I'd personally like to see Novell hire the SAMBA team. It would be pretty cool to see them take back the file and print server space from MS using their name on OSS.
And the point would be??? The main reason current Netware customers have to keep purchasing Netware is file and print services. Novell Linux strategy is to port and SELL Netware services on Linux.
SAMBA is a competitor to them. It does to closed source file and print what Linux does to closed source OS. Do you really want key SAMBA developpers hired by a company that has a vested interest in keeping file and print closed source?
But does Novell have the cash to continue development of all these?
According to their last quaterly report they have $626 million in cash and their loss was $26 million. At this rate they can last a few years before going chapter 11.
Not trying to be a jerk but you should know what is on your network and if you don't, then you're not paying attention and/or trying hard enough.
That depends on the network. When you have 8,000 desktops spread over sixty cities on three continents, 1000 traveling users with laptops that can connect on the local office LAN in any city they go, 500 servers, over 150 staff members authorised to perform moves and changes that report to three different directors, a hot-line that receives 500 support request per day, wireless access points all over the place and VPN connection points for hundreds of remote Internet users; there is no way you can track at all time what is on your network no matter how hard you try or how good your monitoring software is. Just making an inventory that doesn't get obsolete before it is completed can be quite a challenge. Don't laugh, 10,000 employees companies are not that big or uncommon and that is the kind of network they have.
What would be wrong with a class action for slander? The Linux/Debian/OSS hackers have made conscious efforts to develop Linux independently of other OS for the exact purpose of avoiding a situation such as this one. They are willing to promptly fix any infringing code that may have escaped their efforts to keep Linux clean. The hackers have been harmed in their reputation by SCO's unsubstantiated allegations, being depicted as people that do not care about other's copyrights when it is the exact opposite. Reputation is one of the major rewards for the volunteer work of a hacker. Worse even, the hackers' ability to do business as consultants is affected by their customers' fear of lawsuits. All this could be the basis of a class action.
The best defense against slander is often to claim the allegations are the exact truth, meaning that SCO might have to disclose the infringing code. If they don't, they give in to the action. A class action may work as an effective put up or shut up legal action.
The class action could request a temporary injunction to stop them making further allegations until the trial. Either they are shut up by the court, or they try to demonstrate they have a sufficiently good case to make the injunction unnecessary. This may force them to disclose the infringing code even faster.
Once the infringing code is disclosed, the FUD is over. Either the allegations will be refuted, or the infringing code will be replaced. The IBM lawsuit will become a contractual dispute like any others without impacts oustide SCO and IBM.
Remember the whole idea of this lawsuit is in the FUD and the FUD is allowed to exist only because they don't disclose the code. The real battle is in getting them to disclose the code. All the rest is only skirmishes.
Actually SCO doesn't care if anyone can distribute Linux according to the GPL or not. SCO knows that Linux copyright owners just want Linux to be freely distributed. Linus and other contributors will not sue anyone for distributing Linux (except perhaps SCO themselves). They just want to find out which code infringes copyrigths and change it ASAP to put the GPL back on its track. SCO is counting on that to continue their little games.
There was a post a while ago that some companies might have a legal access to SVR V source code because sources were distributed with older Unix systems. Why not have someone with access to such source run a program to compare SVR V and Linux sources to find matching code?
That would provide an exhaustive list of all code that could potentially be infringing copyrights. Of course infringement of SCO's alleged contract rights for so called derivative work would not be included, but that can't impact Linux users. IBM is the only party to the contract isn't it?
The future of wireless security is
802.11i But this standard uses a different encryption scheme than WEP, therefore some hardware upgrade will be required. There is an interim standard called WPA that combines some features of 802.11i with the encryption algorithm of WEP allows only software/firmware upgrades.
He is saying that innovation will no longer come from companies like IBM or Oracle, but from the development of new business processes.
Software organisations like IBM, Oracle and Open Source projects can still innovate on software and features. O'Reilly just says that you can't build new commercial empires based on software and features alone. Open Source competition will make sure there will not be much money to be made this way. The consequence is commercial successes will come from the ability to take advantage of the free (as speech and beer) software.
In fact, if you replace the word commodity with the word marginalization in his interview, you'll better see my point. And as software becomes more and more marginalized, the value of the software as well as those of us who write the software drops.
Don't worry. The new business processes can't exist without code. Peoples that can think new business processes usually can't code by lack of skill or lack of time. Software writer will enter a symbiotic relationship with these guys and the value of the coder will come from its ability to bring the business process implementer where he wants to go. O'reilly really meant commodity, not marginalization.
