Wrong. Linux is nowhere near as IPv6-friendly as the *BSDs. To enable IPv6 in FreeBSD, for example, put 'ipv6_enable="YES"' in/etc/rc.conf
Each FreeBSD CD comes with a bunch prebuilt IPv6-ready apps, like apache, wget, etc -- apps that don't have native IPv6 support.
Wrong yourself. On RedHat 9 for example, to enable IPv6 all you do is type "modprobe ipv6' as root. Rebooting is NOT required.
Most Linux distros contain a LOT of IPv6 support. RedHat comes with IPv6 enabled through a kernel module, and a large number of packages that include IPv6 support out of the box. Examples include Apache, ping6, iproute6, traceroute6, and so on.
The following page describe Linux IPv6 support in a variety of distros.
The whole concept of the 'internet' by means of Tcp/IP is becoming quite dated.
For example, a sample address would be Joe Smith@Earthlink@USA (users within the USA can leave the @USA blank)
this eliminates the need for a domain naming system. takes a lot of power away from ICANN, would help to solve cybersquatting, and provides an infinite number of computer addresses (at no point should the 'name' need to be translated into a numeric address
Not hardly. Computers are going to continue to use binary for the forseeable future. While that continues there will continue to be a need to translate names into a binary representation, something your naming scheme totally ignores. No matter what you name something there is still going to have to be translation between the name and a numerical address.
IPv4 does this, but doesn't provice enough codes - only 4 billion; clearly not enough for a world wide internet on a planet with more than 6 billion people.
IPv6 provides many more addresses - enough to allocate each person on earth more than a billion IP addresses.
Your proposal is a solution to a problem that was solved years ago; all that really remains is deployment of that solution.
Nor does establishment of a new naming scheme change the political and business motivations that lead to cybersquatting and trademark infringements. Some names are going to be more desirable than others. And there will need to be a body to settle disputes.
In fact, your naming scheme of joe@joeville@michigan@us is already in place. Countries already administer TLDs ending on country codes. It has solved none of the issues you refer to.
It looks like SCO might have stolen code from Linux
And thus question becomes - does the code SCO feels was lifted from UNIX actually start in Linux, and migrate to SCO, thus generating the matches that SCO is claiming as evidence of code migrating from UNIX to Linux?
While IANAL, I hold 12 patents and have been involved in a variety of legal wrangles involving patents.
It would be very interesting to see the letters that SCO sent out. If they weren't worded very carefully, and they include assertions of IP rights that SCO in fact doesn't own they could definitely trigger a rash of lawsuits.
When I was working in this field we were VERY careful when we went trolling for license fees. Something like:
Dear Sirs:
It has come to our attention that you may want to consider licensing the following patents (list numbers here).
Signed XYX Patent Attorney.
No claims of infringement etc. Just a word to the wise. The recipient would then decide what sort of position they were in and respond with something like:
Dear XYX:
We are interested in #47, and would like to offer a license to our #53 in exchange.
(In other words, yeah, we might be doing #47, but we think you are doing #53)
-or-
We are not interested. (Prove it).
-or-
We invented that long before you patented it and here is a copy of our documentation of the fact.
You don't have to be a tin-foil-hat conspiracy nut to see an obvious connection there.
Perhaps not, but there is the fact that Microsoft has used BSD code from time to time in Windows. Nothing wrong with that UNLESS the BSD code base is polluted with proprietary code. MS may just be covering their butt after reading a nastygram from SCO lawyers.
I think that anyone with sufficient brains in Microsoft would realize that the SCO lawsuit business is better off playing out on its own with out Microsoft being involved unless there was some other problem, like a threat from SCO to drag Microsoft into it. The reason that Microsoft may have taken a license may simply to be to prevent SCO from suing Microsoft too. This situation, i.e. a suit aimed at Linux and not Microsoft is definitely in MS's best interests.
Broadband was built for p2p and other bandwidth intensive applications. That's why you buy a broadband connection!
Well, Cable Broadband was certainly not designed for the traffic patterns present in P2P. The technical standard in wide use for cable is DOCSIS 1.0. This was developed before P2P took off and the down/up traffic ratio was typically 25/1. With P2P that ratio is typically more like 2/1 which means cable nodes are getting saturated on the up link WAY before they are getting saturated on the downlink. As a result Cable companies are having to cap uplink rates either when they detect heavy upload traffic, or establish various other controls.
