The KHTML port is a total dead end; maintaining it is a nightmare. I hope we'll have a port of Gecko within the next 12 months, which will hopefully be much easier to maintain as it is designed to be portable. Personally I think we need stronger debugging tools before anyone tackles a large codebase like Gecko, so I intend to work on the development toolchain some more and then maybe tackle Gecko.
I would have also liked to see information about what would be involved in developing applications for Syllable. Is there a well-documented API? What about support for multiple languages?
All the information you'd like is on the website but we're Slashdoted, so I can only ask you to try again in a few days time. The API is documented, there are some tutorials, example code is abundent and we're happy to answer questions in the forums and mailing lists. Multi-lingual support is currently in the CVS version which can be compiled if you want it, and will be officially available in the next release of Syllable.
Sorry, the page is hosted on Sourceforge. I didn't know you could Slashdot Sourceforge like that. We can't fix it because we're not Sourceforge administrators.
AtheOS was written in 100% assembler as a pet project by the guy who wrote it
The kernel is written in C. The high level stuff is written in C and C++.
He (and others) later built some POSIX, KDE and GTK API mappings..
The AtheOS kernel has always been about 95% POSIX compliant. There are no KDE or GTK API's for Syllable; it has always had it's own C++ API and appserver.
Question: Is there any way to use Linux device drivers with this os?
Almost all of the drivers are Linux drivers, originally. They have been ported to use Syllable API's, but they're not that different. If you know enough about device drivers you can port a driver from Linux in a couple of days. Some people have ported drivers in a matter of hours.
Posting as AC is just a normal thing for me these days, but if it offends you then I can forgo it this time. Voting and voters rights is one of those things that I believe in and I don't mind being passionate or even downright rude and offensive about it if I have to. People with their heads in the sand wind me up.
So while you're here please tell me, what is it with people who dismiss anything an AC says? Are the opinions any less valid? Are the facts less truthful? Do the words become blurry? Is it maybe just that it is easy to dismiss a valid opinion that you might not like by waving your hand and saying "Bah! 'tis but an Anonymous Coward. Off with his head!" I don't get it.
Thats actually the solution we use with Syllable. Packages go into their own directories under/usr/ Syllable has an extra step of running a tool which creates symlinks from/atheos/autolnk/ into the various files and folders within the package directory. That way PATH only needs to contain the directories in/atheos/autolnk/ instead of all those package directories, which speeds up searches.
Re:While not exactly a clone..
on
Ten Years of BeOS
·
· Score: 5, Informative
Just to point out a few things that a lot of people might not be aware of.
AtheOS is no longer developed, and the codebase has not been updated in several years.
Syllable is our community-driven fork of AtheOS, which was started two years ago.
We're halfway through development of Syllable 0.5.4, which like all previous releases of Syllable, will rock. We support a whole bunch of hardware, have developed the codebase heavily and for those of you who were familiar with Kurt Skuans style of working with AtheOS, we have a far more open development model. All are welcome to contribute. You can even download a LiveCD if you want to give it a spin.
I think it depends on the store (And time of year. December, for example) I know that the store we get our Tesco deliveries from can and does deliver upto 10pm, and we've personally taken deliveries at 9:30pm.
Don't forget Sainsburys and Iceland, as well as some smaller outfits. Asda [Walmart] are very slowly also re-implementing store-based internet shopping, after their disastrous start with the big-warehouse model several years ago.
We currently use Tesco's but would probably switch if the Asda store in our area was up and running.
I think you mean "In America, Internet Grocery Shopping.." because certainly here in the U.K, it's pretty big business. My wife and I have been using Tescos once a week for two and a half years with no problems. Tesco and several other large stores who provide online-ordering and delivery are also turning a tidy profit in addition to their regular brick & mortar stores.
Heres some simple Bash scripts to do almost exactly that They need a re-write to make them more robust and easier to maintain, but they work. I don't know how much work you'd need to do to adapt them to Linux or a more traditonal SysV style system though.
Syllable is a mismanaged kludge? Care to back that up?
There are exactly four people with CVS write access to the Syllable CVS server. Patches come into Syllable via. Henrik, myself or Arno. I get the final say if something gets checked into CVS, and I have at several points in the past rejected or refused changes.
If you could point out how that isn't exactly identical to how SkyOS manages their development I'd like to know.
The only part of Syllable that is a problem is the desktop & launcher, and only because they're stopgap measures which will be replaced when we're ready. It's not like we've removed funtionality from the desktop in AtheOS 0.3.7
Please, if you have specific examples of "kludges", fuckups or mismanagemtn please tell me so I can fix them and make sure they don't happen. However if you've just got a grudge over something then fine, but why not just come out and say so?
