stability;) really, 3.5 series have been very, very stable. while 3.0, 3.1 had some crashes, some high memoery usage cases, latest versions have concentrated on bugfixes and minor features, so this is a very pleasant desktop to use without nasty surprises.
oh, and quite a lot of apps aren't yet fully ported to qt/kde4, so you would have to run kde3 versions - which increases memory usage.
...and they aren't allowed to develop their program on the free qtlib libraries first, then switch to the commercial libs.
On top of that, at least with regular GPL'd programs, the sources as well as the sources of the version required for GPL'd libraries to run the application can be distributed. Not so, with Qt. Instead, the user of any program relying on a GPL'd version of qt has to download their own copy qt (including trying to get the correct version of it). The Author of the GPL'd program is not permitted to redistribute any GPL'd versions (aka as the free-qt version) of Qt with the program. troll alert. not trolltech one, no, just plain troll. it is impossible to be so uninformed but still convinced of oneself. oh, ot would be kinda cool if you could quote gpl license parts that prohibit distribution, all linux distros will have to remove qt now, *horror*.
For instance: WinFS, MinWin, capability to operate with less than half a terabyte of RAM, users...... Sadly, I only have bloat on that list so far... is that bloated users announcement worldwide ?
right, the widespread piracy (in the new sense of the word) illustrates your point perfectly... not. seeing how many people consider such sharing a normal activity speaks for itself.
one more thing - decriminalise current copyright infrigements - and maybe even make personal use legal, only allowing to capitalise on commercial use. this actually is happening on it's own, where publishers (of any material) find new business models. so music is distributed for free (to become recognised), but all other stuff is where the money comes in - including commercial use of the works. and such a model seems to align well with ethics of most people, as opposed to the current system. remember, laws are any good only if majority of people do not consider them wastly stupid.
distro people will know what to whitelist. As a developer, how do I get my package into a distro? usually, if it is not picked up already, you would contact packagers, find out who is packaging similar packages to yours and help them to get it up and running, so that the packager can easily wrap it all into package[s]
and with different mandatory access control mechanisms, each application can b restricted even more to only access resources it is intended to, prventing some exploits in the app itself from working as well. The customized Linux distribution on One Laptop Per Child's XO laptops uses such capability restriction instead of whitelisting. See Bitfrost. well, maybe not instead - they can be easily used together. though current package signing can already can be considered whitelisting for some vectors, it (the whitelisting) still could be made more pervasive.
this could work quite nicely for most home users who are satisfied by what linux distributions offer, actually. distribution packages are pgp signed, and are the only things whitelisted. there, no need for the user to know anything, distro people will know what to whitelist. and with different mandatory access control mechanisms, each application can b restricted even more to only access resources it is intended to, prventing some exploits in the app itself from working as well.
i don't think it can be called "hardware" prevention here. pulling out the cable, that would be hardware prevention, but in this case you have software solution, only you have pushed it to another device. this changes the layout, but the approach stays the same.
why, yes. switch underlying platform, that way you get away with antivirus requirement (except clamav on the mailserver to keep crapmail volume low), and get patch distribution system as well;)
actually, there's a quite neat approach. set some release date in future and just declare "what's ready at that point gets released, other features can be developed after that point in time". and i believe that's a quite nice approach, especially for stable branches.
but you can have most of that redundancy with cheaper boxes, for example, from supermicro. and supermicro support has been far superior than hp in my experience...
right. when converting an old document. but converting to what ? to the new format, i guess. so WHY, oh why does this new format need such nosense ATA ALL ? that complicates the spec, and i can really see only one reason - to ease life for microsoft office developers, who don't have to properly express the formatting (with indentaion, spacing and other parameters). which is how it should be done. and this has been repeated a lot of times. such things should have never ever went into the specification itself. oh, i haven't been following too closely, did they get rid of those pages where they explicitly listed pictures to be used for page borders ?