Yes, but given back very little in comparison to what they've got. The projects Apple has "borrowed" from would exist without apple, OS X most likely would not without them.
The rule for contribution is you give the output of your brain(s) and you receive the output of all contributing brains. Open Source contributors are bound to receive way more than they give no matter how much they contribute. There would be a problem if Apple were a leech, but as long as they contribute something, it is OK.
But then again, they've chosen their licenses so that they allow this, and it's entirely okay if Apple takes everything and never gives anything back.
The distribution of derivative works is not the only way to contribute. Ethically, the fact a license allows a proprietary derivative work does not dispense them from contributing otherwise.
If Boies is their lawyer in this matter, and he's on contingency, what other legal fees do they have to worry about?
Big trials like that don't work with just one lawyer. They must have a big team to do all the legal research and case preparation before going to court. Even if Boies is on contingency, the other lawyers may not be and they would cost a lot.
As pointed out in the parent post (well the parent of the parent of this post), it is possible SCO hired Boies on contingency just to show in court and make the news headlines and give the brunt of the work to the other lawyers.
This could mean Boies is not that involved in the case after all. He may have been hired for his notoriety and ability to publicly plea in court while other lawyers are doing the bulk of the work from the shadows.
Note the deliciously ambiguous statement:
the company revealed Wednesday that it doesn't have to bear the brunt of much of its legal costs.To pursue its case against IBM, SCO hired high-profile attorney David Boies, famous for his antitrust victory over Microsoft as well as his loss in the vote-counting controversy representing Al Gore in the 2000 presidential election.
SCO's legal costs are being paid under a contingency arrangement, McBride said. In such cases, lawyers typically are paid not by the hour, but with a percentage of whatever money they can win for their clients in the case.
This only means that SCO's won't have to bear the brunt of Boies's contract without talking about the other legal fees.
If this theory is confirmed, IBM can still bleed their cash to death by dragging the case.
Aside from commenting in Slashdot (and hoping that IBM is reading), is there a better way to share this useful research?
IANAL but here are my two cents.
Someone could set-up a "site of SCO factoids" where all intesting links are collected in one place for reference to the volunteers researchers. As more factoids are discovered, they are added to the site. With a little chance, this will lead to a more effective research.
Having all the evidence at one place would also help counter SCO's FUD, especially if the page becomes known to industry analysts and press. If the site shows for instance how SCO's claim change overtime, their FUD campaign will become much more apparent to CIOs and CEOs that are considuring using Linux. Some essay explaining basic concepts such as trade-secrets, copyrights, patents and derivative work would help debunking many misconceptions.
If geeks makes a habit of immediatelyposting new SCO claims on the factoid site and immediately explaining what the claims are, then it will become harder for SCO to increase the FUD pressure by confusing issues.
All OSS homepages should link to the SCO factoid site home page. This will increase the likelihood this site turns up as a result on a Google search for SCO.
IBM has acces to both Linux an System V code. They can compare to find all potentially litigious code and then call upon geeks to determine to origin of "code of interest". If they want stay on a legally safe ground, they may even include with such requests some code that doen't match System V. This could help avoid giving ground to claims they indirectly disclose SCO's alleged trade-secret.
If IBM dare identify publicly the Linux code that matches System V code, the list could be posted on the site of SCO's factoids. The result of research on the code origin could be posted as well. The general public would then be able to watch SCO's claims deflate day after day. If some code is found to be of dubious origin, the the public will be able to see how fast the Linux community can correct the issue. Tracking dates when information is posted on the page will be extremely useful.
Linux could also be compared with the ancient Unixes for matching code. This portion of Linux will then be cleared from any trade-secret claims.
Again the result of this search could be posted on the SCO factoids page.
Actually, Oyez is the second person of the imperative for the French verb ouir and it is now fallen into disuse. Although the correct English equivalent is "hear", Oyez still survive in the French culture in the sense of listen because heralds used to precede public annoucements with the shout Oyez! Oyez!. We can still sometimes hear that in French period films.
The only thing this will achieve is Microsoft will raise their price 10% and blame it on the regulation. There will be no impact on their sales.
If a government really wants to enforce Open Standards, they must specify the selected standards in some regulation akin to the building code for construction. Then they should create an independent agency to certify compliance through a public and non discrimatory procedure. If governements don't want to go that road, then they should leave the market play its role.
