Humans have escaped the phenomenon of Natural Selection, for the most part.
All of us who wear glasses? We should have been culled. [etc.]
I disagree. Yes, the simple "the organism dies because of some trait, so there are less offspring with that trait in the next generation" is somewhat true today, as you say. But natural selection extends to much more than that.
For starters, sexual selection. Nowadays people have LOTS of choice in selecting a partner; it is very common to have many partners and even children with multiple partners. In addition, the number of children that people have is very variable - more and more elect to have none, while some people have quite a lot. All of this allows a few 'sexually preferred' individuals to contribute to the next generation's gene pool in a significant way, first in that certain individuals have more opportunity to have children - say, because of attractiveness - and second, that the number of kids is a personal choice that varies greatly.
Now, you talk about wearing glasses, diabetes, etc. - these are 'objective' issues that should be selected against, supposedly: "in the wild, such people would never survive". But the fact that we live in a different environment doesn't mean that OTHER selection pressures, perhaps just as strong, don't exist. They are just different.
Google is expected to release an entire operating system that's supposed to compete heads on with Windows Vista, which is the result of years and years of work and billions of dollars invested? Say what you will about Windows, but it certainly is a massive behemoth with sh*tloads of functionality. You don't just shake something like that out of your sleeve in a few months.
You don't, but you don't need to. You can create your own Linux distro in no time.
they list Novell as part of the OIN, but any patents that Novell has filed are now void because they have signed a promise not to sue.
The promise not to sue refers to customers only (this is how it 'gets around the GPL', supposedly). Novell can still sue Microsoft, so their patents in the OIN are still viable. How much Novell can be trusted is another matter, of course.
What? a choice that would mean that tey wouldn't get paid by people using their framework to earn dosh?
They can still make money, by (1) selling support, (2) charging for a license to extend Qt - i.e. you can link to it for free, but creating a derived work will cost you (or, you must release the derived work under the GPL).
Their expectation to earn money off of every single commercial application that uses their framework - and if KDE were the dominant DE on Linux, then every single commercial application on Linux - is a bit 'optimistic'. On the one hand, there are alternatives (GTK+), and on the other, expecting to monetize all commercial applications in this way is a bit Microsoft-ish - we may as well call it the 'Qt tax'.
>> The commercial license is a bit rediculous, though.
> Why? It seems pretty popular (Adobe, Skype, Google, Opera, etc..) Trolltech's modell is an excellent example of how you can make money on free software.
First, note that all of those companies are not 'open-source' companies (Google standing out somewhat, but still). When you compare Qt to important open-source projects like GNOME and the Linux kernel, you find that both allow commercial development - GNOME is LGPL, and the kernel lets you run whatever you want in userspace. Qt, on the other hand, wants to be paid for commercial development. This is a big deterrent for commercial developers to use Qt. (Especially since you need to decide at the beginning of development which license you want, commercial or free.)
As I see it, the LGPL license was designed for exactly this sort of thing, and I think Trolltech could have made a better choice here. Especially since cross-platform development can be done without paying them (WxWidgets, and also GTK+ can run on various platforms, even Windows).
Yes, TFA was in humor. But is there no basis in fact? In the US, only the ultra-rich run for (and win) the presidency. Bill Gates in the richest man in the US. Now, it's true not every rich candidate can get elected (as nicely pointed out in the book Freakonomics), but still, the possibility can't be discounted. Like Scott Adams says, give Bill a decade or so of charity work, and his popular image will look pretty voteworthy. A chilling thought.
If the submitter RTFP they would have seen that it was Intel, not M$.
Indeed. But if the patent office doesn't read the proposals, why should the Slashdot submitter? (Granted, the other option is that the patent office did read the proposal, but the patent office worker was so ignorant it sounded like a new idea to him/her.)
Currently there are so many variations on this theme already in existence (Skype, Jajah, even Ekiga etc., in a sense), that it boggles the mind such a patent was granted. Yet, TFP says that the filing date was "February 25, 2000". How many of those were around circa 2000? Skype only began around 2002-2003 IIRC...
