“I would like to announce a VMware Player image for Syllable 0.6.0a
available in the normal VMware images location. This works fine with VMware Player on Linux. I haven’t tried it yet on Windows, but I assume it works fine there, too. This is a default install from the ISO, just like the rest, so you’ll have to install Developer’s Delight and whatnot by yourself.” And before people complain: Syllable is an open-source (GPL) operating system aimed at the home/office user. It is the continuation of AtheOS.
Does this mean I can use this image to run Syllable 0.6.0a as a guest OS under VMWare Workstation for Linux? (I have version 5.5 of VMWL). Thanks.
I believe so, yes.
Yes, you’re right. The link page says as much (I should have checked first), and I’ve just run it successfully with VMWare 5.5 for Linux.
Thanks.
Wow, impressive, it boots impressively fast, looks quite promising indeed. Thanks for making the image available, it pushed me over the inertial barrier to trying syllable.
This image also works fine in the Windows version of VMware Player.
Doesn’t start up with GSX 3.1, error “11”. Hmm.
It’s ok I guess. I don’t see it doing anything great for the OS world. Because of the Windows monopoly hobby operating systems like Syllable have no chance in getting any broad use. Developers really should be focusing on the only hope to open up the operating system market, ReactOS.org . Then we can really get somewhere.
Its not about gaining market share, but one of the most impressive features about Syllable is that it does not rely on X-windows for a windowing system, unlike Linux and Unix. As a hobby or research OS, this is of great importance. I’m am pretty tired of the bloated memory requirements of X and many applications and appreciate a deviation in windowing/rendering managers.
The original project, Athene, eventually gave up on trying to maintain the underlying operating system, bought into a commercially available SDK and created its windowing product for Linux (and windows too).
Not all hobby and research OS’s want to be the next Microsoft or OSX, although it would be nice to make money. Most of these ideas are usually derived from the dissatisifaction of what is currently available and the can-do attitude of the programmers, testers, and hobbyists.
“it does not rely on X-windows for a windowing system, unlike Linux and Unix. As a hobby or research OS, this is of great importance. I’m am pretty tired of the bloated memory requirements of X and many applications and appreciate a deviation in windowing/rendering managers. ”
A few corrections:
– Linux and Unix don’t rely on X for a windowing system, but X is the most popular windowing system. Nothing stops you from using something else on these systems.
– X doesn’t need a lot of memory, in fact I’ve seen it run with as little as 16MB. What usually eats up memory is what runs on top of X, and you also can run anything you want there.
If you find Syllable an interesting OS by all means use it, but you have a few misconceptions on Unix and X.
“Linux and Unix don’t rely on X for a windowing system, but X is the most popular windowing system. Nothing stops you from using something else on these systems. ”
Examples please.
Direct FB for one
http://directfb.org/
Well, Mac OS X is a Unix-like system that doesn’t use X.
Other options that should work on GNU/Linux are DirectFB and the Y Window System.
“- Linux and Unix don’t rely on X for a windowing system, but X is the most popular windowing system. Nothing stops you from using something else on these systems.”
You’re right; even an offshoot of AtheOS, Cosmoe, runs on Linux. However, this proved technically very difficult for them. In practice, Linux is pretty much married to X.
Kaj
“You’re right; even an offshoot of AtheOS, Cosmoe, runs on Linux. However, this proved technically very difficult for them. In practice, Linux is pretty much married to X.”
No it isn’t. GNU/Linux is a completely modular system (that started with Emacs
, you can replace X easily if you write a replacement for X (which is not easy).
It shouldn’t even be that much hard to port the desktop environments and development platforms to another system. For example Qt, GTK and GNUstep can run without X today.
“”You’re right; even an offshoot of AtheOS, Cosmoe, runs on Linux. However, this proved technically very difficult for them. In practice, Linux is pretty much married to X.”
No it isn’t. GNU/Linux is a completely modular system (that started with Emacs
, you can replace X easily if you write a replacement for X (which is not easy).
It shouldn’t even be that much hard to port the desktop environments and development platforms to another system. For example Qt, GTK and GNUstep can run without X today.”
You may want to reread my post. I said Linux is married to X IN PRACTICE, which can easily be observed by looking around you and doing a count. Of course, if a juggernaut like EMacs is your idea of modularity and you would classify porting the Linux desktops to X-less systems as not that hard, you’re living in a totally different world than Syllable’s. Creating this world is the entire point of our journey.
Kaj
“You may want to reread my post. I said Linux is married to X IN PRACTICE, which can easily be observed by looking around you and doing a count. Of course, if a juggernaut like EMacs is your idea of modularity and you would classify porting the Linux desktops to X-less systems as not that hard, you’re living in a totally different world than Syllable’s. Creating this world is the entire point of our journey. ”
I never said anything about Emacs being modular, please read again. FYI when RMS started writing the GNU system the first module that he wrote was the text editor (which he would used to write the other modules).