If someone assume user has software X for their business then software X is a platform.
If software X is a platform, then it is a competitor.
Defeating a competitor is making sure it gets no share of the business.
The rest is a brief overview of how they proceed. Basically they don't do the job themselves. They convince others, especially software developpers, to do it for them. This works best when the others are not aware of the alternatives and consequences of their choices. When they get enough people working their way, a critical mass is achieved and Microsoft's platform becomes an unavoidable standard.
This is extremely efficient because of the number of people they can get working for their goal without having them on the payroll.
The power of this article is not in the novelty of the story, it is in illuminating how the whole thing works.
So they are saying the SysV license gives them rights to all derivitive works.. sort of a proprietary version of the GPL's "viral" nature.
So if they win, the GPL also gains strength via precedent, but free software loses credibility as a development model.
And if they lose, the GPL/loses/ credibility as a solid license, and it's enforcability comes into question.
It is the opposite. SCO's argument is if you create something that adds to System V, then it is a derivative work and you are tied to SCO's terms. The GPL states everybody owns the rigth to create derivative work as they want provided the same terms will apply to the derivative work.
SCO's viral argument is nobody is free from them because of the definition of the derivative work. The GPL viral nature is everybody must be free because of the terms of the GPL.
If SCO's view prevails, Proprietary software will lose credibility because you don't know when someone will claim your work based on this software is a derivative. Free software will be a refuge because you are guaranteed by the terms of the license to own the rights to derivative works.
If SCO looses, then only the definition of derivative work is affected. The terms of the GPL will still be valid.
can I point out that underneath it all e-mail has always been considered an UNRELIABLE technology, yet it is the most succesful internet application ever to this point in history
People care about voice reliability. They don't mind if data has the reliability of Windows. Bring VoiP in the picture and it is another story.
I understand why there are people who have trouble grasping the costs associated with non-physical goods (like bandwidth)
Bandwidth is a physical good, at least as physical as CPU Mhz, memory MB or disk GB. The OSI model has a physical layer. Don't you purchase bandwith when you buy LAN gear?
What make you believe wireless is a natural follow up to land lines? Before the telcos installed fiber, they used wireless microwave links in their backbones, many of them are still in operation. Satellite are also in essence wireless. Both wireless and wired technology have existed concurrently for decades.
1. The rapidly increasing bandwidth and range of WiFi and its derivatives. In less than 5 years, we have seen WiFi move from a fringe technology to mainstream deployment, with the 2nd generation (802.11g, just ratified as a standard) increasing bandwidth by 5-fold.
This means nothing. The spectrum is a shared medium. The bandwidth is pooled among all its user. In comparison, each wire owns its own spectrum. If we lack bandwidth on wireless we are stuck unless the regulation frees more spectrum, and this means it is no longer available for other usage. If we lack bandwidth on a wire, we pull another wire. Telcos have dropped microwave in favour of fibre partly because a fibre equipped with DWDM and OC-192 can carry over terabits/sec of data.
2. The increase in applications that exploit peer-to-peer or networked models. The problem with developing networked, distirbuted applications is that they take a different mindset than the ones used to create a single app with a single purpose for a single user on a single machine. As more and more applications adopt these more sophisticated network modesl (e.g. Napster, Gnutella, Jabber, Groove Networks, JINI, JXTA, etc.), the technology will get better.
Do not confuse applications and networks. I know, many applications like to call themselves "networks" (e.g. the so called "Novell" networks) because they enable and manage some form of communication between distributed devices. They are no substitute to hardcore network devices such as switches and routers. The P2P "networks" you refer to are applications that run on OSI layer 7. They require a well engineered physical network to run.
3. The number of people who depend exclusively on their cell phones (a related wireless technology), rather than home phones. Such a cultural change will cut into telco revenue--already has.
Yes. This means the network model from the original question assumes the peering devices are mobile to a large extent. Have you thought of the impact on the routing tables? You can't engineer the traffic that way. The network has no stable state.
4. The number of people who use cable for broadband, not DSL (especially in urban areas). Same as the above, for telcos.
And the point is...? A cable company is a telco. Some of them even offer telephony services in countries where regulation allow it. Remember also that broadcast video is a standard telco offering that can be obtained from the large incumbent telcos. When a TV network broadcasts a hockey game, they often lease a land line from a telco to bring the feed to the TV station. Telcos land lines also often carry signals from the station office to the antennas on some towers or mountain tops. And conversely many of the larger cable companies have offered data services for years, competing head-to-head with the incumbent. This is definitely the case with Videotron and Rogers/Shaw here in Canada.
