That's the Top Fifteen according to ISP Planet which cites numerous other sources. Check my math, but my calculations show that to be about 65.5%.
Well this is more or less what I wrote, that the top 10 control 66% of the market.
"The ISP business is a very specific business for the broadband access. Why? Because it has very high set up costs: the lines themselves, the line equipement. So no, you cannot decide randomly: "hey, this month I will join Cablevision - hey this month I will join Comcast!" - each ISP is limited by geographical area."
It was the same when Dial-Up ISPs were in their infancy. The set up costs were high, and the service and equipment were expensive. Prices go down.
No. You missed the fundamental problem of and the fundemental cost of broadband: the lines. The physical lines. They cost a premium and they are worth a fortune. Most of the cost is physical digging, laying out the cables, and fill the holes back, all this on the property of others, or of the municipality. This is a low-tech job.
Actually, in my area, I have several choices for ISPs, including [to the best of my knowledge] at least three sources for broadband. ProLog/Service Electric, Shen-Heights, Verizon, all offer broadband. There are several ISPs in the area that offer Dial-Up, and since Dial-Up has new subscribers every day, and is still a significant portion of the market, they still count.
This looked strange to me: are you claiming that right now in your home, you have one wire going to ProLog, one wire going to Shen-Heights, and one wire going to Verizon? So I checked, and I saw the reason why there is competition at all is that the governement (the FCC) forced the cable/telco companies to let their competitors use the previous line to the customer. So your quote of a marvellous thousands of ISPs is only due to regulation. Besides that regulation was slashed in 2002, so the hundreds of ISP were poised to disappear (except the deregulation was overruled by some courts, so that's why your ISPs are not here).
People regularly buy up businesses across the country, or across the world. Your basic premise seems to be that new businesses are bought out when a particular company is looking to form an oligopoly. Explain then, how businesses, including ISPs, regularly open up shops or buy out businesses far away from their main office, including in other countries?
First, I know absolutly no small ISP (say 10,000 customer), who has business in both say, New Delhi, and San Francisco. It makes no sense - it is still possible if there is an economic reason for that. You have to have a pretty good reason to go abroad, and small international companies are the exception. If you are not in the top 10 market leaders going abroad is a question mark. Again, I stand that exactly what I said: if you are a small ISP, it makes no sense to buy another ISP at the other end of the country, because there is no business logic there, you are not creating any business synergy or nothing. You're not increasing the shareholder value. You may as well buy a bakery in St Petersburg instead. It may make sense, however, to create an "association of small ISPs". It may make sense to buy an ISP in the town next to yours - because then you can share backbone access. Different businesses have different business logic.
Second, monopolies or oligopolies make more profit than competitive markets. It's a smart CEO move, to buy his competitors to create a monopoly, or an oligopoly, all things being equal. It's just one possibility, not a necessity. Sometimes it's smarter to expand the market, or invest in innovation or whatever.
Being bought out is a safety net, no matter how much you try to convolute the issue.
Well, yes it is a safety net. But only part of the eq
"Yeah, and the top 10 control 66% of the market now (AOL controls 25%). The "AOL-Time Warner" attempt is a prime example of attempt of concentration of power in the hands of corporations."
Yes, uh, a decopoly is such a horrible thing
First, we went from your marvellous world of thousands of ISP to a decopoly. Wow! And you still aren't convinced?
Ok, let dig the matter about this wonderful world of ISP free market: you forgot, that, classical dial-up ISP is dying ; in South Korea, 80% of households have broadband already, and all the US top 25 ISP are doing broadband. The ISP business is a very specific business for the broadband access. Why? Because it has very high set up costs: the lines themselves, the line equipement. So no, you cannot decide randomly: "hey, this month I will join Cablevision - hey this month I will join Comcast!" - each ISP is limited by geographical area. That's why, right now, you don't have a choice of 1000 ISPs, nor even a great choice about the top 10: in practice, for broadband, at a given place, you have oligopoly or monopoly.
"Yeah, instead of bankrupcy, they sell all their assets for a huge bargain - which is what happens in bankrupcy anyway, so what?"
I don't think it is the same thing as bankruptcy. I think you took a bit of a logical leap there, or you're just attacking another straw man.
Huh? Just try this experiment: create your ISP. Then try to sell it, to say Comcast, Cox, or Qwest. First if it is not broadband, they are not interested. Go away. Worse, once broadband comes in your town, you are virtually on the path to bankrupcy. Second if it is broadband, then you are probably relatively large already.
But ok, let's ignore those show-stopper problems for now (which happen in practice), and let's go further. Then, third, because of the geographical property someone who has an ISP somewhere is not interested in buying an ISP at the other end of the US, because it will only create problems (travel expenses, medium range phone calls, little possibility of exchanging employee): at best he would be interested in an ISP in the next town, if not too far. The only case where he would be really interested is buying a local competitor to set up a smaller oligopoly, or a monopoly.
