The patent on Wind expired. Wind technology is now in the public domain.
How about patenting the patent system itself? Everyone except the holder of the meta-patent would be prevented from filing a new patent. If they did, they'd be infringing the patent on the patent system.
Unfortunately, it wouldn't work, because there is prior art. But will the patent officers at the USPTO notice?
Or use a variation of the same: Patent on a new way of abusing the patent system to make money. Just rename it so something more technical, and peng, you may be lucky to have this patent granted. Then start spreading FUD:-)
Better yet, why don't we just block all inbound email from foreign countries?
Wow, that would help a lot, considering the huge amount of spam generated in the US, and thrown at the rest of the world!
Just don't forget to block all outbound email too. Besides spam, emails may contain sensitive intelligence which could be used by the bad world out there. Homeland security, where are you when we need you?;-)
With dynamic IP address allocation, many IP addresses will eventually end up leading to a Win9x user in some point in time. A few minutes later, the same addy will point to a Linux box...
It's actually funny that static IPs are generally pointing to servers or routers, all of them not running Windows at all!
Tsk, tsk,... they are valid IP addresses. It just happens that they are not routable. And why are they not routable? Because network operators configure their routers in such a way as to null-route (drop) packets with those addresses. Furthermore, routers are configured not to advertise those routes over BGP. That's the only reason why those addresses are not routable. But, believe it or not, they are still full-fledged, first-class IP addresses.
You can't use patches that others have contributed, of course, but you can fork off of your original version plus your own enhancements.
Yes, that's exactly the point. But why should I disclose the code if I were prevented from using the community's work? Isn't the main reason for the original submitter to let others participate in enhancing code [so that they can use it?]
Another important reason to stay clear of the GPL are NDAs.
True, but BSD-style licencing doesn't help here. Openness and freedom is opposed to secrecy.
BSD-style licensing helps tremendously here! You may release a library under BSD, and use it in your own closed-source programs which are bound by the NDAs. Were the library GPL (and you didn't own all the patches/code therein), the programs that use it must be open sourced (in violation to the NDAs). Of course, LGPL-ing the library would be acceptable in this case. But not everything is a library. Remember the bison vs. (Berkeley) yacc comparison? All this is very tricky, and for most companies who plan to open source their code, quite dangerous.
Just remember that you own the original code, not the enhancements nor the contributions by the community.
Without a project which holds all the rights, every single programmer would hold rights to his/her own diffs, and those diffs would be licensed under e.g. GPL. This will prevent you from using anything beyond your initial code in closed-source environments.
If a project is founded to support the code, and if every programmer transferred their rights to the project (i.e. to a valid legal entity), then the project would own the rights. If the project now decides to release the code under GPL, it must do so for everyone, including the original (main?) code submitter. Same problem as before.
Isn't it safer to use BSD in the first place and avoid all this trouble?
Because you are the copyright holder. You hold all the rights.
Thanks for clarifying this!
But beware: you are the copyright holder of your own code, but you don't own the diffs. Therefore, you can only use your initial submission without problems. Every contribution from the community would belong to whomever wrote it (including diffs), and here you're bound by the same terms as everyone else.
A common scenario is to transfer the code to a project, and the project will either GPL or BSD (or whatever) it. In this case, even the original submitter has waived all rights to the code (that's what a transfer of [copy]rights is all about), and, in the case of GPL, will be prevented from using it in a closed-source environment.
So I assume that BSD licensing is safer for the original submitter.
GPL or LGPL doesn't prevent me from reusing my own code in a proprietary fashion - it prevents others from doing so.
If it prevents others from doing so, why wouldn't it prevent me from doing so? It actually prevents both me and others from reusing my own code in a proprietary fashion. I'd be breaking my own licensing terms if I did it. Hmmm.
You may want to maintain two different versions of some components: one open sourced, and another which is not. Now suppose that you licensed the basic technology under GPL (or LGPL), you won't be able to fork off a closed-source branch from it. You'll be effectively locked into open sourcing every enhancement you made to your own product, or products that you want to derive from them.
Since those people want to open source their own product, it's only fair for them if they avoided locking themselves into a mandatory open source license which will prevent them from using their own code in current or future products they want to develop. GPLing code would effectively prevent those companies from using it (this code) in further closed-source environments.
Now guess what they will do? They will either release their code under a (for them) more liberal license like BSD, or they won't open source at all (which is unfortunately more common). After all, the main reason for most companies to open source their code is twofold: have the community work on it, and use the improved code in future products (which they don't want to open source). This is a fair arrangement for the open sourcing company, but also for the community.
