you have demonstrated that the ammount of menial labour needed to get a level of service/productivity is decreasing but backhoe drivers and factory workers (making those toilet bowl additives) don't need degrees either. Neither do most jobs in things like retail.
what i'm not convinced of is increasing the ammount of subsidized education (and afaict getting unsubsidized education has always been relatively easy if you have the money, though you may have to go abroad in some countries at the university level) and thus leading to more highly educated people than can be absorbed (china already has this problem, thats why people are fighting so hard to get into its top universities rather than just into university at all) is benificial economically.
given that they are on a shelf and given how unstable they are when standing alone on end its a pretty safe bet that they are going to end up either packed tightly or leaning on each other. so no pack them less tightly is not a soloution as they will still end up touching and catching.
using a specially designed rack is an option but this both increases the cost (though possiblly no more so than using DVD cases will) and means you can't mix them with other items on the shelf.
i've seen those but don't like em because they don't seem to have anything in place to keep the CD away from scraping on the box, nor do they seem to grip the CD very well.
Slimline CDR jewel cases while not as good from a labelling point of view seem far more substatial.
The point is though that the user should not be told that a self signed certificate is secure either. It should just enable the use of encryption.
this is partly a UI design issue, browser UIs seem to be built arround the idea that ssl=secure with the warnings as an afterthought. Though there is a school of thought that self signed certs on the web should be discouraged anyway as they really give little more than a false sense of security (sure its encrypted but anyone can mitm it).
another thing is, i damn well wan't to know if https://www.halifax-online.co.uk/ had a proper cert yesterday and doesn't today because its dns has been hijacked (admittedly some people type the url without the https losing this type of protection).
it *was* but with 2.6.x linus has decided not to fork off a new development branch yet. IIRC the reason for this is he belives most changes for the time being will be confined to subsystems and only major reworking of the kernel as a whole deserves a new branch.
However unfortunately some people are finding that this results in a new kernel version fixing support for one peice of hardware they own at the same time as breaking support for another.
Nobody seems to care because those operating systems don't run all the current games. That is changing.. not that they are running that many more games, but games are moving to consoles
some categories of gamer are very poorly served by consoles, this includes anyone who likes custom content, any games that require a keyboard and mouse (yes its possible to use one with a console but consoles are usually used with the TV not on a desk) for proper play (point and click stratagy games and FPS games that require rapid and accurate aiming). Online gaming is another area (yes consoles offer it but usually only at extra cost, with PC games its usually free).
Unfortunately PC game houses have been in a suicidal race to get better graphics resulting in games that only play well on very expensive PCs.
Other than specialized software that a minority are required to run Yes individual apps may be required by only a small number of people, thats not really the issue at hand though.
Most people have Something that ties them to windows Autocad, Altium designer (formerly protel), MPlab, Xilinx ISE, Photoshop, Internal software (either as native windows apps or as macro heavy msoffice stuff), IE (still needed for many internal sites).
2. With a generational garbage collector, you don't usually have to scan pages that haven't been modified since the last scan (as these are in old generations and their pointers are already represented in the set of roots for the newer generations) But surely sooner or later those older generations still have to be scanned in a long running application. What happens then? loads of swapping in i bet!
and don't forget debian strongly advise against (and don't test) skipping releases when upgrading so your likely to end up doing woody-sarge-etch anyway even if you do the two upgrades at the same time (i'm reffering to after etches release ofc)
hmm will the do not call registry apply if the call never touches pots?
and ofc you can always call from out of country while sticking within the same voip provider (generally making the call both free and hard to identify as international)
the idea of css done right is to produce a fancy layout for your pr department to see but have it degrade gracefully to a simple but perfectly usable layout with anything else.
i thought new apache was also incompatible (at least according to the fsf but most distros seem to listen to them over the vendors of the other license)
you already mentioned openssl, add to that anything else under licenses similar to the 4 clause BSD.
and then there is the mpl and its variants............
compiling against winelib means you have to use a compiler setup that can produce elf and means you have to ship different binaries for windows and linux users and there isn't much point in doing it, once the binary is loaded i doubt it makes any difference whatsoever to performance.
for desktop and server applications in the current climate portability to AMD64 is going to become imporant but not to other architectures.
But being able to run the same OS on your desktop and on your embedded equipment certainly carries its advantages. You can run the vast majority of linux software on your embedded box (performance permitting) with a simple recompile.
The point is, if you're going to be waiting 50ms for disk access, why bother shaving 2ms of processing time by running in a native compiled language? Nobody will ever notice. what they will notice is when the gc decides it needs to scan a memory area that has been swapped out crowding out any other IO on the system.
Average performance only matters for a few time consuming tasks (and they do still exist), what matters far more in end user apps is any apparent hangs, if a button takes 100ms to get a response i probablly won't notice unless i'm gaming. if a button takes 10ms 99% of the time and 1 second the rest then i damn well will notice despite the better average performance, app startup time is also a killer in terms of percived perfomance (and languages like java are terrible for this especilly the first run on boot).