Also, porting GNOME and KDE to X-less systems is *a lot of work*, but it’s not necessarilly hard. Just build the X replacement and port Gtk or Qt to it, for a start.
IMO it’s less work and less hard than it is to write an OS from scratch, but that’s not relevant here.
Well, we can argue this, what’s modularity and what’s bootstrapping and all that, until the cows come home. Syllable is a highly modular OS, but what we observe time and again is that Unix people have very different ideas about modularity. It’s not very productive to compare notes based on different understandings of the same words.
If I am being defensive, I apologize. It’s because we get attacked a lot when we present our present to the world.
Kaj
“The original project, Athene, eventually gave up on trying to maintain the underlying operating system, bought into a commercially available SDK and created its windowing product for Linux (and windows too).”
No, you’re mixing up AtheOS and Athene. The latter is a totally different system, always meant to be capable of running as a portability environment on other systems. Although both have strong Amiga roots.
Kaj
If the best thing about Syllable is that it doesn’t have X windows, I could not tell the difference anyways. The window redraw sucks just as bad. I have nothing against this operating system in general, it’s on par with a lot of the alternative operating systems I’ve tried. It’s just that it doesn’t matter how much better it might be, it will still only ever get used by a handful of OS enthusiasts until the Windows monopoly is brought down. Look at BeOS, it was far ahead of it’s supposed competition (Windows 98 and Me) but they where still crushed by a monopoly. What I’m getting at is NO ONE will be able to be innovative and have it really mean anything until Microsoft is brought down. The only thing that will do that is a free Windows clone.
“Look at BeOS, it was far ahead of it’s supposed competition (Windows 98 and Me) but they where still crushed by a monopoly.”
Huh? BeOS could get crushed because it was a proprietary product from a money-making company. Microsoft can destroy those. However, it cannot destroy a bunch of volunteer programmers writing an open source operating system. Your arguments are mixed-up and don’t apply.
A free Windows clone? Remember when the Win2k source was ‘leaked’ a few years ago? Wait until ReactOS starts to take off, and Microsoft could simply ‘do a SCO’, claim there’s Win32 source in ReactOS, and hold back the project in legal tangles for YEARS. And it doesn’t matter whether it’s true or not — MS has the legal and financial muscle to severely impact ReactOS in many, many ways.
So while ReactOS is cool, it’s better to have a completely unique open source desktop OS than getting into potential legal quagmires with a notorious giant.
everyone has a chance.. maybe not for you, if syllable, reactos or whatever gains 10 more users.. the whole marketing is done right.. as there is no TCO or any other capitalist shit to compute OK?
cu in bloody 2006 ::|
“capitalist shit”? You have really missed the boat. Don’t be a tool. Communism or fascism, I’ll pick the lawful free market. Too bad neither of the former do.
“It’s ok I guess. I don’t see it doing anything great for the OS world. Because of the Windows monopoly hobby operating systems like Syllable have no chance in getting any broad use. Developers really should be focusing on the only hope to open up the operating system market, ReactOS.org . Then we can really get somewhere.”
So because of the ubiquity of Neanderthals, us Cro Magnons should have rolled over and died?
Not to take anything away from ReactOS; their progress is impressive. Maybe someday they’ll implement a Syllable subsystem.
Kaj
“So because of the ubiquity of Neanderthals, us Cro Magnons should have rolled over and died?”
You missed my point completely.
You are not the first to predict the demise of Syllable or bemoan our “impossible” task. I have no doubt you will not be the last. Needless to say, we’ll continue to code and release. As will ReactOS. That’s how you create a market for yourself, not by worrying about the big boys like Microsoft.
“You are not the first to predict the demise of Syllable or bemoan our “impossible” task. I have no doubt you will not be the last. Needless to say, we’ll continue to code and release. As will ReactOS. That’s how you create a market for yourself, not by worrying about the big boys like Microsoft.”
No, that’s how you waste your time like everyone before you. Many other OSes had a much better chance of getting anywhere than Syllable, they all failed or got nowhere because of the MICROSOFT MONOPOLY. The only way to get anywhere is to band together for a time to defeat the true enemy of real innovation, Microsoft.
Oh please, take your rhetoric someplace else. I find it funny that you seem to believe the way to compete against Microsoft is to create a clone of Windows. All that would do is tie developers and users even tighter to Microsoft.
“I find it funny that you seem to believe the way to compete against Microsoft is to create a clone of Windows.”
The biggest downside to any new OS is not having the applications, games or drivers that are programmed for Windows. A Windows clone (ReactOS) makes sense as it will be able to use the device drivers already provided by hardware manufacturers, and run all the programs & games for Windows.
Also, it gives a 3rd party the chance to come up with a better version of Windows (improving on some areas).
