The approach taken in several European countries is simply to not give any benefits to registered "religions" if the organizations aren't charitable. In the case of Scientology, several countries have refused to consider them charitable because they can't show audited accounts that clearly show they are not channeling profits anywhere.
Uhm, the primaries isn't the presidential election. Why do you have ANY expectation that your vote in the primaries will be binding? It's not a vote restricted to the members of the party selecting their candidate.
If you want your choice to count you should be a member of the party and vote for a chance in the system. Why should people that aren't a member of the party have ANY say in what candidates the party put forward for an election?
But then I'm not from the US - to me the entire system of primaries is bizarre to the extreme.
Line shredders stopped being enough ages ago. The Vietnamese for instance are known to have manually reassembled lots of documents that had been shredded as the last US officials were evacuated.
Unless you use a crosscut shredder all you are doing is raising the bar from casual snooping to requiring some dedication.
You ask "why is copyright law a bad thing for society", and then answers why it isn't a bad thing for the creator of a work.
And at the same time you pointed out exactly why it can be a bad thing for society if not properly balanced: It places restrictions on the public dissemination of works of the mind.
I'm not opposed to all use of copyright. However, the stated intent of modern copyright law is to promote the arts and sciences by making it attractive to publish new works. However, if the publics access to those works are severely restricted, and the ease of obtaining copyright (merely publishing something) reduces the volume of freely available works below what it would otherwise be, then one can legitimately ask the question at what point copyright, at least in it's current form, becomes a net negative for the public.
Personally I think a drastically reduced term (say 10 years), possibly combined with a requirement for renewal after an even shorter initial term, would be a huge improvement.
At the very least if you are going to call them names, use the right names.
Patent rights are government granted rights that re only granted to promote innovation for the benefit of their public. If Brazil isn't seeing the benefit, it's every bit within their right to take away the rights they granted.
There is no such thing as "intellectual property". Calling it property only perpetuates the fantasy of the rights holders that there is. There is a reason we have copyright law and patent laws - property law doesn't cover it. Copyrights and patents are just tools to promote innovation for the benefit of the public. If the public doesn't benefit, they've failed, and the government is very much within it's right to take action.
You can't steal something which isn't property. And you aren't violating patent rights when you are acting in the accordance with your laws.
Merck has no inherent right to patent protection. The ONLY right to patent protection they have in Brazil is that which is GRANTED THEM by the Brazilian parliament, which requires them to follow certain rules or face the consequences. They now have to face the consequences.
The other thing is that it's entirely up to Brazil whether or not they want their legal system to even recognize these patents in the first place. Patents aren't a "natural right" in any sense. As copyrights, patents are tools used by the government to encourage innovation for the benefit of the public. If they are being used in ways that harms the public, then it stands to reason that any government that cares about it's people would take action to correct it.
True. But at the same time that means that they are considering 20% on R&D the best factor to give them a good chance of earnings growth (if not, their management is incompetent for not spending more or less...). If they thought only 20% put the company at risk, then surely they'd spend more.
Viagra doesn't save lives. AIDS does. And plenty of drug companies are doing good business in Brazil. Merck just managed to demonstrate what happens when you try to fuck them over when there are lives at stake.
Since when did the drug companies give a fuck about tropical countries anyway? This is one of THE major obstacles in the fight against malaria, for example: The funds allocated to malaria research by drug companies is miniscule, because the main potential customer base doesn't have much money.
In large parts of Europe, the Democrats are considered right wing, and the Republicans are considered far right wing comic relief, though the political parties on the left wing in Europe generally support the Democrats as the lesser of two evils, and the conservative parties tend to support the more moderate parts of the Republican party.
Why would you want to do this? It does not buy you anything, as you can already set your system up to create LVM volumes of all your block devices and forget the non-LVM volumes exist. The existence of "raw" block devices that don't allow you access to the LVM features doesn't take away any functionality from you.
Yes you can. But it depends on the app developers how easy it is. They can statically link their apps, in which case there's no problem. They can dynamically link their apps against versions of the libraries included in the package, in which case there also is no problem. They can dynamically link against library versions in the default linker paths. If so, it depends on whether you have a package with the right library versions easily available.
In effect, you usually see the latter for software that is provided with source, since rebuilding is usually an option - when it's not you can install older versions of the libraries you need -, and the former variations are fairly common with closed source apps as it makes it easier to provide more generic packages.
More importantly, the right way of making it nice is to review the available API's, determine how to fix any problems with the API's to the various layers, and optionally wrap a nice utility around those API's.
Both Yum and Apt have been available for years, though I don't remember when they started including them in the base distro's. APT for RPM has been available since around 2000.
I have both Yum and Apt installed on my Fedora Core 3 box - haven't had any need to upgrade to a newer distro since the stuff I use it for is fairly basic. I had Apt since before I upgraded to Redhat 6.2. In fact, in a moment of masochism I used apt-get to upgrade to 6.2 from a previous version (5.2? can't remember) by pointing at a repository for one of the new versions - that was a bit on the risky side, though it worked despite having to re-run it a few times.