Another important reason to stay clear of the GPL are NDAs. Some products are intimately tied to some proprietary interface which falls under a non-disclosure agreement (Yuck!). Now, if the company released, say, a library under GPL, they will have a conflict w.r.t. their products which use this library: the GPL force them to disclose the product's source code, but the NDA prevents them from showing the interface. The obvious, and only possible way of dealing with this situation is to avoid GPLing the code altogehter. [Well, avoiding NDAs would be better, but unfortunately not always possible.]
Actually, many companies in the embedded hardware market (and certainly other markets as well) are avoiding [L]GPLing their code for exactly this reason. They're often ready to give back to the community by open sourcing their projects, but they are afraid of locking themselves out of business. This is exactly the reason for the popularity of BSD-style licenses among companies going open source.
You should avoid the GPL if you want to reuse components (libraries for instance) in further commercial products. BSD license is better suited for this.
It is not illegal to listen to anything that comes accross the airwaves. Decrypting it, thats another story.
Not long ago, using a satellite dish was illegal in many non-democratic countries (it actually still is illegal in some parts of the world). Yet it didn't stop stop people from doing it anyway. Any law that is being perceived as unjustly repressive will eventually be broken.
Breaking the law in dicatorships requires a lot more courage as in democracies. Are we already so intimidated by our own governments to allow them to restrict our freedom that much? Are we less courageous than millions of people living in much more repressive coutries? Uh oh...
you'd want to be able to mount an emitter on potential targets (like, oh say, a commercial airliner)...
Funny that you mentioned this. Unfortunately, it goes both ways. A ground emitter directed at an airliner would be just as deadly. Very few airplanes are hardened enough against this type of attack, and protection is not likely to happen on a wide scale any time soon.
but I believe that inserting his political beliefs into his work project is less than elegant.
Agreed. GNU Privacy Guard did the same as they opposed the Iraq war. Even now, they're still featuring a PACE button at the bottom of their home page. I actually had trouble convincing some more conservative companies to adopt GNUPG as a replacement to PGP after this.
Why do lemmings commit suicide? Because they have no choice, because they don't know any better, and because they follow their leader. Perhaps because their destination is so popular?
Does popularity really matter? After all, Windows is still hugely popular on the desktop, but we use *NIX anyway. And why do we do such a silly thing? Because we can, because its technically better, and because it helps us achieve what we want.
This netcraft survey just proves that webmasters at those domain parking companies are no lemmings (anymore?). That's a good thing. But does it really matter? Would we switch to IIS if its popularity increased? The lemmings would, be we'll resist. We'd rather watch all those lemmings heading to their final destination.
I used RedHat and SuSE distros too, then I switched to FreeBSD altogether. Never regretted this move. The best support is still the user- and developer community.
So long, Red Hat, thanks for all the development support and good luck in the heavily contented enterprise market. Your move to End-of-Life Red Hat 9 will help smaller, but smarter Linux-supporting companies and -vendors. That's what healthy competition is all about, after all...:)
I'm actually asking if anyone knows of a free, OSS or not alternative.
snort is quite useful on *NIX machines. Quoth FreeBSD's security/snort ports description:
Snort is a libpcap-based packet sniffer/logger which can be used as a
lightweight network intrusion detection system. It features rules based logging
and can perform content searching/matching in addition to being used to detect
a variety of other attacks and probes, such as buffer overflows, stealth port
scans, CGI attacks, SMB probes, and much more. Snort has a real-time alerting
capability, with alerts being sent to syslog, a separate "alert" file, or even
to a Windows computer via Samba.
Packets are logged in their decoded form to directories which are generated
based upon the IP address of the remote peer. This allows Snort to be used as
a sort of "poor man's intrusion detection system" if you specify what traffic
you want to record and what to let through.
For instance, I use it to record traffic of interest to the six computers in
my office at work while I'm away on travel or gone for the weekend. It's
also nice for debugging network code since it shows you most of the Important
Stuff(TM) about your packets (as I see it anyway). The code is pretty easy
to modify to provide more complete packet decoding, so feel free to make
suggestions.
Some companies don't accept HTML-formatted emails anymore, because the BULK of all SPAM is HTML nowadays. I implemented for many companies an autoresponder which, upon recognizing HTML mail sends back a reply asking for users to
either respond to the challange, by sending the message back a la ASK (actually it is a modified ASK-like system)
or configure their E-Mail client in such a way as to disable HTML mail at all. This option lists the reasons why HTML mail is bad, and provides a link to a page with illustrated instructions for popular M$ and other MUAs.