But let's imagine, you can still manage to find someone: then if he buys you, this someone has to deal with a different organization, different equipements, different architecture, different software, different business methods, different contracts with clients, different hotline,... heck everything is different: dealing with this mess has a huge cost (especially if he buys not 1 ISP, but 30 ISP) - and this huge cost will be substracted from the price at which he can buy your ISPs. So no, the price you'll get in the unobvious event that someone is interested at all is not necessarily greater than the price you'll get after the normal sale of your assets after bankrupcy.
Then from the point of view of the seller, whether he sells at a huge bargain or after bankrupcy is not necessarily different, as he may well not be able to recoup his investments.
For the record the broadband ISP is one of the less "free markets" in the world, because the hypothesis of free entry is plain wrong: you cannot set up a broadband ISP, without huge investments. We were talking about ISP, but I remind you that the Sprint/MCI merger attempt would have created a company controlling 80% of the long-distance phone calls. National and international network providers (Bandwidth barons) are already concentrating, like rail companies were concentrating. Wait 20 years from now on, and you'll see "free market" in action.
Also you took a market at a point of time where it was in exponential growth, and everyone and his dog was investing like crazy into any "Internet" related stuff. Now wait, and see what happens in 20 years from now on.
Sure enough, the big dogs started buying up the little ones, but do you know what happened next? A year later, there were some 1100 ISPs.
Yeah, and the top 10 control 66% of the market now (AOL controls 25%). The "AOL-Time Warner" attempt is a prime example of attempt of concentration of power in the hands of corporations. Somehow it is likely that the top ten don't bother buying like 300+ small ISPs because, frankly, they aren't worth the trouble.
If they, themselves, don't make it big, they have the extra safety net of being bought out -- the risks aren't as great.
Yeah, instead of bankrupcy, they sell all their assets for a huge bargain - which is what happens in bankrupcy anyway, so what? The power is still concentrated in the top 10 ISPs.
And some of you accused the BBC of making an unfounded claim when they said this:
"If anyone's anger has no measure, it is the wrath of internet zealots who believe that code should be free to all (open source). So, it seems likely that the perpetrators of the MyDoom virus and its variants are internet vandals with a specific grudge."
Except that in this case, it is not vandalism, because it is a extremely precise and accurate assessment. Vandalism != exact truth.
1) Are they litigious? Definitly. SCO expects 10-15 millions revenue per quarter, and to spend 4-5 millions per quarter on legal expenses. 33 to 50% of their revenue is spent on legal expenses by their own expectations (reality can be worse).
2) Are they bastards? Definitly. First SCO was initiated as Caldera, trying to earn money on hard work of others (who were ok for that), and then suddenly, it has the nerve to tell the world those Linux contributors are a bunch of thieves acting contrary to US constitution, and that they, SCO, are entitled payment on each Linux installation. Not only that, but they engaged a PR campaign, producing lie after lie, slander after slander, in which they dragged in the mud at each single turn the very Linux contributors on the work of which they had capitalized and still plan to capitalize. Not to mention that obviously they knew they were telling lies all the way (no code thief produced as evidence at the trial, SCO contracts with IBM, SCO dispute with Novell). And that they still tried to pressure the random Joe Linux user by threats.
Their actions earn them a well deserved title of "bastards" (following the definition
of Worldnet: bastard, [...] n 1: insulting terms of address for people who are stupid or
irritating or ridiculous [syn: asshole, cocksucker,
dickhead, shit, mother fucker, motherfucker, prick,
whoreson, son of a bitch,
SOB]
In conclusion, I would say that calling SCO litigious bastards is not only an exact truth, but also if one unsuspecting person wanted to find an example of litigious bastards in the IT industry, it would only be fair that Google would return them in first position.
If you have continuations you can create loops, exceptions, even function calls, just with continuations and conditionals.
Which is insanetly cool, unless you try to actually use that in an actual program. Same as the Turing-complete assembly language which has only one opcode.
As rail speeds increase, so does the damage that can be done by a terrorist. A 650km/h maglev sounds interesting at first sight - but how much damage could be done by a well placed bomb? Although the thing contains no fuel on board, the combination of released kinetic and magnetic energy would, I guess, be pretty destructive.
This isn't as gloomy as you depict it. Yes the train would have incredible cinetic energy, by because of this it is incredibly harder to derail, because in order to make the train turn you have to reflect this energy from one direction to another. In anycase, I doubt terrorism is very much a concern in China.
And in China, where most people are still desperately poor, the government has even more responsibility to make the economic decision rather than the vanity decision.
Alternatively, people live relatively decently , so instead to saving about 1 dollar per Chinese, governement is better consider technology that would improve significantly their life in 30 years, plus provide another high technology that the country could export.
