Fair enough. The original post just mentioned that the printer was "the only thing" keeping him on Windows:D
For the record, i'm still using Windows (vista, no less) on my desktop because Linux isn't usable for what i want to do just yet (if Transgaming ever pull their finger out and support Falcon: Allied Force, maybe I'll reinstall and re-subscribe to Cedega)... haven't tested my hardware with a recent Linux distribution (Intel DP35DP mobo, Core2 Q6600, geforce 8800gts, sb-xfi), really should pull my finger out this weekend...
Anyone find it amusing that one of the chief proponents of Gnome (the project started because KDE USED TO use a "non-free" toolkit) is pushing a non-free standard?
Can you say more about what the ideology of Gnome is, and what the flaw in the ideology is?
To paraphrase Linus himself: "the idea that users are idiots and need to be protected from functionality".
Short story is, KDE helps me get shit done. Gnome obscures commonly used UI functions, all in the supposed interests of "usability", no less.
Gnome is currently brain-damaged by design and needs to die - or at least, the "usability" fucktards need to wake up and realise that crippling software to make it less intimidating is no way to progress...
Considered ditching the printer for something that works? yes, it would be nice to not have to do so, but given that the cost of the printer is far less than a Windows license... next time you upgrade O/S, get a new printer instead of paying for Windows?
As I said, I haven't run destop linux for a little while, however, some of the responses you've made to my points have conveniently avoided some of the pitfalls compared to windows. Eg, on windows, sure there's different sound standards. In practice, it all "just works". Linux? Oh, if you have sound daemon X running it fucks up the sound for apps that run using API Y...
As I said, it is stabilising, but if you'd been around long enough to remember the a.out and c library changes, you'd know what i'm talking about when I'm referring to compatibility issues. Back in the days of redhat 5.0 for example, the switch to glibc caused huge numbers of apps to break both in terms of binary compatibility and *source* compatibility.
Yes, there's various packages that may or may not be present on a windows box, however, any app can rely on win32 to be present, along with directX 7 or later. As vista is pushed out, at least some version of.net is present as well. However, the package specifics aren't relevant - what is relevant is that there's a usable base desktop OS present on any windows box, with a known set of components. All the vendor has to put on the box is "needs windows *shiny verson*" The differences between each linux distribution (and customisation by the user of each of those distributions) mean that this is not true for linux. MacOS has the same advantage as Windows here - there's a known "base platform" that is actually useful for desktop applications - Linux does not have this, and needs it, if it's to become more than a hobby OS for the mainstream - on the desktop. Server space is another matter, i've been running Linux boxes in the enterprise since 1996:D
I mean for a start, if i'm building a desktop app right now for Linux, I pretty much have a choice of either KDE or gnome, and whichever I choose probably half (or more) of the linux systems out there will not have the required components installed (some with have KDE some will have gnome, some will have neither). It's something as simple as a fucking toolkit - and there's no "known quantity" that every linux system will have.
I'm not too keen on running outdated applications just to get the correct documentation
If you look at the FreeBSD ports (which is where most of your "applications" come from, you'll probably find that they're as up to date or more than any linux distribution.
The base system however is fairly solid and the *interface* on the included base apps remains consistent. as a random example: you don't get 3 different boot loaders with their own terminology and documentation. You get the freebsd boot loader, which works just fine, and has remained consistent from a user perspective forever. You don't get, for example the deprecation of an entire firewalling toolset, to be replaced by a new toolset with incompatible syntax. Sure, FreeBSD has 3 different firewalling toolsets (why isn't exactly clear to me) - but even though for example, there have been major improvements to pf over the years - the interface has remained consistent - the new version is a superset of the old. You don't have the situation where, for example, in the linux world, you throw away everything you know about ipfwadm, and learn ipchains. Then you throw all that away and need to learn iptables. etc.
Pulling the old "freebsd has outdated sofware" shit just shows you either have no idea or are just being deliberately antagonistic.
BTW, on slackware, KDE is the standard desktop. Gnome packages aren't even included:D
The big thing I like about KDE is the ioslaves. Sure, some of the user interface stuff might appear a little cluttered - but clutter i can deal with once I understand it. What really shits me about gnome is that in the interests of "user-friendliness" they've castrated UI so that fairly common tasks require some extra steps to perform. Why? Is making the user spend 5-10 minutes to learn a UI and then be productive in it worse than making it look nice, but crippled for anyone with an IQ above 50? Am I on crack to think that catering to the user for the *other* 99.999% of their computing use after the first 10 minutes is a bad thing?
