OMG
An article about computing, software and Linus. OMG where was RMS in this?
Is the GPL dead?
Torvalds on patents: 'Sane people know it's bullsh*t but making real change is difficult'
This was the week when a piece of software (supposedly) passed the Turing Test, giving the world all the “robot overlord” headlines they could possibly need in a slow, sunny news week. Facts were hard to come by in the tale of Eugene Goostman, the 13-year-old-boy-simulating software that apparently hit the controversial …
-
Friday 13th June 2014 11:57 GMT Reg T.
Bullpucky patents until...
given valuable IBM stock, and then all those IBM patents which produced the profit which allowed Mr. "F*** you" to "prosper" are forgotten.
Doesn't matter to me but "don't bite the hand..." and all that. All of the corporations given access to the kernel hold patents and also copyrights and trademarks. And, one of the largest kernel partners is the NSA and its "security" software contributions.
All you LInuxheads read all the code - right? Of course you do.
Or, is it that most of you simply follow the herd and use the easiest_distro_for_newbies which is magically dropped down from binary heaven?
Consequently, we observe that Mr. Torvalds is quite adept at working with (and for) the "insane". For instance, the good folk from Microsoft. Rumor has it that Win 9 will also include a free vaccine package from Mr. Gates which will sterilize the recipients and also give them several dread diseases. And, there will be free tickets also for Mr. Ballmers' new "plantation" enterprise/franchise in the NBA.
However, there will be no apologies from Microsoft for corrupting Turkish government officials so that Pardus linux (one of the finer KDE examples then) would be defunded and Windows embraced.
Anyone who has worked with the "insane" has seen firsthand that the most seriously deluded of them are also in a state of denial, claiming that it is others who are nutz.
So, in light of the correct appellation "GNU linux", who of the two "fathers" actually shuns patents and the patent crowd?
-
Friday 13th June 2014 14:39 GMT Sean Healey
IP Numbers
"Currently IPv6 uptake is slow and IPv4 space is running out"
I though that IPv4 technically already was exhausted. RIPE announced that they had begun issuing their very last block 2 years ago ... http://www.ripe.net/internet-coordination/ipv4-exhaustion
The internet has delayed the end of IPv4 through turnover and trading of existing allocated address spaces, and of course NAT.
I'm suprised to see that there are still 15 million IPv4 addresses left in the unallocated pool ... http://www.ripe.net/internet-coordination/ipv4-exhaustion/ipv4-available-pool-graph ... which has actually _increased_ in recent weeks!
-
Saturday 14th June 2014 07:04 GMT pdxbrit
The IPv6 address space is *much* larger than you think it is.
"Sadly, Fry was a bit off the mark there. IP addresses are allocated globally by the Internet Assigned Numbers Authority, which does delegate to local registries, but new addresses are not created with domain names. There is a cavernous amount of space in IPv6 – but it will run out one day."
Umm, no, not even close.
Check out RFC 2373, "IPv6 Addressing Architecture". In the original spec, 1/8th of the 128-bit address space is reserved for "Aggregatable Global Unicast Addresses" (and 5/8th more are still available/reserved).
That means you have 2^125 addresses available.
Imagine you hand out 1,000,000 addresses per second. You will exhaust that 1/8th of the address space in ~1,347,862,190,569,539,760,087,009 years.
The IPv6 address space is vastly larger than any conceivable use. While your point about the allocation of IPv4 addresses is entirely valid, it's rather marred by a wholly inaccurate characterization of IPv6 (the address space will never run out in any conceivable use case).
-
Monday 16th June 2014 07:43 GMT Anonymous Coward
Re: The IPv6 address space is *much* larger than you think it is.
"The IPv6 address space is vastly larger than any conceivable use. While your point about the allocation of IPv4 addresses is entirely valid, it's rather marred by a wholly inaccurate characterization of IPv6 (the address space will never run out in any conceivable use case)
"
Not quite. Only the first 64 bits are used for the network prefix for unicast addresses so at most 2^64 ~ 10^19 neworks could be setup. The number of stars in the Observable Universe is estimated to be 1 billion trillion ie 10^21. Hence there is at least one conceivable use case - assigning a network to every star in the Observable Universe which would exhaust IPv6 as currently conceived.
Far fetched but not inconceivable - who knows what our distant descendants might want to do ?
This topic is closed for new posts.
Biting the hand that feeds IT