That simply isn't true. You seem to confuse the development model of the Linux kernel with the development model of an entire distribution. There are many different distributions of Linux, but there are very few versions of the Linux kernel itself. You can see the same model at work in BSD. There are not hundreds of different BSDs.
The "Linux" development model works and works well, and it can easilly work across an entire Operating System
Development of AtheOS stalled for various reasons. Syllable was created over a year and a half ago by Henrik, Rick & myself to take the AtheOS codebase and open it up for further development. So while AtheOS itself is dead, the code and its spirit lives on quite nicely in Syllable. Some more developers have joined us (Arno and Kaj, with submissions from other developers such as Micheal Krugger and Hilary Cheng) We've developed the original AtheOS 0.3.7 codebase rapidly and released Syllable 0.5.2 only last week. I'd suggest you take a look at the Changelogs or just try it out to get an idea of how much work we have put into it.
But I think one thing that killed Atheos is the same thing that killed almost any alternative to X: inability to support any modern graphics cards at any resolution higher than VESA. Unfortunatley this information is locked up in X drivers that are so tightly integrated with internal complexities of X that it is impossible to extract and reuse it, despite the open source nature.
Syllable has drivers for the following graphics cards with full 2D acceleration, and the ones marked with an asterisk also support video overlays (Xv in XFree86)
S3 Virge
S3 Savage IX/MX
Trident video (VLB & PCI)
Matrox Millenium & Gx00 cards
ATI Mach64*
SiS 3xx/Xabre*
nVidia TNT/GeForce*
nVidia GeForceFX*
In fact the only notable omisions are the ATi Radeon, S3 Trio & Intel Extreme (i810), and I'm confident we'll have some support for those chipsets soon.
Porting drivers from XFree86 is not that difficult and lack of specs is a problem, but not as bad as you might think.
It's very different now. I did, once, have Syllable 0.4.5 running in Virtual PC 5 on Windows 2000, but it's complicated and I havn't had time to create the necasary patches for newer versions to fix the problems.
All I need to do is add a workaround for the broken PCI controller in VPC and identify & fix some stability issues with the installation CD. I can say that once Syllable was installed it was very stable and quite fast.
Syllable works great on VMWare, by the way. Network and audio are supported and there is now a driver for VMWare video which will be in the next release. Not much use on a PowerBook, of course.
No, because he immediately explains that it isn't an exception, it's a clarification:
Um, yeah. So in other words: There is no exception. If you read the text you quote you can see that Linus specifically states that "The kernel _is_ GPL. No ifs, buts and maybe's about it. As a result, anything that is a derived work has to be GPL'd." He does not make any exeptions for kernel modules (Which is the point of the entire discussion on the LKML)
While Linus gets tied up arguing about derived works and wether the code was originally written for Linux, he doesn't actually solve the problem of the "grey area". I don't believe Linus has thought it through totally.
Think about it in terms of the nVidia binary modules. The kernel is GPL and the nVidia wrappers are written specifically for Linux. Therefor the wrapper must be under the GPL. If the wrapper is under the GPL and the nVidia driver itself is linked with this wrapper code..what licence is the core part of the driver under? If it isn't the GPL then it can be argued that the "viral" nature of the GPL is fluff, and cannot be enforced or is easily worked around. If it is GPL, then nVidia are going to be concerned because they don't want their driver under the GPL.
This "grey area" exists because there is no clearly defined boundary defining the seperation between the kernel and the drivers. Modules are parts of the kernel which have not been linked yet. When they're required, they are loaded and linked with the kernel.
The fact that Linus states that there is no exception must worry a lot of companies out there who are producing binary drivers for Linux E.g. nVidia, or SciTech (Who started the LKML thread, after all!) Are nVidia's kernel modules under the GPL? If the possibility exists that they are then I would expect them to suddenly get cold feet over Linux.
If the kernel had a proper boundary with E.g. a set of API's that the kernel and drivers can use to communicate with each other then it would help to solve the issue of what is and isn't "the kernel". For example in Syllable drivers are ELF images which are loaded by the kernel ELF loader. The drivers are loaded under the kernels memory space but there is a very well defined API between the two, and a very clear seperation between them. Under this model I can argue that the kernel is actually being linked to the driver, so the driver can be under any licence while the kernel remains under the GPL. There is no "cross pollenation" between the driver and the kernel. Which is a good thing IMHO, if it avoids issues like the ones being raised on the LKML.