It makes perfect sense, but Microsoft may try to stop ReactOS when they get further ahead. If Linux could run all the Windows programs, @ full speed – no emulation, then many more people would leave MS Windows for Linux. Since this is unlikely to ever occur – this then leaves ReactOS as the next best thing for using Windows based software. Lots of people are interested in running Windows software, ReactOS will give them the chance to do this without having to buy MS Windows.
“Lots of people are interested in running Windows software, ReactOS will give them the chance to do this without having to buy MS Windows.”
Right on!
Plus other X-windows implementations than XFree86/Xorg like for example XDirectFB (runs on top of directfb), Xserver ( http://freedesktop.org/Software/Xserver ), KDrive and so on. Of these I’ve only tried KDrive, it comes with DamnSmallLinux. In addition, there AFAIK exist a few commercial X-servers.
X isn’t GNU/Linux’s only drawback.
Drivers is another issue. Building and installing them on Linux I mean. On Syllable, it’s supposed to be as obscenely easy as it was with BeOS.
Maybe others here can point out other ugly areas of Linux where Syllable shines.
“X isn’t GNU/Linux’s only drawback. ”
It’s not a drawback at all, quite the contrary.
“Drivers is another issue. Building and installing them on Linux I mean. On Syllable, it’s supposed to be as obscenely easy as it was with BeOS. ”
What kind of drivers? Kernel, display (X), printing…? In most cases the driver is included in the OS by default, and where it’s not installation can be simple if the driver distributor does it’s job. Installing a driver is the same as installing an application.
“”Drivers is another issue. Building and installing them on Linux I mean. On Syllable, it’s supposed to be as obscenely easy as it was with BeOS. ”
What kind of drivers? Kernel, display (X), printing…? In most cases the driver is included in the OS by default, and where it’s not installation can be simple if the driver distributor does it’s job. Installing a driver is the same as installing an application.”
I’d advise you to try out Syllable to see how this works. What the original poster meant is that installing a driver in Syllable – any driver, including plug-ins for several applications – is as easy as dropping one file in the right folder.
Kaj
“I’d advise you to try out Syllable to see how this works. What the original poster meant is that installing a driver in Syllable – any driver, including plug-ins for several applications – is as easy as dropping one file in the right folder. ”
That sounds like a nice way to install drivers, assuming the user knows what the right folder is.
Another way of doing it, which is my favorite btw, is this:
1) You double-click a PACKAGE file in a website. The file download begins.
2) Download is finished. You’re asked if you would like to install PACKAGE, and to provide your password in order to do so.
3) Enter password, driver or application installed!
Actually, if you want to use some driver, you just copy it to /system/drivers/dev/some_other_category. If you e.g. open /dev directory, the new driver will initialize immediately (drivers for non-hotpluggable devices) There’s no need for tons of megabytes of .inf files like in c:windowsinf – drivers load extremelly quickly at boot.
“Maybe others here can point out other ugly areas of Linux where Syllable shines.”
Who cares? So long as Windows dominates the market any innovation, and this is where true OS innovation comes from as I’m sure everyone will agree, will be for not.
It’s running very smoothly. There are a couple of bugs, but that’s to be expected. The GUI is actually quite nice, especially the way that they’ve gotten rid of the stupid “File” menu and replaced it with “Application”
Running Syllable is like going back in time. You could might as well install Red Hat 2.0 or something similar. I know, it’s not a finished product, but why starting it in the first place? Obviously, these guys are extremely talented programmers. So why didn’t they spend their effort and knowledge on improving Linux, or even BSD? Is it really worth beginning from scratch with the whole OS when they will, eventually, end up with the same bloated mess as any other modern OS? Or are they truly trying to make something better than what’s available today, for example, a mini-kernel with most stuff running in user-space?
About the UI, I don’t know where to start. I said it already: it’s like going back in time. Why on earth does this OS put the maximize button to the left of the minimize button? What’s the rationale behind such a decision? I mean, someone must have made the decision to switch places of the damn buttons, and, to say theh least, I would find it interesting to hear his reasons. It’s little things like that that makes me wonder who the OS is targetting. Which business would prefer a completely new, unsupported OS in favor of e.g. a fully supported Linux distribution?
Obviously, the OS boots very quickly. That’s because it doesn’t do much yet. Mark my words: Assuming the project will continue to evolve, Syllable will eventually become as bloated as Linux is. And before that happens, it will probably not be useful.
On a more on-topic note, the Virtual Machine worked once, then it doesn’t boot anymore. I’ve tried removing the memory file and resetting, but it won’t start anymore.
Edited 2005-12-30 22:20
Ah, the old “Why?” chestnut!
..why didn’t they spend their effort and knowledge on improving Linux, or even BSD?
