I don't know if the energy is net positive or not, but even without reading any studies I can tell you NO ONE is including the sun's energy in the equation because OF COURSE the balance can't work if you count that. Such a study would be absolutely pointless and I can't believe anyone is that stupid.
This news requires an update to the/. hypocrisy list.
Old hypocrisy:
Bug found in software product. if MSFT, say "This shows how buggy proprietary software development is!" if Mozilla, say "This shows how great OSS is at responding to bugs!"
New hypocrisy:
Company adds support for linux. if IBM, say "Don't be fooled, they're just trying to make money!" if Mozilla, say "Why shouldn't they get paid for all the hard work they put into OSS?"
It seems completely analogous to the launch of the internet. Initially, some people talked as though it was going to be this egalitarian Nirvana where the people would have a voice and no one would be "in control," etc. etc.
To some extent this has come to pass, but more and more governments and jurisdictions and the physical location of servers have come into play, what with the desire to control spam, prevent auctioning Nazi memorabilia, mediate trademark infringement and cybersquatting, protect intellectual property, and so forth.
It's the Wild West on the internet no more. I expect the space race to follow the same pattern: first the land grab (so to speak), then the desire to impose stability through social and legal mechanisms.
A bit far afield, but: in the old days, the carbon paper in the cc slips I used (as a cashier) would tear in half when you pulled it apart. You would then tear off the customer's receipt (which stuck out between the shorter pieces of carbon paper) and give it to him; the carbons in that half would go in the trash under your register. The other two slips and carbon halves would go in the register; the money-counters would throw away those carbons the next day. That was the 'security' to keep dumpster-divers from getting entire cc numbers.
Of course, you had to pull the charge-slip stack apart with just the right snap or else the carbons wouldn't tear down the middle; but I rarely saw anyone complain or notice. (About one in twenty customers would ask for all the carbons.)
Further afield: I had to look up the cc number in a little pamphlet of bad numbers before ringing it through. If the charge was over a certain amount ($50, I think), I had to call it in and read the number off to someone who would give me an approval code to write on the slip.
And I had to walk uphill, in the snow, both ways, to get to school.
Very bizarre, that the future found it easier to build a time machine than to port time_t to 64 bits.
Maybe all the bits get used up in the intervening time, and are therefore scarce in 2036. That would also explain the need for an old bit-conserving IBM 5100.
Yes, it's a "scientific" measurement, but the implication (and the explicit statement in the/. blurb) is that Linux is getting more secure because the survival time of an unpatched system is getting longer. That just ain't a valid conclusion.
As a counter-example, perhaps there are getting to be more and more unpatched windows boxes, so windows exploits become that much more attractive to zombie-network maintainers. That doesn't make Linux "more secure," in the sense of a native attribute of the O/S; it just means people are less interested in writing exploits or there are fewer unpatched machines to propagate them.
Actually, I'd like to see ISPs monitoring for port-scanning activity or other outward evidence of infection. Infected sources go into a "sandbox" that only has proxied http access to the internet. Most infected shlubs wouldn't notice, and the rest of us can enjoy the internet in peace. (Those who do notice would have to go through a process of cleaning their machines and getting reinstated.)
That said, people running listservs and such would need to be flagged by the ISP to not get caught in the dragnet, and I'm not sure how that would work.
I don't understand why this is an issue for startups.
A 10-person company is going to be privately held, therefore there are no stockholders to punish the company when they report a huge stock-option expense (which they won't, because they don't have reporting requirements).
Take, for example, codec software. Let's say it takes 3 man-years to develop an improvement to the compression algorithm, to decrease compressed size by 40%.
The problem, without patents, is that it's easy for a competitor to watch the compressor (in running software) and copy it. Hence, they reap the benefits without the costs. Why would anyone work on improving codecs, then? You'd get undercut and never be able to recoup your development costs.
To me, software (algorithms) should be patentable if (1) they take significant development time and (2) they can be copied with much less effort when given a running copy.
Thus, stupid things like one-click and cookies and downloaded active controls can't be patented, because you can implement them with the same amount of effort as the original implementer, without ever having to actually observe any particular implementation. Algorithms, though, would still be protected, unless you can "reasonably" duplicate them without stepping through the developer's implementation.
Slashdot Mirror
That's a pretty neat trick, given that Google Chrome was only released in September 2008.
I call bullshit.
I don't know if the energy is net positive or not, but even without reading any studies I can tell you NO ONE is including the sun's energy in the equation because OF COURSE the balance can't work if you count that. Such a study would be absolutely pointless and I can't believe anyone is that stupid.
Sheesh.