The macro that is included as a VBA demo in the article http://www.linuxtalk.co.uk/ooo_vba/hypocycloid-dem o.xls works with the OpenOffice that is included with Fedora Core 6. I just tried it on this computer. So the "evil patch" is apparently already present in Red Hat's OpenOffice.
As TFA said, the patch is already included in several distributions. For example, Ubuntu. I tested it on Ubuntu Edgy here and indeed it is included.
Sure, he's an idiot. But the Summary could have mentioned one nice quote of his:
"Apart from Linux, he claimed that most open source software is often abandoned and not developed, and leads to a lot of low-quality software with lots of bugs."
So at least he recognizes that Linux is high-quality...
[...] for anybody wanting something faster than Python and higher level than C++, the choices come down to Java and C#.
More or less, yes, but there is a better option - combining Python and C++. I am currently working on such a project myself. By coding select time-intensive areas in C++, the app can be very fast, while the majority of the code is in convenient Python. And, importantly, both languages used are 100% free.
The cultish Scientology/EST offshoot known as The Landmark Forum
To be precise, Landmark Forum is a direct offshoot of EST (which is the source of a lot of self-improvement 'cults'), but it has - to the best of my knowledge - no relation to Scientology. Unless you go by the claim that EST itself is an offshoot of Scientology. Now, Werner Erhard, creator of EST, admitted influence by Scientology, but that was among a long list of other things. Really the man took whatever he wanted from a lot of sources.
So EST, sure; Scientology, I don't think so. 'Cultish' - yeah, that's been my experience with Forum graduates.
In addition, brain size has little to do with intelligence. Einstein didn't have a gigantic brain. A smallish idiot savant can multiply numbers better than a 'normal' person twice his size. Elephants have larger brains than humans (partially because of body mass; you might say 'Elephants are a different species' - but Neanderthals aren't humans either). etc. etc.
In programmer terms: having a larger executable doesn't make the software better; the choice of algorithm is more important.
Only Novell specific/grown code will be afforded the "convant not to sue" protection, as it probably won't be released under the GPL.
Novell stated specifically (I believe in the FAQ or related document on their site, but I don't have the time to find it again) that Novell's interoperability code with Windows will be open-sourced, and Microsoft's interoperability code will (naturally) not be open-sourced. So, Novell's code will be open-source, no need to worry about that. However, it will not be under the 'covenant not to sue', that part is true. But as many others have explained already, the lawsuit issue is just FUD anyhow.
I see your point, and in fact developers may be wary of Novell code. But I again think that this is mostly psychological, and not justified. If Novell distribute GPL code, then they are necessarily distributing along with it a license to all relevant patents (and again, if they aren't, then Novell are in violation of the GPL). As for non-GPL code from Novell, well, I am wary of non-GPL code in general. The license in such cases needs to be carefully inspected (and if it doesn't contain a patent clause, then that would be dangerous). But if I recall correctly most of Novell's contributions, such as Mono, are in fact GPL or LGPL.
But still, your point is valid, this is a problem for Linux. I tend to think that it will turn out to be not a very serious one, but I admit I may be missing some angle here. Time will tell.
That's hilariously naive. A little good PR? Well, you forget about the BIG BAD PR for the entire linux community. I will spell out for you what exactly happened: Novell (the holder of Unix copyrights btw, and a linux distributor) acknowledged that Microsoft might have valid patent claims in linux. If you haven't realized: that's pretty bad.
I agree that it is bad PR, but disagree on the amount. Now, if the actual patents were disclosed, this would be far worse (but disclosing them would put Novell in GPL violation). As it is, it is a general statement. Yes, some managers will buy Novell products because of a vague fear caused by the claim. But those same managers can buy Red Hat as well, or Oracle for that matter, since they are guaranteed indemnification (Red Hat clarified that matter just yesterday on their website). So in the end, I don't see how such managers' decisions will greatly change.
That is for commercial distros. As for noncommercial distros, nothing has really changed. For the reasons in my previous post, I am wagering they won't get sued.
[SNIP - arguments against Mono]
For those reasons, and others, I develop in Python (and C++, when I must).
So people who use Novell and Microsoft products are safe, but what about those that choose other products? Will they get sued?