To be blunt, because both Linux and BSD suffer from fundemental design brain damage that limits their use as user friendly desktop operating systems. Traditional UNIX systems have always made great servers, but have always sucked on your desktop. Neither Gnome nor KDE solve many of the real problems, and helpfully introduce many of their own.
Why on earth does this OS put the maximize button to the left of the minimize button? What’s the rationale behind such a decision? I mean, someone must have made the decision to switch places of the damn buttons, and, to say theh least, I would find it interesting to hear his reasons.
If you don’t like it, pick a different window decorator. Come to that, where is this universal law that says maximize must be to the right of minimize? What has been “switched”, other than your expectations?
Which business would prefer a completely new, unsupported OS..
Why not ask why businesses are dropping commercial UNIX variants in favour of Linux? I find this blinkered attitude that no new OS can possibly succeed amazing, especially when people use Linux of an example of why it can’t succeed!
Obviously, the OS boots very quickly. That’s because it doesn’t do much yet.
You’re the second person to say almost exactly the same thing, and you’re still both wrong. What do you think Syllable does when it boots? What do you think it doesn’t do that is should be doing?
Mark my words: Assuming the project will continue to evolve, Syllable will eventually become as bloated as Linux is.
What evidence do you base this on? Have you considered that Syllable might just do things differently to Linux for a good reason?
“To be blunt, because both Linux and BSD suffer from fundemental design brain damage that limits their use as user friendly desktop operating systems. ”
Brain damage? Maybe you should try using different words. You one, ones that don’t make you look like an ass.
” Traditional UNIX systems have always made great servers, but have always sucked on your desktop.”
Yeah, tell that to Apple.
Apple has what, 1% of the desktop market? I see how you’ve proven your case with that! Linux is absolutely brain damaged, and it will be until you don’t need to worry with kernel patches to support AMD64 chips (look for notimercheck or timerhack, finally fixed in .13) or an ACPI compiler to compile your own DSDT (yeah joe user will jump right on this!) or driver re-installation every other kernel install (yeah yeah, go ahead and use linux-restricted-modules see how it supports new chipsets like the ATI x200 when it locks up solid the moment you move the mouse).
Linux will suck until the fundemental problems with drivers etc are fixed and it doesn’t take 5+ patches to get your shiny new computer working.
Sure it’s cake for us until we get tired of constantly keeping up with it.
Joe user is going to tell everyone they know that Linux sucks the first time the installer hangs on “ACPI whatever it is” like some of the newer HP laptops do.
Whatever though, live in your little world. Modded down in 3 .. 2 .. 1 ..
Apple has more than 1% of the market, but I wasn’t talking about market share at all.
The argument was that Unix systems traditionally “sucked” on the desktop, and I was just reminding the poster about Mac OS X, which is regarded by many as the best desktop operating system available and it is a Unix-like OS. You can claim Windows XP is the best, and doesn’t suck because it’s not Unix and because it has the higher market share. You can also claim IE is the best browser because it has the higher market share. Syllable and BeOS must absolutely suck, because they have no significant market share at all.
That’s fine, in your own little world things might work like that.
About GNU/Linux and your problems with it, they’re mainly caused by poor support from hardware manufacturers. You and your friend Joe User should learn that GNU/Linux can’t support all PC-compatible computers when hardware specs are closed and some manufactures don’t play nice. You can either install GNU/Linux on supported hardware only (if kernel compiling is mentioned, it’s unsupported and by compiling the kernel you’re just trying to work around that fact), or complain to the manufacturer of the hardware you bought why it’s not supported.
There’s not much the developers can do about it. Unfortunately over 90% of the market is controlled by a monopolist, and a lot of hardware companies don’t seem to give a damn about it.
On the bright side, GNU/Linux supports most of the PC’s on the market, and market share is rising. Yay!
“There’s not much the developers can do about it. Unfortunately over 90% of the market is controlled by a monopolist, and a lot of hardware companies don’t seem to give a damn about it.”
Linux could start by offering a driver ABI. While I agree with your general feelings here, I can full well understand why hardware manufacturers (or software manufacturers, for that matter) don’t want to support a minority platform that thinks it has the luxury to act like quicksand. The competition offers both overwhelming market share and rigorous backward compatibility.
Kaj
Yes, I agree with you that a stable ABI should be offered. Even better would be convincing Linus that 2.6 is supposed to be the stable series. But he’s a dumb nazi.
Brain damage? Maybe you should try using different words. You one, ones that don’t make you look like an ass.
Why? I call it as I see it. There are design flaws in both Linux and BSD that limit their usefulness as a desktop operating system. Some of them go all the way back to AT&T UNIX, some of them are more modern issues.
Yeah, tell that to Apple.