Personally I'd really like to see OpenStep and objective C programming take off on Linux. If it in a usable state, and was part of a standard insallation, writing cross platform stuff between OS/X and Linux/Nix would be a lot more feasible.
Being open is all well and good, but one of the reasons Linux is so hard for commercial software houses to deal with is that it's a moving platform.
It's stabilising gradually, but there's a lot to be said for the "windows way" of maintaining backwards binary compatibility and having a fixed API for drivers, etc.
QT? GTK? Which version? DRI? Which threading library? Mono? Java? Which version? Which sound library, OSS or ALSA? E sound daemon? Which package format? Which C library? Some of those aren't necessarily a problem any more (i haven't run desktop linux on a regular basis for a couple of years), but the general problem of no fixed standard platform is still there.
I love free software, but certain apps (in particular, games for example - the "boring" sofware code-wise is another - eg, business accounting software, tax software) just aren't well catered to by Free software - the projects are bigger now-adays than some kid in his bedroom can manage. Until a company can write for a set platform and expect that his app will work pretty reliably on most machines out there, they'll be reluctant to spend too much time on it.
One of the reasons I like FreeBSD is that FreeBSD = FreeBSD. It's less fragmented, documentation from 2-4 years ago still usually works (even if there's been code updates, most of the time the "interface" side of things remains consistent - I've no idea how many times i've gone through linux howtos over the years to find that they're a year old and everything has changed), and the base platform is a known quantity. However, the same problem exists to an extent - the base platform isn't comprehensive enough...
Ubuntu has gone some way to help out, but there's still some way to go...
Xbox? Open? Since when? PS3 runs linux *with sony's permission* no less - the xbox doesn't.
Back in the day, the megaCD was aimed at the sort of market MS is aiming at now - remember CD+G?
As to the loss of PS2 support in the PS3 - this isn't as bad as you might think. The PS2 is still on sale, and cheap. Most potential PS3 owners already own a PS2. Since I've owned my PS2, i think i've perhaps used the PS1 backwards compatibility maybe 3 times - i have an unused library of 40-50 PS1 games, too...
Agreed, but fact remains in real world performance terms, i have noticed no difference (XP vs vista, on hardware purchased since 2002), so the overhead of what actually gets run, unless you're dealing with DRM content, appears to me, to be minimal.
It's not that developers don't care - well, maybe some don't, it's about controlling software development and maintenance costs.
Programming/debugging/maintaining stuff written in assembler is expensive. Easiest way to go about it is to write in a higher level language, optimise the hot-spots and wear the memory footprint. Memory is cheap.... programmer time is not.
Wasteful? Sure. But if we were still writing everything in hand optimised assembler, we wouldn't be using the internet, downloading ISOs and building home media centres, etc - on free software. We'd more likely have the open-source equivalent of Windows 3.1 for the cost of vista ultimate...
Is Windows bloat free? Of course not. However, when RAM costs I'd much rather be running FreeBSD full time, but it's not because of the bloat - it's because of the user environment - windows treats you like a fucking retard, and it's irritating... but for the apps most people want to run/develop, it's a fairly usable platform.
What i meant to say was "when ram costs $100gb, disk costs $200/500gb, and cpu is cheap - what's the point" but it went missing...
Seriously, just how big do you think the DRM subsystem is in terms of code?
Given that all it does is check encryption keys and decrypt data - i would wager it could be done in the equivalent of 1-2 lines of Perl. I've noticed no real speed difference between XP and vista on an old 2.4ghz non-ht PC. That machine is 5 years old.
The "bloat" in windows is things like:
compatibility with 16 bit windows apps
window toolkit
activeX object library
.net runtime environment
directX
etc
I like linux/bsd as much as the next guy, but you'll notice that as they begin to get feature parity with windows, the "bloat" is going up in them as well.
When putting out an OS you have a choice: do you provide just the bare minimum of services (useful for embedded apps), or do you provide a complete OS including graphics libraries, 3d graphics libraries, various programming widgets, etc?
Is Windows bloat free? Of course not. However, when RAM costs
I'd much rather be running FreeBSD full time, but it's not because of the bloat - it's because of the user environment - windows treats you like a fucking retard, and it's irritating... but for the apps most people want to run/develop, it's a fairly usable platform.
Watch it, you'll get marked "troll" posting support for vista here. Really,/. is not what it used to be when you could actually have an intelligent discussion on any particular topic. Sure, there's gems to be found still, but fuck - the signal:noise ratio is pretty abysmal.
