desired product that is completely forbidden (drugs, for example), a highly taxed item (alcohol, tobacco or even gasoline in places) or an artificially imposed monopoly pricing right such as copyrights and patents.
As the profitability is a function of enforcement, stricter enforcement merely leads to higher profitability. The only ways to actually remove the illicit profitability is by reducing demand or lowering the difference between production cost and end-user pricing.
Reducing demand is, perhaps, possible and desirable in some cases, such as drugs. With media on the other hand you hardly want to limit demand, and lowering the price seems pretty much the realm of p2p sharing.
Either way, it's the artificially inflated pricing that's creating the profit opportunity, and as long as that remains, yes, criminals will profit from it, so it's pointless to whine about it and at the same time support the inflated pricing. Switching the incentives to a levy on final sales going to the creators would be a much saner approach, and it would also lower the difference between legal and illicit profitability by cutting out the margins of anyone between.
Your problem is that you effectively argue for abolition of copyright and patents
Well, no, and I should perhaps been more clear about that, but this was more regarding marketing as it pertains to monopoly protected sectors.
I'm arguing for the abolition of the _monopoly_ aspect of copyright and patents, to be replaced with a _revenue right_ aspect for such materials.
It's the monopoly aspect that is responsible for close to every problem with both copyright and patents, while the redeeming feature is, as you say, revenue (altho a pitiful percentage of it) being directed towards desirable activity.
Perhaps I'm missing the point you're trying to make
Sort of. I'm objecting to the particular implementation that increases the value of marketing to the disadvantage of the incentive to the creators.
then you're going to have to come up with a workable alternative.
Usually I advocate the easiest option; an end-point levy on revenue similar to what's done with radio, but over all end-sales of material. Ie, put a 25-75% 'sales tax' on books, music, games, etc, and hand that money directly to creators and artists. Let the bookstores, music shops, etc, copy for what they're worth, but the creator gets their share, effectively bypassing the whole problem area of todays system (owner-distributors-marketers-channel).
My best bet would be we'd get prices at between 10-50% of todays (indicated by subscription type systems ranging from bookclubs to emusic to radio to cable tv), with more total revenue going to creators and efficiency of transfer of spent funds increasing from 5-10% today up towards 70%.
Your saying that the interests of Big Advertising can somehow outweigh
Not at all. I'm saying advertising is very effective, but most effective when you have more and better than others who are advertising.
Perhaps marketing costs so much because it works so well.
It costs so much because others advertising works so well. And vice versa. It essentially becomes a self-sustaining increasing spiral.
Firstly, there is competition in copyright systems.
Read up on the concept of monopolistic competition. The broadest kinds of competition you'll always find, but that's a far step from free market direct competition. Monopolistic competition does not lower prices or create more efficient production; it'll follow lockstep with consumer disposable income.
then the creator is not going to get any money
Not necessarily. The easiest model I usually suggest is simply levying a sales-tax off any end sales, pretty similar to the model for radio, with a cut of the proceeds from any revenue generated from duplication going directly to the creators. A model that easily bypasses the whole problem with advertising, as the maximum revenue derived in channel becomes set in competitive terms again.
It's not a monopoly, yet it still "suffers" from the same problem.
Last I checked Coca Cola was trademarked, so you may want to rethink that example. But at least trademarks don't prevent substitutes so they're not exactly that damaging.
I think you're blowing this all way out of proportion
I think you misunderstand my objection. I have nothing against marketing in itself. It's the combination with monopoly rights that I have a problem with.
If you accept the most naive explanation for copyright and patents, their sole intent is to encourage production of more useful arts and inventions. If you look at the macroeconomic effects of their implementation, their function is to take money from other places in the economy and funnel into creation and research, essentially a form of taxation and subsidizing system. That is, again, fine with me.
But then when you analyze the financial flows you realize that as much of what is essentially taxpayers money (by merit of the state legislating and enforcing the incentive system) is spent on marketing as on R&D. These systems have a stated end, and when more money goes elsewhere than towards that end it's not a good thing. In some it's merely sad, but in others like pharmaceuticals, it means our money is spent on irrelevant byproducts (hopefully people aren't getting sick from marketing, so marketing isn't what drives sales there) when the primary end goal is a very important one.
Slashdot Mirror
I already have copyright on all possible combinations of 140 characters. You will soon be able to buy an anthology of Znork's collected works on Amazon. It'll be a few trucks and in really fine print, but for the true connaisseur it's definitely worth it.
