Syllable 0.5.5 is now available. This release includes new and updated drivers and a large number of long overdue kernel enhancements. The full changelog is here, and it can be downloaded from Sourceforge.
Syllable 0.5.5 is now available. This release includes new and updated drivers and a large number of long overdue kernel enhancements. The full changelog is here, and it can be downloaded from Sourceforge.
so much potential…keep it up
frankly seems like just anotherLinux distro: change-log mentions importing from the Linux kernel, I recognize the icons from RedHat 8, and I remember configuring the border style for KDE once. really i have trouble seeing this as a seperate os.
How about the fact that it isn’t in any way Linux?
It would be easier to modify it to run BeOS apps than it would to run Linux apps.
It’s much more similiar to BeOS than Linux….and it’s neither of the 2.
Why try and reinvent the wheel? By utilising the other opensource projects they can save heaps of time, allowing them to concentrate on the more important things, things that are specific, that define Syllable.
As for the GUI of Syllable, that comes later. The 0.5 series is meant to create a stable and workable base.
but my test system has an intel i810 graphics chip and for some reason it fails the vesa2 test and the installer never gets launched. Too bad is it looks like a neat os.
“frankly seems like just anotherLinux distro: change-log mentions importing from the Linux kernel, I recognize the icons from RedHat 8, and I remember configuring the border style for KDE once.”
Some drivers are based on Linux ones, primarily because Linux drivers are easy to port and usually very well tested. Similarly, the BlueCurve icon style is clean, simple and approachable — good reasons to use it.
That doesn’t remotely make it Linux. It’s a standalone OS with its own kernel, display system, filesystem layout, applications and more.
“really i have trouble seeing this as a seperate os.”
Then look into it more 🙂 Screenshots don’t tell the whole story. Syllable includes superb design and features which overcome some of the problems Linux has been having, namely unified driver and software installation.
In Syllable, you just drop a driver into the appropriate directory, reboot and you’re sorted. No faffing around. Same with software — copy into place and use, without dependency hassles or working out where all the files go.
Various Linux distro vendors are working to solve these problems, but it’s complicated and Syllable offers many compelling features right now. Sure, it’s still young, but has a great deal of potential — plus it boots at an outrageous speed, is low on resources and zips along.
Give it a try!
Outside of GNOME I really don’t like the BlueCurve icon set. But as this is a growing project, and the full potential of it hasn’t been explored graphically, I am sure the Syllable developers will get some cool icons.
Cap’n Pinko wrote:
> frankly seems like just anotherLinux distro
Nope. Totally different kernel and OS SDK.
> I recognize the icons from RedHat 8
Yeah, I guess they like BlueCurve.
timh wrote:
> It’s much more similiar to BeOS than Linux
Yup. Though, I get the impression from the Syllable folks that they’re a little BeOS-phobic (though many are likely ex-BeOS’ers). It seems like the BeOS community really wants BeOS back, and the Syllable folks don’t want to be the ones bringing it back (even though Syllable is very similar to BeOS in many ways).
anon wrote:
> Why try and reinvent the wheel?
They’re not. It’s a fork of Atheos for one thing. For another, have a look at all the GPL/LGPL code and libs that go into Syllable.
Michael wrote:
> In Syllable, you just drop a driver into the appropriate
> directory, reboot and you’re sorted. No faffing around.
*This*, IMO, is a huge deal. Folks have told me, “Oh, drivers aren’t a problem with Linux. Just copy the appropriate files to the appropriate directories and modprobe it, well, if you have the right kernel version”, but installing drivers on BeOS was how nature intended.
I’d love to be:
– running an OS,
– plug in a new device,
– need the driver,
– go to the proper website to download a .tar.bz2 file,
– unpack it and drop it into some driver directory,
– and (possibly after a reboot) BOOM — the device is now recognized and working.
I tried Syllable at 0.5.3 (I think) but since I’m no OS dev I’m waiting for more polish. I remember fonts not looking all that great (though, they did look *appropriate* for a 0.5.3 release I suppose).