I don’t need to, because they recognised exactly the same problems. Why exactly do you think they invested in Quartz & display PDF instead of using X? Another example; the file system hierachy in OS X and Syllable are actually pretty similiar, and they aim to solve the same issues. You don’t honestly believe OS X is a traditional UNIX system, do you? It’s the complete opposite!
“I don’t need to, because they recognised exactly the same problems. Why exactly do you think they invested in Quartz & display PDF instead of using X? Another example; the file system hierachy in OS X and Syllable are actually pretty similiar, and they aim to solve the same issues. You don’t honestly believe OS X is a traditional UNIX system, do you? It’s the complete opposite!”
So let me get this straight: the “design flaws” of GNU/Linux are the X Window System and the file system hierarchy? These are the only two you bothered to point out, but you still failed to explain why they are brain damaged design decisions.
The file system hierarchy doesn’t matter, because desktop users don’t have to browse the file system to use the system; all the user has to browse is his own personal directory.
Quartz can do stuff that X can’t, and vice versa. The X developers (more specifically, the people working on X.org) are working on some of the stuff that X can’t do, but they are hardly showstoppers when using X in a desktop system.
If you wish to reply, please at least elaborate on your complaints. I don’t think it’s useful for anyone to engage in a discussion where the arguments are “this is brain damaged” or “never gonna happen” without actually any content or thoughtful explanation.
You’ll have to excuse me, but I’ve gone over the exact same arguments and questions a hundred times before. It gets tiresome.
First of all, X & the file system are simply two obvious things that I pointed out as examples of why OS X is not a traditional UNIX and as a way to compare OS X to Syllable. They’re not the only issue I have with UNIX and Linux. My opinion on the matter has been documented many times elsewhere, so excuse me if I don’t do so yet again.
You say the file system hierachy doesn’t matter, I say this is a common falacy and it does. I think it does, Apple think it does. Users must administer their home computers, so why should we place artificial barriers in there way? The traditional UNIX method of installing software by scattering it around sub-directories within /usr might work for centrally administiered servers where updates are few and far between, but it certainly doesn’t work for a user who wants to try that hot new game demo, or just wants to install that FTP client without having to worry about that the installer wants to put it in /usr/local/bin but that isn’t in their $PATH. The attitude that this sort of stuff doesn’t matter has its roots back in the days when users sat at workstations where /usr was NFS mounted and a system adminstrator would take care of it for you, but it just doesn’t wash in todays world of personal computers.
My issues with X are long and varied. The complete lack of a default toolkit is a big problem on many levels. The fact that it solves the wrong problems faced by desktop personal computers is another. The fact that it is decoupled from the underlying OS is, in my opinion, a problem.
On a more general level, the lack of vertical integration is a big, big problem for me. It’s a problem for others, too. Projects such as Freedesktop.org and Portland have been created, recently, in an attempt to fix many of these issues. Is it enough? Probably not. The LSB doesn’t seem to have helped. Linux still has no fixed driver ABI, and it’s staying that way.
These, and more, basically boil down to the fact that Linux sucks as a desktop Operating System. I’ve been using it for nearly eight years now, and I don’t see much improvement. Things seem to change for the sake of it, but don’t ever seem to solve anything. Everything grows but functionality. That’s why I believe these systems suffer from “brain damage” and that’s why I work on Syllable.
Well, I’ll say it again: the file hierarchy doesn’t matter. If the user wants to install new FTP client, he/she downloads a package, installs package, and the FTP client appears in the applications menu. No messing with file structure necessary.
X doesn’t lack a default toolkit, it just doesn’t have a toolkit. That’s not for X to provide. For example, Ubuntu is an operating system that has a default toolkit, GTK+. Kubuntu is a different system, with a different toolkit (Qt). Mac OS X, just like Ubuntu, has it’s defaults (Carbon or Cocoa) but you can also install Gtk+ and Qt on it (and I’m not talking about X11 under OS X, these toolkits have native ports).
What you’re failing to understand is that X is just a part of a GNU/Linux system. The vertical integration you mention is something that the operating systems that use X have to do, and some are actually starting to do it now. Some GNU/Linux distributors are now (finally) starting to see their products as an operating system that a person should use, rather than a binary distribution of a lot of free software.
Btw, here’s a link: http://www.osnews.com/story.php?news_id=5215
An interview with 5 developers. If your OS of choice has at least one developer at least as capable as one of them you should consider yourself very lucky.
If the user wants to install new FTP client, he/she downloads a package, installs package, and the FTP client appears in the applications menu.
It does, if the user downloads the correct version of their package for their distribution. What if there isn’t a suitable package for their system? Will the Redhat RPM work on Mandriva? What about applications which don’t automatically create menu shortcuts; the vast majority of the executables on my Linux machine don’t have a KDE menu entry in fact. Gmplayer is one very useful GUI application that doesn’t creates on automatically. What about applications which don’t come as packages? Acrobat Reader, Firefox and OpenOffice.org: there are plenty of examples. What if something goes wrong and the user has to fix their machine? Believing that package management is a cure-all is ignoring the real problem.
