Don't shoot off your mouth telling them they should go write drivers. That's the wrong attitude.
Re-read what I said:
How about this: if iRiver doesn't work in Linux, complain to iRiver.
Linux supports every piece of hardware I own. It wasn't an accident: I buy hardware from companies that make compatible quality hardware. I suggest others do the same.
Linux needs to support more devices than it does,
So what?
and this article brings that fact to light.
No, it shits on the hard work of thousands of people, and tells people to be happy with what they've got. Instead, it should be telling people not to buy iRiver because iRiver doesn't care about them. iRiver doesn't care about Macintosh users, and they don't care that much about Windows users either.
Fact is, iRiver doesn't care about me, and they don't care about you, and furthermore, I think it's positively the stupidest thing I ever heard to buy their hardware in spite of that.
Btw, iRiver works great on Mac OS X. Simply use a program like XNJB and you're off to the races.
No, iRiver works great with XNJB which uses libmtp- written by and for Linux users. If the original poster had installed an LIBMTP-based application, they would be able to use their iRiver but because they said this is a Linux problem and because you obviously agree, you are not allowed to install XNJB to reverse my claim: iRiver does not work with Mac OS X because you have to install third party (although Free) software to give Mac OS X support.
The difference is that none of these manufacturers of music players are seeking to support Linux as they do other OSes - we as a community have taken on that yoke ourselves for now - and so, as a result, if a particular player doesn't play well with Linux, it is a failure of those who have been trying to improve support for the players on Linux. It's fine to say that companies should provide Linux support for their hardware - but generally that just doesn't happen. If you want to use Linux you need to accept that and deal with it.
That might've been true ten years ago, but these days its completely bogus. Linux has overall better hardware and software support than any other platform. If companies go out of their way to make their products incompatible with Free Software, I tell people to not buy from those companies. I don't say, it's Free Software's fault, try again in a few years, and I am thoroughly disgusted with anyone who would seriously suggest that.
but nevertheless, the set of software that makes up the "Linux Desktop" can't do what you want, so it has failed.
That's a terribly useless definition that makes the Linux Desktop fail because Windows and Macintosh fail just as equally.
There is no reason to use Windows anymore: The fact that "iRiver" doesn't have particularly good drivers on Linux is a good reason to not use iRiver and not a good reason to use Windows.
With Windows, you don't have those options...
How so?
There are no NT or Vista drivers for some STB-based cards. People who bought them cannot use them unless they use Linux. ATI won't make drivers for them because they're old, and Microsoft won't let you make drivers for them because if you could make drivers for them, you might also be tempted to circumvent Digital Restrictions Management software.
Microsoft is completely irrelevant anymore: Free software means that companies have to make their customers happy instead of just making sure that nobody else can make them happier. Since I'm not using them in the workplace, I don't have to use them at home.
Unless you use an iPod, this appears to be a real weak point in the Linux desktop.
iRiver doesn't work very well on a Macintosh either, so I bet you consider it a weak point in the Macintosh desktop as well.
How about this: if iRiver doesn't work in Linux, complain to iRiver.
With Linux, you could also fix the problem yourself. You could also pay someone to fix the problem. If the iRiver is popular enough, you could also wait and someone else will fix it for you.
With Windows, you don't have those options, so I consider that a weak point in the Windows desktop.
The point is that jumping through some hoops that it does work on the Toshiba.
So don't buy a Toshiba?
With a little bit of work Ubuntu could be made smart enough to tell me that the only way my machine will work completely is to possibly install the restricted drivers.
Yeah, and Toshiba could do that work as well. They do that work for windows- certainly you can't install a non-OEM version of Windows on that same laptop without as much - if not more work. Windows won't tell you "you need to go to toshiba.com, download something for a XYZ" someplace.
And certainly makes for a better experience for the novice.
Right, the novice I tell to go buy a dell. Installing the operating system yourself introduces complications that "the novice" shouldn't have to deal with.
Right now, novices have an easier time installing Ubuntu than they do Windows, so I don't have the foggiest idea what you are possibly complaining about.
You can be as defensive as you want about Linux and going about finding just the right hardware for the operating system.
No, that's exactly what systems integrators do.
When I go buy another 30 servers, I don't purchase whatevers the newest or the cheapest, but the hardware I have found the most reliable and that I am the most comfortable with. The fact that AMD makes both compatible and incompatible hardware for my network is irrelevent. I do some of the duties of a systems integrator to make certain that I'm happy.
If you're going to do some of the duties of a system integrator, I commend you. Comparing your lack of experience doing that with Ubuntu with Toshiba's great experience doing it with Windows is ridiculous.
When I solve problems with hardware I try to solve my problem, by hardware that works for me, that is right for me. My computer choice should be based on my needs not the operating systems limitations.
You already noted that Ubuntu does support your hardware with third party drivers, and are probably aware that Windows does support your hardware with third party drivers, and yet somehow you think the way Ubuntu is doing it is "wrong" by letting you download it using synaptic, versus Microsofts way where they send you to a different manufacturer's site for each piece of hardware.
You won't say it's Windows's fault that broadcomm shipped unsigned drivers and needs to disable driver signing as part of their bluetooth installation, but you'll say it's Linux's fault that Toshiba used the winmodem-equiv of wifi.
Real users aren't like you. They're more than happy to blame Linksys for making it confusing for them to find hardware support information, but people like you confuse them and say it's Linux's fault.