And you may find the more modern and high-level design of the interpreted language's library allows you to write faster performing IO code more easily than the simple & low level libraries that are supplied with most compiled languages, at which point you may get better results for the same programming effort for using that language. java.io really sucks for some types of apps as it basically forces you to have one thread per socket and the new java.nio isn't really any higher level than bsd sockets. I don't know what the situation is like over in.net land though maybe its better there.
java and.net are certainly taking vbs share of the desktop market (primerally internal apps that users are forced to use) as well as a large chunk of the server side but they don't seem to be touching the major commerctial apps!
right now my trillan contact list is pretty much the height of my screen (i used to dock it but i dicided the screen real estate was more usefull for other things) and thats with small icons and not including the "offline contacts" group (which is several screenfulls)
so the eastern authodox church adopted a calender thats similar but not identical to the standard gregorian calender just to avoid saying the catholics were right. Sounds about right for a religous institution.
9x to 2K/XP was a major wrench i agree but still not as big a one as windows-linux, all reasonablly modern apps and quite a few old ones continued working just fine accross the switch, most settings were in the same place, the way the programs menu was handled stayed the same etc (the top level start menu changed in XP but it was easy as heck to switch back and the all programs link did exactly what the programs entry in the old start menu).
personally i can't stand kde/gnome as they fill up nearly the entire taskbar with all kinds of crap, icewm has no desktop which is a dealbreaker for me for a main machine (i do run it on the linux box i keep arround for expermenting and acting as a home server). XPDE looks nice but i don't think many distros include it.
also there is the whole large icon obsession, gaim for example is barely usable because of this (my trillian contact list is already problamatically large even with its small icons). Maybe there is a way to change it hidden somewhere but i never found it. Yes large icons are nice in some situations but they are horrible in others windows tends to give you the choice.
Slashdot Mirror
you have demonstrated that the ammount of menial labour needed to get a level of service/productivity is decreasing but backhoe drivers and factory workers (making those toilet bowl additives) don't need degrees either. Neither do most jobs in things like retail.
what i'm not convinced of is increasing the ammount of subsidized education (and afaict getting unsubsidized education has always been relatively easy if you have the money, though you may have to go abroad in some countries at the university level) and thus leading to more highly educated people than can be absorbed (china already has this problem, thats why people are fighting so hard to get into its top universities rather than just into university at all) is benificial economically.
given that they are on a shelf and given how unstable they are when standing alone on end its a pretty safe bet that they are going to end up either packed tightly or leaning on each other. so no pack them less tightly is not a soloution as they will still end up touching and catching.
using a specially designed rack is an option but this both increases the cost (though possiblly no more so than using DVD cases will) and means you can't mix them with other items on the shelf.
i've seen those but don't like em because they don't seem to have anything in place to keep the CD away from scraping on the box, nor do they seem to grip the CD very well.
Slimline CDR jewel cases while not as good from a labelling point of view seem far more substatial.
disable swapping or run it through an encrypted loopback device?
of course things may be harder on windows than on operating systems that just let the user get on and do what they wan't.
The point is though that the user should not be told that a self signed certificate is secure either. It should just enable the use of encryption.
this is partly a UI design issue, browser UIs seem to be built arround the idea that ssl=secure with the warnings as an afterthought. Though there is a school of thought that self signed certs on the web should be discouraged anyway as they really give little more than a false sense of security (sure its encrypted but anyone can mitm it).
another thing is, i damn well wan't to know if https://www.halifax-online.co.uk/ had a proper cert yesterday and doesn't today because its dns has been hijacked (admittedly some people type the url without the https losing this type of protection).
it *was* but with 2.6.x linus has decided not to fork off a new development branch yet. IIRC the reason for this is he belives most changes for the time being will be confined to subsystems and only major reworking of the kernel as a whole deserves a new branch.
However unfortunately some people are finding that this results in a new kernel version fixing support for one peice of hardware they own at the same time as breaking support for another.
i've found if using the quick login fields you can just ignore the captcha.
Nobody seems to care because those operating systems don't run all the current games. That is changing.. not that they are running that many more games, but games are moving to consoles
some categories of gamer are very poorly served by consoles, this includes anyone who likes custom content, any games that require a keyboard and mouse (yes its possible to use one with a console but consoles are usually used with the TV not on a desk) for proper play (point and click stratagy games and FPS games that require rapid and accurate aiming). Online gaming is another area (yes consoles offer it but usually only at extra cost, with PC games its usually free).
Unfortunately PC game houses have been in a suicidal race to get better graphics resulting in games that only play well on very expensive PCs.
Other than specialized software that a minority are required to run
Yes individual apps may be required by only a small number of people, thats not really the issue at hand though.