In short, a cable company is a telco with a large stake in video, a different local loop technology and a different regulatory status. But the technology in their backbones is the same.
5. The recurrence of hotspots and "free community networking" as a meme in techno-cultural discourse. Good ideas that don't die prove they have a germ of truth, and only add momentum over time. The final outcome is rarely what everyone would expect (e.g., free wireless for everyone, everywhere), but any good idea that won't go away has proven itself. And who does that impact? The ISPs. When every node is able to negotiate its own entry into the network--who
...with the management of routing tables. Ever heard of convergence time and route flapping?
WiFi speed is a red herring
on
Is 3G Irrelevant?
·
· Score: 3, Interesting
I have not checked all carriers, but the carrier I know of doesn't pull fibre to install a Wifi hot spot. They connect it to a DSL line instead. Guess where the bottleneck is.
Although WiFi speed is irrelevant its existence still cause trouble to 3G deployments. WiFi hot spots just eats the bulk of the users in high population density areas and divert associated revenues from 3G. The business case for 3G is severely weakened.
From a speed point of view, it would make sense for carrier to skip 3G and go directly to 4G. But speed is not enough. They need an attractive application to get customers. Mobile Internet could be it provide they stop billing by usage as other posters have mentionned. They must also understand people will want to use the full potential of real Internet, not just the subset available through AvantGo and other WAP services.
There is also a need for a cheap PAN that can connect the PDA, the laptop and the mobile phone and also other portable devices such as digital cameras and camcoders as well as MP players. Customer would then move pictures, video and audio recordings over the net.
I wonder if the Open Source movement could try to weaken the "patent standard" business model. It would go as follow:
Develop an open source version of the patent crippled standard.
Publish the source under some nearly free licence, that is a license that is equivalent to a free license when there is no patent but require the user of the software to acquire proper licinse otherwise.
Develop an publish a free specification that could serve as a reference for a "free de facto standard" to compete with the patent crippled standard. An open source project could serve as a "standard body" for that purpose.
Develop a free software according to the free specification.
Develop a common API for both the fre and patent crippled standard.
Distribute to whole thing to OEMs and other corporations at no cost.
That would achieve the following:
The nearly free implementation of the patent crippled standard would be attractive to OEM because of its low cost. Therefore would likely become widely used.
The OEM would get the free standard as well at no extra cost.
Because of the common API, the OEM would incur almost no additional effort to support both standards, making the free standard widely used as well.
The free standard would be the only one available on GPLed software.
Once the free standard become widely used, the usefullness of the patent crippled standard will decrease. As GPLed software become more prevalent on the marketplace, the crippled standard become less and less attractive. With a little chance the revenue stream of the patent holder will dry out. Will even more chance, the open specification will end-up approved by an official standard as an official standard.
If the Open Source movement does this systematically, standard bodies will think twice before approving patented standards.
XviD and Theora may be an opportunity to test that idea.
But then again GPRS is a pretty slow channel, and you'd probably want to browse in text only mode anyway. On that scale, 1M is a lot of content.
You are right. 1M per hour is for pretty intensive usage. Other people may see something more like 500K/hour. I have tried browsing without images. Doesn't work well when you explore for sites using Google because there are too many sites where the text is coded into the images, especially for links. You can't find what you want unless you see them. Turning off images works when you go to favorites though.
Fido also includes with its service a proxy that does image compression to optimise link speed. I haven't tried it yet.
Slashdot Mirror
SAMBA is a competitor to them. It does to closed source file and print what Linux does to closed source OS. Do you really want key SAMBA developpers hired by a company that has a vested interest in keeping file and print closed source?
Sorry, I couldn't resist.
The best defense against slander is often to claim the allegations are the exact truth, meaning that SCO might have to disclose the infringing code. If they don't, they give in to the action. A class action may work as an effective put up or shut up legal action.
The class action could request a temporary injunction to stop them making further allegations until the trial. Either they are shut up by the court, or they try to demonstrate they have a sufficiently good case to make the injunction unnecessary. This may force them to disclose the infringing code even faster.
Once the infringing code is disclosed, the FUD is over. Either the allegations will be refuted, or the infringing code will be replaced. The IBM lawsuit will become a contractual dispute like any others without impacts oustide SCO and IBM.