I run Ubuntu Linux on my Toshiba laptop. I had to install two proprietary drivers to get accelerated video and support for the wifi. Ubuntu does not install these by default. If I were a complete novice I'd have no idea that this would solve my problem nor would I know how to do it.
Since Windows doesn't install on any Toshiba currently being sold, I fail to see what your fucking point is.
You are confusing OEM Windows with Windows. Toshiba went and installed those drivers for you when they installed Windows. If they preinstalled Linux you quite obviously, wouldn't have to install those drivers there, either.
Buying a laptop that works with Retail Windows is very difficult. Most require special drivers just to install it.
The next thing that I consider a shortcoming to using Wifi on Linux is that if I don't have the Wifi radio switched on when I boot the machine Linux does not detect this and allow me to network automatically. I still don't know the solution but whatever it is on Windows and Mac OS X this is a simple no brainer operation. It should be completely automatic and transparent.
So don't buy unsupported hardware?
Seriously, if Toshiba made Linux laptops, they'd integrate WIFI hardware that sucked less.
I have been trying Linux off and on for many years and still see areas where if it were "just a little better" I could replace my Windows with it. I'm looking forward to that day.
FWIW, I've been using Linux as my only desktop operating system since 1994, and the trick is to buy hardware that is supported by Linux.
This isn't that difficult, as Linux has far superiour hardware support over every other operating system, but it means that if you're installing your own operating system, you're doing the job of the integrator, so it's your job to make sure the hardware works.
If you don't want to play integrator, don't: Dell is preloading Ubuntu today, and there are quite a few other OEMs that will sell you Ubuntu preloaded- including laptop makers.
But every time you blame this on Linux, to some you're just making yourself look more and more like a idiot, and to people who know less than you, you're just confusing.
why not just run windows 2000 on this thing? I've gotten 2000 to run acceptably on machines with far lower minimum specs than this.
Uh, no you haven't.
Windows 2000 needs things like a hard drive, lots of non-volatile storage, and a BIOS. It also costs more than the target price.
Windows 2000 is also designed to be difficult to use and discover: It doesn't include a development environment, a word processor, any wifi support, or introspection tools.
In contrast: the users of the OLPC are encouraged to extend the system, and write software for it, and to share that software.
Granted it's not as 1337 as Linux, but a lot easier for non IT personel to run...i.e. middle school and K-school teachers
The OLPC can make critical thinkers and sharp engineering minds out of these kids who simply don't have enough engineering challenges in which to learn these things.
Since Windows 2000 - and all other versions of Windows ever lack the ability to engage and challenge its users to make a better system, it certainly cannot answer that call without writing a whole operating system on top of Windows.
I also fail to see how the fact that you might not have to retrain a group of users who aren't even the target audience of the OLPC is a good thing, and since it means giving up all the other things that are good about the OLPC, I can easily see how it is a very bad thing.
1. successful companies do not cause a raise in taxes.
Of course they do. Poorer people have to pay the taxes because successful companies don't. I don't know of any respectable economist that would even think of arguing that. The real question is whether or not it's a bad thing for the largest population to pay the most in taxes, and there is some significant divide in that camp.
2. buying a 360 puts more money in billg's pockets, which puts more clean water in africa. think about it.
But you have to hold back the state of the art and destroy jobs in order to do it.
If you think that the only way to put clean water in Africa is to stop innovation and development of technologies in the west, then buy your Xbox.
It can take longer than 2 years for a script to go from paper to cinema. What you are suggesting means that your copyrighted work (the script) could be out of copyright before the edit is even started.
So what?
The vast majority of Disney's fortune is built on scripts and stories that are in the public domain (or that they just ouright stole). In fact, I can only think of five "original" theatrical works by Disney, and in that there might only be one original idea in there.
Now Disney isn't the only one to make movies from public domain stories- The Little Mermaid and the Jungle Book for example, have been made into movies by lots of people. Nevertheless, Disney's own is still the most recognizable, and the most profitable.
The objection is that Disney is ruining the public domain for everyone. Our country's founders certainly believed it was wrong, and any moral person can see that it must be the same for Disney to use the public domain, so must the public be allowed to use Disney.
Disney has lobbyists. And money. And confused individuals like youself. So videos like this are designed to get people talking about this, so that you'll understand exactly how it is that you are wrong.
A lifetime's work with no means of sustenance
Copyright doesn't make people pay for your work; it doesn't make you a better writer, or grant you greater knowledge. Knuth's TAOCP stand on the sholders of other mathematicians and programmers, and are great not because of copyright but in spite of it: By rewriting the ideas formulated by other authors and collecting them into one place Knuth has done others a great service. If those copyrights had expired and that work was in the public domain, Knuth could've simply spent more time on other things.
3. Finally, MS doesn't make much money on things 360.
Microsoft hasn't made any money on the 360. Microsoft's "gaming division" (that's XBOX) has lost a billion dollars every year according to Forbes, and the reason they can do that is because they obtained an illegal monopoly in another market.
Competing with a monopoly is impossible, which is why we have antitrust laws.
It's not like they're using the money to burn down forests and put lead in grade school water supplies.
But it is like that. Having Microsoft (or a company like it) means that your taxes are about 10 billion dollars a year higher than they would otherwise have to be. That means that your government can either test for lead, or they can raise your taxes.