Last time I was at Wal-Mart, I was thinking: Gee it's sure a shame that China hasn't benefitted from trade agreements. They only produced a token 80% of the stuff in this store. Clearly, we need to do more.
True, one couldn't be more correct. This means that the 100-150 other countries had little to zero stuff in that store,
especially the poorest from the Fourth World. How many goods manufactured in Ethiopia?
I (we) uniformly disagree on the theory that everything IBM added to AIX must not also be added to Linux.
Not true. If IBM had signed a non-compete agreement with SCO, IBM most likely won't be allowed to do that, even if IBM owns the code - IBM would still be allowed to use it in-house. They certainly didn't sign a non-compete agreement (or an NDA), but that means, you have to look to their contract to see if they are actually allowed to backport modification from AIX to Linux, and/or to have people working previously on AIX, contribute to Linux.
Device, filesystem drivers used with Linux may be considered derivative works, even if not shipped with the Linux distribution, and therefor subject to GPL.
But if someone who had signed an NDA contributed code to Linux, one judge may very well demand that Linus cease the distribution of Linux until that code is removed. Actually the judge not only may but really should decide that (in addition to the dommages paid by the NDA violator).
This makes it clear that SCO is not talking about old code. They aren't claiming that IBM put ancient SysV code into Linux. They're claiming that they own code that IBM wrote and they never saw (and don't have a copy of).
Not really. They are claiming their agreement with IBM prevents IBM to put their (IBM's) code into Linux. Two examples of such situtation actually occuring: if IBM signed an NDA, that would be true ; if IBM signed a non-compete agreement, that would be true as well.
Clearly their interpretation is that anything IBM ever wrote related to UNIX is covered by their new UNIX copyright.
Nope, their interpretation is that anything IBM ever wrote related to UNIX is covered by their contract, that their contract prevents IBM to release this code in Linux, and that Linux which integrate illegally copied code from IBM (BUT owned by IBM), should be stopped being distributed.
Their bold step 2 (as in "2. ???") would probably be to ask the judge for allowing them to sell "licenses", allowing a user to still use legally the same Linux code which incorporate code from IBM which was illegally put into Linux. However this is likely to conflict with the GPL, so they need the judge to take a big magic decision there, as in "SCO has been harmed, so they can release code under the GPL ignoring such and such provision, but only once they have agreed with Linus about the exhaustive list of sections of code for which IBM illegally contributed code".
Does George Lucas own the copyright to every Star Wars book ever published--say, the Timothy Zahn trilogy?
That's exactly the point. Georges Lucas owns the trademarks associated with Star Wars, so even if Timothy Zahn owned 100% of the copyright of his works, it would be illegal for him to distribute it without an agreement with Georges Lucas.
And it's the same problem can happen with OSS with respect to patents - you can be forbidden to distribute code which violate a patent you don't own.
60 pages surely doesn't cover the millions of lines of code they were talking about in the press.
They once acknowledged there was no code owned by SCO that was copied into Linux, unlike what they say in press release. I think their most precise claim would be that code was illegaly copied into Linux.
So, while they have complied in legal terms, they have weakened their case in a significant order of magnitude. Not only that, but they have also weakened their case for any of their prosprective targets in their scoSource shakedown fiasco
Not really. If they managed to prove that IBM violated its contract by integrating some of their code into Linux, the judge may decide that Linux distribution should be stopped unless SCO agrees. Worse, she may decide that distribution should be stopped until it is figured out exactly what code, illegally put in Linux by IBM, should be removed.
What they did in their PR releases is put the entire OSS community on a false track. In reality, it always have been about a contract with IBM, as Linus noted.
This makes it pretty darned clear that IBM does in fact own their own "modifications and derivative works", which covers JFS and all the other features IBM added to AIX (and later Linux).
This is correct. SCO claims that IBM has all the rights, but at the same time, the contract they have with them prevent them to release it into Linux.
Just because Linux has JFS, NUMA... and AIX has the same support does not mean the code is the same. However they could possibly get away with what you described above if they accompanied it with reason(s) why they think the Linux and AIX code is the same (even binary comparison would be good).
Well lastest SCO legal info show they did it:
"17. In some additional cases it was also possible to infer with reasonable certainty from comments in the source code that the engineer who implemented that code had experience and knowledge of the methods, sequence and structures used in either or both of Dynix/ptx and AIX. Confirmation of this would require depositions from the IBM individuals involved in programming the actual Linux modules in question."
Also:
"18.1 The "Read Copy Update" (RCU) code contributed to Linux by IBM was copied substantially verbatim with only relatively minor changes from Dynix/ptx."
You said they needed to access IBM code to support their claim, and that is the very problem. You cannot claim if you do not have some evidence in the first place.
