yes your altitude will change if your velocity changes
an orbit is a situation in which the central accelleration due to gravity interacts with your velocity to give a circular path if you accellerate whilst in orbit you will raise in altitude. If you accelerate sufficiantly you will end up on a course that is no longer orbital and fly off into space.
yes i said sneaky i didn't say right or wrong i just said sneaky
lots of projects are now basing on java and if/when sun (or in the worst case whoever buys it from thier liquidators) start tightning the screws there could be some real pain.
imo java could really do with a free (as in stallman or freer) implementation that actually works properly. BUT much of the drive to create one is removed by the fact that java and its source are a free (as in beer) download.
java apps can eat up all memory quite easilly if they say forget to remove entries from a global list (in some ways gc can make life worse because you have to make sure all references to an object are gone before the gc will even think about freeing it)
reimplementing stuff that people need but that isn't availible as free software is what made it possible to have a completely free software os.
sun has imo played a very sneaky trick by making java not truely free software but just free/open enough to keep the demand for a clone at a fairly low level.
the great thing about access is the self contained nature of the mdb you can put it on a network drive a floppy or whatever and its a single file that you can use it on any machine that has access with no special setup needed.
most of the good databases for linux seem to be database servers. Whilst theese certainly are better for many applications i would not consider them a replacement for access.
(unless otherwise stated apache here reffers to 1.x)
1: apache on unix doesn't use a new process for each request. True each child only serves one request at a time but children lasted for many requests one after another.
2: apache 1.x on win32 WAS multithreaded (iirc it used a single parent and a single child) unlike the unix versions
ofc if you use cgi then it has to start a new process for each cgi request and that is going to be slow on windows
iirc the BBC held some kind of competition to design a computer for an introductory programming course and acorn won
the BBC model B was the original the model A was basically a stripped model B It had the same pcb but half the ram and a lot of perhiperal chips missing. It could be upgraded quite easilly to a model B (if you just wanted to run model B software you could upgrade without soldering but if you wanted the ports you had to solder)
then there was the acorn electron which had as much ram as the BBC B but had a slower clock and a very different physical design (whilst the BBC had space for things like a disk interface on the board the electron had an edge connector on the back).
then there was the B+ and the master series which added yet more to the line
i belive some of the archimedies machines had the BBC owl badge on them as well though they didn't have BBC in thier names
i don't think serious coders ever really used the likes of basic
also unless the compiler is very good at optimising stuff things like array bounds checking are going to carry quite a high penalty. Not to mention it doesn't work too well in conjunction with C's idea that pointers and arrays are more or less the same thing.
the real issue is that C doesn't really have a language level string type (afaict C++ doesn't either but C++ has features that allow you to create one that seems like a primitive to the user) so it has to be done manually with pointers and helper routines which can be error prone to say the least.
isn't hong kong a special administrative region that whilst technically part of china is administered in a ver different way with different rules to the rest of china
at least thats the impression i got correct me if i'm wrong
i seem to remember that some fairly high end routers are not really capable of doing that with any speed (ie you can do it but it forces the packet through the cpu)
you also have to be carefull what you drop when if you have any customers on thier own ip blocks peers etc
i've only ever seen adaptors for sale for putting 2.5 inch hard drives in PCs never ones for putting 3.5 inch hard drives in systems designed for 2.5 inch.
not saying they don't exist but i would guess they are only availible through suppliers for retro computing etc rather than suppliers who aim at pc modding.
they are NOT going to move before sarge release the x.org issue simply happened too late in the sarge develeopment cycle (remember debian supports a lot more than just i386/powepc/ia64/x86-64) which they must ensure that all the crucial system stuff behaves on.
sad as it is such phases seem to be almost a natural part of a countries development.