Buying a 360 makes it easier for Microsoft to continue to do that.
Oh, and X11 isn't that efficient over a network of course!
No, X11 is extremely efficient: X servers need very little memory, and is completely productive on a LAN.
What you probably mean is other remoting protocols are more resistant to latency, which is an entirely different thing. RDP and VNC are almost as usable on a 64K circuit as they are on a LAN, but neither is particularly productive on a LAN. Moreover, they both require gobs more memory in the display than X11.
Given different requirements, I can just as easily (and usefully) state that RDP and VNC aren't that efficient over a network. It doesn't _mean_ anything though.
The only reason for the 'do it in the browser' meme seems to be set up for web this and web that and aren't set up for giving you Windows TS or Unix Shell accounts....
No, it's because "web apps" feel so much faster and more productive than VNC and TSC- which almost everyone is used to. The problem that everyone seems to miss is that "web apps" are written in such a fundamentally different way that they make an advantage out of a rich server and a rich client. Right now, this takes completely reengineering the application, and that's terrible.
However, the author seems to really just want to use Linux, but thinks that "it's just not ready" for him, so he "expertly recommends" that people make it so he can use Linux, without "giving up Windows", but this just aims to keep Microsoft from continuing to become irrelevant, and doesn't actually make anybody more productive.
More than that. I debugged window managers and other XGrab*-using applications _exactly_ this way. It's what happens by default when you use a tiling window manager and run Xnest.
Congratulations Microsoft: You've just now invented the desktop I used in 1992.
And totally illegal as well. All Microsoft EULAs for their free stuff (or stuff which is not sold, like DirectX) forbid installation on non-Windows platforms.
They can forbid it all they want, but Vault v. Quaid, 847 F.2d 255 (5th Cir. 1988) says that it is not illegal. "Shrinkwrap licenses" (also called "EULA") are unconstitutional because your rights cannot be asserted to be taken away by anyone except you, and only by way of a signed contract and a meeting of the minds.
I don't see how this is different from pirating Windows in the first place.
Then you're a fucking idiot. Running software is completely different than distributing copies of someone elses' copyrighted works, and before you said that, I would've thought anyone could see that. Heck, even the US Government can see that [Galoob v. Nintendo, 780 F. Supp 1283 (N.D. Cal. 1991), 22 U.S.P.Q.2d 1587 (9th Cir. 1992), and Foresight v. Pfortmiller, 719 F. Supp 1006 (D. Kan. 1989)], and it's painfully obvious that there's a lot that they miss.
Here's the gist: When one person makes a web page, it is entirely reasonable to assume that I can download that web page, and save a copy on my computer. It is further reasonable to edit it as I would like, and protected under law that I can even distribute my changes (if separate from the original work).
However, I cannot redistribute that web page in whole. Even unchanged.
Copyright protects the redistribution of copies of the work, and nothing more. It doesn't make it "intellectual property", or protect the medium of the work, and it certainly doesn't grant convicted criminals the ability to categorically revoke your rights just because it's on a EULA.
and far better than Slackware's pkgs, which don't manage dependencies at all...
Slackware's packages are truly horrible, but dependency tracking isn't very important.
Maintenance actually gets easier, the more machines you have. If you need to build from ports for some reason, you only have to do it once, and can distribute the generated packages across as many systems as you want.
Of course it should get easier. If maintenance got harder then nobody would use FreeBSD. You're missing the point. If I build the packages from source, then there's exactly _one person_ doing QA for my packages, and that's me. Obviously, if I use someone elses' packages, then I get to share their QA.
If I distribute the generated packages, and I am doing the QA, then I am doing things that almost every other packaging systems do for me. At least slackware users can share the QA to make sure the packages work.
Sharing cpu-time isn't the only think Linux users are sharing.
You're not even a good troll.
And you're the problem with FreeBSD. You keep saying "everything's fine here- no wait- everything is _superior_ here. No problems whatsoever. we're ready for prime time", when FreeBSD has some serious catching up to do.
Fortunately, other FreeBSD developers aren't as stupid as you, and are actually working to fix their flaws.
This sounds like a total hassle. What's wrong with proper package management? (I'm not trying to troll, I'd really like to know!)
The only thing wrong with proper package management is that OpenBSD doesn't have it, so you're going to get lots of touchy-feely responses about how it feels better, or is about some matter of taste to do extra work that someone else has already done.
Fortunately, FreeBSD has something _almost_ as good as Slackware's packaging system (which isn't very) so it shows that at least a few *BSDers recognize that there's a problem, any requests for packages in FreeBSD-land are met with "just compile it yourself", and "disks are cheap", and my all time favorite, "your time isn't worth anything".
Besides, ports and cvsup are "good enough" so long as every machine maintains itself, so people who use a *BSD as their "desktop os" aren't noticing the problems anyway. Two or three machines aren't that bad, and anyone maintaining fifty machines has rolled their own solution long ago.
``That's why IBM and Google are big and profitable. Because they aren't run by you.''
It'll either get you fired or promoted. I wouldn't want to work for that asshole either- no halfway decent manager is ever going to make you waste time and money challenging heresay.
For example, is IBM or HP one day going to submit a patch that brings some sort of DRM into the kernel?
The fact that no one is writing up an analysis for one of the BSDs doesn't mean they do not benefit from the same type of contributions.
No, it means that nobody cares. Work on the *BSDs is charity work.