Yes, that's why they had probably given a handful of examples of developpers likely to have worked on both UNIX version of a code, and a released version of Linux of the same code. According to their personnal legal theory, I think they don't even have to claim that's it's the same person who wrote the same code in both OSes, just something like having worked on UNIX internals before contributing on Linux code could be problematic. But that would be the weaker claim, they'd go first for developpers having worked on the same code (say JFS) for both AIX and Linux. Given that newer internal versions of AIX were ditched, I'm sure at least some developpers might have been realocated from AIX teams to Linux teams.
SCO have repeatedly claimed that Linux also is guilty of copyright infringement directly on SCO products, not just AIX.
As I initially said, SCO claims in PR releases have been stupid, but they of course carefully avoided doing them boldly in front of the court, hence the title "They ARE NOT stupid scum..."
The questions say "with specificity", so the answer cannot be as curt as the one the parent gave.
"All code contributed by IBM" isn't specific enough, but anyway their claim would be: "All Linux code contributed by IBM employees who have worked once on UNIX source tree". They have lots
of choice and probably they can list the files
in fs/jfs (that is:
akefile jfs_dmap.h jfs_inode.c jfs_txnmgr.c jfs_xtree.h
file.c jfs_dtree.c jfs_inode.h jfs_txnmgr.h namei.c
inode.c jfs_dtree.h jfs_lock.h jfs_types.h resize.c
jfs_btree.h jfs_extent.c jfs_logmgr.c jfs_umount.c super.c
jfs_debug.c jfs_extent.h jfs_logmgr.h jfs_unicode.c symlink.c
jfs_debug.h jfs_filsys.h jfs_metapage.c jfs_unicode.h xattr.c
jfs_defragfs.h jfs_imap.c jfs_metapage.h jfs_uniupr.c
jfs_dinode.h jfs_imap.h jfs_mount.c jfs_xattr.h
jfs_dmap.c jfs_incore.h jfs_superblock.h jfs_xtree.c
), in entirity, for starters.
They have claimed that mathematicians and engineers have determined that lots of source code is contaminated in this way.
They did not claim it in front of the court though. In front of the court, they claimed they needed to access IBM development history to support their claims, which may be perfectly true, because I don't see them have a simple way to identify all/most of the IBM employees who worked on both Linux and UNIX, on what and when.
It is not unfair to let a defendant see what he has to defend against.
However, they had no way initially to know exactly the details about who developped what at IBM. I suppose they just gave one or two examples of a IBM developper who worked on both UNIX and Linux, and ask the judge, on the basis of this evidence, to access IBM information in order to give a more precise list.
is obviously a brainless Linux zealot. My points were more correct than 95% of the posts here.
Apparently saying "SCO are stupid scum" gets you moderated "+3 Insightful", while being one of the few posts giving logical reasons why there are not as stupid as people think, gets you mod-down as a troll.
In case you've forgotten, here are some of the questions that SCO must answer before they get a shot at IBM
But the answers are very simple:
Interrogatory 1: all code contributed by IBM employees in Linux.
Interrogatory 2: a) UNIX owner, b) rights as per SCO contract by IBM, c) none. "SCO is giving up the trade secret and patent claims, as it requires unreleased information from IBM".
Interrogatory 3: none. see interrogatory 2.
SCO not only charges that SCO's IP got into Linux against their wishes, but that IBM did it. IBM wants to see SCO's evidence that is had to be IBM and couldn't be someone else.
And? IBM actually contributed to Linux kernel, there is no question about this. SCO is claiming that code, at least some of it, according to the IBM contract, couldn't be put in Linux. They need details of IBM development process to prove it - even if it were actually true, they would need it.
As far as what SCO wants this case to be about, SCO has contradicted itself on so many occasions that it's impossible to say with any certainty what SCO is suing over.
If you ignore press releases, SCO position is consistent. The trial as always been about a breach of contract by IBM which is making IBM contributions to Linux illegal - although they tried to muddy the waters at last hearing, in order to force the judge to compel IBM to release evidence "we don't know what kind of IP violation it is yet, we must see a log of IBM actions first".
So if you will excuse me, I will continue to believe that SCO are stupid scum, because they've not shown any evidence to the contrary.
Not so much. If IBM had signed a non-compete clause with SCO for instance, IBM would lose this trial instantly, and Linux distribution would become problematic. The whole crux of this story is in the detail of IBM contract with SCO (and SCO with Novell), and nowhere else.
Press releases from SCO have always been deceiving bullshit, as that's the case for many companies, although SCO has raised this to an unmatched level.
The term "Anti-Sionist" needs some definition. In many circles, the term "Zionist" merely indicates a belief that Israel should be allowed to exist.