X doesn’t lack a default toolkit, it just doesn’t have a toolkit. That’s not for X to provide.
Yes, and in my opinion, that’s bad. It’s bad for developers and it’s bad for users.
Interestingly you touch on another important point but seem to have missed it yourself; there is no such Operating System as “Linux”. There are hundreds of Operating Systems all based on Linux. It isn’t good for developers and it isn’t good for users.
The vertical integration you mention is something that the operating systems that use X have to do..
No, the vertical integration I’m talking about is something the developers have to do. I’m talking about reducing redundency and implementing features at the correct level in the software stack. I’m talking about producing useful APIs that mean that software can talk to the correct components they need without having to go through five layers of indirection to get there. I’m talking about communicating amongst themselves effectivly so that the right hand knows what the left hand is doing. I’m talking about reducing redundency and flattening the software stack. Agreeing standards and sticking to them. I see this, others see this and the KDE & Gnome developers see it too. What do you think Portland is all about?
An interview with 5 developers. If your OS of choice has at least one developer at least as capable as one of them you should consider yourself very lucky.
I’m not sure if that was an attempt at some sort of silly personal attack, but I have complete faith in any of the core developers who hack Syllable. Personally I’d happilly place Arno Klenke right alongside David Zeuthen and Waldo Bastian. If we had two Arnos we’d be done by now.
“It does, if the user downloads the correct version of their package for their distribution. What if there isn’t a suitable package for their system? Will the Redhat RPM work on Mandriva? ”
Well, if there’s no suitable package, there’s no suitable package. The user can ask the developer for a package for their system or pick a more popular system, where a package is available (yes, installing a non-popular system like Puppy won’t make things exactly easy for the user). I’m sure there are a lot more RPM packages pre-compiled and working for Fedora than there are for hobby OSes, but still less than OS X and Windows. This doesn’t mean these GNU/Linux systems are not ready for the desktop, at worst it means that they’re still not very popular. Two words, chicken, egg.
“What about applications which don’t automatically create menu shortcuts; the vast majority of the executables on my Linux machine don’t have a KDE menu entry in fact. Gmplayer is one very useful GUI application that doesn’t creates on automatically.”
You should point out the freedesktop SPEC to the Gmplayer developers so they can get with the program. If Gmplayer came with your system, it would be nice to CC the report to them, so they can fix it until the Gmplayer developers do.
“What about applications which don’t come as packages? Acrobat Reader, Firefox and OpenOffice.org: there are plenty of examples. ”
These applications are probably thinking of “Linux” as a system, instead of thinking of “Fedora Core”, or “Ubuntu” or whatever. This is why they don’t provide packages, and they’re failing: it’s their fault for creating a bad experience for their users. What they do, however, is provide a package that should work on a wide variety of systems that use GNU and Linux at their core. It’s great that they also provide that, but that’s no excuse for not supporting systems like Fedora and Ubuntu specifically, with RPM and DEB packages.
“What if something goes wrong and the user has to fix their machine? Believing that package management is a cure-all is ignoring the real problem.”
If something goes wrong the user will likely contact the software or driver provider for a solution, or a list about his system that he heard about (like ubuntuforums.org, etc). I don’t think changing the directory hieararchy will help at all, as most desktop users absolutely hate fixing their computers. For example, I think the Windows structure makes it harder to fix the system for the advanced users that fix their system themselves.
“Yes, and in my opinion, that’s bad. It’s bad for developers and it’s bad for users. ”
No it’s not. (Yay! It’s so much fan to do this, not post arguments at all. Less typing).
“Interestingly you touch on another important point but seem to have missed it yourself; there is no such Operating System as “Linux”. There are hundreds of Operating Systems all based on Linux. It isn’t good for developers and it isn’t good for users. ”
Linux is not an operating system, it’s a kernel used by several operating systems, used in video games, mobile phones, handheld devices, supercomputers, servers, desktop computers and more. From a developer’s point of view supporting all these OSes is impossible, but supporting the two or three most popular desktop OSes is not impossible. And it’s only getting easier as time goes by.
And by the way, this is not a problem with FreeBSD. You mentioned the BSD’s are also not ready for the desktop, but when it comes to software installation it has a standard way of doing it for many years now (just like Debian and Fedora).
“Agreeing standards and sticking to them. I see this, others see this and the KDE & Gnome developers see it too. What do you think Portland is all about?”
You’re never gonna force people to stick to standards. There are thousands of packages for OS X that don’t, for example see darwinports.
The goal is to make sure more and more developers follow the standards. And this is something that some serious companies and people are working on.