In many ways corporations are more amenable to BSD.
BSD culture and development wouldn't change in the slightest if the *BSDs were in the public domain because corporations like the attribution and advertising clauses.
I met a sales guy who sold 60,000$US imaging systems on the promise that they weren't a four man shop but instead- "had licensed all this software and intellectual property" blah blah blah- which made them seem a lot bigger than they were.
The real fact is that in the long-term, BSD benefits the dishonest and amoral. If consumers realized this, they might avoid doing business with dishonest companies.
Embedded systems have to support multitasking. It just isn't practical to have one CPU per application in most cases. So all embedded OS's support threads.
Threads aren't the only way to multiprogram. Coroutines and cooperative multitasking are both excellent ways to avoid rapid context switches- one for intra-task communication, and one for inter-task communcation.
You had no issues because a) performance was good enough
If performance is good enough, then you're done. Why would you bother wasting money and time working when the software is done?
b) the rate of incoming client requests was relatively small compared to a loaded webserver.
How do you know this? He didn't say he was talking about webservers, and he didn't say his load was small. I wrote several forking nameservers that handle tens of thousands of requests per second, and threads would only introduce cache contention and only slow it down.
Now if you had thousands of client requests-per-second, the fork will show why-you-should-not-use-it-in-such-situations.
No, it'll mean he can simply buy more computers. fork() means no data is shared; he can run his server on multiple machines and get load balancing for nearly free.
For example, apache gives you the ability to change 'worker' modules...and you can experiment with that to get an idea of all these request processors.
Apache is bad engineering. Nobody said he had to fork when the request came in. He could increase parallelism cheaply by simply forking off a thousand children, and having them all block in accept(). If Apache didn't crap all over the heap, it wouldn't waste megabytes per process.
btw, threads don't complicate your code as long as you minimize data-sharing between threads and write the thread the same way as you were writing a forked process, except that you put all global variables within the thread's local data area. Kashif
Did you ever think that the reasons threads don't complicate your code is that it had already achieved maximum complexity?
Threads don't have local data areas. Several implementations of threading packages have thread-local arenas, but we're not talking about implementations of threading packages, we're talking about good engineering.
Threads are a policy of share-everything. this means more code to audit, more code to check, more places for things to go wrong. Processes are a policy of share-nothing. Share-nothing gives you an awful lot of cool things for free (like scalability).
The startup of fork() is rarely the bottleneck- you bring up Apache which is an excellent example. Apache isn't any faster when threaded. You want a fast webserver? Try and-httpd. It uses fork() and beats the crap out of Apache even when using a threaded mpm. It even outruns IIS. It wins because it's better designed.
Just an FYI, fork isn't always available. The best example I can think of is uClinux, a linux distro for embedded computers. With an MMU, a process cannot be copied, therefore, only threads can be used.
On an embedded system, threads will always decrease your performance. You should be using a different development model.
On Windows, there is a much higher penalty associated with spawning a child process than on Unix. This makes using threads much more attractive - they are faster.
No, it doesn't mean threads are faster, it really just means Window's CreateProcess()+setjmp()+longjmp() slower and uglier than fork().
Windows threads and pthreads are functionally equivalent, and for some metrics in some circumstances, the actual implementations might be better at some tasks versus others. In general, a program written for pthreads ported to windows will be slower on windows for plenty enough other reasons, that it is difficult to say exactly which threaded system is "faster".
I don't know why the Windows equivalent of fork() is slower than the Unix fork().
Because Windows doesn't have an equivalent of fork(). It has a CreateProcessEx() which is like a fork()+exec() -- one of the common uses of fork, but by no means the only use (nor the one parent is talking about).
... perhaps Unix fork() is efficient because it is frequently used and has therefore been optimised in various ways (e.g. memory is only copied if there is a write on Linux). Whereas on Windows, those optimisations are not necessary.
No, it's that Windows developers think differently: Unix people think the Windows way of thinking is wrong. Windows developers think the Unix way of thinking is wrong (or "unnecessary").
fork() means data isn't shared. This means that you don't need complicated locking structures, semaphores, etc. This also means that you know what channels need auditing for security, and which channels need extending to increase parallelism (scale).
No matter how careful you are, threads never make auditing easier and they never make an application scale better (well, not significantly better).
This isn't to say threads don't solve some problems, it is to say that they don't solve the same problems as fork(), and I hope at this point it goes without saying that threads are not a "more efficient fork()".
The difference is, and its a massive difference is that Google can deny you access to your data at any time they choose. With MS you still have physical access to your files.
The big problem with your argument is that Google has a great track record of trying to make it easy for me to work with the data that they save for me: They give me pop3 and smtp access to my mailbox, and let me mirror my docs with relative ease.
Microsoft on the other hand has used every opportunity they can to make using my data as difficult and financially burdensome as possible.
Those free converters work so long as Microsoft doesn't change their formats and only work well enough "most of the time"- not well enough that anyone sane can rely on them.
WiX is fully open source, so if you wanted, you should be able to figure out how to create MSIs from that. In any case, you could try running it under Mono, maybe it'll work under Linux...
You should, but you can't. That's because WiX uses DLL-calls into a non-open-source component to make the MSI.
Most of those EXEs are actually MSIs wrapped with a stub installer that can install the Windows Installer if it's missing.
