I would place it along with the Linux kernel, Mozilla, Samba, and Apache. All of these have a vast number of supporters and end-users. They are the crown jewels of Open Source and get a good bit of ink from the media.
I've seen quite a few comparisons to Mozilla. It should be remembered that at the start of the Mozilla project, virtually all the developers were Netscape employed. Gradudally the code was properly licensed, and the code became easier to work, and everybody got into the swing of it. Nowadays it has lots of developers. I'm sure OOo will do the same thing.
I'd love to see the open source langauges get together and start sharing class libraries. I'm secretly hoping that Parrot will make this possible, but how do you consolidate all the Python/Perl/PHP/Ruby code libraries together?
One of the best things about.NET is that there is 1 class library, so you don't get people constantly reinventing the wheel in language X. I think it's actually pretty ironic that despite virtually all the code in Linux being open, we still lack a good way of sharing that code between languages other than binding it to C and then to other languages manually (yuck!).
You mean you don't know anybody owning 500 CDs ?
I certainly do !
Wierd but true, I've never met anybody with that number of (legal) CDs. You learn something every day. Next question though - do you listen to them all? I don't have anywhere near that number of CDs, and still I have favourites and ones I haven't listened to for ages and don't really care about anymore.
DRM will happen. Deal with it, Michael. What other solution would you offer to deal with the rampant piracy and IP theft that escalates every single day?
I want something that works good with Linux/OpenBSD.
Yes, it's expensive. No there isn't Linux support.
Sounds like just what he wants;) But seriously, a few points to remember if you're planning on getting an iPod:
It requires FireWire. Most PCs don't have this, so you'll need to buy a FireWire card. This will add about $35 for a cheap one to the price of the player itself. The speed increase may be worth it to you though.
You might get burnt. They've probably finished for now, but a friend of mine was given an iPod for his birthday and 3 weeks later it had been obsoleted by a new model, and the one he got had the prices slashed. On the other hand, this is the same for any portable players.
You can't change the headphones. At least on the model I tried (please correct me if that's changed now) you had to use the Apple in ear headphones, you couldn't just jack in your own. Not sure why this was the case, so I may be wrong here....
And last but by no means least, Apple are using them to leverage Mac sales. They are absolutely not interested at all in Linux or OpenBSD - and no, MacOS is not OpenBSD. The support for it is provided by third parties from which you must buy a plugin for a music player that is hardly used (xtunes). Do you really want to buy a product from a company that is absolutely not interested in supporting you at all? Remember the Linux software has been built basically through reverse engineering.
Finally, just out of interest, how many people here actually own 20gig of legal music? That would be 31 CDs of uncompressed digital audio, let alone MP3s. Really, I imagine that the only people who need that sort of capacity are those who have built up huge collections of music off the P2P networks.
See, this is why it's such a bloody good thing that Apple moved over to Open Source. Instead of being a bunch of weirdos with proprietary everything, the fortunes of a large constituency are now tied in with the fortunes of free software. Unlike the masses of clueless Windows users, the masses of clueless Mac users will be affected, will be restricted.
This would be the case if MacOS was open source. Let me remind you however that it is only the UNIX core that is open. MacOS is defined if anything by its GUI. So now Hollywood come banging on Apples door - hey Jobs! Stop your Mac users from copying our stuff.
"Certainly", he would say, "we'll just build it into Quicktime, and stop things at the user interface level".
The core is open, but nobody buys a Mac for its kernel. They buy it for the GUI. If the option of doing "illegal" things is removed above the kernel level, they're still stuffed.
Re:Perhaps they got something right!
on
Linuxworld Fun
·
· Score: 2
I know this is going straight against the mantra on slashdot "choice-is-good", but normal users have no base on what to make a "choice", and there inflexibility is good:
I have to disagree with this. I feel the reason some users are confused by choice is simple - they are not used to it.
Computers are complex and powerful devices it is true, but then so are cars, hi-fis and so on. People seem to cope with choice and flexibility in these markets just fine, I know some people who actually enjoy browsing around car shops going "ooh" and "aah" at all the different makes and models.
Computers are different of course, because there has never been any competition in computing, thanks to Microsoft. As far as the public is concerned, MS is the way, and has been from day 1, ever since the PC arrived. Yes, there was the Mac, but there's a reason Apple lost the platform war so miserably: given a choice between a completely single vendor solution and a semi-open solution, they chose the semi open one. And so we have Microsoft.
It annoys me when people say, or imply, that Linux cannot succeed because people don't want choice, or are too stupid to deal with it. This is doing a disservice to both Linux and computer users. There are users who are confused by choice, but they can and will adjust. The total lack of training many people have doesn't help here of course, far too many people have learnt computing by trial and error and the attitude that "The user must not have to make any effort to learn this thing" is pervasive in computing.
To some extent this is good, because it means we can get up to speed with new software quickly, but it can also backfire in the case of people who never really understood what they were doing in the first place and so can't generalise their learning to something else.
Some users are only inflexible because they are scared by choice, because they have never had it before and don't understand it. They are scared by it, because they know they never really learnt how to work their computer in the first place, only that if you click here, then there, and then on this picture you can access the web. For instance, I once asked somebody to visit a website I'd made and tell me what they thought. What they saw was all messed up, so I asked what browser they were using (I was pretty sure it was ie5 but wanted to check). The answer: "Sorry, I'm not technical, what's a browser?" shocked me. How can anybody use something yet not know what it's called? Simple - she hadn't been taught how to use a computer, she had learnt by rote.
If somebody doesn't want to switch to something better than Outlook, sit down with them and teach them how to use Outlook - and then teach them how to use MozMail or whatever. And if they don't want to know.... well, next time they get a virus, let them fix it themselves. I've done this, and it works wonders.
We tried this in England. Back in the mid nineties, when digital TV was being planned out, the plan was to have 1 decoder box with plugin modules in the back for Sky/Cable/Digital Terrestrial decoder systems. By sharing the boxes, economies of scale would be reached, the hardware would be cheaper and so uptake would be faster.
