um, i'm not. i'm not really afraid sticking whatever cd or usb drives to my computer - and i haven't run antivirus for 7 or so years. ok, the same goes for windows...
oh, finally somebody with my usage pattern:) i recently went from 1024x768 to 1600x1200. i had seen screenshots with people using tiled desktop with smaller resolution than my new one, so i expect myself to adapt to such a thing. nah. i still use windows maximised, because i rrrrreally want to reduce any scrolling. now, i see that sometimes making a small window always on top in some cases is easier to use than before... but still, maximised windows ftw:)
companies already do enormous amounts of custom development. there's a LOT of code developed that way. lately i've seen a slow trend towards development that fully or partially lands in the opensource area, so company gets some expenses reduced (can use existing opensource code, reduced maintenance/testing burden etc). as applications become more of a commodity, this trends becomes stronger. in the government area, which has been a stronghold of inhouse custom apps that are are so overpriced and created again-and-again-and-again in each country, there's a slight push inside european union to work at least somewhat together and reduce double work - there i haven't noticed huge improvements, because everybody still thinks they're soooo special - and, of course, are afraid to lose control & funding. in general, this is probably described a lot over teh internets about how opensource model works financially:)
It's necessary for there to be an economic incentive to develop software. Nobody is going to donate millions of man-hours to write the software for the F-22 out of the goodness of their heart. Nobody is going to donate the man-hours to write the software for my insurance agency or hospital.
nobody says that everybody is prohibited from paying for opensource software.
um. leave a copy on the phone and don't mention any uploads ? if a person is dumb enough to shout "i uploaded them" when there are such threats possible, then that person should go nowhere near such places.
heh. i was talking about http://wiki.openwrt.org/OpenWrtDocs/Hardware/Asus/WL500GPV2, which, at the point when i was planning my purchase, stated that several components are not supported. it has moved forward, but still claims that "There are many opportunities for this to not work. You should be wise in the ways of building and debugging software. You should engineer a console tty connection. You will probably need a lot of patience.".
as already mentioned, wrt54gl (note the l, of course) is a nice router. wireless, 4 wired ports, can run openwrt.
another good one - asus wl500gp v1. additionally has two usb ports.
i've set up several wrt54gls, and i now have the asus one at home. with openwrt, only problems might be cli usage for some of the more advanced features:)
how does this work with different screen/terminal sizes ? attaching to a session that was previously run on different terminal size shows garbage & requires repainting/resizing it. yeah, i'm too lazy to try this myself;)
because then you'd be raving about "why do i have to put in some fucking email address to fucking authenticate";>
what's wrong with url ? it's not like those are usually any longer than..
besides, why propose some email address ? how would that be _any_ better ? you would only tie email provider to openid provider. currently i can choose whether i use single provider for those services or not. if you tie openid to email, all openid validator would know - i have to try to connect that _email_ server, which, sort of, would have to provide openid.
as for understanding openid, i don't know where you see the huge problem. i saw openid logien field on sf one day, decided to read about it. i did, went "hmm, this is pretty cool" - and got an account. i could try explaining openid to some people next week and check whether it is as easy to understand for non-it persons as well:)
your comments look like ranting without much substance, actually - all you do is shout about some mythical email address and claim nobody understands openid. for me, it's the other way - i understand (the basic scheme) of openid, and i don't understand how your proposed email address scheme is supposed to work in today's internet:)
well, i guess that's up to the to change this requirement then;) or one could set up a fake persona in openid provider and feed some crap data in the ms system...
actually, you can choose not to send any metadata for openid logins - just your id. so any differences in metadata can't be a reason for not accepting openid logins.
this might be my bad english understanding, or some translation issue - but the text you cited somehow didn't make much sense. what's the main point of that citation and what does the second half of the first paragraph say (i lost my way around the 'prosecutor' part) ?