Except in the case of the examples I gave, the software doesn't run on machines that didn't come shipped with the Windows Installer, so exactly why would Microsoft do that?
Well..... yes. It is. Sort of. It allows arbitrary binary code to be run - but then again, it's an installer. It could run arbitrary code anyway. Complaining that MSI is "another executable format" would be like complaining that RPM is "another executable format" - it's equally true. (Although I think RPM generally runs shell scripts, not direct binary code. Although I could be wrong.)
No it wouldn't be equally true. I can install RPMs into a system besides the one that I am running. I can run them on a system that isn't fully installed yet, or I can install them into a system over the network. No, MSI isn't even remotely as capable as RPM, and it only goes to further demonstrate that it isn't just those singing the praises of MSI that don't understand the problems.
Right, nobody does that because it's practically impossible to get it working. Instead they discover that it's not in the apt repository and give up. That's the problem with the Linux packaging system - if it's not there already, it's basically impossible to install unless you're a Linux expert.
In your mind, how exactly does this differ from the Windows world? If it hasn't been ported to Windows XP SP 2 yet, or patched or whatnot, how exactly is a non-expert to get it to work?
Worse, from the developer point of view, they might create some great new software that they want to release in an easy to install form - and, really, can't. Maybe they'll generate an RPM or a DEB for their own installed Linux distro, but beyond that, it's just too much effort to try and make an installer for everyone.
People don't make "their own" RPMs for distribution, they make RPMs so that they can track the package in the rpmdb. This makes things easier from the developer's point of view because the developer doesn't have to worry about distribution or collision tracking and dependent upgrades.
Oh, and MSI is horrendously overcomplicated compared to RPM and DEB, in case you wanted to know. Nothing like having to generate a GUID for every single file you might want to install. (Although to be fair, that's likely because most Windows programs don't contain all too many files.)
MSI doesn't solve any problems. Microsoft introduced it because every Linux show they go to people rave about how awesome APT and YUM and that there simply aren't any version maintenance nightmares. Microsoft doesn't realize that the power of APT and YUM is in the distributions that use them, and has nothing to do with third-parties actually distributing software.
As you've noticed, software developers can't reasonably target each distribution in each distribution's native format. The few that try (Opera) end up making things very difficult by comparison for Linux users. They don't really understand the issues either.
Because the debian group is spending their efforts minimizing conflicts and recording overlap, I can merge in and merge out software on any debian system- even when the system isn't running, or do it over a network. Its painless- and it doesn't really matter what other software might be wanted/needed on that system. Installing flash is downright painful because Adobe insists on doing all the distribution themselves. It causes a degree of discomfort that Windo
Which is why, as it currently stands, this year will not be Year Of The Linux Desktop. Consumers won't just accept that they can't install software X because it's an RPM and alien doesn't work
Now my daughter just received a "game" on Windows- brand new (2007) game that insisted on running in some "compatability" mode in Windows, and in a resolution that her LCD display couldn't cope with. The fact is that Windows users have run into this problem attempting to install software that isn't for their particular operating system, and failed on the Internet for a few hours. They just assume that Linux users have run into the same problem.
They don't. Linux users install software out of their software catalog. Occasionally the brave ones go to the author's website, and download the software from there.
This is also one reason people are willing to pay for an operating system that has a standardized and dependable way of doing things.
Bzzt. Wrong. Nobody is willing to pay for Windows, that's why Microsoft doesn't let OEM's give you a choice. Duh, I'll use the Windows I already bought. And don't spread that Lie about how I don't have a License to.
Microsoft even released the WiX toolkit that allows anyone to create MSI installer packages.
But not the MSI format specification. That would allow me to cross-compiler into an installable package. As it stands, my users who run Windows have to deal with no installer.
MSIs are one of the best ideas for Windows in a while... No more dealing with poorly-written homebrew installers or 10-year old, 16-bit InstallShield programs.
You're wrong, and you want proof? Look how many programs- nay, look how many programs come from Microsoft that are still distributed as exe files. That shiny new Zune's software comes in exe-form.
Once that 16-bit installshield program was written, it's forever supported. You can't put the setup.exe genie back in the bottle, and you have to live with that. With Free Software, we can take our software library with us, which is why Free Software always gets better, and non-Free software atrophies.
Instead you have a fully scriptable installer that's transaction-based and has near 100% support coverage.
You are wrong on all counts. Pull the power plug while installing and you'll see just how transactional it is. I don't even think you know what coverage means: Microsoft Support will tell you to reinstall your operating system if a broken/corrupt/poorly-written MSI breaks your system. Even if they make it.
I like apt, but downloading a gzipped file of source or a deb that complains about dependencies still can't compare to an MSI package.
No of course not, but that's why you used a straw man. MSI is an executable, and just made Microsoft's security problem worse: it introduced yet another executable file format. Nobody downloads "gzipped file of source or a deb that complains about dependencies" ever. They say "apt-get install xyz" and it goes and figures out the dependancies itself.
It doesn't have to- Linux users could waste disk space by including the dependencies with every program- and some Linux distributions even do this(!), but it makes upgrades very difficult. For example, when libz had a vulnerability discovered, only one copy needed to be upgraded on most Linux systems. On Windows, almost every program that dealt with gzip or deflate-compressed data (like png or zipfiles) needed to be upgraded. Worse still, that library or program can be anywhere on your hard drive, and you might never know it.
and there are plenty of things that I want to do and buy that I can't (or don't) - it's all about defining what an acceptable standard of living is.