A lot. When AM2 boards were new I went through a bunch of bad RAM (memory manufacturers hadn't quite gotten their act together yet) and RAM voltage would significantly change the number of bits that were 'bad'. 1.9 V and there were a few bits bad, 1.85, some more, 1.8 and memtest would light up all over.
So certainly, if any component is subpar, even a slight power fluctuation could trigger a borderline bad bit.
Yep, there isn't a big company available that would be capable of running Redhat; most of its value both to customers and community (and through that to shareholders) is derived from the strategic consistency and reliability. IE, the huge amounts of goodwill available to the people that make up and control the company. It's unlikely that that would survive through a takeover, and both customers and employees could easily move along to a derivative.
I hope and think IBM is smart enough to understand that Redhat is worth more to them as an independent than as an assimilated component; one might very well hire IBM consultants to work with Redhat related stuff, but buying an actual distro from them and relying on IBM support? Hmm. No.
Oracle might not be quite smart enough, but the amount of egg on Ellison's face after first starting his own Redhat ripoff, then buying Redhat, then losing those customers would probably be a bit too much to bear.
Microsoft? Well, Microsoft might buy them to put them down, but that'd be pointless due to the derivatives anyway.
In the end, Redhat is largely protected by the ultimate poison pill; the fact that the company intanglibles are it's main worth and part of them are already available to anyone who wants them for free, and the rest can walk any day. Anyone who thinks they could run a better Redhat is free to do it; if they don't think they can there's little purpose in paying good money for something whose value they'll probably lower.
Well, the good thing with Redhat is that as they've been very consistent with avoiding proprietary components, the employees could easily quit and start a new company off CentOS or something, and the customers would probably come along with them.
Redhat is rare in the computing field in the sense that the customers they have are their customers completely by choice. Many other companies wouldn't survive a day in that situation; their customers would slam the door with the sales guys head still inside.
The incidence of lab screwups with DNA is fairly high; I've seen numbers as high as 10%. There have been examples where lab procedures had the same sampler needle used for multiple samples, ie, first suspect DNA, followed by the crime scene sample. Leading, of course, to lots of crime scene samples testing positive for the suspect DNA.
So I have to agree with a previous poster, something like CSI: Botched Jobs with every episode ending with an innocent being put in jail while the perpetrator goes on to the next crime would be a valuable public service.
Blank media taxes are common elsewhere too. Not to mention that copyright in itself is fundamentally a delegated taxation form only called otherwise to keep control in the various publishing industries and for propaganda purposes.
Not that handling the issue as a taxation form is necessarily a bad solution; the simplest way to solve the whole economic part of the copyright issue would be to just slap a sales-tax of, say, 50% on any revenues made from copying/publishing/performing specific material with the proceeds going directly to the artists and then freeing up copying to whoever wants to copy anything.
Of course, that probably wont happen, as it would hand money to artists and songwriters instead of the media companies. Which makes it the exact opposite of what they want.
Don't let your backup system have access to your main system.
Ah, but this is a large company, and presumably they use 'enterprise' products. Some (most?) of those tend to be secured with rexec equivalent security (nsrexecd and .nsrhosts, coincidence eh?) for jobs executed from the backup server, and any better security is an afterthought. But they have good salesmen, and they're 'enterprise', and they cost a lot of money so they must be safe to use right...
The sad state of affairs is that at many large corporations it's probably as easy as sticking a laptop into the right network port, configuring the same IP address as the backup servers and you've got root access on every machine they back up, either through script facilities or through the simple capability of pushing a restore of a (completely unchecked) home-made passwd or any other file.
Perhaps it's changed in the last few years, but last I checked there were no host keys to authenticate servers and clients, there were at best simple passwords for the server to make the client do whatever it wanted and there were no checksums to verify that only previously backed up files got pushed to the client. I may also have missed some more capable system.
Then again, many large companies separate responsibility for backup and storage systems from system administration, and/or the people deciding what backup system to use have little experience with system security. Leading to exactly this kind of problem, with huge gaping security holes. Monitoring software is another one of those huge holes.
Preventing being hacked from the backup system is, of course, fairly trivial. The first part of which would be, don't use any backup system that creates huge security holes on your servers, or at least don't put the security hole, or 'backup client' on the servers.
For the second part I'm leaning towards multi-stage backups, particularly if you have separate areas of responsibility; have the server admins being responsible for delivering production data securely to an intermediate storage area (perhaps coalesce the data into a single compressed encrypted, signed file), then let the people responsible for backups dump it to tape/archive it.