i was reading the tutorial into git for svn users, and it seemed understandable to me;) what i saw as a large svn strength - ability to checkout only parts of a repository. imagine a repository that has several large directories, but most users would ever want only one. with git, you'd have to carefully plan and make each of these directories a separate branch - and i'm not even sure that would work so well:)
also, git seems to be very, very source code centric - it was even mentioned that changing a binary file a bit (like slightly changing an image) can make it's history split because git wouldn't recognise it as the same file.
but there is no need to have something like that in _filesystem_. this could be implemented with existing userspace software - browser could ask this question, then it would write this data to some control file that is checked by a cron job periodically. additionally, file manager could be used to manipulate these proprties. i don't think i'd use something like that, though. the thought of some files being automatically deleted... i'm sure some video file i have downloaded would get deleted a day before i'd get to watching it (because i would have not thought about having no time to see it during the set period;) ). as an even easier method you could try out setting up directories like download/{day,week,month,year}, downloading things in those and just run a cronjob daily or so and delete files older than {period} in each directory... then you could see how useful such a feture is to you.
i guess zfs way does not allow reuse of most of the functionality by other filesystems. thus while being very efficient in the short run, it could hinder long term development of new filesystems (not only as replacements for it but also as specific niche products)
to be fair, reading about brtfs, the design also mentions things like snapshots, mirroring, included raid stuff. http://btrfs.wiki.kernel.org/index.php/Main_Page i guess, some middle ground will be reached with brtfs for code reuse and functionality.
Slashdot Mirror
um, i'm not. i'm not really afraid sticking whatever cd or usb drives to my computer - and i haven't run antivirus for 7 or so years. ok, the same goes for windows...
oh, finally somebody with my usage pattern :) :)
i recently went from 1024x768 to 1600x1200. i had seen screenshots with people using tiled desktop with smaller resolution than my new one, so i expect myself to adapt to such a thing. nah. i still use windows maximised, because i rrrrreally want to reduce any scrolling.
now, i see that sometimes making a small window always on top in some cases is easier to use than before... but still, maximised windows ftw
companies already do enormous amounts of custom development. there's a LOT of code developed that way. :)
lately i've seen a slow trend towards development that fully or partially lands in the opensource area, so company gets some expenses reduced (can use existing opensource code, reduced maintenance/testing burden etc).
as applications become more of a commodity, this trends becomes stronger.
in the government area, which has been a stronghold of inhouse custom apps that are are so overpriced and created again-and-again-and-again in each country, there's a slight push inside european union to work at least somewhat together and reduce double work - there i haven't noticed huge improvements, because everybody still thinks they're soooo special - and, of course, are afraid to lose control & funding.
in general, this is probably described a lot over teh internets about how opensource model works financially
It's necessary for there to be an economic incentive to develop software. Nobody is going to donate millions of man-hours to write the software for the F-22 out of the goodness of their heart. Nobody is going to donate the man-hours to write the software for my insurance agency or hospital.
nobody says that everybody is prohibited from paying for opensource software.
um. leave a copy on the phone and don't mention any uploads ?
if a person is dumb enough to shout "i uploaded them" when there are such threats possible, then that person should go nowhere near such places.
they're still quite expensive, though. some 'mass order discount' would help.
oh, and they don't ship to my country as well, even though we are only 500 or slightly more km apart.
bah. i previewed my post and still missed the error. last link should have been http://wiki.openwrt.org/Hardware/Asus
heh. i was talking about http://wiki.openwrt.org/OpenWrtDocs/Hardware/Asus/WL500GPV2, which, at the point when i was planning my purchase, stated that several components are not supported. it has moved forward, but still claims that "There are many opportunities for this to not work. You should be wise in the ways of building and debugging software. You should engineer a console tty connection. You will probably need a lot of patience.".
also, http://wiki.openwrt.org/OpenWrtDocs/Hardware/Asus/WL500GPV2 claims that v1 has a faster cpu.
as already mentioned, wrt54gl (note the l, of course) is a nice router. wireless, 4 wired ports, can run openwrt.
another good one - asus wl500gp v1. additionally has two usb ports.
i've set up several wrt54gls, and i now have the asus one at home. with openwrt, only problems might be cli usage for some of the more advanced features :)
i personally like openwrt the best.