Why everyone is not building on top of the Linux Kernel instead of writing a new kernel from scratch?
You could save time and you would just use the Linux kernel and write everything else as you want.
You could agree or not that Linux distributions or some program suck, but the kernel does not suck at all and if you think so, anyway you are not going to write one better.
Michael Saunders wrote:
In Syllable, you just drop a driver into the appropriate directory, reboot and you’re sorted. No faffing around. Same with software — copy into place and use, without dependency hassles or working out where all the files go.
This would be great for Linux distros.
There is nothing wrong with using the Linux kernel, except that any OS based on it would be unavoidably just another Linux distro.
Well it would be another Linux distro correct. But if you build a new operating system upon it it would not be just another GNU/Linux system and that is what most people think of if the word linux is mentioned.
Why everyone is not building on top of the Linux Kernel instead of writing a new kernel from scratch?
For the same reason hobby painters don’t automatically reach for paint by numbers kits. Some people like starting from scratch, as well as not dealing with other people’s assumptions/decisions, and get more pleasure from the hobby that way even if it takes longer to achieve something.
>> but installing drivers on BeOS was how nature intended.
I’d love to be:
– running an OS,
– plug in a new device,
– need the driver,
– go to the proper website to download a .tar.bz2 file,
– unpack it and drop it into some driver directory,
– and (possibly after a reboot) BOOM — the device is now recognized and working.
What so nature = windows. LOL. 🙂 Noooooooooooo. Just playing. I Love BeOS/Windows/Linux/BSD/Solaris/Netware…..if I can make money out of them. But it seems like a very good proyect. I would like to know the requirements for it. How heavy is it. Can I try it in an old box?
Because the Syllable developers have already said they don’t agree with the Linux philosophy and are trying to do things a different way.
primarily because Linux drivers are easy to port and usually very well tested
I thought NetBSD had the best and most portable drivers?
Easy, you can take code and use it however you want but the Linux kernel has major archetecture design flaws.(Which is why I support GNU/Hurd over GNU/Linux) Millions of dollars have been spent to correct the problems that Linux had. As we all know, Linux was origionally a hobby OS that was just something better than Minix. It’s very different from what it was at first. There’s also A LOT of BSD code in Linux (all you bsd & linux fighters should not be fighting as linux utilizes bsd technology)
Furthermore.. a system can be tweaked can run in a particulay way.(Yeah, I’ve criticised toasters running linux) This is why they have embedded companies that take linux, heavily tweak the kernel and sell it to others. Basicly I think they are just using the trademark as a selling point and getting free work. I would not be surprised if Red hat creates a commercial environment and uses the linux kernel or vice versa.
Basicly, it may be more difficult to take linux and mold it to what you want rather than just building your own OS for your specific needs.
I hope I did not rant too much or gone off into other subjects..I really hope Windows dies. (I don’t want to get microsoft certification basicly)
the linux kernel sis being mentiond alot so here is my view on it. i dont htink thre shoul dbe one kernel….i think thye should divide it up for diferent aims….one foucosing on desktop…..one on embeded…..one for servers….and so on and so forth…..insetead of taking the one and patching the hell out of it ot do all this things……jsut my opinion
my test system has an intel i810 graphics chip and for some reason it fails the vesa2 test and the installer never gets launched
Many i810 systems do not include a Vesa BIOS. The only way to drive an i810 chipset is with an i810 video driver, and Syllable doesn’t have one yet (The i810 is not the same as an i830/i855 so that driver won’t help)
frankly seems like just anotherLinux distro: change-log mentions importing from the Linux kernel, I recognize the icons from RedHat 8, and I remember configuring the border style for KDE once.
Both Syllable and Linux are GPL kernels. It makes sense for us to import code and drivers from Linux where appropriate, but if you read the ChangeLog the imports are small, things like primitive types and inline assembler E.g. the atomic functions. The actual drivers themselves are good quality and tested and with some modification to work with the Syllable driver management and busmanagers they are a very reliable source of hardware support.