Eventually I think that the answer has to be tiered or metered plans. It is not fair to the average user to make him subsidize the heavy user.
A post above had a good point - should TV's cost more because they use up a significant portion of the electricity - after all, more electricity, more damge to the environment and more cost to the Power company!
That is not a good point - it's nonsense. You don't pay a flat rate for your electricity consumption, you pay per kilowatt hour. Buy a big TV, your electrical bill will go up.
The article states that the retail value MAY reach $1 billion/yr, but Microsoft feels that it will be more likely 1/2 that, or less.
And what would be the level of the tax benefit claimed, considering that the cost to Microsoft is roughly $0.00 per piece
Tax law is quite clear that the deduction allowed for corporate in-kind donations is the cost of production. So given the Microsoft estimate of the actual retail value, and typical MS profit margins, this looks more like a $20-$40 million dollar per year cost to MS, and perhaps a $10 million dollar reduction in taxes.
A flea bite as far as Microsoft goes, and something that will not affect stockholder value in any way. So all those cynics claiming that Microsoft HAS to be doing this out of corporate greed are in fact probably full of it.
That "potentially promising" non-profit market is in reality only 1-2% of the software market, and a very financially limited one at that.
I think that it is very hard to see that this is a strategically important market segment for Microsoft. It seems to me that the primary benefits to Microsoft are good will (something intangible but important to corporations nonetheless), increasing use of computers in all segments of society as a general good thing for MS, and a tax writeoff.
Bottom line: Corporations give gifts, not out of concience, or goodwill, but from a perception of self benefit of some sort.
Perhaps, however self-interest can be a pretty broad proposition that includes benefitting people other than the corporation in addition to the corporation itself. Examples include funding scholarships at universities that provide well-educated employees, hospitals in areas that the company operates which make that area more attractive to employees, and so on.
No way this article is about software testing. This is about an evaluation lab where customers bring in their applications to show to Microsoft. It's a marketing puff-piece, that's all.
Where is the description of the test methodologies used? The bug escalation and change control systems? What sort of configuration control is used?
And I just don't buy the argument that running Linux at the desktop is going to reduce call or problem volume.
Given that most calls are going to be things like 'I forgot my password' or 'How do you sort email' I would agree that call volume is not going to change drastically. But I don't think most organizations are handling these with real admins any more. Costs to handle these should be independent of OS.
Incident (in ITIL terms) volume maybe, but I don't see it cutting down on normal issues, which means I still have to have 400 people running around.
How many are going to be doing real administration vs. level 1 activities ? 40?
Guess what, we are in a nasty recession. ALL capital spending is way off. Businesses spend money to increase capacity when growth is strong. No growth, capacity underutilization etc. means no need to increase capacity through either productivity or staff improvements. IT accounts for something like 20-30% of capital investments.
When the economy starts growing again people will start being in short supply (unemployment overall is a relatively low 6% given the severity of the recession) soon. The only way businesses will be able to increase capacity is through capital investment. During all this Moore's law has been chugging along and those new systems will be 10x+ faster (and getting 64bit CPUs in commodity hardware allowing manipulation of massive data sets cheaply) than what the business has been using, and software to take advantage has been getting better too, making upgrades very cost-effective. Businesses have also been amassing mammouth quantities of data in electronic form in the meantime because of their previous investments.
Moore's law doesn't tell the whole story by far. Data storage capacity and network bandwith are increasing at rates far faster than Moore's law. CPUs are relatively mature by comparison, ONLY doubling every two years or so. Hard disk bytes/$ doubles every 9 months, and network bandwidth is tripled every year.
The confluence of trends is obvious, and somebody is going to make a lot of money tying all this together.
The IT good times will roll again soon enough. When it happens you won't be able to outsource to China or India because they will be caught up in it too. Companies who outsourced their IT will find that the new crown jewels of productivity will be beyond their grasp and they will relearn the hard lesson that you DO NOT outsource core business activities if you want to maintain control of your fate.
And you're more expensive than any Microsoft admin, so it costs the business more.
