Sigh. I imagine you use some of this Open Source software sometimes. Please try to get your head around the fact that it would not be possible for such software to exist and for folks like you to benefit from it, unless it was developed. And it would not be developed without a developer community, and that community would not be able to do their work unless they had the right to modify and redistribute the software. Thus, Open Source must be more than just visible source code - it has to include the right to distribute and modify, and it also needs the right for you to use it. So, that's 4 things - source, use privilege, distribution privilege, modification privilege and there's a bit more. Years ago, I wrote down what was necessary for software to be Open Source, and OSI uses that Open Source Definition to classify licenses. It is not an arbitrary thing.
The trademark was applied for years ago, and then OSI abandoned it on, in my opinion, really bad legal advice (sorry, Larry, but that's how I call it). It's still a common-law mark. Maybe they should try to re-register.
Sorry, your comment doesn't make sense. This has nothing to do with who wrote the code, or the GPL. It has to do with the license of the code being compliant to the Open Source Definition, and the source code being available. If we don't demand that of people who brand their products as "Open Soucrce", anything - whether it has source code or not - will be called "Open Source".
The problem is that if nobody enforces standards for the "Open Source" brand, then anything, whether or not it even includes permission to modify the code or redistribute it, will be called Open Source. OSI acts to certify that software conforms to the Open Source Definition. This definition is available for anyone to read and craft a license to fit it. However, we certainly don't need more licenses - there are so many now that the combinatorial problem causes developer pain.
Badgeware, software that requires that you place a badge on your web site, is the problem. SugarCRM is one of the offenders.
In composing the Open Source Definition, my aim was that there would be NO legal burden upon users at all. A user should be able to put OSI Certified Open Source software on a computer and go, without reading the license. That's one reason the OSD prohibited the then-popular "Educational Use Only" provision. Requiring the user to perform something - like transmit an advertising badge - is similarly unacceptable because it would place a legal burden on the user.
In practice, issues like patents and even Sarbanes-Oxley create a legal load for some users, but that is not under our developer's control. We had a license back then that imposed an attribution-related legal load on distributors: the original Apache license required attribution in some advertising. It was awkward, there was no way to implement it automatically, and it would have presented a scaling problem if hundreds of authors required attribution in - for example - an operating system distribution's advertising. The community eventually convinced Apache to drop the advertising provision.
That old Apache license has obvious parallels to more current "badgeware" proposals, but badgeware isn't a new issue for Open Source. Almost a decade ago, I reviewed the first Zope license for SPI. Zope's original submission required a web badge. I convinced their investor, Hadar Pedhazur, to make that a nice request rather than a requirement. Most people honored the request. The GPL contained an attribution provision, and still does. OSD was written to fit GPL. But the burden of the GPL's requirement was only upon developers, and it was the minimum necessary amount of attribution to communicate the author identification and the fact that the user had rights.
A developer can be especially irked when someone else labels that developers work as his own. But protecting that developer's business and taking the commercial value out of such relabeling does not require attribution. It could be done using any share-and-share-alike (GPL-like) license that closes the ASP loophole. But it's still fair for a developer to want to be attributed. I think the minimal practical goals for attribution are:
1) The developer should be able to find it.
and
2) The developer should be able to show to others.
This does not mean that the developer should require others to blow his advertising horn. Some companies already use attribution that isn't visible unless you are looking for it - it's embedded in HTML comments. In contrast, the draft Affero flavor of GPL 3 closes the ASP loophole by implementing the minimal necessary requirement for a developer to preserve a visible message informing the user of their rights. The visibility is to implement a separate goal of FSF: that users appreciate their rights and learn to demand them. Should OSI approve this new Affero license, or should it ask that FSF - rather than demanding a direct channel to users, search for that information and present the result on its own sites?
Open Source software has licensing that conforms to the Open Source Definition and in addition the source code must be available. It is not just anything you think of. Long ago, OSI decided to abandon registration of "Open Source" alone, in my opinion on bad advice from their then legal counsel. However, it's still a common law trademark. But regardless of its trademark status, the community should not tolerate just anything being called Open Source, or you will find Microsoft using it for stuff that includes no freedom to modify, etc.
I wasn't talking about leaky coax, but TV coax - RG-6, -57, -59, etc. Some of the "antenna distribution systems" not only use that, they might even use the same wire that is used to distribute television around the building. Obviously a solution that does not require wire pulling is attractive to these properties, that's why so many of the hotel-room wired Ethernet devices are really a sort of short-haul DSL piggybacked on the room phone line. I bet they feed the WAPs with that pseudo-DSL, too.
When you said "zero-managment access points", were you referring to Cisco's LWAPP architecture
Cisco has a nice product in this niche, maybe with better reliability than consumer equipment and maybe that is worth the price. But most consumer equipment can be put into "LAN Bridge Mode" and would achieve the same thing.
Oh sure, Slingbox is going to switch from being P2P to a multicast provider of copyrighted content in order to let Marriott go cheap on bandwidth.