Remember the whole idea of this lawsuit is in the FUD and the FUD is allowed to exist only because they don't disclose the code. The real battle is in getting them to disclose the code. All the rest is only skirmishes.
As always IANAL.
Actually SCO doesn't care if anyone can distribute Linux according to the GPL or not. SCO knows that Linux copyright owners just want Linux to be freely distributed. Linus and other contributors will not sue anyone for distributing Linux (except perhaps SCO themselves). They just want to find out which code infringes copyrigths and change it ASAP to put the GPL back on its track. SCO is counting on that to continue their little games.
That would provide an exhaustive list of all code that could potentially be infringing copyrights. Of course infringement of SCO's alleged contract rights for so called derivative work would not be included, but that can't impact Linux users. IBM is the only party to the contract isn't it?
The future of wireless security is 802.11i But this standard uses a different encryption scheme than WEP, therefore some hardware upgrade will be required. There is an interim standard called WPA that combines some features of 802.11i with the encryption algorithm of WEP allows only software/firmware upgrades.
Big trials like that don't work with just one lawyer. They must have a big team to do all the legal research and case preparation before going to court. Even if Boies is on contingency, the other lawyers may not be and they would cost a lot.
As pointed out in the parent post (well the parent of the parent of this post), it is possible SCO hired Boies on contingency just to show in court and make the news headlines and give the brunt of the work to the other lawyers.
Note the deliciously ambiguous statement:
This only means that SCO's won't have to bear the brunt of Boies's contract without talking about the other legal fees.
If this theory is confirmed, IBM can still bleed their cash to death by dragging the case.
For a corporation that has thousands of Intel boxes and wants to switch to Linux, OSX is not an option.
Someone could set-up a "site of SCO factoids" where all intesting links are collected in one place for reference to the volunteers researchers. As more factoids are discovered, they are added to the site. With a little chance, this will lead to a more effective research.
Having all the evidence at one place would also help counter SCO's FUD, especially if the page becomes known to industry analysts and press. If the site shows for instance how SCO's claim change overtime, their FUD campaign will become much more apparent to CIOs and CEOs that are considuring using Linux. Some essay explaining basic concepts such as trade-secrets, copyrights, patents and derivative work would help debunking many misconceptions.
If geeks makes a habit of immediatelyposting new SCO claims on the factoid site and immediately explaining what the claims are, then it will become harder for SCO to increase the FUD pressure by confusing issues.
All OSS homepages should link to the SCO factoid site home page. This will increase the likelihood this site turns up as a result on a Google search for SCO.
IBM has acces to both Linux an System V code. They can compare to find all potentially litigious code and then call upon geeks to determine to origin of "code of interest". If they want stay on a legally safe ground, they may even include with such requests some code that doen't match System V. This could help avoid giving ground to claims they indirectly disclose SCO's alleged trade-secret.
If IBM dare identify publicly the Linux code that matches System V code, the list could be posted on the site of SCO's factoids. The result of research on the code origin could be posted as well. The general public would then be able to watch SCO's claims deflate day after day. If some code is found to be of dubious origin, the the public will be able to see how fast the Linux community can correct the issue. Tracking dates when information is posted on the page will be extremely useful.
Linux could also be compared with the ancient Unixes for matching code. This portion of Linux will then be cleared from any trade-secret claims. Again the result of this search could be posted on the SCO factoids page.
Actually, Oyez is the second person of the imperative for the French verb ouir and it is now fallen into disuse. Although the correct English equivalent is "hear", Oyez still survive in the French culture in the sense of listen because heralds used to precede public annoucements with the shout Oyez! Oyez!. We can still sometimes hear that in French period films.
If a government really wants to enforce Open Standards, they must specify the selected standards in some regulation akin to the building code for construction. Then they should create an independent agency to certify compliance through a public and non discrimatory procedure. If governements don't want to go that road, then they should leave the market play its role.
- If someone assume user has software X for their business then software X is a platform.
- If software X is a platform, then it is a competitor.
- Defeating a competitor is making sure it gets no share of the business.
The rest is a brief overview of how they proceed. Basically they don't do the job themselves. They convince others, especially software developpers, to do it for them. This works best when the others are not aware of the alternatives and consequences of their choices. When they get enough people working their way, a critical mass is achieved and Microsoft's platform becomes an unavoidable standard.This is extremely efficient because of the number of people they can get working for their goal without having them on the payroll.
The power of this article is not in the novelty of the story, it is in illuminating how the whole thing works.