That's a load of crap. You don't get to define what an acceptable standard of living is when you're living above what you consider it to be.
The man who works three jobs knows the ``acceptable standard of living'' is above where he presently lives. If he could physically work four, he would.
Someone made up the poverty limit, someone else made up the minimum wage. They will always be low (hence "poverty" and "minimum"), but there is a fine line between making sure people are alive, and living their lives on behalf of them (I live in a "high-welfare, high-tax society". Unemployment is ridiculous, but not in official numbers, since they are not counting welfare-receipients.)
Is the job of an economy to make sure their working class is alive? Who cares if you get three hours of sleep, walk 17 miles a day between jobs, and vommit blood. At least you're alive.
The rich are rich because there are poor, not despite them.
Slashdot Mirror
Fact is, iRiver doesn't care about me, and they don't care about you, and furthermore, I think it's positively the stupidest thing I ever heard to buy their hardware in spite of that.No, iRiver works great with XNJB which uses libmtp- written by and for Linux users. If the original poster had installed an LIBMTP-based application, they would be able to use their iRiver but because they said this is a Linux problem and because you obviously agree, you are not allowed to install XNJB to reverse my claim: iRiver does not work with Mac OS X because you have to install third party (although Free) software to give Mac OS X support.
There is no reason to use Windows anymore: The fact that "iRiver" doesn't have particularly good drivers on Linux is a good reason to not use iRiver and not a good reason to use Windows.There are no NT or Vista drivers for some STB-based cards. People who bought them cannot use them unless they use Linux. ATI won't make drivers for them because they're old, and Microsoft won't let you make drivers for them because if you could make drivers for them, you might also be tempted to circumvent Digital Restrictions Management software.
Microsoft is completely irrelevant anymore: Free software means that companies have to make their customers happy instead of just making sure that nobody else can make them happier. Since I'm not using them in the workplace, I don't have to use them at home.
How about this: if iRiver doesn't work in Linux, complain to iRiver.
With Linux, you could also fix the problem yourself. You could also pay someone to fix the problem. If the iRiver is popular enough, you could also wait and someone else will fix it for you.
With Windows, you don't have those options, so I consider that a weak point in the Windows desktop.
Right now, novices have an easier time installing Ubuntu than they do Windows, so I don't have the foggiest idea what you are possibly complaining about.No, that's exactly what systems integrators do.
When I go buy another 30 servers, I don't purchase whatevers the newest or the cheapest, but the hardware I have found the most reliable and that I am the most comfortable with. The fact that AMD makes both compatible and incompatible hardware for my network is irrelevent. I do some of the duties of a systems integrator to make certain that I'm happy.
If you're going to do some of the duties of a system integrator, I commend you. Comparing your lack of experience doing that with Ubuntu with Toshiba's great experience doing it with Windows is ridiculous.You already noted that Ubuntu does support your hardware with third party drivers, and are probably aware that Windows does support your hardware with third party drivers, and yet somehow you think the way Ubuntu is doing it is "wrong" by letting you download it using synaptic, versus Microsofts way where they send you to a different manufacturer's site for each piece of hardware.
You won't say it's Windows's fault that broadcomm shipped unsigned drivers and needs to disable driver signing as part of their bluetooth installation, but you'll say it's Linux's fault that Toshiba used the winmodem-equiv of wifi.
Real users aren't like you. They're more than happy to blame Linksys for making it confusing for them to find hardware support information, but people like you confuse them and say it's Linux's fault.
You are confusing OEM Windows with Windows. Toshiba went and installed those drivers for you when they installed Windows. If they preinstalled Linux you quite obviously, wouldn't have to install those drivers there, either.
Buying a laptop that works with Retail Windows is very difficult. Most require special drivers just to install it.So don't buy unsupported hardware?
Seriously, if Toshiba made Linux laptops, they'd integrate WIFI hardware that sucked less.FWIW, I've been using Linux as my only desktop operating system since 1994, and the trick is to buy hardware that is supported by Linux.
This isn't that difficult, as Linux has far superiour hardware support over every other operating system, but it means that if you're installing your own operating system, you're doing the job of the integrator, so it's your job to make sure the hardware works.
If you don't want to play integrator, don't: Dell is preloading Ubuntu today, and there are quite a few other OEMs that will sell you Ubuntu preloaded- including laptop makers.
But every time you blame this on Linux, to some you're just making yourself look more and more like a idiot, and to people who know less than you, you're just confusing.
Windows 2000 needs things like a hard drive, lots of non-volatile storage, and a BIOS. It also costs more than the target price.
Windows 2000 is also designed to be difficult to use and discover: It doesn't include a development environment, a word processor, any wifi support, or introspection tools.
In contrast: the users of the OLPC are encouraged to extend the system, and write software for it, and to share that software.The OLPC can make critical thinkers and sharp engineering minds out of these kids who simply don't have enough engineering challenges in which to learn these things.
Since Windows 2000 - and all other versions of Windows ever lack the ability to engage and challenge its users to make a better system, it certainly cannot answer that call without writing a whole operating system on top of Windows.
I also fail to see how the fact that you might not have to retrain a group of users who aren't even the target audience of the OLPC is a good thing, and since it means giving up all the other things that are good about the OLPC, I can easily see how it is a very bad thing.