That was when it was just the engineers. Then the lawyers got involved. Oh yes. The lovely EU Competition Commision discovered this bunch of engineers from all the major players working together and decided that wouldn't do, and split them up. They forced ITV and the cable companies to eject Sky, and pay Murdoch a few hundred million pounds in compensation. From that point on things just went downhill. The idea of the universal box was killed the moment the managers, lawyers and marketroids got hold of it.
The humble cable box, for years a mere channel-flipper, is in for a multimedia makeover: Beefed-up boxes of the future could let you play video games online, share digital photos with friends, and maybe do other things people in the business haven't even imagined yet.
They need to come over to my house and play with Sky Digital if this is what they think. Playing games on your TV? Been there, done that, it doesn't work needless to say. For one, the games take forever to load - even with the gigabits of bandwidth they have coming off of Astra it can take several minutes to load games - it's like being back with a Commodore 64. Then, when the games do load, they are extremely primitive and loaded with advertising. Realtime games are out as the latency involved from the handset is huge - I tried a simple top down racing game one time, it was almost unplayable as the car responded almost a second after I hit the button. Finally, there aren't many of them, as interactive TV applications are far more expensive to produce than computer apps. Interactive TV is basically dead as a from of entertainment. Where it does shine is in getting information - BBC News Interactive and Sky News Active are great. It also does simple interactive additions to programs quite well. The multiple football camera angles are rare however due to the large amount of bandwidth required.
Believe me, interactive TV on a universal cable box? It's a TV engineers dream but in the real world, we've done it, and the PC kicks its ass in almost every respect. It's a big (expensive) white elephant.
Apparently Sun will be revealing its first general-purpose, low-end Linux machine, and its own version of Linux
Huh? Another distro? What could Sun's edge over the others be is what I'm wondering. Companies like SuSE/Mandrake already have the low end, RedHat does the middle and Solaris has the high end. So is there going to be a high end Linux?
Oh! Here's an idea. Perhaps McNealy sees this as a chance to return to the Network Computer idea (not a bad one in itself) by having thin client machines running XFree. Suns Linux could be like the LTSP distro.
1) NO GOOD SOUND API's. Cups is moving in the right direction, but nothing matches ASIO on windows or mac for sample-accurate synchronization across multiple input and output devices
I think you are confused. CUPS is a printer API, it has nothing to do with sound. In fact, there is no need for an ASIO equivalent as Linux with the kernel pre-empt and low latency patches has far lower latency normally than Windows even with ASIO. I don't know if they are sample accurate, but I know you can get extremely low audio latencies with this sort of setup.
2) Inconsistent platform. We do not have a uniform development platform. Each distro ships a different version of GCC with its own inconsistencies, different GLIBC, same issue, and so on down to GTK or QT, XFree86, etc. With a product like Cubase taking at least 6 months to port, no doubt the platform will keep changing underneath them. What solution would they have other than supporting ONE distro only or shipping their own?
I don't understand you. Currently Linux is going through a switch of the C++ ABI. If your audio app is written in C++ then you will need to provide 2 binaries for the duration of the switchover. They do not require porting, just a recompile. The widget toolkits only break compatability every few years, and you can always use compatability libs as they can be installed side by side. Moonlight for instance uses its own version of Qt 2.2. X has been backwards compatible for over 10 years. They can do what all the other Linux companies do, and write their software to be distro neutral. It's not hard, the biggest challenge is the installer.
3) Lack of hardware support. Yes, vendors like my personal favorite, Echo, aren't releasing enough specs to the community for a free driver. But on the other hand, the community isn't providing them a stable platform to develop on, see #2.
NVidia manage it. There are even drivers now for WinModems (which use proprietary drivers). It's more work to support all kernel versions, and I agree that Linus should stop breaking compatability, however it is perfectly possible to write drivers that will compile on any kernel version. If Echo don't release the specs, or write their own drivers then yes you are locked in to Windows. Congratulations.
4) Lack of unchanging commercial plugin standards. VST works. The API doesn't have a million tiny revisions, and any VST plugin works great on anything from Logic to Fruity Loops to the latest Cubase SX. The API was published WHEN IT WAS FINISHED, and NOT CHANGED.
VST plugins are the standard because of the popularity of Cubase. This is hardly the fault of Linux - if one sequencer gets more popular than the rest then I'm sure we'll see something similar. How often the plugin API changes is entirely up to the developer: remember that APIs change more frequently on Linux because most stuff is open source, so changing the API to get a better system is less painful than on a closed source system.
No matter how they do it, we will have people trying this CD and saying: "Linux boots slower than the OS on my HD, Linux requires more ram than the OS on my HD, and Linux generally perform poor compared to the OS on my HD."
The SuSE Live CD tells the user about slow bootup when starting. Also, it creates a couple of files on c:\ (they can be deleted afterwards) that contains your home directory and a swap file, so they don't have to worry about it using up all the RAM and they can even save documents with it! SuSE definately has this done very slickly indeed.
You pretty much write HTMl pages the same way you would write a standard GUI application and.NET takes care of the rest. Its a great way to program HTML UIs. And I think that was part of his point.
But surely you get a lot less flexibility? I seem to recall MS tried to rewrite Outlook in DHTML, the project bombed, as HTML (even with loads of IE extensions) just wasn't suitable for user interfaces. That's why Mozilla created XUL (which is). How does it deal with things like snap-off toolbars, popup menus, zigzagged underlines and all the other stuff that's either impossible or extremely hard to do with HTML?
It sounds like you'd end up with an extremely limited and rather slow widget set - a bit like AWT in fact. On the other hand, I haven't tried it, so what do I know?
Uh oh.... I just bought an nVidia card for my PC because I thought it'd work well with Linux - and I have an AMD Duron. Could you please tell me more about this problem?
Bottom line: ITV Digital collapsed because some suits wanted to rid themselves of a less than profitable contract that they and clubs both signed in good faith.