Nowadays, Zionism is understood as "all of Palestine should belong to Israel". Because 1) Israel already exists and always will (a few atomic bombs are there to ensure this), 2) it is this form of Zionism is the initial official program of Likud: "
Self-Rule:
The Government of Israel flatly rejects the establishment of a Palestinian Arab state west of the Jordan river
[...]
The Jordan River as a Permanent Border:
The Jordan Valley and the territories that dominate it shall be under Israeli sovereignty. The Jordan river will be the permanent eastern border of the State of Israel.
The Kingdom of Jordan is a desirable partner in the permanent status arrangement between Israel and the Palestinians in matters that will be agreed upon.
"
This was illustrated by the vote in 2002 by the Likud, against Sharon will, of the "refusal to accept any Palestinian state".
Anti-Zionists support the Palestinian thug-ocracy, and by extension, their desire to eliminate Israel altogether by whatever murderous and terroristic means necessary
No, anti-zionits simply said go back to 1967-borders, sometimes even to 1947 borders, and that Israel isn't entitled to more. They denounce settlements.
An anti-Zionist assigns moral equivalence between the tactics of the IDF and Hizbollah.
Indeed there is moral equivalence between the implicit wish to drive all Palestinians out of Palestine, and fund the Great Israel, and the Hizbollah. Ethnic cleansing, at the heart of settlement and annexation plans when implemented. Sometimes of course, Zionists will settle for gaining
only half of Palestinian territory, or, when they don't support ethnic cleansing (by moving all Palestinians to Jordan) for an appartheid state, where arab citizen will be have only limited rights so that they don't "threaten Israel existence or safety", which obviously can only mean limited/no rights to vote or be elected.
I'm missing the fundamental difference between anti-semitism and anti-zionism.
Slashdot Mirror
Well this is more or less what I wrote, that the top 10 control 66% of the market.
"The ISP business is a very specific business for the broadband access. Why? Because it has very high set up costs: the lines themselves, the line equipement. So no, you cannot decide randomly: "hey, this month I will join Cablevision - hey this month I will join Comcast!" - each ISP is limited by geographical area."
It was the same when Dial-Up ISPs were in their infancy. The set up costs were high, and the service and equipment were expensive. Prices go down.
No. You missed the fundamental problem of and the fundemental cost of broadband: the lines. The physical lines. They cost a premium and they are worth a fortune. Most of the cost is physical digging, laying out the cables, and fill the holes back, all this on the property of others, or of the municipality. This is a low-tech job.
Actually, in my area, I have several choices for ISPs, including [to the best of my knowledge] at least three sources for broadband. ProLog/Service Electric, Shen-Heights, Verizon, all offer broadband. There are several ISPs in the area that offer Dial-Up, and since Dial-Up has new subscribers every day, and is still a significant portion of the market, they still count.
This looked strange to me: are you claiming that right now in your home, you have one wire going to ProLog, one wire going to Shen-Heights, and one wire going to Verizon? So I checked, and I saw the reason why there is competition at all is that the governement (the FCC) forced the cable/telco companies to let their competitors use the previous line to the customer. So your quote of a marvellous thousands of ISPs is only due to regulation. Besides that regulation was slashed in 2002, so the hundreds of ISP were poised to disappear (except the deregulation was overruled by some courts, so that's why your ISPs are not here).
People regularly buy up businesses across the country, or across the world. Your basic premise seems to be that new businesses are bought out when a particular company is looking to form an oligopoly. Explain then, how businesses, including ISPs, regularly open up shops or buy out businesses far away from their main office, including in other countries?
First, I know absolutly no small ISP (say 10,000 customer), who has business in both say, New Delhi, and San Francisco. It makes no sense - it is still possible if there is an economic reason for that. You have to have a pretty good reason to go abroad, and small international companies are the exception. If you are not in the top 10 market leaders going abroad is a question mark. Again, I stand that exactly what I said: if you are a small ISP, it makes no sense to buy another ISP at the other end of the country, because there is no business logic there, you are not creating any business synergy or nothing. You're not increasing the shareholder value. You may as well buy a bakery in St Petersburg instead. It may make sense, however, to create an "association of small ISPs". It may make sense to buy an ISP in the town next to yours - because then you can share backbone access. Different businesses have different business logic.
Second, monopolies or oligopolies make more profit than competitive markets. It's a smart CEO move, to buy his competitors to create a monopoly, or an oligopoly, all things being equal. It's just one possibility, not a necessity. Sometimes it's smarter to expand the market, or invest in innovation or whatever.
Being bought out is a safety net, no matter how much you try to convolute the issue.
Well, yes it is a safety net. But only part of the eq
He's abandoned any pretense of fairness or objectivity in his columns.
Indeed, but then he relies on actual facts and data, as an economist.
Yes, uh, a decopoly is such a horrible thing
First, we went from your marvellous world of thousands of ISP to a decopoly. Wow! And you still aren't convinced?
