Talking of thinking differently in natural langauge and programming, one think I noted is when doing X programming, I'll happily write
XAllocColour(...)// allocate a pretty colour without even realising I'm spelling colo(u)r differently. It's just automatic.
First thing that occurs to me is how close this is getting to Rev. 13:16-18. How long before we need this kind of ID to go buy things, gain access to our workplaces, homes, etc...
How soon before people start being killed for their digital IDs? Even for those who don't believe the bible, consider how lucrative a target for muggers this will be, if you can easily gain access to someone's money and possesions just by killing them and extracting a tiny piece of kit from their skull.
This is perhaps the worst idea I have heard in a long, long time.
I looked at Freedows with great interest at least 18 months ago, maybe two years. Joined the mailing list, etc, all eager to see what was going on.
What I actually found was lots of bickering, and people spending hours and hours writing bureaucratic documents about coding style (indenting, variable naming), arguments about how the kernel group were more l33t than the UI group, and generally lots of vapourware hype.
The fact that now, nearly two years on, the web-site still has nothing to show for itself, and in fact, none of the dates have been updated for 10 months, suggests it's never going to happen.
Personally, I think Wine has most chance in this area - at least the kernel is already written, but witness there how much of an uphill struggle there is - and this is with lots of support. Re-inventing the wheel again is not the answer.
It seems to me, that in many cases decentralisation of TLDs would be a good thing. At present, everyone in the world shares root DNS servers for the top-level. These have pointers to other servers, generally different for each TLD, so that other naming entities can divide up the namespace as they see fit.
But it isn't really necessary for everyone to use the same root nameservers. For instance, here in.uk, Nominet could tell every ISP here to use their nameservers for TLD resolution, rather than the root nameservers, and those TLDs that aren't UK based can have pointers to the relevant countries nameservers.
This requires that all countries running TLD servers (or the new root servers) work together to ensure that each TLD in their space map to each others root servers, but it does mean that each country is then responsible for their own DNS completely.
This could allow several nice effects, for example a TLD for.int could have different resolutions depending what country you are in. Obviously, this would again need a centralised naming system for.int, but would allow companies to have a universal name that maps to different hosts depending where in the world you are performing the lookup. No more having to guess the closest mirror to you - it would all be taken care of behind the scenes by DNS.
Also, it would lead to a more logical ability to have national URLs. For instance, within the UK, I might be able to refer to mega-pizza.ltd, and have this available to non-UK users (and of course, UK users) as mega-pizza.ltd.uk. This simply requires that each company avoids TLD codes in their own TLD.
The only problem I see with my idea is that every user is then at the mercy of their country's root nameservers to resolve to the correct nameserver for each TLD. This could possibly lead to governments denying access to nameservers in other countries for political reasons, but I can't really see this being a problem in practice. And in any case, if you feel really strongly, you could always choose to resolve from other countries domains if you choose. And I can see some governments preferring not to trust US companies for their nameserver backbone.
Slashdot Mirror
Talking of thinking differently in natural langauge and programming, one think I noted is when doing X programming, I'll happily write XAllocColour(...) // allocate a pretty colour without even realising I'm spelling colo(u)r differently. It's just automatic.
How soon before people start being killed for their digital IDs? Even for those who don't believe the bible, consider how lucrative a target for muggers this will be, if you can easily gain access to someone's money and possesions just by killing them and extracting a tiny piece of kit from their skull.
This is perhaps the worst idea I have heard in a long, long time.
What I actually found was lots of bickering, and people spending hours and hours writing bureaucratic documents about coding style (indenting, variable naming), arguments about how the kernel group were more l33t than the UI group, and generally lots of vapourware hype.
The fact that now, nearly two years on, the web-site still has nothing to show for itself, and in fact, none of the dates have been updated for 10 months, suggests it's never going to happen.
Personally, I think Wine has most chance in this area - at least the kernel is already written, but witness there how much of an uphill struggle there is - and this is with lots of support. Re-inventing the wheel again is not the answer.
It seems to me, that in many cases decentralisation of TLDs would be a good thing. At present, everyone in the world shares root DNS servers for the top-level. These have pointers to other servers, generally different for each TLD, so that other naming entities can divide up the namespace as they see fit.
.uk, Nominet could tell every ISP here to use their nameservers for TLD resolution, rather than the root nameservers, and those TLDs that aren't UK based can have pointers to the relevant countries nameservers.
.int could have different resolutions depending what country you are in. Obviously, this would again need a centralised naming system for .int, but would allow companies to have a universal name that maps to different hosts depending where in the world you are performing the lookup. No more having to guess the closest mirror to you - it would all be taken care of behind the scenes by DNS.
But it isn't really necessary for everyone to use the same root nameservers. For instance, here in
This requires that all countries running TLD servers (or the new root servers) work together to ensure that each TLD in their space map to each others root servers, but it does mean that each country is then responsible for their own DNS completely.
This could allow several nice effects, for example a TLD for
Also, it would lead to a more logical ability to have national URLs. For instance, within the UK, I might be able to refer to mega-pizza.ltd, and have this available to non-UK users (and of course, UK users) as mega-pizza.ltd.uk. This simply requires that each company avoids TLD codes in their own TLD.
The only problem I see with my idea is that every user is then at the mercy of their country's root nameservers to resolve to the correct nameserver for each TLD. This could possibly lead to governments denying access to nameservers in other countries for political reasons, but I can't really see this being a problem in practice. And in any case, if you feel really strongly, you could always choose to resolve from other countries domains if you choose. And I can see some governments preferring not to trust US companies for their nameserver backbone.