It is a fork of AtheOS. Syllable uses it's own kernel and drivers, but most drivers can and have been easily ported from Linux. It has its own GUI, toolkit, filesystem, filesystem layout and applications. It does use the GNU tools and Glibc underneath but there is no reason to use them if you don't need them.
The GUI isn't complete (Close, but not 100% complete) and the desktop is just there until we actually write a proper one, but the hardware support is pretty good (For an "alternative" OS it certainly is). The kernel is almost complete but still has a few bits to complete & tweak.
It also has all sorts of cool stuff, just to satisfy your last requirement.
Matrox were also amoung the few companies which actually provided full specs for their hardware. You can register with them and download complete spec sheets from everything from the older-than-Millenium to the G450. I can't understand why they have suddenly clamed up when it comes to the Parhaliea. Apparently there are binary only, "beta" drivers for X/Linux but no specs or open drivers. Not much use.
Any OS which requires VESA 2 support (Darwin, for instance, or SkyOS), will not work in VMWare.
Actually you're wrong. I know for a fact that SkyOS and Syllable both run fine under VMWare using Vesa2. They just run faster with a dedicated VMWare video driver, like any video hardware.
You may be thinking of Bochs, which doesn't do Vesa2 unless you patch it. Someone has done this, but running Syllable on Bochs was glacial.
No, you can manipulate a very limited, pre defined and hard coded set of meta data.
What other meta data? How about a MIME type, or icon data, or a log listing the last ten accesses to the data (Not just the last changed date). Store any data you want, in fact. Is there any real reason to store things like ID3 tags inside the MP3 file itself? It is after all just meta data, as is the filesize, the creation date and the persmissions. Why not store all the meta data in one place?
People tend to rubbish the idea of arbitary meta data simply because they personally do not see a use for them. Admittedly, the entire concept of meta-data is a little arcane and I guess it just takes a little mental gymnastics to fully grasp the concept. Either way, arbitary meta data is useful and should be encouraged.
The data is just meta-data. Your question is like asking "Why do I want to store the filesize and file permissions with the file?" The data is meta-data, or data about data. It is useful and in many cases, essential.
Consider two common scenarios for the uses of meta data. A filesystem that can store the MIME type of the file with the file itself, instead of relying on a filename extension or a file type magic lookup every time the file is accessed. You can also store things like icon data along with the file; no more Mystery Files with the generic "File" icon.
Then we get into things such as author, copyright data, archival information...anything in fact that you might want to know about a file.
So it is in fact, very very useful. SGI realised this too, which is why XFS has the ability to store arbitary meta data, too.
Slashdot Mirror
The KHTML port is a total dead end; maintaining it is a nightmare. I hope we'll have a port of Gecko within the next 12 months, which will hopefully be much easier to maintain as it is designed to be portable. Personally I think we need stronger debugging tools before anyone tackles a large codebase like Gecko, so I intend to work on the development toolchain some more and then maybe tackle Gecko.
I would have also liked to see information about what would be involved in developing applications for Syllable. Is there a well-documented API? What about support for multiple languages?
All the information you'd like is on the website but we're Slashdoted, so I can only ask you to try again in a few days time. The API is documented, there are some tutorials, example code is abundent and we're happy to answer questions in the forums and mailing lists. Multi-lingual support is currently in the CVS version which can be compiled if you want it, and will be officially available in the next release of Syllable.
Sorry, the page is hosted on Sourceforge. I didn't know you could Slashdot Sourceforge like that. We can't fix it because we're not Sourceforge administrators.
AtheOS was written in 100% assembler as a pet project by the guy who wrote it
The kernel is written in C. The high level stuff is written in C and C++.
He (and others) later built some POSIX, KDE and GTK API mappings..
The AtheOS kernel has always been about 95% POSIX compliant. There are no KDE or GTK API's for Syllable; it has always had it's own C++ API and appserver.
Question: Is there any way to use Linux device drivers with this os?
Almost all of the drivers are Linux drivers, originally. They have been ported to use Syllable API's, but they're not that different. If you know enough about device drivers you can port a driver from Linux in a couple of days. Some people have ported drivers in a matter of hours.
Posting as AC is just a normal thing for me these days, but if it offends you then I can forgo it this time. Voting and voters rights is one of those things that I believe in and I don't mind being passionate or even downright rude and offensive about it if I have to. People with their heads in the sand wind me up.
So while you're here please tell me, what is it with people who dismiss anything an AC says? Are the opinions any less valid? Are the facts less truthful? Do the words become blurry? Is it maybe just that it is easy to dismiss a valid opinion that you might not like by waving your hand and saying "Bah! 'tis but an Anonymous Coward. Off with his head!" I don't get it.