There was also:
Possibly illegal anti-competitive behaviour from Sky (the government are looking into this)
Large amounts of government regulation. ITV Digital was very expensive to run, as ITV had to do regional and sub-regional programming. This brought what should have been 1 channel to more like 40, each of which needed their own encoding infrastructure. ITV was also regulated by the ITV and had to meet targets for closed captioning etc, which Sky didn't.
The box wars. This was the killer - Sky has almost bottomless pockets but can afford it as it's riding on the back of News Corp. Sky has lost far more money than ITVD did, but ITV couldn't afford it, and Sky could.
Management stupidity/incompetence - why rename it from ONdigital????
Limited bandwidth: Sky has gigabits, ITVD had very little in comparison
Spineless politicians, who wouldn't allow them to boost transmitter power (it would inconvenience the larger base of analogue viewers). This meant that a lot of people couldn't receive it, or like us, were on the edge of the broadcast zone and got huge amounts of corruption.
In short there were many reasons why ITV Digital fell, most of which were political rather than technical.
Further, It looks like the Ruby folk are more interested in cooperating than the Python folk. But in my mind it would be amazing if Python, Perl6, Ruby, and some Lisp variations all had Parrot backends.
I was just researching Parrot when I came across this story. The problem with Python isn't so much that they're not interested, it's more lack of volunteers. All the "core team" (only about 4/5 people) is too busy maintaining the language. Python should be fairly easy to retarget to Parrot though.
I do think this would be cool. We need a truly open cross-language VM. Mono would be OK, but unfortunately there are just way too many people (including me) who just don't trust Microsoft with.NET
What especially interests me is the possibility that Parrot binaries could share objects and such - this would for the first time allow Perl to use Python code and vice-versa (as well as any other languages targetted to Parrot). The merge between Python and Perl was an April Fools joke, but I'm wondering - would it not make sense to allow code to be easily shared between them and to construct a.NET style "shared" set of APIs? The amount of code duplication is staggering with these two languages, if it could be cut down that'd be fantastic.
I'm sorry, but there's a reason why there are multiple Mozdev projects to build browsers without Mozilla's cumbersome interface, why Dave Hyatt [mozillazine.org] and mpt [phrasewise.com] have savaged the current interface.
Here is his list of usability problems with Mozilla
From what I recall, the main criticisms of MPT boiled down to "I don't like it". For instance, he makes a big deal of the fact that the Home link is on the Bookmarks toolbar, rather than the main toolbar. This immediately leads of course to flamewars between people who believe it "belongs with the reload button" or people who thinks it makes more sense to have it with your other links. This is hardly a usability issue (remember neither Hyatt or MPT have any usability training at all - no disrespect to them, but it's true). It's just personal preference.
He talks about speed as well - that's hardly as much of a problem as it was. Especially on Windows, Mozilla feels just as snappy as IE (no, really, and I have a PIII/500).
Text editing bugs : these are bugs, not usability problems.
Message Display: he doesn't like the fact that headers are in their own section. Personally I don't mind this at all, but clearly he feels otherwise.
The list goes on and on. Some of his points are good. Many are simply pet peeves on his part. This is often the problem with "usability", it's a very vague concept and the science of usability is still in its infancy. Therefore a "usability" review often degenerates into a case of the UI reviewer picking on things they don't like. For instance, the "I don't think this feature is useful, so it's preferences bloat". There is a grain of truth to this sometimes, but often it just ends up pissing off the people who worked on something only to be told it's "unusable" without any scientific backing for this assertion at all. I've had some dialog boxes of mine put through an UI review. Some of the points made were good, but some were for instance "There shouldn't be a horizontal line there, it looks unprofessional" which is not usability review, it's just irritating.
I have yet to find any major problems with the Mozilla UI - where I define major as being, I notice a big usability problem and get annoyed because of it. Saying, I can't drag toolbars around is valid, but that'd merely a feature request rather than a statement about the underlying design of the product.
Oh and finally, for those who like to bash XUL, remember one thing: if it wasn't for that, Mozilla probably wouldn't be cross platform, and as a result, would only exist on Windows.
And while you may consider that a disadvantage that you cant move apps around, I consider it a huge advantage. Say I want to rearrange my OS X box's/Applications folder, I create new folders there, drop the apps in, and all of them work. On Windows, update registry hell. On Linux, find conf files, edit them, if you can.
Huh? On Windows/Linux you never move apps around, because you have menus that can be reordered instead. The fact that on the Mac there are no apps menus simply means you have to perform filing system manipulations as well - by the way, how do you deal with multi-user scenarios here? Does everybody share the same menu/Applications folder?
MS Office X isnt a bundle, so using it as an example isnt even relevent. Its an old CFM app, an OS 9 app.
Dunno about this, maybe you're right. When I used Office X it seemed like a pretty tight rewrite to me, certainly it used Appfolders and was all Aquified.
But MS made installation a little too easy, and stuck their registration information in the Office folder, so you could drag and drop the folder.
Or maybe it was because otherwise if you uninstalled apps, you'd end up with a load of hidden files just lying around not being used in the Prefs folder?
And the Installer systems exist for programs that need to stick crap in Systemland, such as configs in the/Library rather in the ~/Library. They dont exist for any other reason.
Most Installers do way more than this. They check registration codes, let you customise the components of the app... the list goes on and on.
Wow. This is what I love about slashdot - this is the kind of thread that shows people what a debate really is: a set of well constructed and eloquently expressed arguments. But anyway....
What seems to "bother" you is that Stallman has advanced persuasive arguments in favor of an idea that conflicts with your existing world view. Rather than rethink that world view in light of the new information, you emotionally reject it as "propaganda."
Agreed 100% with your post by the way, however I think the reason many reject Stallmans philosophies is because to fully accept them they would have to fully dissociate themselves from the current world economy. foobar104 says things like "I believe personal gain is legitimate". What he means is, in a capitalist country, personal gain is seen as legitimate and encouraged. This is not the same thing as "personal gain is good", which seems to be what he is implying.