Ok, let dig the matter about this wonderful world of ISP free market: you forgot, that, classical dial-up ISP is dying ; in South Korea, 80% of households have broadband already, and all the US top 25 ISP are doing broadband. The ISP business is a very specific business for the broadband access. Why? Because it has very high set up costs: the lines themselves, the line equipement. So no, you cannot decide randomly: "hey, this month I will join Cablevision - hey this month I will join Comcast!" - each ISP is limited by geographical area. That's why, right now, you don't have a choice of 1000 ISPs, nor even a great choice about the top 10: in practice, for broadband, at a given place, you have oligopoly or monopoly.
"Yeah, instead of bankrupcy, they sell all their assets for a huge bargain - which is what happens in bankrupcy anyway, so what?"
I don't think it is the same thing as bankruptcy. I think you took a bit of a logical leap there, or you're just attacking another straw man.
Huh? Just try this experiment: create your ISP. Then try to sell it, to say Comcast, Cox, or Qwest. First if it is not broadband, they are not interested. Go away. Worse, once broadband comes in your town, you are virtually on the path to bankrupcy. Second if it is broadband, then you are probably relatively large already.
But ok, let's ignore those show-stopper problems for now (which happen in practice), and let's go further. Then, third, because of the geographical property someone who has an ISP somewhere is not interested in buying an ISP at the other end of the US, because it will only create problems (travel expenses, medium range phone calls, little possibility of exchanging employee): at best he would be interested in an ISP in the next town, if not too far. The only case where he would be really interested is buying a local competitor to set up a smaller oligopoly, or a monopoly.
But let's imagine, you can still manage to find someone: then if he buys you, this someone has to deal with a different organization, different equipements, different architecture, different software, different business methods, different contracts with clients, different hotline, ... heck everything is different: dealing with this mess has a huge cost (especially if he buys not 1 ISP, but 30 ISP) - and this huge cost will be substracted from the price at which he can buy your ISPs. So no, the price you'll get in the unobvious event that someone is interested at all is not necessarily greater than the price you'll get after the normal sale of your assets after bankrupcy.
Then from the point of view of the seller, whether he sells at a huge bargain or after bankrupcy is not necessarily different, as he may well not be able to recoup his investments.
For the record the broadband ISP is one of the less "free markets" in the world, because the hypothesis of free entry is plain wrong: you cannot set up a broadband ISP, without huge investments. We were talking about ISP, but I remind you that the Sprint/MCI merger attempt would have created a company controlling 80% of the long-distance phone calls. National and international network providers (Bandwidth barons) are already concentrating, like rail companies were concentrating. Wait 20 years from now on, and you'll see "free market" in action.
Also you took a market at a point of time where it was in exponential growth, and everyone and his dog was investing like crazy into any "Internet" related stuff. Now wait, and see what happens in 20 years from now on.
Yeah, and the top 10 control 66% of the market now (AOL controls 25%). The "AOL-Time Warner" attempt is a prime example of attempt of concentration of power in the hands of corporations. Somehow it is likely that the top ten don't bother buying like 300+ small ISPs because, frankly, they aren't worth the trouble.
If they, themselves, don't make it big, they have the extra safety net of being bought out -- the risks aren't as great.
Yeah, instead of bankrupcy, they sell all their assets for a huge bargain - which is what happens in bankrupcy anyway, so what? The power is still concentrated in the top 10 ISPs.
"If anyone's anger has no measure, it is the wrath of internet zealots who believe that code should be free to all (open source). So, it seems likely that the perpetrators of the MyDoom virus and its variants are internet vandals with a specific grudge."
Except that in this case, it is not vandalism, because it is a extremely precise and accurate assessment. Vandalism != exact truth.
In conclusion, I would say that calling SCO litigious bastards is not only an exact truth, but also if one unsuspecting person wanted to find an example of litigious bastards in the IT industry, it would only be fair that Google would return them in first position.
Huh? There is a very high probability that the average customer will think they are associated, and as such, both produced by Microsoft.
Which is insanetly cool, unless you try to actually use that in an actual program. Same as the Turing-complete assembly language which has only one opcode.
This isn't as gloomy as you depict it. Yes the train would have incredible cinetic energy, by because of this it is incredibly harder to derail, because in order to make the train turn you have to reflect this energy from one direction to another. In anycase, I doubt terrorism is very much a concern in China.
And in China, where most people are still desperately poor, the government has even more responsibility to make the economic decision rather than the vanity decision.
Alternatively, people live relatively decently , so instead to saving about 1 dollar per Chinese, governement is better consider technology that would improve significantly their life in 30 years, plus provide another high technology that the country could export.
True, one couldn't be more correct. This means that the 100-150 other countries had little to zero stuff in that store, especially the poorest from the Fourth World. How many goods manufactured in Ethiopia?