If you think that the only way to put clean water in Africa is to stop innovation and development of technologies in the west, then buy your Xbox.
The vast majority of Disney's fortune is built on scripts and stories that are in the public domain (or that they just ouright stole). In fact, I can only think of five "original" theatrical works by Disney, and in that there might only be one original idea in there.
Now Disney isn't the only one to make movies from public domain stories- The Little Mermaid and the Jungle Book for example, have been made into movies by lots of people. Nevertheless, Disney's own is still the most recognizable, and the most profitable.
The objection is that Disney is ruining the public domain for everyone. Our country's founders certainly believed it was wrong, and any moral person can see that it must be the same for Disney to use the public domain, so must the public be allowed to use Disney.
Disney has lobbyists. And money. And confused individuals like youself. So videos like this are designed to get people talking about this, so that you'll understand exactly how it is that you are wrong.Copyright doesn't make people pay for your work; it doesn't make you a better writer, or grant you greater knowledge. Knuth's TAOCP stand on the sholders of other mathematicians and programmers, and are great not because of copyright but in spite of it: By rewriting the ideas formulated by other authors and collecting them into one place Knuth has done others a great service. If those copyrights had expired and that work was in the public domain, Knuth could've simply spent more time on other things.
Competing with a monopoly is impossible, which is why we have antitrust laws.But it is like that. Having Microsoft (or a company like it) means that your taxes are about 10 billion dollars a year higher than they would otherwise have to be. That means that your government can either test for lead, or they can raise your taxes.
Buying a 360 makes it easier for Microsoft to continue to do that.
What you probably mean is other remoting protocols are more resistant to latency, which is an entirely different thing. RDP and VNC are almost as usable on a 64K circuit as they are on a LAN, but neither is particularly productive on a LAN. Moreover, they both require gobs more memory in the display than X11.
Given different requirements, I can just as easily (and usefully) state that RDP and VNC aren't that efficient over a network. It doesn't _mean_ anything though.No, it's because "web apps" feel so much faster and more productive than VNC and TSC- which almost everyone is used to. The problem that everyone seems to miss is that "web apps" are written in such a fundamentally different way that they make an advantage out of a rich server and a rich client. Right now, this takes completely reengineering the application, and that's terrible.
However, the author seems to really just want to use Linux, but thinks that "it's just not ready" for him, so he "expertly recommends" that people make it so he can use Linux, without "giving up Windows", but this just aims to keep Microsoft from continuing to become irrelevant, and doesn't actually make anybody more productive.
More than that. I debugged window managers and other XGrab*-using applications _exactly_ this way. It's what happens by default when you use a tiling window manager and run Xnest.
Congratulations Microsoft: You've just now invented the desktop I used in 1992.
Here's the gist: When one person makes a web page, it is entirely reasonable to assume that I can download that web page, and save a copy on my computer. It is further reasonable to edit it as I would like, and protected under law that I can even distribute my changes (if separate from the original work).
However, I cannot redistribute that web page in whole. Even unchanged.
Copyright protects the redistribution of copies of the work, and nothing more. It doesn't make it "intellectual property", or protect the medium of the work, and it certainly doesn't grant convicted criminals the ability to categorically revoke your rights just because it's on a EULA.
If I distribute the generated packages, and I am doing the QA, then I am doing things that almost every other packaging systems do for me. At least slackware users can share the QA to make sure the packages work.
Sharing cpu-time isn't the only think Linux users are sharing.And you're the problem with FreeBSD. You keep saying "everything's fine here- no wait- everything is _superior_ here. No problems whatsoever. we're ready for prime time", when FreeBSD has some serious catching up to do.
Fortunately, other FreeBSD developers aren't as stupid as you, and are actually working to fix their flaws.
Fortunately, FreeBSD has something _almost_ as good as Slackware's packaging system (which isn't very) so it shows that at least a few *BSDers recognize that there's a problem, any requests for packages in FreeBSD-land are met with "just compile it yourself", and "disks are cheap", and my all time favorite, "your time isn't worth anything".
Besides, ports and cvsup are "good enough" so long as every machine maintains itself, so people who use a *BSD as their "desktop os" aren't noticing the problems anyway. Two or three machines aren't that bad, and anyone maintaining fifty machines has rolled their own solution long ago.
Just say:
``That's why IBM and Google are big and profitable. Because they aren't run by you.''
It'll either get you fired or promoted. I wouldn't want to work for that asshole either- no halfway decent manager is ever going to make you waste time and money challenging heresay.
I met a sales guy who sold 60,000$US imaging systems on the promise that they weren't a four man shop but instead- "had licensed all this software and intellectual property" blah blah blah- which made them seem a lot bigger than they were.
The real fact is that in the long-term, BSD benefits the dishonest and amoral. If consumers realized this, they might avoid doing business with dishonest companies.
You really should read Knuth.
Threads don't have local data areas. Several implementations of threading packages have thread-local arenas, but we're not talking about implementations of threading packages, we're talking about good engineering.
Threads are a policy of share-everything. this means more code to audit, more code to check, more places for things to go wrong. Processes are a policy of share-nothing. Share-nothing gives you an awful lot of cool things for free (like scalability).
The startup of fork() is rarely the bottleneck- you bring up Apache which is an excellent example. Apache isn't any faster when threaded. You want a fast webserver? Try and-httpd. It uses fork() and beats the crap out of Apache even when using a threaded mpm. It even outruns IIS. It wins because it's better designed.