Yes the icons are Bluecurve because they’re good icons but that isn’t the final look of Syllable because it’s only at version 0.5.5. We have other things to do than make pretty screenshots. As for using the Photon or BeOS window decorator on KDE, that would have been one mean trick. You might have configured a KDE theme that looks similiar to the decorators in Syllable but Syllable has it’s own display engine and GUI (The appserver and libsyllable).
You are judging an entire codebase on the basis of a screenshot. A 1024×768 JPEG doesn’t tell you anything at all about what is underneath. Why not grab the LiveCD and try Syllable instead?
Why everyone is not building on top of the Linux Kernel instead of writing a new kernel from scratch?
Well first of all Syllable didn’t write the kernel from scratch, we inherited it from AtheOS where Kurt did write most of it from scratch. AtheOS was started in 1999.
There are many good reasons not to use Linux. Architectural differences are one. Philosophical are another.
You could save time and you would just use the Linux kernel and write everything else as you want.
You could save time yes, but it would be a trade off: You’d have to live with the many braindead architectural decisions of the Linux designers. The Linux driver model sucks on purpose, for example. There is no stable ABI that kernel modules can use.
The Syllable kernel is intended to be used to run a desktop machine. It is lightweight. It is multi-threaded (Even some drivers run in multiple kernel threads) and it is fully re-entrant. The ELF loader is part of the kernel, which in conjunction with a careful design gives us our stable ABI that drivers can use. Linux takes the view that functionality can be pushed up to user-space deamons that do the work. Syllable runs tasks like these in kernel threads; there is very little reason not to and it keeps things much simplier.
So there are reasons why we prefer our kernel over Linux, even if ours needs some care and attention.
“I thought NetBSD had the best and most portable drivers?”
Most portable, perhaps, but best? NetBSD’s drivers are – in general – very robust and well behaved, but simply because of the smaller userbase, they’re not tested by as many people and in as many configurations as Linux drivers.
>>Why everyone is not building on top of the Linux Kernel instead of writing a new kernel from scratch?
Why??? Why is not everyone using windows? LOL
What kind of question was that. For the same reason that Linux started. If somebody has another approach, why does it have to be Linux? We like linux, but that doesnt mean that everybody must use it, or that they cannot come with their own ideas. That sounds more like a Microsoft theme. Freedom to innovate, but only if is windows. What the hell. And please dont come with the argument of reinventing the wheel. Of course it is necessary. If not radical and better ideas will never be developed.
As long as it’s POSIX compatible and is able to use GNU tools, there really shouldn’t be too big an issue in creating new OSes. The great thing about Linux’s large open source repository is that any Unix-like system can benefit from it, including MacOSX. Each new OS that use open source adds new development mindshare to open source.
Even if Syllable goes nowhere, it can be an inspiration for some ideas in Linux. For instance, NextStep isn’t X-based, but it proved that Unix could be User friendly. BeOS and AmigaOS also provided numerous inspirations.
Dropping devices into the device folder to have them work may seem like a “fluff idea”, but I remember not long ago that “auto mount” was a fluff concept. Now it’s a part of virtually every Linux distribution.
Every single release of this thing has never worked for me. I look at the changelog and it mentions VMWare. So I download the ISO and tried to run it in VMWare. GRUB boots and loads the kernel. A bunch of driver info rolls by and we freeze forever on the line “0:init::init(-1) : Mount boot FS: @boot iso9660” Do these guys test any of this stuff before releasing. I would think that, because VMWare presents pretty standard hardware to the OS, it will at least be able to boot. My guess is that some guy, using an existing install created through some obscure method, got it to boot in VMWare and everyone said “Yeah, that’s gold baby! Go with it!”
Why everyone is not building on top of the Linux Kernel instead of writing a new kernel from scratch?
Ask Mr. Tanenbaum
So because you fail to read the FAQ which contains an entire section on how to use Syllable with VMWare we don’t test anything? Plenty of people are using Syllable on VMWare; how do you think the VMWare video driver came to be? For your future reference: http://www.syllable.org/faq.php?cat_id=3&cat_name=Vmware The last FAQ is relevent to you.