Of course, this scheme is not perfect, because it doesn't catch every SPAM, and, more importantly, because it rejects otherwise legitimate email. However, the advantages clearly outweigh the shortcomings, at least for those business users, so they decided to use this method out of sheer necessity.
Any email address which has been around for some years is bound to become a regular SPAM victim. This is especially true for business email addresses dissiminated by biz cards or ads in print media, which cannot be changed easily.
BSD is fading out. It is not in the mainstream. It is like the Amiga.
Linux is fading out. It is not in the mainstream. Unfortunately, M$ is (still?) mainstream. So why should we hack on Linux (or BSD)? They're just used by a tiny amount of the total PC loosers^Husers population. We're getting saddled with a hobby that no one else wants. Right?
Actually, BSD is gaining in popularity, even at the expenses of Linux, which corporate users confidence was being badly shaken by FUD campaigns from SCO and the like. Now, I wounldn't want to take advantage of this, but it is fact that many corporate environments (and I'm not just referring to the ISP industry) quitely switched gear and is migrating towards BSD. Of course, this would probably not have happened without Linux conversion first; but once you've got a Unix-based IT environment, differences between flavors like Linux, BSD, Solaris,... are so small, that they really don't matter that much.
In Germany, one has to pay approx. 60 Euro/Month for an 768/128 ADSL connection with static IP. A perhaps typical setup there is:
20 Euro for T-DSL (german telecom)
30 Euro for unlimited traffic through an independent DSL-ISP (e.g. KAMP DSL)
10 Euro extra for static IP
This ISP doesn't block ports, has no objections of you operating your own servers, and is well-connected to major backbones.
Bigger ISPs are generally more reluctant to give you full access to their backbones, but smaller ISPs compete for customers and this is a very important point for many of them.
The biggest issue with this is T-DSL, which disconnects you every 24 hrs, so you have to immediately reconnect if you operate an HTTP, SMTP or DNS server over their link. Most ppp daemons or ppp routers will do this for you automatically. With a static IP, programs won't even notify the disconnect.
While it is trivial to operate your own domain(s) over an ADSL link, I wouldn't recommend it for more than just home usage purposes: You need to provide redundancy through alternative links, multiple servers (at least hot swappable), uninterruptible power supplies, RAID arrays, etc... You also need backup MXes and would need one or more external DNS servers, e.g. at zoneedit.com or dyndns.org. For a very small number of domains/users, this is too costly. You'd prefer to leverage economics of scale by renting a physical or virtual server in a well-connected data center. Of course, YMMV.
mod_throttle and other non-core modules are normally only available to the 1.3.x series (for now).
Another reason is laziness: If the average admin can't find a package/port/... for an apache 2.x module in their favorite Linux distro or BSD ports collection, they'll normally go for 1.3.x, instead of porting the module to 2.x.
Slashdot Mirror
Wind is already patented?
The patent on Wind expired. Wind technology is now in the public domain.
How about patenting the patent system itself? Everyone except the holder of the meta-patent would be prevented from filing a new patent. If they did, they'd be infringing the patent on the patent system.
Unfortunately, it wouldn't work, because there is prior art. But will the patent officers at the USPTO notice?
Or use a variation of the same: Patent on a new way of abusing the patent system to make money. Just rename it so something more technical, and peng, you may be lucky to have this patent granted. Then start spreading FUD :-)
Better yet, why don't we just block all inbound email from foreign countries?
Wow, that would help a lot, considering the huge amount of spam generated in the US, and thrown at the rest of the world!
Just don't forget to block all outbound email too. Besides spam, emails may contain sensitive intelligence which could be used by the bad world out there. Homeland security, where are you when we need you? ;-)
government of the people, by the people, for the people
That was version 1.0. Govt. 2.0 is: Government of the people, by the burocrats, for the burocrats.. You just missed the update.
All they need for this is a list of IP blocks. Available at any regional registry (RIPE, ARIN, ...).
With dynamic IP address allocation, many IP addresses will eventually end up leading to a Win9x user in some point in time. A few minutes later, the same addy will point to a Linux box...
It's actually funny that static IPs are generally pointing to servers or routers, all of them not running Windows at all!
Tsk, tsk, ... they are valid IP addresses. It just happens that they are not routable. And why are they not routable? Because network operators configure their routers in such a way as to null-route (drop) packets with those addresses. Furthermore, routers are configured not to advertise those routes over BGP. That's the only reason why those addresses are not routable. But, believe it or not, they are still full-fledged, first-class IP addresses.
You can't use patches that others have contributed, of course, but you can fork off of your original version plus your own enhancements.
Yes, that's exactly the point. But why should I disclose the code if I were prevented from using the community's work? Isn't the main reason for the original submitter to let others participate in enhancing code [so that they can use it?]