Anyone can get the "clean" version of RealPlayer demanded by the BBC for their customers by downloading via the BBC link:
l
http://www.bbc.co.uk/radio/audiohelp_install.shtm
Bless the beeb.
(I hesitate to publish the deep link into RN's site, because they'd probably switch it around.)
This news requires an update to the /. hypocrisy list.
Old hypocrisy:
Bug found in software product.
if MSFT, say "This shows how buggy proprietary software development is!"
if Mozilla, say "This shows how great OSS is at responding to bugs!"
New hypocrisy:
Company adds support for linux.
if IBM, say "Don't be fooled, they're just trying to make money!"
if Mozilla, say "Why shouldn't they get paid for all the hard work they put into OSS?"
It seems completely analogous to the launch of the internet. Initially, some people talked as though it was going to be this egalitarian Nirvana where the people would have a voice and no one would be "in control," etc. etc.
To some extent this has come to pass, but more and more governments and jurisdictions and the physical location of servers have come into play, what with the desire to control spam, prevent auctioning Nazi memorabilia, mediate trademark infringement and cybersquatting, protect intellectual property, and so forth.
It's the Wild West on the internet no more. I expect the space race to follow the same pattern: first the land grab (so to speak), then the desire to impose stability through social and legal mechanisms.
A bit far afield, but: in the old days, the carbon paper in the cc slips I used (as a cashier) would tear in half when you pulled it apart. You would then tear off the customer's receipt (which stuck out between the shorter pieces of carbon paper) and give it to him; the carbons in that half would go in the trash under your register. The other two slips and carbon halves would go in the register; the money-counters would throw away those carbons the next day. That was the 'security' to keep dumpster-divers from getting entire cc numbers.
Of course, you had to pull the charge-slip stack apart with just the right snap or else the carbons wouldn't tear down the middle; but I rarely saw anyone complain or notice. (About one in twenty customers would ask for all the carbons.)
Further afield: I had to look up the cc number in a little pamphlet of bad numbers before ringing it through. If the charge was over a certain amount ($50, I think), I had to call it in and read the number off to someone who would give me an approval code to write on the slip.
And I had to walk uphill, in the snow, both ways, to get to school.
I type in my account number and password at my bank's website (over the internet!) all the time.
Calling someone stupid for doing it yet again isn't quite fair.
Very bizarre, that the future found it easier to build a time machine than to port time_t to 64 bits.
Maybe all the bits get used up in the intervening time, and are therefore scarce in 2036. That would also explain the need for an old bit-conserving IBM 5100.
"I'm not exactly sure what, if anything, compu-global hyper-mega /. does, but rather than compete I'm going to simply buy you out.
/. has become a windows fanboy site.
"Buy 'em out, boys!"
That said, this is the first time I've heard the suggestion that
WINDOWS == QWERTY
I leave it to the reader to work out the details of the analogy.
Yes, it's a "scientific" measurement, but the implication (and the explicit statement in the /. blurb) is that Linux is getting more secure because the survival time of an unpatched system is getting longer. That just ain't a valid conclusion.
As a counter-example, perhaps there are getting to be more and more unpatched windows boxes, so windows exploits become that much more attractive to zombie-network maintainers. That doesn't make Linux "more secure," in the sense of a native attribute of the O/S; it just means people are less interested in writing exploits or there are fewer unpatched machines to propagate them.
Actually, I'd like to see ISPs monitoring for port-scanning activity or other outward evidence of infection. Infected sources go into a "sandbox" that only has proxied http access to the internet. Most infected shlubs wouldn't notice, and the rest of us can enjoy the internet in peace. (Those who do notice would have to go through a process of cleaning their machines and getting reinstated.)
That said, people running listservs and such would need to be flagged by the ISP to not get caught in the dragnet, and I'm not sure how that would work.
I don't understand why this is an issue for startups. A 10-person company is going to be privately held, therefore there are no stockholders to punish the company when they report a huge stock-option expense (which they won't, because they don't have reporting requirements).
There is some software that should be patentable.
Take, for example, codec software. Let's say it takes 3 man-years to develop an improvement to the compression algorithm, to decrease compressed size by 40%.
The problem, without patents, is that it's easy for a competitor to watch the compressor (in running software) and copy it. Hence, they reap the benefits without the costs. Why would anyone work on improving codecs, then? You'd get undercut and never be able to recoup your development costs.
To me, software (algorithms) should be patentable if (1) they take significant development time and (2) they can be copied with much less effort when given a running copy.
Thus, stupid things like one-click and cookies and downloaded active controls can't be patented, because you can implement them with the same amount of effort as the original implementer, without ever having to actually observe any particular implementation. Algorithms, though, would still be protected, unless you can "reasonably" duplicate them without stepping through the developer's implementation.