The world as Stallman describes it would be quite close to Utopia - a world in which everybody shares, a perpetual gift economy in which everybodies intelellectual contributions to society are maximised. Unfortunately, capitalism is the exact opposite of this in many ways, and so people say "I could not do this in the 'real' world, therefore Stallman is an extremist who should be ignored". Far from it - Stallman is an extremely smart guy whos ideas are unfortunately prevented from being realised by the world in which we live.
An analogy just occurred to me. Capitalism is the Windows of economics. We virtually all use it, there is a lot of pressure on those who don't to adopt it (see IMF/World Bank etc), if you don't use it you are locked out of world trade flows. It's a long way from being perfect, but the alternatives are not as developed so people ignore them.
Stallman has started one revolution already - Linux is ticking along nicely, and because of that he'll go down in the history books. But he is angry, because the world is not living up to his ideal vision: there is still non-free code (don't confuse that with non-GPL), and people still ignore him: they have to in order to survive.
If Stallman isn't careful, he'll die a bitter old man, convinced that Eric Raymond "stole" his revolution, when in reality he just faced up to the implications of our economic system. Stallman has changed the world once, now I think he should try again, but this time he should create an economic system that would allow his original vision (a good one) to be realised without huge economic pain.
I don't know how far along Hans Reiser is with his vision [namesys.com], but if he can pull it off, that'd be a perfect foundation..
He says 31st December:) The big problem I see with this is that it's specific to ReiserFS - not a problem in and of itself, but the desktop environments get hung up about running on Windows (!) so I think it unlikely there'd be a mass redesign of all the software to make it specific to one filing system. The most interesting thing with Reiser4 is that the kernel has been changed to allow files to also be directories. I can see sooooooo many uses for this
Mac OS X seems to have overcome this hurdle with a workaround called "bundles". The user experience is not provided solely by "screen furniture". This is an elitist idea. The OS has to have the desktop user in the picture from the start.
Elitist? Hardly. For your information, distributing applications as bundles has a set of disadvantages as well as advantages:
Advantages : Simple, and programs can be easily manipulated by the user. Programs can be easily copied, downloaded, etc. Uninstallation/installation is drag and drop.
Disadvantages : No dependancies (MacOS, being proprietary, has very little code sharing compared to Linux, and virtually all deps are based in the OS upgrades, so it doesn't really need them). No installation scripts - you can't present EULAs [1], request copy protect keys, remove generated data such as configs on uninstall and so on. Makes piracy much easier, as applications can simply be dragged onto a disk. Witness the case of a kid who was caught copying MS Office X onto his iPod - this would be much harder with Linux, and almost impossible with Windows. No install time customisation: for instance, with Mozilla you can opt to not have the Mail program, Chatzilla etc if you don't want them. This is not an option with the appfolders system.
Because of all these disadvantages, Apple provides the Apple Installer service. Some programs use this, I believe XDarwin does. However, even this is not advanced enough for some apps, so Wise is making a tidy packet out of providing InstallShield style stub programs to deal with all the features that some people need, but appfolders lack.
In short get off your horse. This idea that the desktop must be "designed with the user in mind", or that the Linux packages system is "elitist", is, frankly, bollocks. On what basis do you assume you are an expert in this area? [rant] I find a lot of Mac users are like this. They assume because they bought a Mac, they are instantly usability experts. They are not of course. [end rant].
I never understood why the open source crowd decided to hop on the Unix horse. Proprietary Unix is no better than proprietary Windows, or proprietary Mac OS.
The reasons for why UNIX was chosen rather cloning anything else or creating something new from scratch are well documented in the GNU Manifesto. You should read it, it's very interesting. If you can't be bothered however, here is the reasoning:
Our aim is to create an open/free platform. It is not to create a "cool" research OS, it is freedom. Therefore we will recreate something that we know will work, and that we know is good. Most importantly, as the internet is still in the early stages of development, we must be able to work with minimal communication between the teams. Therefore we will use UNIX, as it's highly modular, there is a lot of shared experience here, and although is it not perfect, it is not bad (Stallmans words, not mine). UNIX is also very well standardised through POSIX and so on, and was well designed to begin with. This way, we will avoid endless flamewars over how to design the "perfect" OS, by uniting people behind a common goal they can work towards largely in isolation.
So that is why Linux is based on UNIX and not something else entirely.
[1] I had somebody tell me you can put them into the background of a.dmg file - not good enough I'm afraid, they must be click through. The best time for this is at installation. You can put it on first execution if you like, but how do you record that the user accepted? By using a config file that will be left hanging around when the user uninstalls the app?
In addition, it would require complete re-writes of existing monolithic applications with no benefit to the companies save additional competition. Since it was a Mac-only technology, it would have made porting software nigh impossible as well.
Actually, OpenDoc was cross platform (in theory). However, it was unfortunately done at the time when Apple rather arrogantly believed OpenDoc was good enough to reinforce falling Mac marketshare. As a result, the Windows version (produced by IBM if I recall correctly) was nigh on impossible to find: I should know, I spent many weeks searching for it.
The official website was no help - it was hosted by Apple and made no mention of a Windows version, instead choosing to trumpet the Mac. IBMs site was so huge searching through it all was extremely difficult and the best search engine available was AltaVista. Enough said.
The overwhelming numbers of BSD users do [apple.com].
Two things you clearly don't understand.
1) MacOS is not BSD. FreeBSD is BSD, OpenBSD is BSD, MacOS X is... well, MacOS. That's why it's called MacOS and not Apple BSD. The fact that they used large parts of FreeBSD and Linux as the core is irrelevant. It is MacOS.
2) There are at least 4 times as many Linux desktop users as MacOS X users. According to Apple, MacOS X has 0.5% market share (1 in 10 mac users, mac's have 5%). These figures were produced when under pressure from Microsoft to increase sales, so they're likely to be a slight overestimate. I'm assuming 2% Linux Desktop usage here, which is the estimate I get from IDC (an independant 3rd party). Don't assume that OS X has lots of users. Microsoft is struggling to make money from the premier office suite sales are so slow. What, Office X is too expensive? Is this from the people who are saying that $120 is reasonable from Apple? One quarter of the number of Linux users - this is hardly overwhelming.