CD-ROM is on IDE 1:0, and Virtual HD is on IDE 0:0. Still the same shit. I guess your incredibly technical “IDE driver doesn’t like virtual CD-ROMs” analysis needs more work. I’m sorry for expecting OSS software to work without having to read reams of FAQs, HOW-TOs, erratas, etc.
“Ask Mr. Tanenbaum
”
that would never work. he is obsessed with micro kernels by his own admission and its a well know fact that linus doesnt believe in that approach.
search linux vs tanenbaum in google
Of course, your Windows and Linux have always worked perfectly, haven’t they? And if not, people have always helped you because you asked nicely, didn’t you?
Of course, your Windows and Linux have always worked perfectly, haven’t they? And if not, people have always helped you because you asked nicely, didn’t you?
They’ve all been able to boot.
Looks like that part of the FAQ is out of date. Instead of “ata0=nodma”, one should add “enable_ata_dma=false” to the list of kernel parameters – that worked for me in both VMware and Virtual PC.
Looks like that part of the FAQ is out of date. Instead of “ata0=nodma”, one should add “enable_ata_dma=false” to the list of kernel parameters – that worked for me in both VMware and Virtual PC.
Wow, a non-documented and highly obscure fix to a bug caused by improper default settings. Of course it boots to a terminal window spewing out “/usr/ruby/lib/ruby/1.8/rational.rb:1: Invalid char ‘377’ in expression” over and over until it crashes and I am left with a green screen that does nothing. I feel Linux deja-vu. Time to have another twinkie.
While I find it terribly amusing that you’re so angry about it all I’ll make you a fair offer; if you pay for a licence for a current copy of VMWare so that I can run Syllable upon it I’ll debug the problem and fix it.
No? Then I guess you’ll have to wait until someone else is kind enough to spend their time and effort tracking down the bug and fixing it.
That’ll be a “no” then will it?
Waste your time you say? Don’t make me laugh. What would be so important that you would put it off in order to post angry rants on OSNews and fiddle with an Operating System which is currently at release 0.5.5 I.e. pre-alpha? I am sincerely sorry (I really am) that Syllable does not yet work for you and I hope that we’ll get these bugs fixed over time, but your attitude that we *must* help you and it *must* work is rude and confrontational. We all spend quite enough time working on Syllable as it is and it is time we do not get paid for and do not ask you to compensate us for.
If something doesn’t work then the best you can expect is a friendly ear and a “Sorry that doesn’t work”. Should we actually consider the bug bad enough that it is worthy of our own donated time and effort to find and fix should be considered a favour, not a right. Without access to VMWare nor the time to spend, not considering the fact that many users on the syllable-developer mailing list are happilly using Syllable on VMware means that this bug in particular is way down on our list of priorities. It will be fixed before we release Syllable 1.0 but so will thousands of other things.
As I say, I’m sorry that Syllable didn’t work for you this time.
“I’m sorry for expecting OSS software to work without havi ng to read reams of FAQs, HOW-TOs, erratas, etc.”
Perhaps if you’d read the version number first, ie 0.5.5 (half way through development), you’d realise that it’s NOT intended to work perfectly for everyone. It’s a mid-development system for people to try out, improve and enhance.
“know what you are doing and waste my time.”
Do you realise how arrogant that sounds? Nobody forced you to try Syllable. Nobody said it was a finished, perfected product that would work without any problems. Maybe your time isn’t that valuable if you spend it flaming people on here for their hard voluntary work, instead of making reasonable suggestions and sending bug reports?
I know it’s frustrating when something doesn’t work. Even open source software. But your approach doesn’t help in the slightest, and just sounds bitter. Grow up a little and appreciate that people are doing this for free and ARE willing to help if you’re reasonable.
The LiveCD ran fine on my IBM thinkpad. It is nice.. fast and lean. Boots in a few seconds. And its seems to live completely in memory. It is so nice and fast. The GUI is getting nice. It feels more “solid-state” than any linux distro, if you know what i mean. It doesn’t feel like there are so many layers there slowing things down.