Not true. If IBM had signed a non-compete agreement with SCO, IBM most likely won't be allowed to do that, even if IBM owns the code - IBM would still be allowed to use it in-house. They certainly didn't sign a non-compete agreement (or an NDA), but that means, you have to look to their contract to see if they are actually allowed to backport modification from AIX to Linux, and/or to have people working previously on AIX, contribute to Linux.
Device, filesystem drivers used with Linux may be considered derivative works, even if not shipped with the Linux distribution, and therefor subject to GPL.
But if someone who had signed an NDA contributed code to Linux, one judge may very well demand that Linus cease the distribution of Linux until that code is removed. Actually the judge not only may but really should decide that (in addition to the dommages paid by the NDA violator).
Not really. They are claiming their agreement with IBM prevents IBM to put their (IBM's) code into Linux. Two examples of such situtation actually occuring: if IBM signed an NDA, that would be true ; if IBM signed a non-compete agreement, that would be true as well.
Clearly their interpretation is that anything IBM ever wrote related to UNIX is covered by their new UNIX copyright.
Nope, their interpretation is that anything IBM ever wrote related to UNIX is covered by their contract, that their contract prevents IBM to release this code in Linux, and that Linux which integrate illegally copied code from IBM (BUT owned by IBM), should be stopped being distributed .
Their bold step 2 (as in "2. ???") would probably be to ask the judge for allowing them to sell "licenses", allowing a user to still use legally the same Linux code which incorporate code from IBM which was illegally put into Linux. However this is likely to conflict with the GPL, so they need the judge to take a big magic decision there, as in "SCO has been harmed, so they can release code under the GPL ignoring such and such provision, but only once they have agreed with Linus about the exhaustive list of sections of code for which IBM illegally contributed code".
Does George Lucas own the copyright to every Star Wars book ever published--say, the Timothy Zahn trilogy?
That's exactly the point. Georges Lucas owns the trademarks associated with Star Wars, so even if Timothy Zahn owned 100% of the copyright of his works, it would be illegal for him to distribute it without an agreement with Georges Lucas.
And it's the same problem can happen with OSS with respect to patents - you can be forbidden to distribute code which violate a patent you don't own.
They once acknowledged there was no code owned by SCO that was copied into Linux, unlike what they say in press release. I think their most precise claim would be that code was illegaly copied into Linux.
So, while they have complied in legal terms, they have weakened their case in a significant order of magnitude. Not only that, but they have also weakened their case for any of their prosprective targets in their scoSource shakedown fiasco
Not really. If they managed to prove that IBM violated its contract by integrating some of their code into Linux, the judge may decide that Linux distribution should be stopped unless SCO agrees. Worse, she may decide that distribution should be stopped until it is figured out exactly what code, illegally put in Linux by IBM, should be removed.
What they did in their PR releases is put the entire OSS community on a false track. In reality, it always have been about a contract with IBM, as Linus noted.
This is correct. SCO claims that IBM has all the rights, but at the same time, the contract they have with them prevent them to release it into Linux.
Well lastest SCO legal info show they did it: "17. In some additional cases it was also possible to infer with reasonable certainty from comments in the source code that the engineer who implemented that code had experience and knowledge of the methods, sequence and structures used in either or both of Dynix/ptx and AIX. Confirmation of this would require depositions from the IBM individuals involved in programming the actual Linux modules in question."
Also: "18.1 The "Read Copy Update" (RCU) code contributed to Linux by IBM was copied substantially verbatim with only relatively minor changes from Dynix/ptx."
You said they needed to access IBM code to support their claim, and that is the very problem. You cannot claim if you do not have some evidence in the first place.
Yes, that's why they had probably given a handful of examples of developpers likely to have worked on both UNIX version of a code, and a released version of Linux of the same code. According to their personnal legal theory, I think they don't even have to claim that's it's the same person who wrote the same code in both OSes, just something like having worked on UNIX internals before contributing on Linux code could be problematic. But that would be the weaker claim, they'd go first for developpers having worked on the same code (say JFS) for both AIX and Linux. Given that newer internal versions of AIX were ditched, I'm sure at least some developpers might have been realocated from AIX teams to Linux teams.
SCO have repeatedly claimed that Linux also is guilty of copyright infringement directly on SCO products, not just AIX.
As I initially said, SCO claims in PR releases have been stupid, but they of course carefully avoided doing them boldly in front of the court, hence the title "They ARE NOT stupid scum..."