It's totally possible to rip or pirate content using vista. Every DVD I buy/rent/whatever gets ripped to ISO image and then div-x'ed when convenient. On vista.
All vista does is enable you to play DRM'd content.
So, boycott it all you want because of DRM - all you're doing is fucking yourself over in terms of what content is available to you.
Slashdot Mirror
For the record, i'm still using Windows (vista, no less) on my desktop because Linux isn't usable for what i want to do just yet (if Transgaming ever pull their finger out and support Falcon: Allied Force, maybe I'll reinstall and re-subscribe to Cedega)... haven't tested my hardware with a recent Linux distribution (Intel DP35DP mobo, Core2 Q6600, geforce 8800gts, sb-xfi), really should pull my finger out this weekend...
Got hypocrisy much?
To paraphrase Linus himself: "the idea that users are idiots and need to be protected from functionality".
Short story is, KDE helps me get shit done. Gnome obscures commonly used UI functions, all in the supposed interests of "usability", no less.
Gnome is currently brain-damaged by design and needs to die - or at least, the "usability" fucktards need to wake up and realise that crippling software to make it less intimidating is no way to progress...
Considered ditching the printer for something that works? yes, it would be nice to not have to do so, but given that the cost of the printer is far less than a Windows license... next time you upgrade O/S, get a new printer instead of paying for Windows?
Rather, it depends if you're in one of the vast number of english speaking countries - or an american.
Got nanny state?
*sigh*
Yeah, i have troll/flamebait set to +2. it makes the site much more amusing :)
As I said, it is stabilising, but if you'd been around long enough to remember the a.out and c library changes, you'd know what i'm talking about when I'm referring to compatibility issues. Back in the days of redhat 5.0 for example, the switch to glibc caused huge numbers of apps to break both in terms of binary compatibility and *source* compatibility.
Yes, there's various packages that may or may not be present on a windows box, however, any app can rely on win32 to be present, along with directX 7 or later. As vista is pushed out, at least some version of .net is present as well. However, the package specifics aren't relevant - what is relevant is that there's a usable base desktop OS present on any windows box, with a known set of components. All the vendor has to put on the box is "needs windows *shiny verson*" The differences between each linux distribution (and customisation by the user of each of those distributions) mean that this is not true for linux. MacOS has the same advantage as Windows here - there's a known "base platform" that is actually useful for desktop applications - Linux does not have this, and needs it, if it's to become more than a hobby OS for the mainstream - on the desktop. Server space is another matter, i've been running Linux boxes in the enterprise since 1996 :D
I mean for a start, if i'm building a desktop app right now for Linux, I pretty much have a choice of either KDE or gnome, and whichever I choose probably half (or more) of the linux systems out there will not have the required components installed (some with have KDE some will have gnome, some will have neither). It's something as simple as a fucking toolkit - and there's no "known quantity" that every linux system will have.
If you look at the FreeBSD ports (which is where most of your "applications" come from, you'll probably find that they're as up to date or more than any linux distribution.
The base system however is fairly solid and the *interface* on the included base apps remains consistent. as a random example: you don't get 3 different boot loaders with their own terminology and documentation. You get the freebsd boot loader, which works just fine, and has remained consistent from a user perspective forever. You don't get, for example the deprecation of an entire firewalling toolset, to be replaced by a new toolset with incompatible syntax. Sure, FreeBSD has 3 different firewalling toolsets (why isn't exactly clear to me) - but even though for example, there have been major improvements to pf over the years - the interface has remained consistent - the new version is a superset of the old. You don't have the situation where, for example, in the linux world, you throw away everything you know about ipfwadm, and learn ipchains. Then you throw all that away and need to learn iptables. etc.
Pulling the old "freebsd has outdated sofware" shit just shows you either have no idea or are just being deliberately antagonistic.
BTW, on slackware, KDE is the standard desktop. Gnome packages aren't even included :D
The big thing I like about KDE is the ioslaves. Sure, some of the user interface stuff might appear a little cluttered - but clutter i can deal with once I understand it. What really shits me about gnome is that in the interests of "user-friendliness" they've castrated UI so that fairly common tasks require some extra steps to perform. Why? Is making the user spend 5-10 minutes to learn a UI and then be productive in it worse than making it look nice, but crippled for anyone with an IQ above 50? Am I on crack to think that catering to the user for the *other* 99.999% of their computing use after the first 10 minutes is a bad thing?
Personally I'd really like to see OpenStep and objective C programming take off on Linux. If it in a usable state, and was part of a standard insallation, writing cross platform stuff between OS/X and Linux/Nix would be a lot more feasible.