Most people have Something that ties them to windows Autocad, Altium designer (formerly protel), MPlab, Xilinx ISE, Photoshop, Internal software (either as native windows apps or as macro heavy msoffice stuff), IE (still needed for many internal sites).
i know quite a few people who consider the gameplay of 2K3/2K4 to be inferior to the original UT.
i've never tried them myself parly due to lack of suitable hardware and partly because i don't game anywhere near as much as i used to.
2. With a generational garbage collector, you don't usually have to scan pages that haven't been modified since the last scan (as these are in old generations and their pointers are already represented in the set of roots for the newer generations)
But surely sooner or later those older generations still have to be scanned in a long running application. What happens then? loads of swapping in i bet!
agreed
and don't forget debian strongly advise against (and don't test) skipping releases when upgrading so your likely to end up doing woody-sarge-etch anyway even if you do the two upgrades at the same time (i'm reffering to after etches release ofc)
of course europe and asia is as artificial a split as north america vs south america possiblly more so...
hmm will the do not call registry apply if the call never touches pots?
and ofc you can always call from out of country while sticking within the same voip provider (generally making the call both free and hard to identify as international)
the idea of css done right is to produce a fancy layout for your pr department to see but have it degrade gracefully to a simple but perfectly usable layout with anything else.
i thought new apache was also incompatible (at least according to the fsf but most distros seem to listen to them over the vendors of the other license)
you already mentioned openssl, add to that anything else under licenses similar to the 4 clause BSD.
and then there is the mpl and its variants............
true but for all the vital stuff the likes of debian have done it already, so you can just grab thier patches.
compiling against winelib means you have to use a compiler setup that can produce elf and means you have to ship different binaries for windows and linux users and there isn't much point in doing it, once the binary is loaded i doubt it makes any difference whatsoever to performance.
for desktop and server applications in the current climate portability to AMD64 is going to become imporant but not to other architectures.
But being able to run the same OS on your desktop and on your embedded equipment certainly carries its advantages. You can run the vast majority of linux software on your embedded box (performance permitting) with a simple recompile.
The point is, if you're going to be waiting 50ms for disk access, why bother shaving 2ms of processing time by running in a native compiled language? Nobody will ever notice.
.net land though maybe its better there.
what they will notice is when the gc decides it needs to scan a memory area that has been swapped out crowding out any other IO on the system.
Average performance only matters for a few time consuming tasks (and they do still exist), what matters far more in end user apps is any apparent hangs, if a button takes 100ms to get a response i probablly won't notice unless i'm gaming. if a button takes 10ms 99% of the time and 1 second the rest then i damn well will notice despite the better average performance, app startup time is also a killer in terms of percived perfomance (and languages like java are terrible for this especilly the first run on boot).
And you may find the more modern and high-level design of the interpreted language's library allows you to write faster performing IO code more easily than the simple & low level libraries that are supplied with most compiled languages, at which point you may get better results for the same programming effort for using that language.
java.io really sucks for some types of apps as it basically forces you to have one thread per socket and the new java.nio isn't really any higher level than bsd sockets. I don't know what the situation is like over in
thats NOT written in native code.
.net are certainly taking vbs share of the desktop market (primerally internal apps that users are forced to use) as well as a large chunk of the server side but they don't seem to be touching the major commerctial apps!
java and
right now my trillan contact list is pretty much the height of my screen (i used to dock it but i dicided the screen real estate was more usefull for other things) and thats with small icons and not including the "offline contacts" group (which is several screenfulls)
ok maybe i should have used a quater symbol instead of .25..................
using 365.24 doesn't change the final result when rounded to the nearest week though.
so the eastern authodox church adopted a calender thats similar but not identical to the standard gregorian calender just to avoid saying the catholics were right. Sounds about right for a religous institution.
9x to 2K/XP was a major wrench i agree but still not as big a one as windows-linux, all reasonablly modern apps and quite a few old ones continued working just fine accross the switch, most settings were in the same place, the way the programs menu was handled stayed the same etc (the top level start menu changed in XP but it was easy as heck to switch back and the all programs link did exactly what the programs entry in the old start menu).
personally i can't stand kde/gnome as they fill up nearly the entire taskbar with all kinds of crap, icewm has no desktop which is a dealbreaker for me for a main machine (i do run it on the linux box i keep arround for expermenting and acting as a home server). XPDE looks nice but i don't think many distros include it.
also there is the whole large icon obsession, gaim for example is barely usable because of this (my trillian contact list is already problamatically large even with its small icons). Maybe there is a way to change it hidden somewhere but i never found it. Yes large icons are nice in some situations but they are horrible in others windows tends to give you the choice.
sorry i did the correct calculation but transposed a couple of digits in my post.
what it should say is:
10 full years to the nearest week is 522 weeks assuming 365.25 days/year.