As has been discussed at length in many places in recent days, this doesn't seem likely. To summarize:
1. There is no real 'new' risk to other Linux distros than there was a week ago. Just FUD. Saying "but Novell admit there are patents" isn't a real argument.
2. Microsoft cannot easily crush their competition with patents, because (a) of the risk of antitrust measures, (b) the EU may decide to never allow software patents, once they see such abuse, and (c) other parties, e.g. IBM, which have a great interest in Linux, won't stand by idly, and they have their own patent portfolios (leading to mutually assured destruction for all).
3. Should Microsoft sue (e.g.) Red Hat on some patent, and later on Novell admit that they have a license for the same patent from Microsoft, then they are in violation of the GPL, and Novell's entire Linux line of products is dead in the water. Novell would never put themselves at such risk. So either (a) Microsoft won't sue other people, or (b) Novell won't claim they have a license from Microsoft in the event that they do sue. Note that in the latter case, Novell appear about as protected as Red Hat (which currently indemnify their customers), and no more.
So, in the end, nothing has changed, except for the apparent "gentlemans' arrangement" between Microsoft and Novell not to sue each other. This may appeal to some managers and lead them to choose SUSE, but that is all it is, a little good PR.
To take your point in a slightly different direction, the Microsoft/Novell deal can perhaps be seen as, in large part, being about.Net. Both Microsoft and Novell (with Mono) are invested in.Net, and making it the standard development platform will be of benefit to both. Ensuring interoperability between Windows and Suse will help.Net become a multiplatform solution.
I'm not sure I buy into this theory, but it's interesting. In particular it seems to imply that Microsoft are willing to take a (possibly small) hit on Windows, in order to improve the position of.Net. Whether that can be believed, is a good question.
What's in it for Microsoft? I don't think anyone has the answer, at this point. But here's another bit of speculation: perhaps the patent segment of the Microsoft/Novell agreement is more important that it seems at first. On the one hand, Microsoft have plenty of patents, but if they (ab)use them against their competitors, they may become vulnerable to antitrust action. Novell, on the other hand, can assert their patents more freely. Perhaps Novell have patents that Microsoft are wary of. In that case, Microsoft benefit from a mutual don't-sue pact, since they wouldn't be able to sue anyhow.
These researchers were not Dr. Frankensteins, messing with things man was not meant to know.
No? Well, TFA says
"Richard Ebright, a molecular biologist at Rutgers University in New Jersey, says any study that creates new viruses or activates old ones should be subject to a special review at the national or international level. What's more, he says, because the researchers couldn't be absolutely sure about Phoenix's infectivity, the study should have been carried out under Biosafety level 4 conditions--the best-protected labs available--instead of the level 3 conditions utilized." [emphasis mine]
So, perhaps not "Dr. Frankensteins" in all aspects, but they seem guilty of one of the good doctor's failings - not taking every possible precaution. In such a landmark experiment/study, I would have expected them to.
If anyone is curious, I narrowed down the bug. It seems to only occur under particular circumstances (see comment in the bugreport; basically, it occurs only when your homepage is set to an empty string).
The ability to close and continue sessions later removes a major reason why many people kept their browsers open for long periods of time. Before when you close your browser you had to open your tabs again and get it in the same configuration, now it goes to being the same as before immidiately.
I'm running a fresh install of Ubuntu Edgy here, with Firefox 2.0. The feature you mention doesn't work, sadly; I filed a bug. Of course, this could be some freak issue with my particular installation (although it is a clean install from 48 hours ago), or with Edgy, and not Mozilla's fault.
The real complaint here is "Firefox uses a lot of memory". That may be so, but then it has a pretty high compatibility for rendering all the silly XML and Javascript and crap the web requires these days.
Agreed, but still - Firefox takes (in my experience) more memory than Opera and Konqueror, both of which are also highly compatible.
Slashdot Mirror
Humans have escaped the phenomenon of Natural Selection, for the most part.
All of us who wear glasses? We should have been culled. [etc.]
I disagree. Yes, the simple "the organism dies because of some trait, so there are less offspring with that trait in the next generation" is somewhat true today, as you say. But natural selection extends to much more than that.