It creates a whole other bunch of problems and potential issues of course (and solves another bunch of issues with backup clients that don't understand all file types, file systems, metadata, etc), and you may have to take a serious look at database backups and such, but in the choice between bad and some more bad, at least such a solution won't lube you up for a penetration, it'll be fairly cheap (licenses vs. diskspace), it gives you much flexibility on how to backup which data, and it should keep your data safe. It may cost more man hours on the system side, but it'll probably save them on the storage side with fewer backup clients. The joy the first time you get to do a several-minute restore instead of getting an 'oops-saved-the-data-on-every-tape-in-the-silo' experience waiting a day for less than a hundred gigabytes of small files getting restored will just be another bonus.
a couple of 30 second adverts an hour is a small price to pay.
Don't worry, that wont last. As these services start showing any profitability the music industry will raise the prices and you get either services as expensive as buying the music (spotify arguably already costs as much as emusic for various levels of consumption), or you'll get mostly ads.
See, the fundamental issue is that the music industry is not at all interested in you choosing what music to play. In most places the division of the protection racket money extorted out of resturants, shops, bars, clubs and other venues is decided by radio statistics, and streaming is close to that. With radio, the selection of what to play is largely made by a few companies, through middle-men where legally necessary, and through social and marketing pressure (parties, freebies, etc, the stuff that makes music expensive). IE, the large companies decide who gets that money. Needless to say, it's not the long-tail artists.
If people suddenly start deciding what music to play on their own, that entire revenue stream would risk falling into the broader independent segments and people not beholden to the RIAA corps might get their fair share. Not a desirable outcome from the RIAA corps side.
Bands are going to have to go back to making money off playing live
It's not about the bands. It's never been about the bands. Apart from a few, most have always made any money they get from the live side, the rest gets taken by the label.
So the main interest of the corporations is how they get to continue taking most of the money, and as one part of that is having control over what gets played, that's one thing they care strongly about. And as they control most of the music interest organizations, that's the line that'll continue being driven.
Same here, streaming doesn't appeal to me, I barely ever use the feature. I use it in combination with emusic to get suggestions on what to download.
If he's just going to experiment with RAC and virtualization there won't be any serious performance issues on any modern hardware, so he doesn't even really need a dual core.
As his requirements sounds mostly like he needs something cheap that will do the job, I'd recommend he go with a few AM2+ motherboards (ASUS MB, 4 memory slots), some cheap AMD X2 CPU, and throw 8 GB memory in each.
For the shared storage there's either iSCSI (IET, iSCSI enterprise target) or drbd (unbeatable solution for shared-disk semantics on two-node clusters).
It wont be a high-performance solution, but if experimenting with clustering and shared disks I'd rather have several cheap machines than one expensive one, even if I planned to virtualize the nodes.
It's not so much one-man-one-vote that does it, but districted first-past-post winner takes all that results in two party systems.
Proportional representation systems tend to work slightly better for allowing politics to shift according to voters preferences. But considering that such a gain for democracy would come at a price for the players in the current system it's unlikely to happen; it's simply more profitable both for politicians and lobbyists to have only two choices available.
Of course, today it often works out like this: Company A pays Xa to market a trademark. Company A needs something to sell to recoup marketing costs, and know their campaign generate an average product purchase rate of Za. Company A buys generic goods from production company B for Ya. Company A sets price of product to (Xa/Za)+Ya.
Well, that often comes along with it, but in principle it's not necessary unless the original design cost is so low that it is close to the design cost of copying.
Amortized of the series we're talking about design cost is negligible, even if you actually do it inhouse; design isn't a very people-intensive area. Marketing, lobbying and channel control would be far more expensive.
I suspect trademarks are only slightly less damaging to the economy than patents and copyright, but at least their damage isn't infinite as far as to preventing others from distributing similar goods.
There are also reasonable ways to counter the worst of the damage; one could require goods to be designer, producer and product ID marked to be protected by trademarks, ie, if the trademark owner outsources design and production, then customers can buy the exact same unmarked good by buying no-name product ID referenced goods, basically bypassing the scam.
Legally, many banks cannot loan money at this time.
This hasn't been true for some while if you look at the reserve statistics: http://research.stlouisfed.org/fred2/series/BOGNONBR
There may be some extra reserves held against further depreciation of assets, but investments in solid assets could certainly be made (and are made in, for example, treasuries) without jeopardizing the reserve requirements. The more likely reason that most banks don't lend much money these days is that nobody who wants loans is qualified and nobody who's qualified wants loans.