Have you thought about including a web browser in the livecd? Would it be possible to port over Firefox? Throw some games and demos on this thing too. How about at least the Amiga “spinning ball” demo?
Anyway, looks good. This thing has a ton of potential.
In fact it doesn’t ”live completely in memory”. Only some configuration-files are stored in memory, at this time.
There is a webbrowser. An old one
but it’s there. Look for ABrowse in Applications. Arno has just released a new version, something that everybody though was impossible. That one is not on the LiveCD.
I’ve been thinking a lot, about including some games, but didn’t know if it was something people wanted.
Poor Vanders. I don’t understand why a guy which like to be spoon feed wanted to try Syllable. It is OK if he didn’t bashed others but to guy like him, it is better to buy Windows and use it.
Anyway, I love the boot time of Syllable compared to Linux. I believe Linux should find ways in boot steps rather than depend on current init style which always will increase the boot time when you add the number of application/services to run, whereas supposedly, those services/application can be defered after the user get their desktop, sometimes.
“It feels more “solid-state” than any linux distro, if you know what i mean. It doesn’t feel like there are so many layers there slowing things down. ”
no need for such silly gripes against linux. every software should be built on layers. its a fundamental concept of software development called abstraction.
syllable is improving but its nowhere near the amount of development of maturity of Linux anywhere in the industry. stop dreaming and start working on what you believe.
I will be open to all improvements as long as you guys talk about your virtues rather than attacking other operating systems.
its is especially amusing to see all those hobbyist operating systems evangelists attack Linux while ignoring bigger fish like Windows.
it happens with syllable,haiku and skyos. I hope you guys learn to appreciate the complexity in developing an operating system in a solid state and dont fool yourself into believing that you are just about there. Linux has been bordering on this stage for a long long time. the hard parts are the last parts
Well, actually Linux has no GUI, it is just a kernel. I guess I was comparing it to KDE and GNOME, which seem bloated and slow by comparison on Linux, FreeBSD or any other OS they run on. That’s what I mean by solid-state. The GUI of Syl. is fully integrated into the OS, whereas, x.org + desktop manager seems like a big hack.
But try it if you don’t believe me. This is the first time I’ve seen the project and i just downloaded the livecd today. But I’ve tried maybe 30 distros in the past three years, so i know what I’m talking about.
> the hard parts are the last parts
Yeah, but since both Syllable and Linux are GPL’d (not to mention the GNU stuff), Syllable can benefit from some (much?) of the “last parts” work going into GNU/Linux.
—
Just how much overlap is there between Syllable and GNU/Linux?
“Ask Mr. Tanenbaum
”
that would never work. he is obsessed with micro kernels by his own admission and its a well know fact that linus doesnt believe in that approach.
search linux vs tanenbaum in google
That’s exactly what i mean: by reading that you can understand many flaws in linux kernel implementation. John asked why not using linux kernel, well, Mr Tanenbaum can tell u a bunch of motives for not using it
Just how much overlap is there between Syllable and GNU/Linux?
We can benefit by porting Linux drivers to the Syllable kernel, which gives us very good hardware support (Considering our size and developer base). We also use Glibc, the GNU userland tools (E.g. coreutils) and the GNU development toolchain (E.g. Gcc), but these things are not really Linux.
> You could agree or not that Linux distributions or some
> program suck, but the kernel does not suck at all and if
> you think so, anyway you are not going to write one better.
Ironically, the man who disproved this point was Mr. Torvalds.
Why not make a Xlib/Xt wrapper for Syllable GUI and therefore inherits all X-based Unix/linux programs ?
For me, not use a X server/window manager is reasonable but ignore the X legacy is not smart.
“Why not make a Xlib/Xt wrapper for Syllable GUI and therefore inherits all X-based Unix/linux programs ?”
Sure, go ahead and get coding — it’d make a lot of people very happy 🙂 Syllable is a cohesive system, and native, integrated apps are the priority right now, so nobody’s too eager to have an X layer shoehorned in.
Even if one existed, it almost certainly wouldn’t be part of the base system (I think Vanders talked about this recently).