You can rant and rave about your shiny new toy all you like, the fact is, we don't care. The Mac is proprietary, and that's not the way forward. You're finding this out the hard way with the 10.2 upgrade price and dotMac. Quit the shills, it annoys us people who really do want to get things done.
This is wierd, I can't post using my normal connection - I just get a "You can't post to this page message", and that's even the case when logged out. Anyway:
If Apple really does have a good thing that promotes *nix at some level and provides a bulwark against a known untrustworthy actor, what steps could Jobs take (supplementing making part of the system open source) that would: advance the art, allow Apple to make a decent profit (profit is not a bad thing), give you an solid alternative that you could stomach and gain market share against a bully?
The only thing Apple could do that would be OK in my books is to quite the platform business. They can make as many media players as they like, that's cool, I like the iPod. My problem is with closed platforms. And by the way, there's nothing inherantly great about UNIX. I don't get why people are posting say "UNIX will take over the planet, hurrah". So what? UNIX is just a core, it's quite good, but it's been surpassed many times since then.
Apple can make lots of money doing what it does best: making slick products with slick advertising. However, as long as it continues to link everything to the Mac, I will have big problems with this company and by extension anyone who schills for them (far too many people do this). If they can survive without the Mac then great. Otherwise, tough - they should go out of business.
Apple is no longer a closed system. They have a proprietary GUI, true, but how is that different from a Linux distributor bundling in their proprietary software? The core of the OS, Darwin, is open source, BSD license. But from the way they treat it, you'd think they thought it was GPLed (actually, there is a GNU-Darwin too, if you'd rather run that). Apple has gotten open enough hardware wise that you can take the FAT32 drive from your PC and put it in a G4 tower, and the Mac can read it.
Apple has been, and always will be a closed system. You think it's just the GUI? Wrong. It's everything BUT the kernel. Not just Quartz, or Aqua, but also the dock, the finder, the desktop shell, the artwork, the applets, the bundled applications. Hardware? That's closed too. The fact that they are basically PCs is irrelevant, you can't build Macs, only Apple can. Therefore they are closed.
The "Microsoft Tax" refers to paying for Windows on a new PC that you want to run Linux on. Apple throws their OS on a new Mac for free. You pay for the hardware you get, and if you want to run Linux on it, Apple doesn't mind at all.
LOL! Free! Last time I checked, it cost $120! Just because the price is added on to that of the hardware certainly doesn't make it free. You can't buy a Mac without OS X, and you can't buy a PC without Windows. I fail to see the difference.
This was written on a G4 iMac running OS X, using Mozilla. Emacs was running in the background (I've been an Emacs user since 1990 when I first read the GPL).
Some of us know the issues surrounding the use of Free and Open Source software far better than you think.
On the contrary, if you understood those issues, you'd understand why Stallman wasn't happy with just a free text editor. You'd understand why a free platform is far more important.
Slashdot Mirror
I've seen quite a few comparisons to Mozilla. It should be remembered that at the start of the Mozilla project, virtually all the developers were Netscape employed. Gradudally the code was properly licensed, and the code became easier to work, and everybody got into the swing of it. Nowadays it has lots of developers. I'm sure OOo will do the same thing.
One of the best things about .NET is that there is 1 class library, so you don't get people constantly reinventing the wheel in language X. I think it's actually pretty ironic that despite virtually all the code in Linux being open, we still lack a good way of sharing that code between languages other than binding it to C and then to other languages manually (yuck!).
Wierd but true, I've never met anybody with that number of (legal) CDs. You learn something every day. Next question though - do you listen to them all? I don't have anywhere near that number of CDs, and still I have favourites and ones I haven't listened to for ages and don't really care about anymore.
The Gift Economy
(and yes, I know it looks odd in Konqueror ..... )
Yes, it's expensive. No there isn't Linux support.
Sounds like just what he wants ;) But seriously, a few points to remember if you're planning on getting an iPod:
Finally, just out of interest, how many people here actually own 20gig of legal music? That would be 31 CDs of uncompressed digital audio, let alone MP3s. Really, I imagine that the only people who need that sort of capacity are those who have built up huge collections of music off the P2P networks.
This would be the case if MacOS was open source. Let me remind you however that it is only the UNIX core that is open. MacOS is defined if anything by its GUI. So now Hollywood come banging on Apples door - hey Jobs! Stop your Mac users from copying our stuff.
"Certainly", he would say, "we'll just build it into Quicktime, and stop things at the user interface level".
The core is open, but nobody buys a Mac for its kernel. They buy it for the GUI. If the option of doing "illegal" things is removed above the kernel level, they're still stuffed.
I have to disagree with this. I feel the reason some users are confused by choice is simple - they are not used to it.
Computers are complex and powerful devices it is true, but then so are cars, hi-fis and so on. People seem to cope with choice and flexibility in these markets just fine, I know some people who actually enjoy browsing around car shops going "ooh" and "aah" at all the different makes and models.
Computers are different of course, because there has never been any competition in computing, thanks to Microsoft. As far as the public is concerned, MS is the way, and has been from day 1, ever since the PC arrived. Yes, there was the Mac, but there's a reason Apple lost the platform war so miserably: given a choice between a completely single vendor solution and a semi-open solution, they chose the semi open one. And so we have Microsoft.
It annoys me when people say, or imply, that Linux cannot succeed because people don't want choice, or are too stupid to deal with it. This is doing a disservice to both Linux and computer users. There are users who are confused by choice, but they can and will adjust. The total lack of training many people have doesn't help here of course, far too many people have learnt computing by trial and error and the attitude that "The user must not have to make any effort to learn this thing" is pervasive in computing.
To some extent this is good, because it means we can get up to speed with new software quickly, but it can also backfire in the case of people who never really understood what they were doing in the first place and so can't generalise their learning to something else.