“I’m not sure if that was an attempt at some sort of silly personal attack, but I have complete faith in any of the core developers who hack Syllable. Personally I’d happilly place Arno Klenke right alongside David Zeuthen and Waldo Bastian. If we had two Arnos we’d be done by now.”
No personal attack at all (I can’t even see how that’s how you understood it), I just hoped you’d read a little from the people making the so-called brain damaged design decisions. That Jim Gettys in particular is a real idiot, as he was one of the two people that set the early design principles of X in 1984.
Just one point I’d like to make
“And by the way, this is not a problem with FreeBSD. You mentioned the BSD’s are also not ready for the desktop, but when it comes to software installation it has a standard way of doing it for many years now (just like Debian and Fedora).”
What if I use DragonFly BSD, or OpenBSD? FreeBSD is no better in many regards than Fedora, although at least the BSDs are better at communicating between themselves and making an effort to be compatible as much as possible.
[i]”I just hoped you’d read a little from the people making the so-called brain damaged design decisions. That Jim Gettys in particular is a real idiot, as he was one of the two people that set the early design principles of X in 1984.
“[i]
I think you misunderstand me. I’ve never called any developer Brain Damaged. They’re all great developers, really talented people. What is brain damaged is trying to use these software stacks way outside their original target platforms. X is great if you need a way of displaying a remote GUI from a big UNIX server on your workstation, but it was never designed with a single user personal computer in mind. I really don’t understand this emotional attachment people have to X; it’s just a display server. I fairly simple one at that; the upper layers arn’t even covered by it. Why do people get so offended by the idea that X may not be suitable for what they’re trying to use it for?
“Why not ask why businesses are dropping commercial UNIX variants in favour of Linux?”
You know, most of those businesses are choosing Linux with commercial support, which I’m sure was his point.
“Mark my words: Assuming the project will continue to evolve, Syllable will eventually become as bloated as Linux is. And before that happens, it will probably not be useful.”
Mark your words? Who are you, anyway? But this point is laughable: it’s making the assumption that all projects are coded in the same way. It’s like saying, “any office suite tat gets to OpenOffice.org’s level of functionality will be just as bloated” — absolute nonsense. OpenOffice.org is a big slow hulking mess because of its coding, not because of its features.
Even now, they’re finding massive overengineering and excessive memory usage in Gnome. So you don’t believe anyone can do any better? I hate to break it to you, but there’s still sloppy coding in some of these big projects, and fundamentak design decisions that lead to massive wastage. Syllable can, and is already bypassing that.
“Running Syllable is like going back in time. You could might as well install Red Hat 2.0 or something similar. I know, it’s not a finished product, but why starting it in the first place? Obviously, these guys are extremely talented programmers. So why didn’t they spend their effort and knowledge on improving Linux, or even BSD?”
Because Linux and BSD simply weren’t designed for what they want to do. Can you get sub-10 second boot times in Linux? Sure, if you want to spend alot of time fiddling with the startup scripts.
“Is it really worth beginning from scratch with the whole OS when they will, eventually, end up with the same bloated mess as any other modern OS?”
What makes you assume that they will end up bloated? This is entirely contrary to their goal.
“About the UI, I don’t know where to start. I said it already: it’s like going back in time. Why on earth does this OS put the maximize button to the left of the minimize button?”
Whats the difference? Syllable has themes for its window borders, you can use a win95-like one with minimize button all the way on the left if thats what you want.
“Obviously, the OS boots very quickly. That’s because it doesn’t do much yet.”
No. It boots very quickly because it is designed to boot very quickly. Unlike Linux developers, the Syllable developers realise that computers, especially laptops, need to be turned off every once in a while and they want to get you back to the desktop as fast as they can.
“Because Linux and BSD simply weren’t designed for what they want to do. Can you get sub-10 second boot times in Linux? Sure, if you want to spend alot of time fiddling with the startup scripts. ”
Actually that’s incorrect. The startup scripts used currently can be slow, but it’s not an inherent design flaw of GNU/Linux, it’s a design flaw of the init scripts used at the moment. Work is being done in that area.
“”Because Linux and BSD simply weren’t designed for what they want to do. Can you get sub-10 second boot times in Linux? Sure, if you want to spend alot of time fiddling with the startup scripts. ”
Actually that’s incorrect. The startup scripts used currently can be slow, but it’s not an inherent design flaw of GNU/Linux, it’s a design flaw of the init scripts used at the moment. Work is being done in that area.”
I’ll have to disappoint you. Syllable’s boot speed is not due to its startup scripts, but due to its driver architecture. Something that Linux and BSD simply don’t have, although we’re achieving this speed by making use of the very same drivers, ported to our architecture.