I may be more expensive than most Microsoft admins on an hourly basis. However my skills scale far better and result in more stable operations than you will get from some freshly minted ex-burger flipper now MSCE via the Chubb Institute. That translates directly to lower business costs.
Accountability goes beyond recovery from damages. It mitigates my organization's suport responsibilities to issues specific to our environment.
I fail to see how this has ANYTHING to do with vendor choice. You are simply defining the scope of your organization. Anything else is just not supported. You will support email, but the Computational Fluid Dynamics lab has to find some other way to support their Fujitsu supercomputer.
Microsoft has core-OS problems, just like Linux does. The fact that you choose RedHat over Microsoft does not mean you have to support troubleshooting these.
Even with the cost of licensing, it's still cheaper for me to hire 2 Windows guys than it is one Linux guy.
It's not just the cost of the licenses. It's the cost of being able to prove compliance, and the potential liability from an employee sneaking in some pirated package and risk of having the BSA tie up your operations in an crippling audit.
Err. That seems more like a story of professional incompetence than it does of crap software.
Could be, but I've seen some pretty horrific stuff happen to Exchange servers. I even saw a case where a large company I worked for had to fly two Microsoft engineers out from Redmond to troubleshoot one of these things. It never really ran right despite spending a buttload of money on licenses, hardware and expensive support on it.
Slashdot Mirror
isn't required Linus making money directly by selling, trading, etc... proven intellectual property from SCO to be a sueable candidat?
There is a research exemption, however I would think Linus has at least taken speaking fees at conferences, etc.
and how in hell can SCO afford all this crap anyway?
They can't. This is being handled on a contingency basis. If there are a lot of counter-suits SCO could be in big trouble.
Slashdot SERIOUSLY needs to implement spellchecking as a feature of its software.
The error rate on main page articles is atrocious.
Wrong. Linux is nowhere near as IPv6-friendly as the *BSDs. To enable IPv6 in FreeBSD, for example, put 'ipv6_enable="YES"' in /etc/rc.conf
Each FreeBSD CD comes with a bunch prebuilt IPv6-ready apps, like apache, wget, etc -- apps that don't have native IPv6 support.
Wrong yourself. On RedHat 9 for example, to enable IPv6 all you do is type "modprobe ipv6' as root. Rebooting is NOT required.
Most Linux distros contain a LOT of IPv6 support. RedHat comes with IPv6 enabled through a kernel module, and a large number of packages that include IPv6 support out of the box. Examples include Apache, ping6, iproute6, traceroute6, and so on.
The following page describe Linux IPv6 support in a variety of distros.
We all know that Asian countries should convert to IPv6. The better question is will they?
Yes. All the other solutions have serious drawbacks and cost just as much money, if not more to implement.
The whole concept of the 'internet' by means of Tcp/IP is becoming quite dated.
For example, a sample address would be
Joe Smith@Earthlink@USA (users within the USA can leave the @USA blank)
this eliminates the need for a domain naming system. takes a lot of power away from ICANN, would help to solve cybersquatting, and provides an infinite number of computer addresses (at no point should the 'name' need to be translated into a numeric address
Not hardly. Computers are going to continue to use binary for the forseeable future. While that continues there will continue to be a need to translate names into a binary representation, something your naming scheme totally ignores. No matter what you name something there is still going to have to be translation between the name and a numerical address.
IPv4 does this, but doesn't provice enough codes - only 4 billion; clearly not enough for a world wide internet on a planet with more than 6 billion people.
IPv6 provides many more addresses - enough to allocate each person on earth more than a billion IP addresses.
Your proposal is a solution to a problem that was solved years ago; all that really remains is deployment of that solution.
Nor does establishment of a new naming scheme change the political and business motivations that lead to cybersquatting and trademark infringements. Some names are going to be more desirable than others. And there will need to be a body to settle disputes.
In fact, your naming scheme of joe@joeville@michigan@us is already in place. Countries already administer TLDs ending on country codes. It has solved none of the issues you refer to.
It looks like SCO might have stolen code from Linux
And thus question becomes - does the code SCO feels was lifted from UNIX actually start in Linux, and migrate to SCO, thus generating the matches that SCO is claiming as evidence of code migrating from UNIX to Linux?