Thats actually the solution we use with Syllable. Packages go into their own directories under /usr/ Syllable has an extra step of running a tool which creates symlinks from /atheos/autolnk/ into the various files and folders within the package directory. That way PATH only needs to contain the directories in /atheos/autolnk/ instead of all those package directories, which speeds up searches.
- AtheOS is no longer developed, and the codebase has not been updated in several years.
- Syllable is our community-driven fork of AtheOS, which was started two years ago.
- AtheOS domain lapsed and is now hosting a knock-off website hawking drugs
We're halfway through development of Syllable 0.5.4, which like all previous releases of Syllable, will rock. We support a whole bunch of hardware, have developed the codebase heavily and for those of you who were familiar with Kurt Skuans style of working with AtheOS, we have a far more open development model. All are welcome to contribute. You can even download a LiveCD if you want to give it a spin.I think it depends on the store (And time of year. December, for example) I know that the store we get our Tesco deliveries from can and does deliver upto 10pm, and we've personally taken deliveries at 9:30pm.
Don't forget Sainsburys and Iceland, as well as some smaller outfits. Asda [Walmart] are very slowly also re-implementing store-based internet shopping, after their disastrous start with the big-warehouse model several years ago.
We currently use Tesco's but would probably switch if the Asda store in our area was up and running.
I think you mean "In America, Internet Grocery Shopping.." because certainly here in the U.K, it's pretty big business. My wife and I have been using Tescos once a week for two and a half years with no problems. Tesco and several other large stores who provide online-ordering and delivery are also turning a tidy profit in addition to their regular brick & mortar stores.
Heres some simple Bash scripts to do almost exactly that They need a re-write to make them more robust and easier to maintain, but they work. I don't know how much work you'd need to do to adapt them to Linux or a more traditonal SysV style system though.
Syllable is a mismanaged kludge? Care to back that up?
There are exactly four people with CVS write access to the Syllable CVS server. Patches come into Syllable via. Henrik, myself or Arno. I get the final say if something gets checked into CVS, and I have at several points in the past rejected or refused changes.
If you could point out how that isn't exactly identical to how SkyOS manages their development I'd like to know.
The only part of Syllable that is a problem is the desktop & launcher, and only because they're stopgap measures which will be replaced when we're ready. It's not like we've removed funtionality from the desktop in AtheOS 0.3.7
Please, if you have specific examples of "kludges", fuckups or mismanagemtn please tell me so I can fix them and make sure they don't happen. However if you've just got a grudge over something then fine, but why not just come out and say so?
That simply isn't true. You seem to confuse the development model of the Linux kernel with the development model of an entire distribution. There are many different distributions of Linux, but there are very few versions of the Linux kernel itself. You can see the same model at work in BSD. There are not hundreds of different BSDs.
The "Linux" development model works and works well, and it can easilly work across an entire Operating System
But I think one thing that killed Atheos is the same thing that killed almost any alternative to X: inability to support any modern graphics cards at any resolution higher than VESA. Unfortunatley this information is locked up in X drivers that are so tightly integrated with internal complexities of X that it is impossible to extract and reuse it, despite the open source nature.
Syllable has drivers for the following graphics cards with full 2D acceleration, and the ones marked with an asterisk also support video overlays (Xv in XFree86)
- S3 Virge
- S3 Savage IX/MX
- Trident video (VLB & PCI)
- Matrox Millenium & Gx00 cards
- ATI Mach64*
- SiS 3xx/Xabre*
- nVidia TNT/GeForce*
- nVidia GeForceFX*
In fact the only notable omisions are the ATi Radeon, S3 Trio & Intel Extreme (i810), and I'm confident we'll have some support for those chipsets soon.Porting drivers from XFree86 is not that difficult and lack of specs is a problem, but not as bad as you might think.
I'm glad to hear it. Meet Syllable.
It's very different now. I did, once, have Syllable 0.4.5 running in Virtual PC 5 on Windows 2000, but it's complicated and I havn't had time to create the necasary patches for newer versions to fix the problems.
All I need to do is add a workaround for the broken PCI controller in VPC and identify & fix some stability issues with the installation CD. I can say that once Syllable was installed it was very stable and quite fast.
Syllable works great on VMWare, by the way. Network and audio are supported and there is now a driver for VMWare video which will be in the next release. Not much use on a PowerBook, of course.