yes, and people place things on page with spaces (or the advanced ones, with tabs). yes, and people mix outline numbering, paragraph numbering and manual numbering throughout a long document. yes, and people insert automated table of contents only to edit it manually later, thus rendering it non-updateable ever. yes, and people set paragraph style to heading, then manually format it into a normal paragraph look-alike. yes, and people have no idea what non-breaking space is, so they write date together with the month or initial together with the surname to prevent them from appearing on different lines.
for 99% of msoffice users, kword or abiword would be an overkill, not to talk about openoffice.org (or try to talk about what features oo.org has that msword doesn't). the knowledge general population has about using "basic" office software, the documents that result from this - that's a huge mess. so touting feature count (which is debatable on itself) at this point just sounds a bit silly:)
sorry about the rant, just 10 minutes ago i had a prolonged conversation with a colleague about this very same topic, so i had just recounted most of the usual abuses:)
but if the driver is opensourced, it is very, very likely to end in upstream packages, be it kernel, xorg, cups or sane. so now this driver will be packaged by your distributor and no compiling for most users. the only cases when somebody would want top compile the driver - is using a distribution that for some reason does not include the required package (and then driver being opensourced does not change anything much), and wishing to use some other version of the driver. but i'd hope those still are marginal cases...
i wonder why it would appear as such. it definitely works on linux (on this opensuse installation, i could just install package openvpn-2.0.9-44@i586).
not sure about macos, but i'd expect it to work there as well. actually, i think openvpn is the best vpn solution at this point.
they have always done what ? surely they haven't. i remember the outrage when it was first discovered that some of them started marking drives with 10 base units (or even mixed 2/10 base units when calculating size). granted, that was at the time when drives were still measured in megabytes. yes, megabytes.
i find it amazing that these people have time to follow, notice trends and reply to all of those posts. and yes, i agree it's even more annoying than somebody else arguing with himself. as if all those "it's him again !" were... well, him as well.
An easier way to install drivers would be appreciated as well. Typing a command at the prompt is not my idea of user-friendly even though I'm capable of doing it. Not everyone knows what 'insmod' or 'modprobe' are for and they shouldn't have to know. which is exactly what opensource drivers will provide. they will install just like other opensource drivers currently - and will just work. and upgrading them will be included in distro upgrade process.
Of course, with truly open source drivers users have to go through the extra hassle of *compiling* the driver prior to installation. i just have to ask - wtf ? what are you talking about ? that sounds so incredibly misinformed that a) you do not actually know what modprobe does; b) are trolling. how many of modern linux distribution users compile everything on their boxes ? because, you know, alsmo tall of it is open source. as usual, it will be compiled by distributors, only now those who will have a need, will be able to compile themselves.
what's so annoying about subtitles and recompressing ? decent subtitles are in a text file, with timestamp markers. unless your recompressing process changes movie length, they do not impact recompressing in any way.
yes, one thing is microsoft application internal integration, but do we really want to move from being dependent on one vendor to being dependent on another - even larger vendor ? i'd guess that for companies migrating to linux and other opensource software vendor independence is taken quite seriously, at least in larger companies. aren't there really viable solutions already available in opensource land ? i've heard that http://www.kolab.org/ is something to consider, especially the latest version - but i haven't had a chance to try it out myself. any users of it, maybe ?
the classic, by now. i was thinking about posting that link myself, if nobody would have managed to do that before. too bad i run out of modpoints few days ago:)
well, 'commercial music' should be defined precisely then;) if music is created with intent to sell it but fails - is that commercial music ? if music is created without commercial intent but becomes widely successful commercially - is that commercial music ?
i don't think current 'commercial music' would completely die off - just as with other niches, new business models can and will work. the market will only reshape, and then become more robust (there have been several showcases lately - nin, radiohead etc).
also, one side is the motivation to create, which can adapt, and then there's the insane length of copyright. i think that current piracy is only fueled by the copyright length, as re-selling of the same product for decades only damages its perception (in this case - perceived value of the music) in the eyes of the general public.
actually, lately i am seeing more and more new music created. authors place it for free download, because, well, they want people to hear it. as a result, people tend to go more to gigs and so on. i'm not interested in those sweet boybands that some old producer with weird sexual preferences creates one after another, as those can't adapt to such an environment. so, if we get less "music" like that and more of 'underground' one... hey, go for it:)
actually, this is simply following their own procedures. if you have a law regarding procurements that states in what cases a company can not participate, you sort of are expected to follow it. mostly. in this case the question would be whether a single company should be awarded an exception.