True, but BSD-style licencing doesn't help here. Openness and freedom is opposed to secrecy.
BSD-style licensing helps tremendously here! You may release a library under BSD, and use it in your own closed-source programs which are bound by the NDAs. Were the library GPL (and you didn't own all the patches/code therein), the programs that use it must be open sourced (in violation to the NDAs). Of course, LGPL-ing the library would be acceptable in this case. But not everything is a library. Remember the bison vs. (Berkeley) yacc comparison? All this is very tricky, and for most companies who plan to open source their code, quite dangerous.
Yes, you're right. I was somewhat confused there.
Just remember that you own the original code, not the enhancements nor the contributions by the community.
Without a project which holds all the rights, every single programmer would hold rights to his/her own diffs, and those diffs would be licensed under e.g. GPL. This will prevent you from using anything beyond your initial code in closed-source environments.
If a project is founded to support the code, and if every programmer transferred their rights to the project (i.e. to a valid legal entity), then the project would own the rights. If the project now decides to release the code under GPL, it must do so for everyone, including the original (main?) code submitter. Same problem as before.
Isn't it safer to use BSD in the first place and avoid all this trouble?
Because you are the copyright holder. You hold all the rights.
Thanks for clarifying this!
But beware: you are the copyright holder of your own code, but you don't own the diffs. Therefore, you can only use your initial submission without problems. Every contribution from the community would belong to whomever wrote it (including diffs), and here you're bound by the same terms as everyone else.
A common scenario is to transfer the code to a project, and the project will either GPL or BSD (or whatever) it. In this case, even the original submitter has waived all rights to the code (that's what a transfer of [copy]rights is all about), and, in the case of GPL, will be prevented from using it in a closed-source environment.
So I assume that BSD licensing is safer for the original submitter.
GPL or LGPL doesn't prevent me from reusing my own code in a proprietary fashion - it prevents others from doing so.
If it prevents others from doing so, why wouldn't it prevent me from doing so? It actually prevents both me and others from reusing my own code in a proprietary fashion. I'd be breaking my own licensing terms if I did it. Hmmm.
You may want to maintain two different versions of some components: one open sourced, and another which is not. Now suppose that you licensed the basic technology under GPL (or LGPL), you won't be able to fork off a closed-source branch from it. You'll be effectively locked into open sourcing every enhancement you made to your own product, or products that you want to derive from them.
Since those people want to open source their own product, it's only fair for them if they avoided locking themselves into a mandatory open source license which will prevent them from using their own code in current or future products they want to develop. GPLing code would effectively prevent those companies from using it (this code) in further closed-source environments.
Now guess what they will do? They will either release their code under a (for them) more liberal license like BSD, or they won't open source at all (which is unfortunately more common). After all, the main reason for most companies to open source their code is twofold: have the community work on it, and use the improved code in future products (which they don't want to open source). This is a fair arrangement for the open sourcing company, but also for the community.
Another important reason to stay clear of the GPL are NDAs. Some products are intimately tied to some proprietary interface which falls under a non-disclosure agreement (Yuck!). Now, if the company released, say, a library under GPL, they will have a conflict w.r.t. their products which use this library: the GPL force them to disclose the product's source code, but the NDA prevents them from showing the interface. The obvious, and only possible way of dealing with this situation is to avoid GPLing the code altogehter. [Well, avoiding NDAs would be better, but unfortunately not always possible.]
Actually, many companies in the embedded hardware market (and certainly other markets as well) are avoiding [L]GPLing their code for exactly this reason. They're often ready to give back to the community by open sourcing their projects, but they are afraid of locking themselves out of business. This is exactly the reason for the popularity of BSD-style licenses among companies going open source.
You should avoid the GPL if you want to reuse components (libraries for instance) in further commercial products. BSD license is better suited for this.
Benchmarks are dying
It is not illegal to listen to anything that comes accross the airwaves. Decrypting it, thats another story.
Not long ago, using a satellite dish was illegal in many non-democratic countries (it actually still is illegal in some parts of the world). Yet it didn't stop stop people from doing it anyway. Any law that is being perceived as unjustly repressive will eventually be broken.
Breaking the law in dicatorships requires a lot more courage as in democracies. Are we already so intimidated by our own governments to allow them to restrict our freedom that much? Are we less courageous than millions of people living in much more repressive coutries? Uh oh...
you'd want to be able to mount an emitter on potential targets (like, oh say, a commercial airliner)...