RHEL 3 and RHEL 4 both had up2date. If you bought either of them, then presumably you would be using up2date against RHN. If not, Apt have been available for Redhat for many years, and Yum a bit shorter.
So what are you complaining about? The only reason for you to "spending hours trying to figure out" package dependencies is if you're intentionally being obtuse and insisting on not using the tools available, and that's how it's been for years.
You apparently also either intentionally or out of ignorance keep comparing RHEL, which is meant specifically to be a slow moving distribution targeted at corporate users (who DO NOT want a high update frequency, and want predictable updates) with distro's with completely different goals. RHEL adopts things when they are proven. It on purpose lags far behind distro's like Fedora. If you can't deal with that, then don't use it.
As someone else has pointed out, when you do "yum install somepackage" you may not know what the size of the package is. Further, you also likely don't know what version yum will try to install. It may try to install a version you don't want, or it may only find a 32 bit package when you want a 64 bit package. Asking is a sensible way of making sure you don't accidentally hose the users install by "upgrading" to a version the user considers broken, or starting hours worth of downloads when the user thought he/she was asking for a small package.
If you don't like it, add the right command line option. For ordinary users, asking is likely to be the safer choice.
So what you are saying is that the layered interfaces aren't ideal. In other words, they might benefit from being refactored: Either you push some of the stuff ZFS wants to do down a layer and modify the interfaces accordingly, or you modify the interfaces to expose the capabilities ZFS need. You don't throw away the layering and put it all in one big blob of code.
ZFS solves this problem in two ways, both of which reuires the storage model to be part of the filesystem:
1. Each physical write never overwrites "live" data on the disk. It writes the stripe to a new location, and once it's been completely committed to disk the old data is marked as free.
2. ZFS uses variable stripe width, so that it does not have to write larger stripes than nescessary. In other words, a large write can be directly translated to a write to a large stripe on the sotrage system, and a smaller write can use a smaller stripe width. This can improve performance since it can reduce the amount of data written.
You claim this requires the storage model to be part of the filesystem, but it doesn't.
All it requires is sharing some minimal extra information, such as letting the block layer have a concept of free/used data blocks, and allow the filesystem to give hints about the strip width, for example. Sufficient information to deduce those values might even already be there with the existing API's (I don't know - I haven't looked much at that part of the Linux kernel).
If done right, other filesystems would be able to take advantage of it with little effort, instead of having to duplicate this functionality in every filesystem that wants it.
As for snapshotting, that also at most needs knowledge of which blocks are free and which are not - that's sufficient to expose a virtual block level device to the upper levels while doing copy-on-write transparently on the physical disk as long as there's enough space. You can even do it without understanding free space, but it means you need to preallocate some space for block copies which you can only reclaim if a snapshot gets deleted but that's usually ok for a lot of scenarios (such as using snapshots for hot backups).
You have three alternatives: apt-get, yum, or up2date. All three works. All three resolves dependencies for you and installs them if you want them to. All three are perfectly fine in a corporate environment unless your IT department is a bunch of retards. In fact, in a corporate environment, by preference would be to have a local Yum repository rather than depend on up2date against the Redhat servers (because I DO NOT WANT stuff to get updated without having verified it works in our environment first).
Besides, up2date supports Yum repositories too.
As for your complaints about Redhat Enterprise Linux, updates are slow on purpose. If you run mission critical systems, you don't want the world to keep changing under you. You don't want to be anywhere near the bleeding edge - you let other people bleed for you. At work we're now finally moving to RHEL 4. You might have noticed 5 has been released. We're not touching it until it has probably a year or so behind it. Even then, we have no reason to upgrade until getting updates of packages that work on RHEL 4 starts becoming a problem.
Slashdot Mirror
The approach taken in several European countries is simply to not give any benefits to registered "religions" if the organizations aren't charitable. In the case of Scientology, several countries have refused to consider them charitable because they can't show audited accounts that clearly show they are not channeling profits anywhere.
Uhm, the primaries isn't the presidential election. Why do you have ANY expectation that your vote in the primaries will be binding? It's not a vote restricted to the members of the party selecting their candidate.
But then I'm not from the US - to me the entire system of primaries is bizarre to the extreme.
Unless you use a crosscut shredder all you are doing is raising the bar from casual snooping to requiring some dedication.
I'm an atheist, and I don't hate any religions any more than I hate children for believing in santa claus and the tooth fairy.
And at the same time you pointed out exactly why it can be a bad thing for society if not properly balanced: It places restrictions on the public dissemination of works of the mind.
I'm not opposed to all use of copyright. However, the stated intent of modern copyright law is to promote the arts and sciences by making it attractive to publish new works. However, if the publics access to those works are severely restricted, and the ease of obtaining copyright (merely publishing something) reduces the volume of freely available works below what it would otherwise be, then one can legitimately ask the question at what point copyright, at least in it's current form, becomes a net negative for the public.
Personally I think a drastically reduced term (say 10 years), possibly combined with a requirement for renewal after an even shorter initial term, would be a huge improvement.
At the very least if you are going to call them names, use the right names.