What 'innocents'? The investment banks who didn't research what they did? The fund managers who wanted higher bonuses and stuck their clients money in more risky papers? The speculators who bought high-yield papers without caring why they were higher yield? The supposedly qualified traders who ignored counterparty risks? The average layman who had money in these companies when the half the internet was shouting about these companies being _dead_ and who apparently learned nothing from the dotcom crash?
It certainly wasn't hard to see exactly what was going to happen, and there were warnings everywhere. It's not like it came as a big surprise for anyone who actually cared at all what happened to their money. Of course, for those who were managing someone elses money there may not have been that much reason to care.
So why should the actual innocents (those who didn't participate in the greed carousel, and who learned their lesson the last time) get the crushing instead? They didn't play the game for the upside, why should they get the downside?
This isn't a choice between innocent getting crushed or not. With the bailout the innocent and the perhaps slightly less guilty (or merely unqualified) get to pay and the guilty get away scot free. Without a bailout the slightly less guilty still get to pay, while the thoroughly guilty parties get burned and may actually change somewhat (or at least not remain in control so they can go on and sabotage more companies).
Wipe these companies out. The fratboy and yes-men culture of these institutions make them a liability for the whole financial system of the world, and we simply cannot afford to let them take us all for any more rides.
doesn't have a SAN
What's stopping you? Mine does. Or, to be fair, my server cabinet does, but that's nothing but PC components.
multi-gigabit switch built in
Fair enough. Technically you could stick it in a PC if you wanted to, but 10GBE tech is unfortunately still a tad expensive. Altho that applies wherever you put it; it's not a good price/performance proposition for most cases besides interconnects.
Well, one could argue that Cell is IBM and seriously multi-threaded, but true, they don't have it in a general purpose server line.
As for memory, sure those expensive servers have DDR2 memory and slower at that than what I have in my desktop, but it's very expensive DDR2 memory. And, um, the slot is different.
IBM is able to pick up a good company with solid products
Are there any specific products that Sun sells that IBM doesn't have equivalents of? Sun has some good products, but I'm not sure IBM is after any specific products rather than just buying customers in certain segments and getting rid of some competition as a bonus at a fairly good price.
a good long-term strategy
Sun has a long-term strategy? Not one that's the long-term strategy of the month, but something, eh, more long-term? Having worked with Sun stuff for more than a decade, that's one of the more irritating habits the company has; sudden changes in strategy, often accompanied with a total re-branding of large parts of their product series.
Depends on what you mean with standard PC equipment.
Very low power can be accomplished with, for example, VIA or Atom. You get fast recovery by stripping a linux dist down to the initrd plus whatever you actually need, or possibly even a pre-hibernated image if you're stateless. Stability is best accomplished by using standardized parts tested by millions before you (frankly, I'd say standard PC equipment tends to be significantly more stable than proprietary hardware, but that's just my experience). Less moving parts is solved by moving to SSD disk and using passive cooling and appropriate components.
The equipment that falls into the standard PC range is amazingly wide these days and certainly includes embedded-equivalent systems.
If you go 'right over the head of' the server, network and storage people there's nobody who can tell the CIO that he's paying a 5000% markup on a PC with Linux. It's a great sales strategy.
It doesn't sound much different this time; 'by cramming computer power into the very box that contains storage capacity and the networking'. Uh-huh. Also known as 'a PC'.
I seem to recall a time when 'enterprise' used to mean somewhat of an improvement over what I have at home. These days, not so much.
That's a matter of debate; power in the US congress tends to get wielded by people supported by less than 20% of eligible voters. That's on par with what former one party system rulers tend to get when countries move to democracy.
We should be blaming ourselves for electing politicians
With first past post systems there isn't enough choice to put much blame on the voters. With only two or three realistic choices any candidates with a chance can easily be bought, and any unbuyable candidate smeared enough to not make it.
Perhaps the blame should be for not applying enough tar and feathers to get the system fundamentally changed...
I think those four may be the only people with standing to sue.
Correct me if I'm wrong, but wouldn't the rest of the kernel devs also have standing (for the distribution of the rest of the kernel, not the vfat module specifically)? If the vfat module itself becomes 'encumbered' in TomToms opinion, wouldn't they technically be obliged to distribute it separately from the kernel, ie, the old proprietary driver issue; they can distribute one or the other but not both together?
It's certainly one of those more tricky areas, and as far as I know there hasn't been any specific enforcement, but IIRC I've heard arguments towards that end around the general issue of proprietary drivers distributed together with the kernel.