Funny that you mentioned this. Unfortunately, it goes both ways. A ground emitter directed at an airliner would be just as deadly. Very few airplanes are hardened enough against this type of attack, and protection is not likely to happen on a wide scale any time soon.
but I believe that inserting his political beliefs into his work project is less than elegant.
Agreed. GNU Privacy Guard did the same as they opposed the Iraq war. Even now, they're still featuring a PACE button at the bottom of their home page. I actually had trouble convincing some more conservative companies to adopt GNUPG as a replacement to PGP after this.
I got a bit nervous...
Why do lemmings commit suicide? Because they have no choice, because they don't know any better, and because they follow their leader. Perhaps because their destination is so popular?
Does popularity really matter? After all, Windows is still hugely popular on the desktop, but we use *NIX anyway. And why do we do such a silly thing? Because we can, because its technically better, and because it helps us achieve what we want.
This netcraft survey just proves that webmasters at those domain parking companies are no lemmings (anymore?). That's a good thing. But does it really matter? Would we switch to IIS if its popularity increased? The lemmings would, be we'll resist. We'd rather watch all those lemmings heading to their final destination.
I think he meant crackers ...
Oh, would he be subtle enough to know the difference?
I used RedHat and SuSE distros too, then I switched to FreeBSD altogether. Never regretted this move. The best support is still the user- and developer community.
So long, Red Hat, thanks for all the development support and good luck in the heavily contented enterprise market. Your move to End-of-Life Red Hat 9 will help smaller, but smarter Linux-supporting companies and -vendors. That's what healthy competition is all about, after all... :)
I'm actually asking if anyone knows of a free, OSS or not alternative.
snort is quite useful on *NIX machines. Quoth FreeBSD's security/snort ports description:
Some companies don't accept HTML-formatted emails anymore, because the BULK of all SPAM is HTML nowadays. I implemented for many companies an autoresponder which, upon recognizing HTML mail sends back a reply asking for users to
Of course, this scheme is not perfect, because it doesn't catch every SPAM, and, more importantly, because it rejects otherwise legitimate email. However, the advantages clearly outweigh the shortcomings, at least for those business users, so they decided to use this method out of sheer necessity.
Any email address which has been around for some years is bound to become a regular SPAM victim. This is especially true for business email addresses dissiminated by biz cards or ads in print media, which cannot be changed easily.
BSD is fading out. It is not in the mainstream. It is like the Amiga.
Linux is fading out. It is not in the mainstream. Unfortunately, M$ is (still?) mainstream. So why should we hack on Linux (or BSD)? They're just used by a tiny amount of the total PC loosers^Husers population. We're getting saddled with a hobby that no one else wants. Right?
Actually, BSD is gaining in popularity, even at the expenses of Linux, which corporate users confidence was being badly shaken by FUD campaigns from SCO and the like. Now, I wounldn't want to take advantage of this, but it is fact that many corporate environments (and I'm not just referring to the ISP industry) quitely switched gear and is migrating towards BSD. Of course, this would probably not have happened without Linux conversion first; but once you've got a Unix-based IT environment, differences between flavors like Linux, BSD, Solaris, ... are so small, that they really don't matter that much.
In Germany, one has to pay approx. 60 Euro/Month for an 768/128 ADSL connection with static IP. A perhaps typical setup there is:
This ISP doesn't block ports, has no objections of you operating your own servers, and is well-connected to major backbones.
Bigger ISPs are generally more reluctant to give you full access to their backbones, but smaller ISPs compete for customers and this is a very important point for many of them.
The biggest issue with this is T-DSL, which disconnects you every 24 hrs, so you have to immediately reconnect if you operate an HTTP, SMTP or DNS server over their link. Most ppp daemons or ppp routers will do this for you automatically. With a static IP, programs won't even notify the disconnect.
While it is trivial to operate your own domain(s) over an ADSL link, I wouldn't recommend it for more than just home usage purposes: You need to provide redundancy through alternative links, multiple servers (at least hot swappable), uninterruptible power supplies, RAID arrays, etc... You also need backup MXes and would need one or more external DNS servers, e.g. at zoneedit.com or dyndns.org. For a very small number of domains/users, this is too costly. You'd prefer to leverage economics of scale by renting a physical or virtual server in a well-connected data center. Of course, YMMV.
mod_throttle and other non-core modules are normally only available to the 1.3.x series (for now).
Another reason is laziness: If the average admin can't find a package/port/... for an apache 2.x module in their favorite Linux distro or BSD ports collection, they'll normally go for 1.3.x, instead of porting the module to 2.x.
Absolutely. Since hackers like recursive acronyms, why not invent some kind of "recursive logo"? How would a recursive graphic look like? Hmmm...