No, because he immediately explains that it isn't an exception, it's a clarification:
Um, yeah. So in other words: There is no exception. If you read the text you quote you can see that Linus specifically states that "The kernel _is_ GPL. No ifs, buts and maybe's about it. As a result, anything that is a derived work has to be GPL'd." He does not make any exeptions for kernel modules (Which is the point of the entire discussion on the LKML)
While Linus gets tied up arguing about derived works and wether the code was originally written for Linux, he doesn't actually solve the problem of the "grey area". I don't believe Linus has thought it through totally.
Think about it in terms of the nVidia binary modules. The kernel is GPL and the nVidia wrappers are written specifically for Linux. Therefor the wrapper must be under the GPL. If the wrapper is under the GPL and the nVidia driver itself is linked with this wrapper code..what licence is the core part of the driver under? If it isn't the GPL then it can be argued that the "viral" nature of the GPL is fluff, and cannot be enforced or is easily worked around. If it is GPL, then nVidia are going to be concerned because they don't want their driver under the GPL.
This "grey area" exists because there is no clearly defined boundary defining the seperation between the kernel and the drivers. Modules are parts of the kernel which have not been linked yet. When they're required, they are loaded and linked with the kernel.
The fact that Linus states that there is no exception must worry a lot of companies out there who are producing binary drivers for Linux E.g. nVidia, or SciTech (Who started the LKML thread, after all!) Are nVidia's kernel modules under the GPL? If the possibility exists that they are then I would expect them to suddenly get cold feet over Linux.
If the kernel had a proper boundary with E.g. a set of API's that the kernel and drivers can use to communicate with each other then it would help to solve the issue of what is and isn't "the kernel". For example in Syllable drivers are ELF images which are loaded by the kernel ELF loader. The drivers are loaded under the kernels memory space but there is a very well defined API between the two, and a very clear seperation between them. Under this model I can argue that the kernel is actually being linked to the driver, so the driver can be under any licence while the kernel remains under the GPL. There is no "cross pollenation" between the driver and the kernel. Which is a good thing IMHO, if it avoids issues like the ones being raised on the LKML.
Personally, I use a FAT32 partition to transfer data from one OS to another I know I'm not the only one.
Guess we need to get the ext2 driver working a little better, though. Just in case.
Sounds like Syllable to me.
It is a fork of AtheOS. Syllable uses it's own kernel and drivers, but most drivers can and have been easily ported from Linux. It has its own GUI, toolkit, filesystem, filesystem layout and applications. It does use the GNU tools and Glibc underneath but there is no reason to use them if you don't need them.
The GUI isn't complete (Close, but not 100% complete) and the desktop is just there until we actually write a proper one, but the hardware support is pretty good (For an "alternative" OS it certainly is). The kernel is almost complete but still has a few bits to complete & tweak.
It also has all sorts of cool stuff, just to satisfy your last requirement.
Matrox were also amoung the few companies which actually provided full specs for their hardware. You can register with them and download complete spec sheets from everything from the older-than-Millenium to the G450. I can't understand why they have suddenly clamed up when it comes to the Parhaliea. Apparently there are binary only, "beta" drivers for X/Linux but no specs or open drivers. Not much use.
Any OS which requires VESA 2 support (Darwin, for instance, or SkyOS), will not work in VMWare.
Actually you're wrong. I know for a fact that SkyOS and Syllable both run fine under VMWare using Vesa2. They just run faster with a dedicated VMWare video driver, like any video hardware.
You may be thinking of Bochs, which doesn't do Vesa2 unless you patch it. Someone has done this, but running Syllable on Bochs was glacial.
No, you can manipulate a very limited, pre defined and hard coded set of meta data.
What other meta data? How about a MIME type, or icon data, or a log listing the last ten accesses to the data (Not just the last changed date). Store any data you want, in fact. Is there any real reason to store things like ID3 tags inside the MP3 file itself? It is after all just meta data, as is the filesize, the creation date and the persmissions. Why not store all the meta data in one place?
People tend to rubbish the idea of arbitary meta data simply because they personally do not see a use for them. Admittedly, the entire concept of meta-data is a little arcane and I guess it just takes a little mental gymnastics to fully grasp the concept. Either way, arbitary meta data is useful and should be encouraged.
The data is just meta-data. Your question is like asking "Why do I want to store the filesize and file permissions with the file?" The data is meta-data, or data about data. It is useful and in many cases, essential.
Consider two common scenarios for the uses of meta data. A filesystem that can store the MIME type of the file with the file itself, instead of relying on a filename extension or a file type magic lookup every time the file is accessed. You can also store things like icon data along with the file; no more Mystery Files with the generic "File" icon.
Then we get into things such as author, copyright data, archival information...anything in fact that you might want to know about a file.
So it is in fact, very very useful. SGI realised this too, which is why XFS has the ability to store arbitary meta data, too.