It's stabilising gradually, but there's a lot to be said for the "windows way" of maintaining backwards binary compatibility and having a fixed API for drivers, etc.
QT? GTK? Which version? DRI? Which threading library? Mono? Java? Which version? Which sound library, OSS or ALSA? E sound daemon? Which package format? Which C library? Some of those aren't necessarily a problem any more (i haven't run desktop linux on a regular basis for a couple of years), but the general problem of no fixed standard platform is still there.
I love free software, but certain apps (in particular, games for example - the "boring" sofware code-wise is another - eg, business accounting software, tax software) just aren't well catered to by Free software - the projects are bigger now-adays than some kid in his bedroom can manage. Until a company can write for a set platform and expect that his app will work pretty reliably on most machines out there, they'll be reluctant to spend too much time on it.
One of the reasons I like FreeBSD is that FreeBSD = FreeBSD. It's less fragmented, documentation from 2-4 years ago still usually works (even if there's been code updates, most of the time the "interface" side of things remains consistent - I've no idea how many times i've gone through linux howtos over the years to find that they're a year old and everything has changed), and the base platform is a known quantity. However, the same problem exists to an extent - the base platform isn't comprehensive enough...
Ubuntu has gone some way to help out, but there's still some way to go...
There's some awesome stuff out there now, and there was some real trash out there previously (custer's revenge anyone?).
The 90% law is pretty accurate too - there seems to be an abundance of crap these days simply because the market is so much bigger.
Back in the day, the megaCD was aimed at the sort of market MS is aiming at now - remember CD+G?
As to the loss of PS2 support in the PS3 - this isn't as bad as you might think. The PS2 is still on sale, and cheap. Most potential PS3 owners already own a PS2. Since I've owned my PS2, i think i've perhaps used the PS1 backwards compatibility maybe 3 times - i have an unused library of 40-50 PS1 games, too...
Just ask sega about how well console "upgrades" sell - eg MegaCD, 32X, etc.
Agreed, but fact remains in real world performance terms, i have noticed no difference (XP vs vista, on hardware purchased since 2002), so the overhead of what actually gets run, unless you're dealing with DRM content, appears to me, to be minimal.
Programming/debugging/maintaining stuff written in assembler is expensive. Easiest way to go about it is to write in a higher level language, optimise the hot-spots and wear the memory footprint. Memory is cheap.... programmer time is not.
Wasteful? Sure. But if we were still writing everything in hand optimised assembler, we wouldn't be using the internet, downloading ISOs and building home media centres, etc - on free software. We'd more likely have the open-source equivalent of Windows 3.1 for the cost of vista ultimate...
I've accidentally rebooted a running machine with liquified compressed air before :D
These days, modern hardware seems to die after 3 years or so anyway. no warranty = people need to re-purchase new hardware anyway.
What i meant to say was "when ram costs $100gb, disk costs $200/500gb, and cpu is cheap - what's the point" but it went missing...
Given that all it does is check encryption keys and decrypt data - i would wager it could be done in the equivalent of 1-2 lines of Perl. I've noticed no real speed difference between XP and vista on an old 2.4ghz non-ht PC. That machine is 5 years old.
The "bloat" in windows is things like:
I like linux/bsd as much as the next guy, but you'll notice that as they begin to get feature parity with windows, the "bloat" is going up in them as well.
When putting out an OS you have a choice: do you provide just the bare minimum of services (useful for embedded apps), or do you provide a complete OS including graphics libraries, 3d graphics libraries, various programming widgets, etc?
Is Windows bloat free? Of course not. However, when RAM costs I'd much rather be running FreeBSD full time, but it's not because of the bloat - it's because of the user environment - windows treats you like a fucking retard, and it's irritating... but for the apps most people want to run/develop, it's a fairly usable platform.
Watch it, you'll get marked "troll" posting support for vista here. Really, /. is not what it used to be when you could actually have an intelligent discussion on any particular topic. Sure, there's gems to be found still, but fuck - the signal:noise ratio is pretty abysmal.
All vista does is enable you to play DRM'd content.
So, boycott it all you want because of DRM - all you're doing is fucking yourself over in terms of what content is available to you.
Funny? It's true damn it :D
That would be because XP64 is a dead-end, dead-on-arrival product and Vista 64 is not.
Sound nuts? Yes... but they do. Large clusters of many inexpensive machines set up in a redundant manner...
Drop in the ocean, etc.
Hope their plans are a bit bigger/more ambitious than that, long term :D