While IANAL, I hold 12 patents and have been involved in a variety of legal wrangles involving patents.
It would be very interesting to see the letters that SCO sent out. If they weren't worded very carefully, and they include assertions of IP rights that SCO in fact doesn't own they could definitely trigger a rash of lawsuits.
When I was working in this field we were VERY careful when we went trolling for license fees. Something like:
Dear Sirs:
It has come to our attention that you may want to consider licensing the following patents (list numbers here).
Signed
XYX Patent Attorney.
No claims of infringement etc. Just a word to the wise. The recipient would then decide what sort of position they were in and respond with something like:
Dear XYX:
We are interested in #47, and would like to offer a license to our #53 in exchange.
(In other words, yeah, we might be doing #47, but we think you are doing #53)
-or-
We are not interested. (Prove it).
-or-
We invented that long before you patented it and here is a copy of our documentation of the fact.
And so on.
You don't have to be a tin-foil-hat conspiracy nut to see an obvious connection there.
Perhaps not, but there is the fact that Microsoft has used BSD code from time to time in Windows. Nothing wrong with that UNLESS the BSD code base is polluted with proprietary code. MS may just be covering their butt after reading a nastygram from SCO lawyers.
I think that anyone with sufficient brains in Microsoft would realize that the SCO lawsuit business is better off playing out on its own with out Microsoft being involved unless there was some other problem, like a threat from SCO to drag Microsoft into it. The reason that Microsoft may have taken a license may simply to be to prevent SCO from suing Microsoft too. This situation, i.e. a suit aimed at Linux and not Microsoft is definitely in MS's best interests.
Broadband was built for p2p and other bandwidth intensive applications. That's why you buy a broadband connection!
Well, Cable Broadband was certainly not designed for the traffic patterns present in P2P. The technical standard in wide use for cable is DOCSIS 1.0. This was developed before P2P took off and the down/up traffic ratio was typically 25/1. With P2P that ratio is typically more like 2/1 which means cable nodes are getting saturated on the up link WAY before they are getting saturated on the downlink. As a result Cable companies are having to cap uplink rates either when they detect heavy upload traffic, or establish various other controls.
Eventually I think that the answer has to be tiered or metered plans. It is not fair to the average user to make him subsidize the heavy user.
A post above had a good point - should TV's cost more because they use up a significant portion of the electricity - after all, more electricity, more damge to the environment and more cost to the Power company!
That is not a good point - it's nonsense. You don't pay a flat rate for your electricity consumption, you pay per kilowatt hour. Buy a big TV, your electrical bill will go up.
Theoretically, if they donate enough, they'll eventually wipe out all their taxable income.
Perhaps in some other universe. In this universe the IRS caps the tax benefit at 10% of taxable income.
and they'll take a tax deduction of $1b.
No, they won't. Corp write-offs of donations of products are limited to production cost.
Would that be at full retail price?
The article states that the retail value MAY reach $1 billion/yr, but Microsoft feels that it will be more likely 1/2 that, or less.
And what would be the level of the tax benefit claimed, considering that the cost to Microsoft is roughly $0.00 per piece
Tax law is quite clear that the deduction allowed for corporate in-kind donations is the cost of production. So given the Microsoft estimate of the actual retail value, and typical MS profit margins, this looks more like a $20-$40 million dollar per year cost to MS, and perhaps a $10 million dollar reduction in taxes.
A flea bite as far as Microsoft goes, and something that will not affect stockholder value in any way. So all those cynics claiming that Microsoft HAS to be doing this out of corporate greed are in fact probably full of it.
That "potentially promising" non-profit market is in reality only 1-2% of the software market, and a very financially limited one at that.
I think that it is very hard to see that this is a strategically important market segment for Microsoft. It seems to me that the primary benefits to Microsoft are good will (something intangible but important to corporations nonetheless), increasing use of computers in all segments of society as a general good thing for MS, and a tax writeoff.
Lucent didn't invent Unix, gifted inventors at Bell Labs did.
Lucent is the corporation formed from Bell Labs when AT&T was broken up.
I really don't understand how it is wrong or harmful for anyone, monopoly or otherwise, to give away something.
Tell that to Netscape.