Britan went through it (industrial revoloution anyone?). The USA went through it (railroad companies anyone?) As each country develops it seems to go through a phase where is competes purely on price and things like workers rights be damned.
why do people work in such situations? because they consider it better than the alternatives! I cant see how not buying from such countries helps the workers there.
debian supports many architectures and this is both a blessing and a curse.
blessing: they are one of the few distors that do and it keeps real geeks friendly with them. If debian didn't support theese architectures it is likely no distro would.
curse: it slows everything down trying to keep things working on all architectures.
ubuntu on the other hand is a branch that favours quick release cycles for a few major architectures (very much like fedora/suse/mandrake do)
unstable is where most contibutions are uploaded so it seems perfectly natural for developers to run the system they are compiling and uploading packages to
testing was and will be (its not right now because debian is in final release freeze) the place where the next stable release is built up. Unless pacakges had problems then prior to the recent freeze they should have made it from unstable to testing in less than two weeks. Hardly an eternity.
i'm not sure what you mean by holding unstable hostage? sure in the run up to a release more care is needed with stuff that could affect propogations into a partially and later fully frozen testing (mainly stuff like shared libs and key system components) but i don't see anything wrong with that.
sure devs can test stuff before uploading to expermental but with something as core system as X or APT you can get a really fucked up system if something goes badly wrong with the program so an extra level of checking is needed.
someone fitted a mac mini inside a pc case and hooked up desktop drives which at least solves the issue of the general shittyness of laptop drives
he did however have to make a custom adaptor to do it though (the adaptor plugged in to the hard drive connector on the riser and was then connected to a standard ide cable running to the hdd and cdrom fitted in the pc case).
Slashdot Mirror
why doesn't someone just write a program to copy the bugs before they get restricted?
or he can get hold of grey market kit just like you could still get booze during prohibition
yes your altitude will change if your velocity changes
an orbit is a situation in which the central accelleration due to gravity interacts with your velocity to give a circular path if you accellerate whilst in orbit you will raise in altitude. If you accelerate sufficiantly you will end up on a course that is no longer orbital and fly off into space.
yes i said sneaky i didn't say right or wrong i just said sneaky
lots of projects are now basing on java and if/when sun (or in the worst case whoever buys it from thier liquidators) start tightning the screws there could be some real pain.
imo java could really do with a free (as in stallman or freer) implementation that actually works properly. BUT much of the drive to create one is removed by the fact that java and its source are a free (as in beer) download.
sure but if they can find out who wrote particular peices of code and get permission they can bring it accross.
java apps can eat up all memory quite easilly if they say forget to remove entries from a global list (in some ways gc can make life worse because you have to make sure all references to an object are gone before the gc will even think about freeing it)
reimplementing stuff that people need but that isn't availible as free software is what made it possible to have a completely free software os.
sun has imo played a very sneaky trick by making java not truely free software but just free/open enough to keep the demand for a clone at a fairly low level.
the great thing about access is the self contained nature of the mdb you can put it on a network drive a floppy or whatever and its a single file that you can use it on any machine that has access with no special setup needed.
most of the good databases for linux seem to be database servers. Whilst theese certainly are better for many applications i would not consider them a replacement for access.
couple of factual errors in that post
(unless otherwise stated apache here reffers to 1.x)
1: apache on unix doesn't use a new process for each request. True each child only serves one request at a time but children lasted for many requests one after another.
2: apache 1.x on win32 WAS multithreaded (iirc it used a single parent and a single child) unlike the unix versions
ofc if you use cgi then it has to start a new process for each cgi request and that is going to be slow on windows
also iirc esa images are only free for non-commercial use wheras nasa images are public domain
btw the "BBC" computers were made by acorn
iirc the BBC held some kind of competition to design a computer for an introductory programming course and acorn won
the BBC model B was the original the model A was basically a stripped model B It had the same pcb but half the ram and a lot of perhiperal chips missing. It could be upgraded quite easilly to a model B (if you just wanted to run model B software you could upgrade without soldering but if you wanted the ports you had to solder)
then there was the acorn electron which had as much ram as the BBC B but had a slower clock and a very different physical design (whilst the BBC had space for things like a disk interface on the board the electron had an edge connector on the back).
then there was the B+ and the master series which added yet more to the line
i belive some of the archimedies machines had the BBC owl badge on them as well though they didn't have BBC in thier names
i don't think serious coders ever really used the likes of basic
also unless the compiler is very good at optimising stuff things like array bounds checking are going to carry quite a high penalty. Not to mention it doesn't work too well in conjunction with C's idea that pointers and arrays are more or less the same thing.