For starters, sexual selection. Nowadays people have LOTS of choice in selecting a partner; it is very common to have many partners and even children with multiple partners. In addition, the number of children that people have is very variable - more and more elect to have none, while some people have quite a lot. All of this allows a few 'sexually preferred' individuals to contribute to the next generation's gene pool in a significant way, first in that certain individuals have more opportunity to have children - say, because of attractiveness - and second, that the number of kids is a personal choice that varies greatly.
Now, you talk about wearing glasses, diabetes, etc. - these are 'objective' issues that should be selected against, supposedly: "in the wild, such people would never survive". But the fact that we live in a different environment doesn't mean that OTHER selection pressures, perhaps just as strong, don't exist. They are just different.
I agree TFA is silly. But regarding
Google is expected to release an entire operating system that's supposed to compete heads on with Windows Vista, which is the result of years and years of work and billions of dollars invested? Say what you will about Windows, but it certainly is a massive behemoth with sh*tloads of functionality. You don't just shake something like that out of your sleeve in a few months.
You don't, but you don't need to. You can create your own Linux distro in no time.
they list Novell as part of the OIN, but any patents that Novell has filed are now void because they have signed a promise not to sue.
The promise not to sue refers to customers only (this is how it 'gets around the GPL', supposedly). Novell can still sue Microsoft, so their patents in the OIN are still viable. How much Novell can be trusted is another matter, of course.
What? a choice that would mean that tey wouldn't get paid by people using their framework to earn dosh?
They can still make money, by (1) selling support, (2) charging for a license to extend Qt - i.e. you can link to it for free, but creating a derived work will cost you (or, you must release the derived work under the GPL).
Their expectation to earn money off of every single commercial application that uses their framework - and if KDE were the dominant DE on Linux, then every single commercial application on Linux - is a bit 'optimistic'. On the one hand, there are alternatives (GTK+), and on the other, expecting to monetize all commercial applications in this way is a bit Microsoft-ish - we may as well call it the 'Qt tax'.
>> The commercial license is a bit rediculous, though.
> Why? It seems pretty popular (Adobe, Skype, Google, Opera, etc..) Trolltech's modell is an excellent example of how you can make money on free software.
First, note that all of those companies are not 'open-source' companies (Google standing out somewhat, but still). When you compare Qt to important open-source projects like GNOME and the Linux kernel, you find that both allow commercial development - GNOME is LGPL, and the kernel lets you run whatever you want in userspace. Qt, on the other hand, wants to be paid for commercial development. This is a big deterrent for commercial developers to use Qt. (Especially since you need to decide at the beginning of development which license you want, commercial or free.)
As I see it, the LGPL license was designed for exactly this sort of thing, and I think Trolltech could have made a better choice here. Especially since cross-platform development can be done without paying them (WxWidgets, and also GTK+ can run on various platforms, even Windows).
Yes, TFA was in humor. But is there no basis in fact? In the US, only the ultra-rich run for (and win) the presidency. Bill Gates in the richest man in the US. Now, it's true not every rich candidate can get elected (as nicely pointed out in the book Freakonomics), but still, the possibility can't be discounted. Like Scott Adams says, give Bill a decade or so of charity work, and his popular image will look pretty voteworthy. A chilling thought.
If the submitter RTFP they would have seen that it was Intel, not M$.
Indeed. But if the patent office doesn't read the proposals, why should the Slashdot submitter? (Granted, the other option is that the patent office did read the proposal, but the patent office worker was so ignorant it sounded like a new idea to him/her.)
Currently there are so many variations on this theme already in existence (Skype, Jajah, even Ekiga etc., in a sense), that it boggles the mind such a patent was granted. Yet, TFP says that the filing date was "February 25, 2000". How many of those were around circa 2000? Skype only began around 2002-2003 IIRC...
The macro that is included as a VBA demo in the article http://www.linuxtalk.co.uk/ooo_vba/hypocycloid-dem o.xls works with the OpenOffice that is included with Fedora Core 6. I just tried it on this computer. So the "evil patch" is apparently already present in Red Hat's OpenOffice.