Some users are only inflexible because they are scared by choice, because they have never had it before and don't understand it. They are scared by it, because they know they never really learnt how to work their computer in the first place, only that if you click here, then there, and then on this picture you can access the web. For instance, I once asked somebody to visit a website I'd made and tell me what they thought. What they saw was all messed up, so I asked what browser they were using (I was pretty sure it was ie5 but wanted to check). The answer: "Sorry, I'm not technical, what's a browser?" shocked me. How can anybody use something yet not know what it's called? Simple - she hadn't been taught how to use a computer, she had learnt by rote.
If somebody doesn't want to switch to something better than Outlook, sit down with them and teach them how to use Outlook - and then teach them how to use MozMail or whatever. And if they don't want to know .... well, next time they get a virus, let them fix it themselves. I've done this, and it works wonders.
That was when it was just the engineers. Then the lawyers got involved. Oh yes. The lovely EU Competition Commision discovered this bunch of engineers from all the major players working together and decided that wouldn't do, and split them up. They forced ITV and the cable companies to eject Sky, and pay Murdoch a few hundred million pounds in compensation. From that point on things just went downhill. The idea of the universal box was killed the moment the managers, lawyers and marketroids got hold of it.
The humble cable box, for years a mere channel-flipper, is in for a multimedia makeover: Beefed-up boxes of the future could let you play video games online, share digital photos with friends, and maybe do other things people in the business haven't even imagined yet.
They need to come over to my house and play with Sky Digital if this is what they think. Playing games on your TV? Been there, done that, it doesn't work needless to say. For one, the games take forever to load - even with the gigabits of bandwidth they have coming off of Astra it can take several minutes to load games - it's like being back with a Commodore 64. Then, when the games do load, they are extremely primitive and loaded with advertising. Realtime games are out as the latency involved from the handset is huge - I tried a simple top down racing game one time, it was almost unplayable as the car responded almost a second after I hit the button. Finally, there aren't many of them, as interactive TV applications are far more expensive to produce than computer apps. Interactive TV is basically dead as a from of entertainment. Where it does shine is in getting information - BBC News Interactive and Sky News Active are great. It also does simple interactive additions to programs quite well. The multiple football camera angles are rare however due to the large amount of bandwidth required.
Believe me, interactive TV on a universal cable box? It's a TV engineers dream but in the real world, we've done it, and the PC kicks its ass in almost every respect. It's a big (expensive) white elephant.
Huh? Another distro? What could Sun's edge over the others be is what I'm wondering. Companies like SuSE/Mandrake already have the low end, RedHat does the middle and Solaris has the high end. So is there going to be a high end Linux?
Oh! Here's an idea. Perhaps McNealy sees this as a chance to return to the Network Computer idea (not a bad one in itself) by having thin client machines running XFree. Suns Linux could be like the LTSP distro.
I think you are confused. CUPS is a printer API, it has nothing to do with sound. In fact, there is no need for an ASIO equivalent as Linux with the kernel pre-empt and low latency patches has far lower latency normally than Windows even with ASIO. I don't know if they are sample accurate, but I know you can get extremely low audio latencies with this sort of setup.
2) Inconsistent platform. We do not have a uniform development platform. Each distro ships a different version of GCC with its own inconsistencies, different GLIBC, same issue, and so on down to GTK or QT, XFree86, etc. With a product like Cubase taking at least 6 months to port, no doubt the platform will keep changing underneath them. What solution would they have other than supporting ONE distro only or shipping their own?
I don't understand you. Currently Linux is going through a switch of the C++ ABI. If your audio app is written in C++ then you will need to provide 2 binaries for the duration of the switchover. They do not require porting, just a recompile. The widget toolkits only break compatability every few years, and you can always use compatability libs as they can be installed side by side. Moonlight for instance uses its own version of Qt 2.2. X has been backwards compatible for over 10 years. They can do what all the other Linux companies do, and write their software to be distro neutral. It's not hard, the biggest challenge is the installer.
3) Lack of hardware support. Yes, vendors like my personal favorite, Echo, aren't releasing enough specs to the community for a free driver. But on the other hand, the community isn't providing them a stable platform to develop on, see #2.
NVidia manage it. There are even drivers now for WinModems (which use proprietary drivers). It's more work to support all kernel versions, and I agree that Linus should stop breaking compatability, however it is perfectly possible to write drivers that will compile on any kernel version. If Echo don't release the specs, or write their own drivers then yes you are locked in to Windows. Congratulations.
4) Lack of unchanging commercial plugin standards. VST works. The API doesn't have a million tiny revisions, and any VST plugin works great on anything from Logic to Fruity Loops to the latest Cubase SX. The API was published WHEN IT WAS FINISHED, and NOT CHANGED.
VST plugins are the standard because of the popularity of Cubase. This is hardly the fault of Linux - if one sequencer gets more popular than the rest then I'm sure we'll see something similar. How often the plugin API changes is entirely up to the developer: remember that APIs change more frequently on Linux because most stuff is open source, so changing the API to get a better system is less painful than on a closed source system.
The SuSE Live CD tells the user about slow bootup when starting. Also, it creates a couple of files on c:\ (they can be deleted afterwards) that contains your home directory and a swap file, so they don't have to worry about it using up all the RAM and they can even save documents with it! SuSE definately has this done very slickly indeed.
But surely you get a lot less flexibility? I seem to recall MS tried to rewrite Outlook in DHTML, the project bombed, as HTML (even with loads of IE extensions) just wasn't suitable for user interfaces. That's why Mozilla created XUL (which is). How does it deal with things like snap-off toolbars, popup menus, zigzagged underlines and all the other stuff that's either impossible or extremely hard to do with HTML?
It sounds like you'd end up with an extremely limited and rather slow widget set - a bit like AWT in fact. On the other hand, I haven't tried it, so what do I know?
Uh oh .... I just bought an nVidia card for my PC because I thought it'd work well with Linux - and I have an AMD Duron. Could you please tell me more about this problem?
There was also:
In short there were many reasons why ITV Digital fell, most of which were political rather than technical.
So could parrot also be used by a lower-level, strictly typed language? Would it be possible to target C++ for it for instance?