Kaj
“I’ll have to disappoint you. Syllable’s boot speed is not due to its startup scripts, but due to its driver architecture. Something that Linux and BSD simply don’t have, although we’re achieving this speed by making use of the very same drivers, ported to our architecture. ”
I wasn’t speaking about why Syllable is fast to boot, but why some GNU/Linux are currently slow to boot. With a new init system GNU/Linux systems can boot considerably faster than they boot now was all I was trying to say.
I have tried Syllable with Qemu so I can’t speak about it’s booting speed. Perhaps it will always be faster than GNU/Linux, but there’s a lot of room for improvement in that platform that doesn’t require changing the driver architecture of the kernel (Linux).
You don’t have to be so defensive about Syllable btw. I don’t have anything against it, other than my lazyness to try it for real (I booted the qemu image and played with it for a whole 15 minutes a long time ago). My comments in this thread are all about about GNU/Linux and Unix-like systems, which I guess a lot of people will be comparing with Syllable for as long as it exists.
[re. installing drivers] Another way of doing it, which is my favorite btw, is this:[snip double-click install steps]
A driver “package installer” (for Syllable) is as simple as a one-line shell script to copy the file to where it belongs.
Edited 2005-12-31 03:40
Yeah, tell that to Apple.
I used to use Mac OS X as my desktop at work (for web dev and some remote admin stuff) but recently switched to GNU/Linux + IceWM on older and slower hardware and now have a better desktop for what I need, and am much more productive.
For one thing, I use a programmable ergonomic keyboard (with no “command” key), and Apple’s using their command key for everything Aqua instead of the normal control key was a pain in the neck. I had to keep switching back and forth between key mappings to get my work done.
Also, no virtual desktops on OS X, which was making me much less productive.
The Mac was a dual 1 GHz (or so) G4 with 768 MB memory, but actually feels slower than the rickety old 700 MHz Athlon (512 MB memory) I’m running IceWM on.
Now, what I really can’t wait for is to be able to switch to an even slower machine with even less memory but have it run Syllable, giving me an even snappier desktop experience.
You often speak about BeOS, thinking about the past.
But *remember* that Haiku OS is right behind the corner, ready to kick other Hobby OSes ass!
(Just Trolling guys
I love hobby oses scene!!!)
“Haiku OS is right behind the corner, ready to kick other Hobby OSes ass!”
I sure hope so!
There are only a few mainstream OSes and this won’t change any time soon. They are MS Windows, Linux/BSD, Mac OS X & Zeta. In the coming years, there is a good chance that Haiku-OS & ReactOS will be added to the list.
Syllable, & similar OSes, will only be for the hobby user. Some newer OSes may be faster, or better programmed (ie: Syllable) but will *never* (or at the very least not anytime soon) get widespread popularity. ie: Linux is too popular to be beat by new OSes. Everybody is on the Linux craze, so even if a new OS is better and/or faster overall – it will never dethrown Linux (or be very hard to do).
Isn’t it better to contribute to an OS that many users will use/benefit from (ie: Linux, Haiku-OS or ReactOS) rather than creating a new one which hobbyists will mainly play with?
I see many new OSes springing up and trying to show the world that theirs is the best, but does it matter if they only attract a small userbase? I believe that having the best OS is nothing without getting people to use it.
What I’m saying is that if Syllable developers went & created their own Linux distro (using their knowledge to make it fast, efficient, etc.) that there is a greater chance they would benefit many more computer users out there & achieve greater popularity/recognition (ie: look @ PCLinuxOS). I personally prefer Haiku-OS or ReactOS, as there are too many distros for Linux already.
In all fairness, I tried out Syllable (through VMPlayer) and it truly is fast & responsive. I was impressed with it. The programmers are doing an excellent job creating this OS. There were only a few applications to try out, but hopefully many more will come along. It truly is a terrific OS to play with (more as a hobby OS as of now), maybe one day, with more applications, it will achieve greater popularity. Syllable OS looks very promising and definately got my attention. Good work guys.
Edited 2005-12-31 17:54
My comment about “Popular OSes” was a bit strong. I was going to edit it, but found out that editing only works upto 20 mins. after posting.
So, to clear it up, I’ll say Syllable is a top notch OS. It will certainly interest the hobbyist out there and is worth checking out. Hopefully, one day it will become more popular or gain more attention.
I also just learned that Syllable is a “massively multithreaded C++ design.” Performance should be amazing with dual core cpus or SMP systems (once programs are coded to take advantage of the OS & available cpu power).
I believe BeOS is also a multithreaded C++ design & it rocks. Syllable reminds me of an earlier version of BeOS (look & feel, design, multithreading).
I can see Syllable becoming as good or better than BeOS – but only once applications are ported over (or developed) and there is greater support of hardware. Probably quite some time before this may ever happen. But I’ll stay positive and continue to watch Syllable’s progress.
Edited 2006-01-01 01:12
na
Browser: Mozilla/5.0 (compatible; Konqueror/2.2.2; QtEmbedded/240×320)