I am well aware of the legal issues and did not mean to pose them as being simple. Hotel chains have lucrative relationships with sports enterprises.
Cache the Internet on a local server and connected each room via a quad fiber ATM connection.
This proposal not solve problem, which is limtied by the pipe to the building rather than the pipe within the building. Caching works great for static content, but the main bandwidth hog is streaming.
Placing all access points in a single telecom closet for what are generally rather spacious properties requires that 2.4 GHz signals be carried through coaxial cable that is very lossy at that frequency - it might be fair to expect up to 90% of the signal to be lost in the wire. There is an FCC limit on the transmitted power, and even if you manage to boost that at the antenna you will be boosting noise as well. And this attenuation and noise will of course hurt receiving too. This is in general going to result in lower wireless quality than desired, much lower than possible.
Instead, get zero-management access points that do not do NAT, routing, etc, and treat them just like antennas once you set the SSID. Do the protocol processing in the telecom closet with a higher grade of hardware than consumer equipment. Cache DNS and web transfers there. Work with Slingbox to engineer channel aggregation with multicasting that bypasses the home units while transmitting the same programming, because so many of those folks are watching the same sports game. I can think of some interesting approaches to the possible legal issues with Slingbox aggregating channels, no doubt they can as well. Can an in-house video alternative be made as attractive as Slingbox? That's another solution.
I wrote a GPL software project for Merrill Lynch last year. Yes, the Wall Street bank. I make a living helping companies understand Free Software. There are very many large companies that accept, and now understand, the GPL and will accept GPL3 as well.
If software is released under a license which allows a developer to use it and integrate it into their project while maintaining their OWN license, it's a good thing.
Don't you realize that this is a fast path to bankruptcy for the developer? The developer needs GPL3 so that he can dual-license his product and make money from the folks that don't want to share their modifications. It is only fair that people who don't want to share should pay. The person you are calling "the developer" isn't the developer at all, he's just a free-rider. Make him pay!
Here's the story in translation: We want gifts from you developers, and we don't want to share! GPL3 alienates us because we want free gifts and we don't want to share! GPL3 bad, BSD good, we want you to use BSD so we don't have to share! If Free Software were a forest, we'd want to clear-cut it, because that leads to PROFIT FOR US and screw everyone else! Who the hell cares if there's nothing left of the forest when we are done with it! So, give us stuff, now!
Sorry for the flaming, but I hope you can see it's easy to lose patience with this sort of thing. And I lose patience with Slashdot for running this story over and over again.
Yes, it violates common sense for Microsoft to pay Xandros to take the protection, if you judge the agreement at face value. Which is an important point to bring up in talking about this.
Both versions of linux (regardless of your stand on the patents) are safe from microsoft lawsuits
Not at all. You didn't read the Novell-Microsoft agreement. No use of server-client software is safe. No use of systems that send mail is safe. No use of wine, OpenOffice, several other programs is safe. No use of the software on the desktop is safe. And there are some additional ambiguous exceptions that may well apply to anything Microsoft decides they apply to.
I REALLY need to be able to contact you when your state, nation, whatever is considering a bill that would be harmful to Open Source software. To do that, I need to know where you are. This is much more important than the Xandros and Novell matters.
I think that Microsoft is making sweetheart deals with failing Linux distributions so that it can support its political stance in favor of increases in software patenting. They are using Xandros to drive Microsoft's campaign to create a situation in which any competing product must somehow be infringing of Microsoft, or at least is believed to be infringing of Microsoft.
Xandros didn't have the money to pay Microsoft for this. They were dying anyway.
For those who have never heard of Xandros (which will be a lot of you), it's
a commercial distribution descended from Corel's Linux system, funded by the same VCs that funded Ximian, and derived from Debian at one point, although I don't know how much comes from there any longer. They've been around a long time, although for obvious reasons I can't believe they are very successful.
They took the money that Microsoft offered. That's really all the news there
is here - that Microsoft found another foundering commercial Linux distribution
willing to sign up to the patent covenant and give it publicity. The technical aspects are
irrelevant, as they indeed are in the Novell deal. Xandros is a little fish
without significant technology to offer. Even in the case of Novell, nobody
needed Microsoft's help with virtualization - the only thing Microsoft can
offer is the slight performance increment of paravirtualization for Windowsover the full virtualization that is available now.
There's not much to do about Xandros. They aren't a big player, this isn't
going to make them into one. We should turn away from them as was the casewith Novell, but it seems a bit silly since most of us didn't even know they
existed.
And yet the FSF hasn't gone after hardware manufacturers who provide binary drivers for free OSes even though they're already in violation of GPL2.
The FSF doesn't own the copyright of the Linux kernel, and they have not been asked by the copyright holders to help with their defense. One of those two things needs to happen before FSF can intervene.
We'd like some companies to stop using Free Software: the companies that can't comply with the license in both letter and spirit, and insist on engineering loopholes - be they in hardware (Tivo) or in law (Novell-Microsoft). Those companies work to de-motivate the developers of the software that they are using, who contributed their software on a share-and-share-alike basis and expect that to be respected. We would do better without them.