Slashdot Mirror
stability ;)
really, 3.5 series have been very, very stable.
while 3.0, 3.1 had some crashes, some high memoery usage cases, latest versions have concentrated on bugfixes and minor features, so this is a very pleasant desktop to use without nasty surprises.
oh, and quite a lot of apps aren't yet fully ported to qt/kde4, so you would have to run kde3 versions - which increases memory usage.
...and they aren't allowed to develop their program on the free qtlib libraries first, then switch to the commercial libs.On top of that, at least with regular GPL'd programs, the sources as well as the sources of the version required for GPL'd libraries to run the application can be distributed. Not so, with Qt. Instead, the user of any program relying on a GPL'd version of qt has to download their own copy qt (including trying to get the correct version of it). The Author of the GPL'd program is not permitted to redistribute any GPL'd versions (aka as the free-qt version) of Qt with the program. troll alert. not trolltech one, no, just plain troll.
it is impossible to be so uninformed but still convinced of oneself.
oh, ot would be kinda cool if you could quote gpl license parts that prohibit distribution, all linux distros will have to remove qt now, *horror*.
Sadly, I only have bloat on that list so far... is that bloated users announcement worldwide ?
right, the widespread piracy (in the new sense of the word) illustrates your point perfectly... not.
seeing how many people consider such sharing a normal activity speaks for itself.
one more thing - decriminalise current copyright infrigements - and maybe even make personal use legal, only allowing to capitalise on commercial use.
this actually is happening on it's own, where publishers (of any material) find new business models. so music is distributed for free (to become recognised), but all other stuff is where the money comes in - including commercial use of the works.
and such a model seems to align well with ethics of most people, as opposed to the current system. remember, laws are any good only if majority of people do not consider them wastly stupid.
this could work quite nicely for most home users who are satisfied by what linux distributions offer, actually.
distribution packages are pgp signed, and are the only things whitelisted. there, no need for the user to know anything, distro people will know what to whitelist.
and with different mandatory access control mechanisms, each application can b restricted even more to only access resources it is intended to, prventing some exploits in the app itself from working as well.
i don't think it can be called "hardware" prevention here. pulling out the cable, that would be hardware prevention, but in this case you have software solution, only you have pushed it to another device. this changes the layout, but the approach stays the same.
why, yes. switch underlying platform, that way you get away with antivirus requirement (except clamav on the mailserver to keep crapmail volume low), and get patch distribution system as well ;)
actually, there's a quite neat approach. set some release date in future and just declare "what's ready at that point gets released, other features can be developed after that point in time".
and i believe that's a quite nice approach, especially for stable branches.
but you can have most of that redundancy with cheaper boxes, for example, from supermicro.
and supermicro support has been far superior than hp in my experience...
right. when converting an old document. but converting to what ? to the new format, i guess.
so WHY, oh why does this new format need such nosense ATA ALL ?
that complicates the spec, and i can really see only one reason - to ease life for microsoft office developers, who don't have to properly express the formatting (with indentaion, spacing and other parameters). which is how it should be done. and this has been repeated a lot of times. such things should have never ever went into the specification itself.
oh, i haven't been following too closely, did they get rid of those pages where they explicitly listed pictures to be used for page borders ?
yes, and people place things on page with spaces (or the advanced ones, with tabs).