another good router is asus wl500-gp v1 (v2 has problems) - it has two usb ports, that linksys does not.
how does this work with different screen/terminal sizes ? attaching to a session that was previously run on different terminal size shows garbage & requires repainting/resizing it. ;)
yeah, i'm too lazy to try this myself
oh. could you provide some references to lawsuits against microsoft - and successful ones at that ?
i haven't heard of such a thing happening so far.
they should have used extension .gov
hi. i would like to repeatedly cut the inventor of "hard plastic clamshell casings" with sharp objects. please apply.
thanks, bye.
at least for some distros (primarily redhat heritage, also some suse capabilities) there's http://dag.wieers.com/home-made/mrepo/
there's also http://www.redhat.com/spacewalk/, the recently opensourced satellite spinoff - but it still requires oracle as a backend, so screw that ;)
because then you'd be raving about "why do i have to put in some fucking email address to fucking authenticate" ;>
what's wrong with url ? it's not like those are usually any longer than ..
besides, why propose some email address ? how would that be _any_ better ? you would only tie email provider to openid provider. currently i can choose whether i use single provider for those services or not.
if you tie openid to email, all openid validator would know - i have to try to connect that _email_ server, which, sort of, would have to provide openid.
as for understanding openid, i don't know where you see the huge problem. i saw openid logien field on sf one day, decided to read about it. i did, went "hmm, this is pretty cool" - and got an account. i could try explaining openid to some people next week and check whether it is as easy to understand for non-it persons as well :)
your comments look like ranting without much substance, actually - all you do is shout about some mythical email address and claim nobody understands openid. for me, it's the other way - i understand (the basic scheme) of openid, and i don't understand how your proposed email address scheme is supposed to work in today's internet :)
well, i guess that's up to the to change this requirement then ;)
or one could set up a fake persona in openid provider and feed some crap data in the ms system...
actually, you can choose not to send any metadata for openid logins - just your id. so any differences in metadata can't be a reason for not accepting openid logins.
this might be my bad english understanding, or some translation issue - but the text you cited somehow didn't make much sense.
what's the main point of that citation and what does the second half of the first paragraph say (i lost my way around the 'prosecutor' part) ?
i was reading the tutorial into git for svn users, and it seemed understandable to me ;) :)
what i saw as a large svn strength - ability to checkout only parts of a repository. imagine a repository that has several large directories, but most users would ever want only one.
with git, you'd have to carefully plan and make each of these directories a separate branch - and i'm not even sure that would work so well
also, git seems to be very, very source code centric - it was even mentioned that changing a binary file a bit (like slightly changing an image) can make it's history split because git wouldn't recognise it as the same file.
but there is no need to have something like that in _filesystem_. ;) ).
this could be implemented with existing userspace software - browser could ask this question, then it would write this data to some control file that is checked by a cron job periodically. additionally, file manager could be used to manipulate these proprties.
i don't think i'd use something like that, though. the thought of some files being automatically deleted... i'm sure some video file i have downloaded would get deleted a day before i'd get to watching it (because i would have not thought about having no time to see it during the set period
as an even easier method you could try out setting up directories like download/{day,week,month,year}, downloading things in those and just run a cronjob daily or so and delete files older than {period} in each directory... then you could see how useful such a feture is to you.
i guess zfs way does not allow reuse of most of the functionality by other filesystems. thus while being very efficient in the short run, it could hinder long term development of new filesystems (not only as replacements for it but also as specific niche products)
to be fair, reading about brtfs, the design also mentions things like snapshots, mirroring, included raid stuff.
http://btrfs.wiki.kernel.org/index.php/Main_Page
i guess, some middle ground will be reached with brtfs for code reuse and functionality.
it could have been dual-licensed. and "GPL is almost as bad as proprietary" bit was unnecessary trolling.
ext4 has a shot at 2 years for root :) ;)
i don't see btrfs becoming widespread default fs sooner than some 5 years.
i'd be glad if proven wrong, though