"All code contributed by IBM" isn't specific enough, but anyway their claim would be: "All Linux code contributed by IBM employees who have worked once on UNIX source tree". They have lots of choice and probably they can list the files in fs/jfs (that is: akefile jfs_dmap.h jfs_inode.c jfs_txnmgr.c jfs_xtree.h file.c jfs_dtree.c jfs_inode.h jfs_txnmgr.h namei.c inode.c jfs_dtree.h jfs_lock.h jfs_types.h resize.c jfs_btree.h jfs_extent.c jfs_logmgr.c jfs_umount.c super.c jfs_debug.c jfs_extent.h jfs_logmgr.h jfs_unicode.c symlink.c jfs_debug.h jfs_filsys.h jfs_metapage.c jfs_unicode.h xattr.c jfs_defragfs.h jfs_imap.c jfs_metapage.h jfs_uniupr.c jfs_dinode.h jfs_imap.h jfs_mount.c jfs_xattr.h jfs_dmap.c jfs_incore.h jfs_superblock.h jfs_xtree.c ), in entirity, for starters.
They have claimed that mathematicians and engineers have determined that lots of source code is contaminated in this way.
They did not claim it in front of the court though. In front of the court, they claimed they needed to access IBM development history to support their claims, which may be perfectly true, because I don't see them have a simple way to identify all/most of the IBM employees who worked on both Linux and UNIX, on what and when.
It is not unfair to let a defendant see what he has to defend against.
However, they had no way initially to know exactly the details about who developped what at IBM. I suppose they just gave one or two examples of a IBM developper who worked on both UNIX and Linux, and ask the judge, on the basis of this evidence, to access IBM information in order to give a more precise list.
Apparently saying "SCO are stupid scum" gets you moderated "+3 Insightful", while being one of the few posts giving logical reasons why there are not as stupid as people think, gets you mod-down as a troll.
LOL
But the answers are very simple:
SCO not only charges that SCO's IP got into Linux against their wishes, but that IBM did it. IBM wants to see SCO's evidence that is had to be IBM and couldn't be someone else.
And? IBM actually contributed to Linux kernel, there is no question about this. SCO is claiming that code, at least some of it, according to the IBM contract, couldn't be put in Linux. They need details of IBM development process to prove it - even if it were actually true, they would need it.
As far as what SCO wants this case to be about, SCO has contradicted itself on so many occasions that it's impossible to say with any certainty what SCO is suing over.
If you ignore press releases, SCO position is consistent. The trial as always been about a breach of contract by IBM which is making IBM contributions to Linux illegal - although they tried to muddy the waters at last hearing, in order to force the judge to compel IBM to release evidence "we don't know what kind of IP violation it is yet, we must see a log of IBM actions first".
So if you will excuse me, I will continue to believe that SCO are stupid scum, because they've not shown any evidence to the contrary.
Not so much. If IBM had signed a non-compete clause with SCO for instance, IBM would lose this trial instantly, and Linux distribution would become problematic. The whole crux of this story is in the detail of IBM contract with SCO (and SCO with Novell), and nowhere else.
Press releases from SCO have always been deceiving bullshit, as that's the case for many companies, although SCO has raised this to an unmatched level.
Maybe but it is distributed on and by the mplayer site, just click on "downloads".
You can compile mplayer entirely from source with DivX, mpeg4 and Quicktime support.
Yes. But they still distribute those binary codecs in clear violation of the law.
Nowadays, Zionism is understood as "all of Palestine should belong to Israel". Because 1) Israel already exists and always will (a few atomic bombs are there to ensure this), 2) it is this form of Zionism is the initial official program of Likud: " Self-Rule:
The Government of Israel flatly rejects the establishment of a Palestinian Arab state west of the Jordan river
[...]
The Jordan River as a Permanent Border:
The Jordan Valley and the territories that dominate it shall be under Israeli sovereignty. The Jordan river will be the permanent eastern border of the State of Israel. The Kingdom of Jordan is a desirable partner in the permanent status arrangement between Israel and the Palestinians in matters that will be agreed upon. "
This was illustrated by the vote in 2002 by the Likud, against Sharon will, of the "refusal to accept any Palestinian state".
Anti-Zionists support the Palestinian thug-ocracy, and by extension, their desire to eliminate Israel altogether by whatever murderous and terroristic means necessary
No, anti-zionits simply said go back to 1967-borders, sometimes even to 1947 borders, and that Israel isn't entitled to more. They denounce settlements.
An anti-Zionist assigns moral equivalence between the tactics of the IDF and Hizbollah.
Indeed there is moral equivalence between the implicit wish to drive all Palestinians out of Palestine, and fund the Great Israel, and the Hizbollah. Ethnic cleansing, at the heart of settlement and annexation plans when implemented. Sometimes of course, Zionists will settle for gaining only half of Palestinian territory, or, when they don't support ethnic cleansing (by moving all Palestinians to Jordan) for an appartheid state, where arab citizen will be have only limited rights so that they don't "threaten Israel existence or safety", which obviously can only mean limited/no rights to vote or be elected.
I'm missing the fundamental difference between anti-semitism and anti-zionism.
Yes you are.