As TFA said, the patch is already included in several distributions. For example, Ubuntu. I tested it on Ubuntu Edgy here and indeed it is included.
Sure, he's an idiot. But the Summary could have mentioned one nice quote of his:
" Apart from Linux, he claimed that most open source software is often abandoned and not developed, and leads to a lot of low-quality software with lots of bugs."
So at least he recognizes that Linux is high-quality...
Companies such as Microsoft cannot compete with the hippy OS
Damn it, Linux might be a lot of things, but it isn't fat.
[...] for anybody wanting something faster than Python and higher level than C++, the choices come down to Java and C#.
More or less, yes, but there is a better option - combining Python and C++. I am currently working on such a project myself. By coding select time-intensive areas in C++, the app can be very fast, while the majority of the code is in convenient Python. And, importantly, both languages used are 100% free.
The cultish Scientology/EST offshoot known as The Landmark Forum
To be precise, Landmark Forum is a direct offshoot of EST (which is the source of a lot of self-improvement 'cults'), but it has - to the best of my knowledge - no relation to Scientology. Unless you go by the claim that EST itself is an offshoot of Scientology. Now, Werner Erhard, creator of EST, admitted influence by Scientology, but that was among a long list of other things. Really the man took whatever he wanted from a lot of sources.
So EST, sure; Scientology, I don't think so. 'Cultish' - yeah, that's been my experience with Forum graduates.
In addition, brain size has little to do with intelligence. Einstein didn't have a gigantic brain. A smallish idiot savant can multiply numbers better than a 'normal' person twice his size. Elephants have larger brains than humans (partially because of body mass; you might say 'Elephants are a different species' - but Neanderthals aren't humans either). etc. etc.
In programmer terms: having a larger executable doesn't make the software better; the choice of algorithm is more important.
Only Novell specific/grown code will be afforded the "convant not to sue" protection, as it probably won't be released under the GPL.
Novell stated specifically (I believe in the FAQ or related document on their site, but I don't have the time to find it again) that Novell's interoperability code with Windows will be open-sourced, and Microsoft's interoperability code will (naturally) not be open-sourced. So, Novell's code will be open-source, no need to worry about that. However, it will not be under the 'covenant not to sue', that part is true. But as many others have explained already, the lawsuit issue is just FUD anyhow.
I think it's a hopeful start.
Indeed, a hopeful start. I have plenty of hope in research carried out by people that express themselves in ways like this (from TFA):
"The virus particles that are released are, like, sterile. They are nonpathogenic," June said. [emphasis mine]
I see your point, and in fact developers may be wary of Novell code. But I again think that this is mostly psychological, and not justified. If Novell distribute GPL code, then they are necessarily distributing along with it a license to all relevant patents (and again, if they aren't, then Novell are in violation of the GPL). As for non-GPL code from Novell, well, I am wary of non-GPL code in general. The license in such cases needs to be carefully inspected (and if it doesn't contain a patent clause, then that would be dangerous). But if I recall correctly most of Novell's contributions, such as Mono, are in fact GPL or LGPL.
But still, your point is valid, this is a problem for Linux. I tend to think that it will turn out to be not a very serious one, but I admit I may be missing some angle here. Time will tell.
That's hilariously naive. A little good PR? Well, you forget about the BIG BAD PR for the entire linux community. I will spell out for you what exactly happened: Novell (the holder of Unix copyrights btw, and a linux distributor) acknowledged that Microsoft might have valid patent claims in linux. If you haven't realized: that's pretty bad.
I agree that it is bad PR, but disagree on the amount. Now, if the actual patents were disclosed, this would be far worse (but disclosing them would put Novell in GPL violation). As it is, it is a general statement. Yes, some managers will buy Novell products because of a vague fear caused by the claim. But those same managers can buy Red Hat as well, or Oracle for that matter, since they are guaranteed indemnification (Red Hat clarified that matter just yesterday on their website). So in the end, I don't see how such managers' decisions will greatly change.
That is for commercial distros. As for noncommercial distros, nothing has really changed. For the reasons in my previous post, I am wagering they won't get sued.