:)
:)
yes, and people mix outline numbering, paragraph numbering and manual numbering throughout a long document.
yes, and people insert automated table of contents only to edit it manually later, thus rendering it non-updateable ever.
yes, and people set paragraph style to heading, then manually format it into a normal paragraph look-alike.
yes, and people have no idea what non-breaking space is, so they write date together with the month or initial together with the surname to prevent them from appearing on different lines.
for 99% of msoffice users, kword or abiword would be an overkill, not to talk about openoffice.org (or try to talk about what features oo.org has that msword doesn't).
the knowledge general population has about using "basic" office software, the documents that result from this - that's a huge mess.
so touting feature count (which is debatable on itself) at this point just sounds a bit silly
sorry about the rant, just 10 minutes ago i had a prolonged conversation with a colleague about this very same topic, so i had just recounted most of the usual abuses
but if the driver is opensourced, it is very, very likely to end in upstream packages, be it kernel, xorg, cups or sane. so now this driver will be packaged by your distributor and no compiling for most users.
the only cases when somebody would want top compile the driver - is using a distribution that for some reason does not include the required package (and then driver being opensourced does not change anything much), and wishing to use some other version of the driver.
but i'd hope those still are marginal cases...
i wonder why it would appear as such. it definitely works on linux (on this opensuse installation, i could just install package openvpn-2.0.9-44@i586).
not sure about macos, but i'd expect it to work there as well.
actually, i think openvpn is the best vpn solution at this point.
they have always done what ?
surely they haven't. i remember the outrage when it was first discovered that some of them started marking drives with 10 base units (or even mixed 2/10 base units when calculating size).
granted, that was at the time when drives were still measured in megabytes. yes, megabytes.
i find it amazing that these people have time to follow, notice trends and reply to all of those posts. and yes, i agree it's even more annoying than somebody else arguing with himself. as if all those "it's him again !" were... well, him as well.
and upgrading them will be included in distro upgrade process. Of course, with truly open source drivers users have to go through the extra hassle of *compiling* the driver prior to installation. i just have to ask - wtf ?
what are you talking about ? that sounds so incredibly misinformed that
a) you do not actually know what modprobe does;
b) are trolling.
how many of modern linux distribution users compile everything on their boxes ? because, you know, alsmo tall of it is open source.
as usual, it will be compiled by distributors, only now those who will have a need, will be able to compile themselves.
what's so annoying about subtitles and recompressing ? decent subtitles are in a text file, with timestamp markers. unless your recompressing process changes movie length, they do not impact recompressing in any way.
yes, one thing is microsoft application internal integration, but do we really want to move from being dependent on one vendor to being dependent on another - even larger vendor ?
i'd guess that for companies migrating to linux and other opensource software vendor independence is taken quite seriously, at least in larger companies.
aren't there really viable solutions already available in opensource land ?
i've heard that http://www.kolab.org/ is something to consider, especially the latest version - but i haven't had a chance to try it out myself. any users of it, maybe ?
the classic, by now. :)
i was thinking about posting that link myself, if nobody would have managed to do that before.
too bad i run out of modpoints few days ago
and where are torrent links ? phew, how unprofessional submission...
well, 'commercial music' should be defined precisely then ;)
if music is created with intent to sell it but fails - is that commercial music ?
if music is created without commercial intent but becomes widely successful commercially - is that commercial music ?
i don't think current 'commercial music' would completely die off - just as with other niches, new business models can and will work. the market will only reshape, and then become more robust (there have been several showcases lately - nin, radiohead etc).
also, one side is the motivation to create, which can adapt, and then there's the insane length of copyright. i think that current piracy is only fueled by the copyright length, as re-selling of the same product for decades only damages its perception (in this case - perceived value of the music) in the eyes of the general public.
actually, lately i am seeing more and more new music created. authors place it for free download, because, well, they want people to hear it. as a result, people tend to go more to gigs and so on. :)
i'm not interested in those sweet boybands that some old producer with weird sexual preferences creates one after another, as those can't adapt to such an environment. so, if we get less "music" like that and more of 'underground' one... hey, go for it
actually, this is simply following their own procedures. if you have a law regarding procurements that states in what cases a company can not participate, you sort of are expected to follow it. mostly.
in this case the question would be whether a single company should be awarded an exception.