Windows threads and pthreads are functionally equivalent, and for some metrics in some circumstances, the actual implementations might be better at some tasks versus others. In general, a program written for pthreads ported to windows will be slower on windows for plenty enough other reasons, that it is difficult to say exactly which threaded system is "faster".Because Windows doesn't have an equivalent of fork(). It has a CreateProcessEx() which is like a fork()+exec() -- one of the common uses of fork, but by no means the only use (nor the one parent is talking about).No, it's that Windows developers think differently: Unix people think the Windows way of thinking is wrong. Windows developers think the Unix way of thinking is wrong (or "unnecessary").
fork() means data isn't shared. This means that you don't need complicated locking structures, semaphores, etc. This also means that you know what channels need auditing for security, and which channels need extending to increase parallelism (scale).
No matter how careful you are, threads never make auditing easier and they never make an application scale better (well, not significantly better).
This isn't to say threads don't solve some problems, it is to say that they don't solve the same problems as fork(), and I hope at this point it goes without saying that threads are not a "more efficient fork()".
Microsoft on the other hand has used every opportunity they can to make using my data as difficult and financially burdensome as possible.
Those free converters work so long as Microsoft doesn't change their formats and only work well enough "most of the time"- not well enough that anyone sane can rely on them.
I'm having fun playing games that I expected (Warioware, Excitetruck), and having fun playing games that I didn't expect to enjoy (Madden).
So for the last time, No: I'm not trading you for your PS3.
You should, but you can't. That's because WiX uses DLL-calls into a non-open-source component to make the MSI.
Except in the case of the examples I gave, the software doesn't run on machines that didn't come shipped with the Windows Installer, so exactly why would Microsoft do that?
No it wouldn't be equally true. I can install RPMs into a system besides the one that I am running. I can run them on a system that isn't fully installed yet, or I can install them into a system over the network. No, MSI isn't even remotely as capable as RPM, and it only goes to further demonstrate that it isn't just those singing the praises of MSI that don't understand the problems.
In your mind, how exactly does this differ from the Windows world? If it hasn't been ported to Windows XP SP 2 yet, or patched or whatnot, how exactly is a non-expert to get it to work?
People don't make "their own" RPMs for distribution, they make RPMs so that they can track the package in the rpmdb. This makes things easier from the developer's point of view because the developer doesn't have to worry about distribution or collision tracking and dependent upgrades.
MSI doesn't solve any problems. Microsoft introduced it because every Linux show they go to people rave about how awesome APT and YUM and that there simply aren't any version maintenance nightmares. Microsoft doesn't realize that the power of APT and YUM is in the distributions that use them, and has nothing to do with third-parties actually distributing software.
As you've noticed, software developers can't reasonably target each distribution in each distribution's native format. The few that try (Opera) end up making things very difficult by comparison for Linux users. They don't really understand the issues either.
Because the debian group is spending their efforts minimizing conflicts and recording overlap, I can merge in and merge out software on any debian system- even when the system isn't running, or do it over a network. Its painless- and it doesn't really matter what other software might be wanted/needed on that system. Installing flash is downright painful because Adobe insists on doing all the distribution themselves. It causes a degree of discomfort that Windo
They don't. Linux users install software out of their software catalog. Occasionally the brave ones go to the author's website, and download the software from there.Bzzt. Wrong. Nobody is willing to pay for Windows, that's why Microsoft doesn't let OEM's give you a choice. Duh, I'll use the Windows I already bought. And don't spread that Lie about how I don't have a License to.But not the MSI format specification. That would allow me to cross-compiler into an installable package. As it stands, my users who run Windows have to deal with no installer.You're wrong, and you want proof? Look how many programs- nay, look how many programs come from Microsoft that are still distributed as exe files. That shiny new Zune's software comes in exe-form.
Once that 16-bit installshield program was written, it's forever supported. You can't put the setup.exe genie back in the bottle, and you have to live with that. With Free Software, we can take our software library with us, which is why Free Software always gets better, and non-Free software atrophies.You are wrong on all counts. Pull the power plug while installing and you'll see just how transactional it is. I don't even think you know what coverage means: Microsoft Support will tell you to reinstall your operating system if a broken/corrupt/poorly-written MSI breaks your system. Even if they make it.No of course not, but that's why you used a straw man. MSI is an executable, and just made Microsoft's security problem worse: it introduced yet another executable file format. Nobody downloads "gzipped file of source or a deb that complains about dependencies" ever. They say "apt-get install xyz" and it goes and figures out the dependancies itself.
It doesn't have to- Linux users could waste disk space by including the dependencies with every program- and some Linux distributions even do this(!), but it makes upgrades very difficult. For example, when libz had a vulnerability discovered, only one copy needed to be upgraded on most Linux systems. On Windows, almost every program that dealt with gzip or deflate-compressed data (like png or zipfiles) needed to be upgraded. Worse still, that library or program can be anywhere on your hard drive, and you might never know it.
The man who works three jobs knows the ``acceptable standard of living'' is above where he presently lives. If he could physically work four, he would.Is the job of an economy to make sure their working class is alive? Who cares if you get three hours of sleep, walk 17 miles a day between jobs, and vommit blood. At least you're alive.
The rich are rich because there are poor, not despite them.