I was just researching Parrot when I came across this story. The problem with Python isn't so much that they're not interested, it's more lack of volunteers. All the "core team" (only about 4/5 people) is too busy maintaining the language. Python should be fairly easy to retarget to Parrot though.
I do think this would be cool. We need a truly open cross-language VM. Mono would be OK, but unfortunately there are just way too many people (including me) who just don't trust Microsoft with .NET
What especially interests me is the possibility that Parrot binaries could share objects and such - this would for the first time allow Perl to use Python code and vice-versa (as well as any other languages targetted to Parrot). The merge between Python and Perl was an April Fools joke, but I'm wondering - would it not make sense to allow code to be easily shared between them and to construct a .NET style "shared" set of APIs? The amount of code duplication is staggering with these two languages, if it could be cut down that'd be fantastic.
Here is his list of usability problems with Mozilla From what I recall, the main criticisms of MPT boiled down to "I don't like it". For instance, he makes a big deal of the fact that the Home link is on the Bookmarks toolbar, rather than the main toolbar. This immediately leads of course to flamewars between people who believe it "belongs with the reload button" or people who thinks it makes more sense to have it with your other links. This is hardly a usability issue (remember neither Hyatt or MPT have any usability training at all - no disrespect to them, but it's true). It's just personal preference.
He talks about speed as well - that's hardly as much of a problem as it was. Especially on Windows, Mozilla feels just as snappy as IE (no, really, and I have a PIII/500).
Text editing bugs : these are bugs, not usability problems.
Message Display: he doesn't like the fact that headers are in their own section. Personally I don't mind this at all, but clearly he feels otherwise.
The list goes on and on. Some of his points are good. Many are simply pet peeves on his part. This is often the problem with "usability", it's a very vague concept and the science of usability is still in its infancy. Therefore a "usability" review often degenerates into a case of the UI reviewer picking on things they don't like. For instance, the "I don't think this feature is useful, so it's preferences bloat". There is a grain of truth to this sometimes, but often it just ends up pissing off the people who worked on something only to be told it's "unusable" without any scientific backing for this assertion at all. I've had some dialog boxes of mine put through an UI review. Some of the points made were good, but some were for instance "There shouldn't be a horizontal line there, it looks unprofessional" which is not usability review, it's just irritating.
I have yet to find any major problems with the Mozilla UI - where I define major as being, I notice a big usability problem and get annoyed because of it. Saying, I can't drag toolbars around is valid, but that'd merely a feature request rather than a statement about the underlying design of the product.
Oh and finally, for those who like to bash XUL, remember one thing: if it wasn't for that, Mozilla probably wouldn't be cross platform, and as a result, would only exist on Windows.
Huh? On Windows/Linux you never move apps around, because you have menus that can be reordered instead. The fact that on the Mac there are no apps menus simply means you have to perform filing system manipulations as well - by the way, how do you deal with multi-user scenarios here? Does everybody share the same menu/Applications folder?
MS Office X isnt a bundle, so using it as an example isnt even relevent. Its an old CFM app, an OS 9 app.
Dunno about this, maybe you're right. When I used Office X it seemed like a pretty tight rewrite to me, certainly it used Appfolders and was all Aquified.
But MS made installation a little too easy, and stuck their registration information in the Office folder, so you could drag and drop the folder.
Or maybe it was because otherwise if you uninstalled apps, you'd end up with a load of hidden files just lying around not being used in the Prefs folder?
And the Installer systems exist for programs that need to stick crap in Systemland, such as configs in the /Library rather in the ~/Library. They dont exist for any other reason.
Most Installers do way more than this. They check registration codes, let you customise the components of the app... the list goes on and on.
What seems to "bother" you is that Stallman has advanced persuasive arguments in favor of an idea that conflicts with your existing world view. Rather than rethink that world view in light of the new information, you emotionally reject it as "propaganda."
Agreed 100% with your post by the way, however I think the reason many reject Stallmans philosophies is because to fully accept them they would have to fully dissociate themselves from the current world economy. foobar104 says things like "I believe personal gain is legitimate". What he means is, in a capitalist country, personal gain is seen as legitimate and encouraged. This is not the same thing as "personal gain is good", which seems to be what he is implying.
The world as Stallman describes it would be quite close to Utopia - a world in which everybody shares, a perpetual gift economy in which everybodies intelellectual contributions to society are maximised. Unfortunately, capitalism is the exact opposite of this in many ways, and so people say "I could not do this in the 'real' world, therefore Stallman is an extremist who should be ignored". Far from it - Stallman is an extremely smart guy whos ideas are unfortunately prevented from being realised by the world in which we live.
An analogy just occurred to me. Capitalism is the Windows of economics. We virtually all use it, there is a lot of pressure on those who don't to adopt it (see IMF/World Bank etc), if you don't use it you are locked out of world trade flows. It's a long way from being perfect, but the alternatives are not as developed so people ignore them.
Stallman has started one revolution already - Linux is ticking along nicely, and because of that he'll go down in the history books. But he is angry, because the world is not living up to his ideal vision: there is still non-free code (don't confuse that with non-GPL), and people still ignore him: they have to in order to survive.
If Stallman isn't careful, he'll die a bitter old man, convinced that Eric Raymond "stole" his revolution, when in reality he just faced up to the implications of our economic system. Stallman has changed the world once, now I think he should try again, but this time he should create an economic system that would allow his original vision (a good one) to be realised without huge economic pain.
He says 31st December :) The big problem I see with this is that it's specific to ReiserFS - not a problem in and of itself, but the desktop environments get hung up about running on Windows (!) so I think it unlikely there'd be a mass redesign of all the software to make it specific to one filing system. The most interesting thing with Reiser4 is that the kernel has been changed to allow files to also be directories. I can see sooooooo many uses for this
Elitist? Hardly. For your information, distributing applications as bundles has a set of disadvantages as well as advantages:
Advantages : Simple, and programs can be easily manipulated by the user. Programs can be easily copied, downloaded, etc. Uninstallation/installation is drag and drop.