Patent rights are government granted rights that re only granted to promote innovation for the benefit of their public. If Brazil isn't seeing the benefit, it's every bit within their right to take away the rights they granted.
There is no such thing as "intellectual property". Calling it property only perpetuates the fantasy of the rights holders that there is. There is a reason we have copyright law and patent laws - property law doesn't cover it. Copyrights and patents are just tools to promote innovation for the benefit of the public. If the public doesn't benefit, they've failed, and the government is very much within it's right to take action.
Merck has no inherent right to patent protection. The ONLY right to patent protection they have in Brazil is that which is GRANTED THEM by the Brazilian parliament, which requires them to follow certain rules or face the consequences. They now have to face the consequences.
Show me a single country that have stopped the spread of HIV, and I'll believe you. Problem is you can't.
You've "reversed" the number. The article says 7.9 people per cars, not 7.9 cars per person.
The other thing is that it's entirely up to Brazil whether or not they want their legal system to even recognize these patents in the first place. Patents aren't a "natural right" in any sense. As copyrights, patents are tools used by the government to encourage innovation for the benefit of the public. If they are being used in ways that harms the public, then it stands to reason that any government that cares about it's people would take action to correct it.
True. But at the same time that means that they are considering 20% on R&D the best factor to give them a good chance of earnings growth (if not, their management is incompetent for not spending more or less...). If they thought only 20% put the company at risk, then surely they'd spend more.
Viagra doesn't save lives. AIDS does. And plenty of drug companies are doing good business in Brazil. Merck just managed to demonstrate what happens when you try to fuck them over when there are lives at stake.
In large parts of Europe, the Democrats are considered right wing, and the Republicans are considered far right wing comic relief, though the political parties on the left wing in Europe generally support the Democrats as the lesser of two evils, and the conservative parties tend to support the more moderate parts of the Republican party.
Why would you want to do this? It does not buy you anything, as you can already set your system up to create LVM volumes of all your block devices and forget the non-LVM volumes exist. The existence of "raw" block devices that don't allow you access to the LVM features doesn't take away any functionality from you.
In effect, you usually see the latter for software that is provided with source, since rebuilding is usually an option - when it's not you can install older versions of the libraries you need -, and the former variations are fairly common with closed source apps as it makes it easier to provide more generic packages.
More importantly, the right way of making it nice is to review the available API's, determine how to fix any problems with the API's to the various layers, and optionally wrap a nice utility around those API's.
I have both Yum and Apt installed on my Fedora Core 3 box - haven't had any need to upgrade to a newer distro since the stuff I use it for is fairly basic. I had Apt since before I upgraded to Redhat 6.2. In fact, in a moment of masochism I used apt-get to upgrade to 6.2 from a previous version (5.2? can't remember) by pointing at a repository for one of the new versions - that was a bit on the risky side, though it worked despite having to re-run it a few times.
So what are you complaining about? The only reason for you to "spending hours trying to figure out" package dependencies is if you're intentionally being obtuse and insisting on not using the tools available, and that's how it's been for years.
You apparently also either intentionally or out of ignorance keep comparing RHEL, which is meant specifically to be a slow moving distribution targeted at corporate users (who DO NOT want a high update frequency, and want predictable updates) with distro's with completely different goals. RHEL adopts things when they are proven. It on purpose lags far behind distro's like Fedora. If you can't deal with that, then don't use it.
If you don't like it, add the right command line option. For ordinary users, asking is likely to be the safer choice.
So what you are saying is that the layered interfaces aren't ideal. In other words, they might benefit from being refactored: Either you push some of the stuff ZFS wants to do down a layer and modify the interfaces accordingly, or you modify the interfaces to expose the capabilities ZFS need. You don't throw away the layering and put it all in one big blob of code.
You claim this requires the storage model to be part of the filesystem, but it doesn't.
All it requires is sharing some minimal extra information, such as letting the block layer have a concept of free/used data blocks, and allow the filesystem to give hints about the strip width, for example. Sufficient information to deduce those values might even already be there with the existing API's (I don't know - I haven't looked much at that part of the Linux kernel).
If done right, other filesystems would be able to take advantage of it with little effort, instead of having to duplicate this functionality in every filesystem that wants it.
As for snapshotting, that also at most needs knowledge of which blocks are free and which are not - that's sufficient to expose a virtual block level device to the upper levels while doing copy-on-write transparently on the physical disk as long as there's enough space. You can even do it without understanding free space, but it means you need to preallocate some space for block copies which you can only reclaim if a snapshot gets deleted but that's usually ok for a lot of scenarios (such as using snapshots for hot backups).
Besides, up2date supports Yum repositories too.
As for your complaints about Redhat Enterprise Linux, updates are slow on purpose. If you run mission critical systems, you don't want the world to keep changing under you. You don't want to be anywhere near the bleeding edge - you let other people bleed for you. At work we're now finally moving to RHEL 4. You might have noticed 5 has been released. We're not touching it until it has probably a year or so behind it. Even then, we have no reason to upgrade until getting updates of packages that work on RHEL 4 starts becoming a problem.