SCO's viral argument is nobody is free from them because of the definition of the derivative work. The GPL viral nature is everybody must be free because of the terms of the GPL.
If SCO's view prevails, Proprietary software will lose credibility because you don't know when someone will claim your work based on this software is a derivative. Free software will be a refuge because you are guaranteed by the terms of the license to own the rights to derivative works.
If SCO looses, then only the definition of derivative work is affected. The terms of the GPL will still be valid.
As usual IANAL.
This means nothing. The spectrum is a shared medium. The bandwidth is pooled among all its user. In comparison, each wire owns its own spectrum. If we lack bandwidth on wireless we are stuck unless the regulation frees more spectrum, and this means it is no longer available for other usage. If we lack bandwidth on a wire, we pull another wire. Telcos have dropped microwave in favour of fibre partly because a fibre equipped with DWDM and OC-192 can carry over terabits/sec of data.
Do not confuse applications and networks. I know, many applications like to call themselves "networks" (e.g. the so called "Novell" networks) because they enable and manage some form of communication between distributed devices. They are no substitute to hardcore network devices such as switches and routers. The P2P "networks" you refer to are applications that run on OSI layer 7. They require a well engineered physical network to run.
Yes. This means the network model from the original question assumes the peering devices are mobile to a large extent. Have you thought of the impact on the routing tables? You can't engineer the traffic that way. The network has no stable state.
And the point is...? A cable company is a telco. Some of them even offer telephony services in countries where regulation allow it. Remember also that broadcast video is a standard telco offering that can be obtained from the large incumbent telcos. When a TV network broadcasts a hockey game, they often lease a land line from a telco to bring the feed to the TV station. Telcos land lines also often carry signals from the station office to the antennas on some towers or mountain tops. And conversely many of the larger cable companies have offered data services for years, competing head-to-head with the incumbent. This is definitely the case with Videotron and Rogers/Shaw here in Canada.
In short, a cable company is a telco with a large stake in video, a different local loop technology and a different regulatory status. But the technology in their backbones is the same.
...with the management of routing tables. Ever heard of convergence time and route flapping?
Although WiFi speed is irrelevant its existence still cause trouble to 3G deployments. WiFi hot spots just eats the bulk of the users in high population density areas and divert associated revenues from 3G. The business case for 3G is severely weakened.
From a speed point of view, it would make sense for carrier to skip 3G and go directly to 4G. But speed is not enough. They need an attractive application to get customers. Mobile Internet could be it provide they stop billing by usage as other posters have mentionned. They must also understand people will want to use the full potential of real Internet, not just the subset available through AvantGo and other WAP services.
There is also a need for a cheap PAN that can connect the PDA, the laptop and the mobile phone and also other portable devices such as digital cameras and camcoders as well as MP players. Customer would then move pictures, video and audio recordings over the net.
- Develop an open source version of the patent crippled standard.
- Publish the source under some nearly free licence, that is a license that is equivalent to a free license when there is no patent but require the user of the software to acquire proper licinse otherwise.
- Develop an publish a free specification that could serve as a reference for a "free de facto standard" to compete with the patent crippled standard. An open source project could serve as a "standard body" for that purpose.
- Develop a free software according to the free specification.
- Develop a common API for both the fre and patent crippled standard.
- Distribute to whole thing to OEMs and other corporations at no cost.
That would achieve the following:- The nearly free implementation of the patent crippled standard would be attractive to OEM because of its low cost. Therefore would likely become widely used.
- The OEM would get the free standard as well at no extra cost.
- Because of the common API, the OEM would incur almost no additional effort to support both standards, making the free standard widely used as well.
- The free standard would be the only one available on GPLed software.
Once the free standard become widely used, the usefullness of the patent crippled standard will decrease. As GPLed software become more prevalent on the marketplace, the crippled standard become less and less attractive. With a little chance the revenue stream of the patent holder will dry out. Will even more chance, the open specification will end-up approved by an official standard as an official standard.If the Open Source movement does this systematically, standard bodies will think twice before approving patented standards.
XviD and Theora may be an opportunity to test that idea.
You are right. 1M per hour is for pretty intensive usage. Other people may see something more like 500K/hour. I have tried browsing without images. Doesn't work well when you explore for sites using Google because there are too many sites where the text is coded into the images, especially for links. You can't find what you want unless you see them. Turning off images works when you go to favorites though.
Fido also includes with its service a proxy that does image compression to optimise link speed. I haven't tried it yet.