the real issue is that C doesn't really have a language level string type (afaict C++ doesn't either but C++ has features that allow you to create one that seems like a primitive to the user) so it has to be done manually with pointers and helper routines which can be error prone to say the least.
isn't hong kong a special administrative region that whilst technically part of china is administered in a ver different way with different rules to the rest of china
at least thats the impression i got correct me if i'm wrong
plugwash@ppc-osx2:~$ mount /dev/disk0s9 on / (local) /dev (local) /dev (union) /.vol (read-only) /Network/Servers (automounted) /automount (automounted) /private/automount/home/users (automounted)
/dev there doesn't seem to be a /proc listed there though
devfs on
fdesc on
on
automount -fstab [337] on
automount -static [337] on
fserve:/home/users on
plugwash@ppc-osx2:~$
seems that it certainly does have
osx does hide it from the luser interface by default though.
yes osx is more radial than linux in some areas (mostly gui related ones) but most of the archaic unix stuff is still around.
if they have access to run arbitary code on your machine they can just install a new root cert in your browser
once thats done they can mitm your connections and your browser won't put up any warnings at all.
i seem to remember that some fairly high end routers are not really capable of doing that with any speed (ie you can do it but it forces the packet through the cpu)
you also have to be carefull what you drop when if you have any customers on thier own ip blocks peers etc
i've only ever seen adaptors for sale for putting 2.5 inch hard drives in PCs never ones for putting 3.5 inch hard drives in systems designed for 2.5 inch.
not saying they don't exist but i would guess they are only availible through suppliers for retro computing etc rather than suppliers who aim at pc modding.
well the freeze means that even if the debian X guys wanted to rush xorg into sarge they wouldn't be allowed to.
making big changes to unstable at this point is considered very bad form due to dependency issues.
that post was me btw i forgot to login
http://necrotic.deadbeast.net/svn/xfree86/trunk/de bian/local/FAQ.xhtml#debianplans
they are NOT going to move before sarge release the x.org issue simply happened too late in the sarge develeopment cycle (remember debian supports a lot more than just i386/powepc/ia64/x86-64) which they must ensure that all the crucial system stuff behaves on.
sad as it is such phases seem to be almost a natural part of a countries development.
Britan went through it (industrial revoloution anyone?). The USA went through it (railroad companies anyone?) As each country develops it seems to go through a phase where is competes purely on price and things like workers rights be damned.
why do people work in such situations? because they consider it better than the alternatives! I cant see how not buying from such countries helps the workers there.
also i think a lot of smaller white box vendors are pretty much building every box to order anyway (they let you customise basically any part of it.
ofc the problem with linux for theese white box vendors is making sure they build the boxes with hardware linux likes.
debian supports many architectures and this is both a blessing and a curse.
blessing: they are one of the few distors that do and it keeps real geeks friendly with them. If debian didn't support theese architectures it is likely no distro would.
curse: it slows everything down trying to keep things working on all architectures.
ubuntu on the other hand is a branch that favours quick release cycles for a few major architectures (very much like fedora/suse/mandrake do)
unstable is where most contibutions are uploaded so it seems perfectly natural for developers to run the system they are compiling and uploading packages to
testing was and will be (its not right now because debian is in final release freeze) the place where the next stable release is built up. Unless pacakges had problems then prior to the recent freeze they should have made it from unstable to testing in less than two weeks. Hardly an eternity.
i'm not sure what you mean by holding unstable hostage? sure in the run up to a release more care is needed with stuff that could affect propogations into a partially and later fully frozen testing (mainly stuff like shared libs and key system components) but i don't see anything wrong with that.
sure devs can test stuff before uploading to expermental but with something as core system as X or APT you can get a really fucked up system if something goes badly wrong with the program so an extra level of checking is needed.
someone fitted a mac mini inside a pc case and hooked up desktop drives which at least solves the issue of the general shittyness of laptop drives
he did however have to make a custom adaptor to do it though (the adaptor plugged in to the hard drive connector on the riser and was then connected to a standard ide cable running to the hdd and cdrom fitted in the pc case).