Seriously, for a monopoly to do something like this can be construed as an illegal anti-competitive act.
Bottom line: Corporations give gifts, not out of concience, or goodwill, but from a perception of self benefit of some sort.
Perhaps, however self-interest can be a pretty broad proposition that includes benefitting people other than the corporation in addition to the corporation itself. Examples include funding scholarships at universities that provide well-educated employees, hospitals in areas that the company operates which make that area more attractive to employees, and so on.
No way this article is about software testing. This is about an evaluation lab where customers bring in their applications to show to Microsoft. It's a marketing puff-piece, that's all.
Where is the description of the test methodologies used? The bug escalation and change control systems? What sort of configuration control is used?
even if you explicitly set references to null.
Setting references to null in Java is not what I would expect from a professional programmer.
System load is the average number of blocked processes.
Depends on what tools you are using. Many (uptime on RedHat for example) exclude processes blocked by I/O.
And I just don't buy the argument that running Linux at the desktop is going to reduce call or problem volume.
Given that most calls are going to be things like 'I forgot my password' or 'How do you sort email' I would agree that call volume is not going to change drastically. But I don't think most organizations are handling these with real admins any more. Costs to handle these should be independent of OS.
Incident (in ITIL terms) volume maybe, but I don't see it cutting down on normal issues, which means I still have to have 400 people running around.
How many are going to be doing real administration vs. level 1 activities ? 40?
The sky is falling etc.
Guess what, we are in a nasty recession. ALL capital spending is way off. Businesses spend money to increase capacity when growth is strong. No growth, capacity underutilization etc. means no need to increase capacity through either productivity or staff improvements. IT accounts for something like 20-30% of capital investments.
When the economy starts growing again people will start being in short supply (unemployment overall is a relatively low 6% given the severity of the recession) soon. The only way businesses will be able to increase capacity is through capital investment. During all this Moore's law has been chugging along and those new systems will be 10x+ faster (and getting 64bit CPUs in commodity hardware allowing manipulation of massive data sets cheaply) than what the business has been using, and software to take advantage has been getting better too, making upgrades very cost-effective. Businesses have also been amassing mammouth quantities of data in electronic form in the meantime because of their previous investments.
Moore's law doesn't tell the whole story by far. Data storage capacity and network bandwith are increasing at rates far faster than Moore's law. CPUs are relatively mature by comparison, ONLY doubling every two years or so. Hard disk bytes/$ doubles every 9 months, and network bandwidth is tripled every year.
The confluence of trends is obvious, and somebody is going to make a lot of money tying all this together.
The IT good times will roll again soon enough. When it happens you won't be able to outsource to China or India because they will be caught up in it too. Companies who outsourced their IT will find that the new crown jewels of productivity will be beyond their grasp and they will relearn the hard lesson that you DO NOT outsource core business activities if you want to maintain control of your fate.
And you're more expensive than any Microsoft admin, so it costs the business more.
I may be more expensive than most Microsoft admins on an hourly basis. However my skills scale far better and result in more stable operations than you will get from some freshly minted ex-burger flipper now MSCE via the Chubb Institute. That translates directly to lower business costs.
Accountability goes beyond recovery from damages. It mitigates my organization's suport responsibilities to issues specific to our environment.
I fail to see how this has ANYTHING to do with vendor choice. You are simply defining the scope of your organization. Anything else is just not supported. You will support email, but the Computational Fluid Dynamics lab has to find some other way to support their Fujitsu supercomputer.
Microsoft has core-OS problems, just like Linux does. The fact that you choose RedHat over Microsoft does not mean you have to support troubleshooting these.
Even with the cost of licensing, it's still cheaper for me to hire 2 Windows guys than it is one Linux guy.
It's not just the cost of the licenses. It's the cost of being able to prove compliance, and the potential liability from an employee sneaking in some pirated package and risk of having the BSA tie up your operations in an crippling audit.
Err. That seems more like a story of professional incompetence than it does of crap software.
Could be, but I've seen some pretty horrific stuff happen to Exchange servers. I even saw a case where a large company I worked for had to fly two Microsoft engineers out from Redmond to troubleshoot one of these things. It never really ran right despite spending a buttload of money on licenses, hardware and expensive support on it.