Michael is right, I wouldn’t ship any form of X or legacy X toolkit in the base distribution. The main objection to X are actually the number of toolkits; too many cooks spoil the broth and too many toolkits spoil the OS. I want to encourage the use of the Syllable GUI API as much as possible and give it time to become established as a viable toolkit in it’s own right, becoming the primary toolkit for anyone writing or porting an application for Syllable. Porting an X toolkit would be a sure way of ensuring an early death of the Syllable API, so I’d like to avoid it.
> Just how much overlap is there between Syllable and
> GNU/Linux?
We can benefit by porting Linux drivers to the Syllable kernel, which gives us very good hardware support (Considering our size and developer base). We also use Glibc, the GNU userland tools (E.g. coreutils) and the GNU development toolchain (E.g. Gcc), but these things are not really Linux.
I also see from the FAQ that Syllable is largely POSIX-compatible, so right there that’s a large swath of command-line programs that come along for the ride as well.
Definitely huge potential here. I hope to see the Syllable docs get more and more mature and complete with each new release. Good docs are, IMO, essential for new software projects to succeed. What would be great would be to have a “modify this page” button at the bottom of each web doc page that would serve you up the html (or whatever format) source, let you edit it (similar to a wiki), and then allow you to submit your proposed change back to the documentation maintainer for inspection and possible inclusion.
Holy mackerel that’s a good idea! I’m going to sniff around and see if someone’s implemented it already.
Good docs are, IMO, essential for new software projects to succeed. What would be great would be to have a “modify this page” button at the bottom of each web doc page that would serve you up the html (or whatever format) source, let you edit it (similar to a wiki), and then allow you to submit your proposed change back to the documentation maintainer for inspection and possible inclusion.
That’s actually something we’re hoping to have. It would sort of be a large community maintained and edited Syllable reference book. The Syllable Users Bible ( http://www.other-space.com/sub/ ) does some of this already.
Actually, the Syllable User Bible does all of this right now.
Few years back, there are thread about BeOS replacement here at OSMEWS. I’ve post and wrote that I believe Syllable will be will ahead of BlueeyedOS or OpenBEOS (and maybe Cosmoe – I can’t remember very well). Today, I’m glad to see that my wish came true but at the same time feel sad because OpenBEOS (Haiku) are crawling too slow whereas BlueeyedOS look stagnant.
Anybody know how is the life of original AtheOS developer? Did he remain addicted to flying his plane?
Actually, the more people are experimenting with developing Free/Libre Open Source Operating Systems, the more information there is available on how to improve them, how to make them work best on the sort of hardware that makes a lot of the boundary conditions of the older hardware and older software irrelevant. You know, if you have a CPU that updates the cache every few microseconds, that puts a lot of stress on algorithms that expect the updating to happen every second …
Who knows, there’s nothing set in concrete or carved on stone tablets (quite bad for headaches, or so I’ve heard, and absolutely terrible for stomachaches
that the next great algorithm is bound to occur in any of the major F/LOSS OSes – it could occur in any of the smaller projects that “cluster” so to speak, around the F/LOSS community’s outskirts.
> Actually, the Syllable User Bible does all of this right now.
Wha?!
Did you just do that today Brent?
Nice!
Though, the front “table of contents” page could use the same capability to be edited.
Say, did you write that functionality yourself, or did you use some third-party content management software?
“Anybody know how is the life of original AtheOS developer?”
I think Kurt is writing games now, and doesn’t appear to be in touch with any of the current Syllable coders!
Tried 0.5.4 and that was the first distro. that really worked for me (on my hardware) and I was thrilled! Not ready to commit to it, yet, of course.
Ah! Download finished while writing this… now to burn and try it out.
johnMG wrote, “Did you just do that today Brent?
Nice!”
No, it’s had that capability for months. I’m sure I posted about it on syllable-developer.
“Though, the front “table of contents” page could use the same capability to be edited.”
No. I don’t want people posting “F*CK THE WORLD IN THE A*S” on the front page of SUB. Folks can always request a tutorial on the “Request a Tutorial” page, which I can then integrate into SUB.