Disadvantages : No dependancies (MacOS, being proprietary, has very little code sharing compared to Linux, and virtually all deps are based in the OS upgrades, so it doesn't really need them). No installation scripts - you can't present EULAs [1], request copy protect keys, remove generated data such as configs on uninstall and so on. Makes piracy much easier, as applications can simply be dragged onto a disk. Witness the case of a kid who was caught copying MS Office X onto his iPod - this would be much harder with Linux, and almost impossible with Windows. No install time customisation: for instance, with Mozilla you can opt to not have the Mail program, Chatzilla etc if you don't want them. This is not an option with the appfolders system.
Because of all these disadvantages, Apple provides the Apple Installer service. Some programs use this, I believe XDarwin does. However, even this is not advanced enough for some apps, so Wise is making a tidy packet out of providing InstallShield style stub programs to deal with all the features that some people need, but appfolders lack.
In short get off your horse. This idea that the desktop must be "designed with the user in mind", or that the Linux packages system is "elitist", is, frankly, bollocks. On what basis do you assume you are an expert in this area? [rant] I find a lot of Mac users are like this. They assume because they bought a Mac, they are instantly usability experts. They are not of course. [end rant].
I never understood why the open source crowd decided to hop on the Unix horse. Proprietary Unix is no better than proprietary Windows, or proprietary Mac OS.
The reasons for why UNIX was chosen rather cloning anything else or creating something new from scratch are well documented in the GNU Manifesto. You should read it, it's very interesting. If you can't be bothered however, here is the reasoning:
Our aim is to create an open/free platform. It is not to create a "cool" research OS, it is freedom. Therefore we will recreate something that we know will work, and that we know is good. Most importantly, as the internet is still in the early stages of development, we must be able to work with minimal communication between the teams. Therefore we will use UNIX, as it's highly modular, there is a lot of shared experience here, and although is it not perfect, it is not bad (Stallmans words, not mine). UNIX is also very well standardised through POSIX and so on, and was well designed to begin with. This way, we will avoid endless flamewars over how to design the "perfect" OS, by uniting people behind a common goal they can work towards largely in isolation.
So that is why Linux is based on UNIX and not something else entirely.
[1] I had somebody tell me you can put them into the background of a .dmg file - not good enough I'm afraid, they must be click through. The best time for this is at installation. You can put it on first execution if you like, but how do you record that the user accepted? By using a config file that will be left hanging around when the user uninstalls the app?
Actually, OpenDoc was cross platform (in theory). However, it was unfortunately done at the time when Apple rather arrogantly believed OpenDoc was good enough to reinforce falling Mac marketshare. As a result, the Windows version (produced by IBM if I recall correctly) was nigh on impossible to find: I should know, I spent many weeks searching for it.
The official website was no help - it was hosted by Apple and made no mention of a Windows version, instead choosing to trumpet the Mac. IBMs site was so huge searching through it all was extremely difficult and the best search engine available was AltaVista. Enough said.
Two things you clearly don't understand.
1) MacOS is not BSD. FreeBSD is BSD, OpenBSD is BSD, MacOS X is ... well, MacOS. That's why it's called MacOS and not Apple BSD. The fact that they used large parts of FreeBSD and Linux as the core is irrelevant. It is MacOS.
2) There are at least 4 times as many Linux desktop users as MacOS X users. According to Apple, MacOS X has 0.5% market share (1 in 10 mac users, mac's have 5%). These figures were produced when under pressure from Microsoft to increase sales, so they're likely to be a slight overestimate. I'm assuming 2% Linux Desktop usage here, which is the estimate I get from IDC (an independant 3rd party). Don't assume that OS X has lots of users. Microsoft is struggling to make money from the premier office suite sales are so slow. What, Office X is too expensive? Is this from the people who are saying that $120 is reasonable from Apple? One quarter of the number of Linux users - this is hardly overwhelming.
You can rant and rave about your shiny new toy all you like, the fact is, we don't care. The Mac is proprietary, and that's not the way forward. You're finding this out the hard way with the 10.2 upgrade price and dotMac. Quit the shills, it annoys us people who really do want to get things done.
If Apple really does have a good thing that promotes *nix at some level and provides a bulwark against a known untrustworthy actor, what steps could Jobs take (supplementing making part of the system open source) that would: advance the art, allow Apple to make a decent profit (profit is not a bad thing), give you an solid alternative that you could stomach and gain market share against a bully?
The only thing Apple could do that would be OK in my books is to quite the platform business. They can make as many media players as they like, that's cool, I like the iPod. My problem is with closed platforms. And by the way, there's nothing inherantly great about UNIX. I don't get why people are posting say "UNIX will take over the planet, hurrah". So what? UNIX is just a core, it's quite good, but it's been surpassed many times since then.
Apple can make lots of money doing what it does best: making slick products with slick advertising. However, as long as it continues to link everything to the Mac, I will have big problems with this company and by extension anyone who schills for them (far too many people do this). If they can survive without the Mac then great. Otherwise, tough - they should go out of business.
Apple has been, and always will be a closed system. You think it's just the GUI? Wrong. It's everything BUT the kernel. Not just Quartz, or Aqua, but also the dock, the finder, the desktop shell, the artwork, the applets, the bundled applications. Hardware? That's closed too. The fact that they are basically PCs is irrelevant, you can't build Macs, only Apple can. Therefore they are closed.
The "Microsoft Tax" refers to paying for Windows on a new PC that you want to run Linux on. Apple throws their OS on a new Mac for free. You pay for the hardware you get, and if you want to run Linux on it, Apple doesn't mind at all.
LOL! Free! Last time I checked, it cost $120! Just because the price is added on to that of the hardware certainly doesn't make it free. You can't buy a Mac without OS X, and you can't buy a PC without Windows. I fail to see the difference.
This was written on a G4 iMac running OS X, using Mozilla. Emacs was running in the background (I've been an Emacs user since 1990 when I first read the GPL).
Some of us know the issues surrounding the use of Free and Open Source software far better than you think.
On the contrary, if you understood those issues, you'd understand why Stallman wasn't happy with just a free text editor. You'd understand why a free platform is far more important.