Could a solar near-magnetosphere bow-shock wave be the cause of long-delayed echoes? These are echoes of radio signals that are no multiple of the distances to likely objects. The average ham who is active on HF hears about one a year.
Oh yeah, that GPL stuff never fails. It was a warm, sunny morning in California recently, when Valerie whispered in my ear: Tell me about derivative works, honey! And it just went from there...
Slashdot Mirror
Bruce
Bruce
Bruce
Bruce
In composing the Open Source Definition, my aim was that there would be NO legal burden upon users at all. A user should be able to put OSI Certified Open Source software on a computer and go, without reading the license. That's one reason the OSD prohibited the then-popular "Educational Use Only" provision. Requiring the user to perform something - like transmit an advertising badge - is similarly unacceptable because it would place a legal burden on the user.
In practice, issues like patents and even Sarbanes-Oxley create a legal load for some users, but that is not under our developer's control. We had a license back then that imposed an attribution-related legal load on distributors: the original Apache license required attribution in some advertising. It was awkward, there was no way to implement it automatically, and it would have presented a scaling problem if hundreds of authors required attribution in - for example - an operating system distribution's advertising. The community eventually convinced Apache to drop the advertising provision.
That old Apache license has obvious parallels to more current "badgeware" proposals, but badgeware isn't a new issue for Open Source. Almost a decade ago, I reviewed the first Zope license for SPI. Zope's original submission required a web badge. I convinced their investor, Hadar Pedhazur, to make that a nice request rather than a requirement. Most people honored the request. The GPL contained an attribution provision, and still does. OSD was written to fit GPL. But the burden of the GPL's requirement was only upon developers, and it was the minimum necessary amount of attribution to communicate the author identification and the fact that the user had rights.
A developer can be especially irked when someone else labels that developers work as his own. But protecting that developer's business and taking the commercial value out of such relabeling does not require attribution. It could be done using any share-and-share-alike (GPL-like) license that closes the ASP loophole. But it's still fair for a developer to want to be attributed. I think the minimal practical goals for attribution are:
1) The developer should be able to find it.
and
2) The developer should be able to show to others.
This does not mean that the developer should require others to blow his advertising horn. Some companies already use attribution that isn't visible unless you are looking for it - it's embedded in HTML comments. In contrast, the draft Affero flavor of GPL 3 closes the ASP loophole by implementing the minimal necessary requirement for a developer to preserve a visible message informing the user of their rights. The visibility is to implement a separate goal of FSF: that users appreciate their rights and learn to demand them. Should OSI approve this new Affero license, or should it ask that FSF - rather than demanding a direct channel to users, search for that information and present the result on its own sites?
Bruce
Bruce
Thanks
Bruce
Bruce
Instead, get zero-management access points that do not do NAT, routing, etc, and treat them just like antennas once you set the SSID. Do the protocol processing in the telecom closet with a higher grade of hardware than consumer equipment. Cache DNS and web transfers there. Work with Slingbox to engineer channel aggregation with multicasting that bypasses the home units while transmitting the same programming, because so many of those folks are watching the same sports game. I can think of some interesting approaches to the possible legal issues with Slingbox aggregating channels, no doubt they can as well. Can an in-house video alternative be made as attractive as Slingbox? That's another solution.
Bruce
Thanks
Bruce
Sorry for the flaming, but I hope you can see it's easy to lose patience with this sort of thing. And I lose patience with Slashdot for running this story over and over again.
Bruce
Bruce
Thanks
Bruce
Bruce
Bruce
Not at all. You didn't read the Novell-Microsoft agreement. No use of server-client software is safe. No use of systems that send mail is safe. No use of wine, OpenOffice, several other programs is safe. No use of the software on the desktop is safe. And there are some additional ambiguous exceptions that may well apply to anything Microsoft decides they apply to.
Bruce
Thanks
Bruce
Xandros didn't have the money to pay Microsoft for this. They were dying anyway.
Thanks
Bruce
Those of you who would like to tell Xandros what you think may do so here. You may also tell Novell here. - Bruce
Consider the difference between from and to.
They took the money that Microsoft offered. That's really all the news there is here - that Microsoft found another foundering commercial Linux distribution willing to sign up to the patent covenant and give it publicity. The technical aspects are irrelevant, as they indeed are in the Novell deal. Xandros is a little fish without significant technology to offer. Even in the case of Novell, nobody needed Microsoft's help with virtualization - the only thing Microsoft can offer is the slight performance increment of paravirtualization for Windowsover the full virtualization that is available now.
There's not much to do about Xandros. They aren't a big player, this isn't going to make them into one. We should turn away from them as was the casewith Novell, but it seems a bit silly since most of us didn't even know they existed.
Bruce
Bruce
Bruce
Oh yeah, that GPL stuff never fails. It was a warm, sunny morning in California recently, when Valerie whispered in my ear: Tell me about derivative works, honey! And it just went from there...