This is a fairly general rule; any policy that artificially inflates prices beyond competitive market price creates a profit opportunity for anyone willing to violate the policy.
This is true whether the subject is a desired product that is completely forbidden (drugs, for example), a highly taxed item (alcohol, tobacco or even gasoline in places) or an artificially imposed monopoly pricing right such as copyrights and patents.
As the profitability is a function of enforcement, stricter enforcement merely leads to higher profitability. The only ways to actually remove the illicit profitability is by reducing demand or lowering the difference between production cost and end-user pricing.
Reducing demand is, perhaps, possible and desirable in some cases, such as drugs. With media on the other hand you hardly want to limit demand, and lowering the price seems pretty much the realm of p2p sharing.
Either way, it's the artificially inflated pricing that's creating the profit opportunity, and as long as that remains, yes, criminals will profit from it, so it's pointless to whine about it and at the same time support the inflated pricing. Switching the incentives to a levy on final sales going to the creators would be a much saner approach, and it would also lower the difference between legal and illicit profitability by cutting out the margins of anyone between.
(Additional; I should have replied to this above)
Ah, missed this :)
Your problem is that you effectively argue for abolition of copyright and patents
Well, no, and I should perhaps been more clear about that, but this was more regarding marketing as it pertains to monopoly protected sectors.
I'm arguing for the abolition of the _monopoly_ aspect of copyright and patents, to be replaced with a _revenue right_ aspect for such materials.
It's the monopoly aspect that is responsible for close to every problem with both copyright and patents, while the redeeming feature is, as you say, revenue (altho a pitiful percentage of it) being directed towards desirable activity.
Perhaps I'm missing the point you're trying to make
Sort of. I'm objecting to the particular implementation that increases the value of marketing to the disadvantage of the incentive to the creators.
then you're going to have to come up with a workable alternative.
Usually I advocate the easiest option; an end-point levy on revenue similar to what's done with radio, but over all end-sales of material. Ie, put a 25-75% 'sales tax' on books, music, games, etc, and hand that money directly to creators and artists. Let the bookstores, music shops, etc, copy for what they're worth, but the creator gets their share, effectively bypassing the whole problem area of todays system (owner-distributors-marketers-channel).
My best bet would be we'd get prices at between 10-50% of todays (indicated by subscription type systems ranging from bookclubs to emusic to radio to cable tv), with more total revenue going to creators and efficiency of transfer of spent funds increasing from 5-10% today up towards 70%.
Your saying that the interests of Big Advertising can somehow outweigh
Not at all. I'm saying advertising is very effective, but most effective when you have more and better than others who are advertising.
Perhaps marketing costs so much because it works so well.
It costs so much because others advertising works so well. And vice versa. It essentially becomes a self-sustaining increasing spiral.
Firstly, there is competition in copyright systems.
Read up on the concept of monopolistic competition. The broadest kinds of competition you'll always find, but that's a far step from free market direct competition. Monopolistic competition does not lower prices or create more efficient production; it'll follow lockstep with consumer disposable income.
then the creator is not going to get any money
Not necessarily. The easiest model I usually suggest is simply levying a sales-tax off any end sales, pretty similar to the model for radio, with a cut of the proceeds from any revenue generated from duplication going directly to the creators. A model that easily bypasses the whole problem with advertising, as the maximum revenue derived in channel becomes set in competitive terms again.
It's not a monopoly, yet it still "suffers" from the same problem.
Last I checked Coca Cola was trademarked, so you may want to rethink that example. But at least trademarks don't prevent substitutes so they're not exactly that damaging.
I think you're blowing this all way out of proportion
I think you misunderstand my objection. I have nothing against marketing in itself. It's the combination with monopoly rights that I have a problem with.
If you accept the most naive explanation for copyright and patents, their sole intent is to encourage production of more useful arts and inventions. If you look at the macroeconomic effects of their implementation, their function is to take money from other places in the economy and funnel into creation and research, essentially a form of taxation and subsidizing system. That is, again, fine with me.
But then when you analyze the financial flows you realize that as much of what is essentially taxpayers money (by merit of the state legislating and enforcing the incentive system) is spent on marketing as on R&D. These systems have a stated end, and when more money goes elsewhere than towards that end it's not a good thing. In some it's merely sad, but in others like pharmaceuticals, it means our money is spent on irrelevant byproducts (hopefully people aren't getting sick from marketing, so marketing isn't what drives sales there) when the primary end goal is a very important one.