[SNIP - arguments against Mono]
For those reasons, and others, I develop in Python (and C++, when I must).
So people who use Novell and Microsoft products are safe, but what about those that choose other products? Will they get sued?
As has been discussed at length in many places in recent days, this doesn't seem likely. To summarize:
1. There is no real 'new' risk to other Linux distros than there was a week ago. Just FUD. Saying "but Novell admit there are patents" isn't a real argument.
2. Microsoft cannot easily crush their competition with patents, because (a) of the risk of antitrust measures, (b) the EU may decide to never allow software patents, once they see such abuse, and (c) other parties, e.g. IBM, which have a great interest in Linux, won't stand by idly, and they have their own patent portfolios (leading to mutually assured destruction for all).
3. Should Microsoft sue (e.g.) Red Hat on some patent, and later on Novell admit that they have a license for the same patent from Microsoft, then they are in violation of the GPL, and Novell's entire Linux line of products is dead in the water. Novell would never put themselves at such risk. So either (a) Microsoft won't sue other people, or (b) Novell won't claim they have a license from Microsoft in the event that they do sue. Note that in the latter case, Novell appear about as protected as Red Hat (which currently indemnify their customers), and no more.
So, in the end, nothing has changed, except for the apparent "gentlemans' arrangement" between Microsoft and Novell not to sue each other. This may appeal to some managers and lead them to choose SUSE, but that is all it is, a little good PR.
To take your point in a slightly different direction, the Microsoft/Novell deal can perhaps be seen as, in large part, being about .Net. Both Microsoft and Novell (with Mono) are invested in .Net, and making it the standard development platform will be of benefit to both. Ensuring interoperability between Windows and Suse will help .Net become a multiplatform solution.
.Net. Whether that can be believed, is a good question.
I'm not sure I buy into this theory, but it's interesting. In particular it seems to imply that Microsoft are willing to take a (possibly small) hit on Windows, in order to improve the position of
What's in it for Microsoft? I don't think anyone has the answer, at this point. But here's another bit of speculation: perhaps the patent segment of the Microsoft/Novell agreement is more important that it seems at first. On the one hand, Microsoft have plenty of patents, but if they (ab)use them against their competitors, they may become vulnerable to antitrust action. Novell, on the other hand, can assert their patents more freely. Perhaps Novell have patents that Microsoft are wary of. In that case, Microsoft benefit from a mutual don't-sue pact, since they wouldn't be able to sue anyhow.
These researchers were not Dr. Frankensteins, messing with things man was not meant to know.
No? Well, TFA says
"Richard Ebright, a molecular biologist at Rutgers University in New Jersey, says any study that creates new viruses or activates old ones should be subject to a special review at the national or international level. What's more, he says, because the researchers couldn't be absolutely sure about Phoenix's infectivity, the study should have been carried out under Biosafety level 4 conditions--the best-protected labs available--instead of the level 3 conditions utilized." [emphasis mine]
So, perhaps not "Dr. Frankensteins" in all aspects, but they seem guilty of one of the good doctor's failings - not taking every possible precaution. In such a landmark experiment/study, I would have expected them to.
Hell, even if they don't want to make business with RedHat, at least hire some CentOS developers to put together a decent distro!
Are you sure they haven't?
If anyone is curious, I narrowed down the bug. It seems to only occur under particular circumstances (see comment in the bugreport; basically, it occurs only when your homepage is set to an empty string).
The ability to close and continue sessions later removes a major reason why many people kept their browsers open for long periods of time. Before when you close your browser you had to open your tabs again and get it in the same configuration, now it goes to being the same as before immidiately.
I'm running a fresh install of Ubuntu Edgy here, with Firefox 2.0. The feature you mention doesn't work, sadly; I filed a bug. Of course, this could be some freak issue with my particular installation (although it is a clean install from 48 hours ago), or with Edgy, and not Mozilla's fault.
The real complaint here is "Firefox uses a lot of memory". That may be so, but then it has a pretty high compatibility for rendering all the silly XML and Javascript and crap the web requires these days.
Agreed, but still - Firefox takes (in my experience) more memory than Opera and Konqueror, both of which are also highly compatible.