Finally, if you want more Syllable Desktop progress, then make some! Syllable is a volunteer project.

Creating a new desktop OS is fiendishly hard work and it's all the harder to try and support all of the devices you'll need to support. Syllable Server shows a great way forward: Just slide the Linux kernel in underneath. Yeah it's got all kinds of problems, but if you're really interested in building a nice desktop OS the kernel is the last thing you should be worrying about.

Despite all of the effort the *nix desktop is stillborn even today. Most Linux distributions are 99% the same and share all kinds of problems. I think that there might be a welcoming user base if a Linux-based system that is radically different were introduced. Look at android: The Linux kernel plus stacks of stuff that make it what it is. Who cares about the kernel? Not desktop users! Just build all of your nice things on top of Linux, get stability and hardware support for free, take the FOSS desktop world by storm.

I always find these kind of argument puzzling: it takes *more* effort to start from "scratch" than to modify an existing system.

First, NVidia's driver is a bad example and you know it: it's proprietary, fortunately there are lots of other open source drivers bundled in Linux that you can modify as you wish..

then you're *wrong*, X isn't the only solution: if you had looked at the release notes you'd notice that DirectFB is a part of it.

Plus the components of a desktop Linux system are too much tied together. Try to code a X replacement that does not work like X and is compatible with Nvidia's driver, as an example : I wish you good luck.

"First, NVidia's driver is a bad example and you know it: it's proprietary, fortunately there are lots of other open source drivers bundled in Linux that you can modify as you wish..

Question : if I run several CPU-hogging apps (something in the vein of while(1);) in a terminal, does a DirectFB desktop remain responsive, unlike something based on X ?

Question : if I run several CPU-hogging apps (something in the vein of while(1);) in a terminal, does a DirectFB desktop remain responsive, unlike something based on X ?

I have to agree with this. Syllable seems to be stagnant. I visit the web site and before the server announcement the most recent news was six months old.

I like Syllable (I helped choose the logo) but progress is so slow that when they make a release, it surprises me that they are still around.

I have always said that part of the problem with Syllable is that the developers don't eat their dog food. Do any of the Syllable developers use Syllable as their primary OS? I have a hard time believing that they do because if they did, they would see how much is left to be done.

2) Does this really happen if the X-server is reniced to a higher priority?

" [q]First, NVidia's driver is a bad example and you know it: it's proprietary, fortunately there are lots of other open source drivers bundled in Linux that you can modify as you wish..

" Question : if I run several CPU-hogging apps (something in the vein of while(1);) in a terminal, does a DirectFB desktop remain responsive, unlike something based on X ?

The keyword here is "seems", and an amount of exaggeration. People have very fixed obsessions (I don't mean to single you out). People surf the web all the time, so they tend to take website news items as the measure of all things, but really, this is very superficial. There are many channels through which things happen in a software project, and we make them publicly available, but few people bother to consult them. That would be fine, but people still tend to form opinions that are much stronger than warranted based on the limited information they are willing to consider. And then once they have formed an opinion, they tend to magnify it and discard opposing new information. Our news item from before Server 0.4 was halfway January, and it's now early June, so there were four and a half months between them, not six.

(...)

This happens when you measure a software development project by its news releases. The news slowdown you talk about was exactly caused by the frantic work to produce Server 0.4. Surely one can complain that we should publish more than we already do, but we're a software development project, not a news production project. We have to set the right priorities with the extremely limited resources we have.

Yet, we spend a large amount of our time on communication, and the larger slowdown in both Syllable Desktop and Server releases was caused by a necessary shift of focus to our communication and management systems, in particular the development of my CMS. When you have those extremely limited resources that we do, a strategic hopping between goals means that you can't develop all goals equally at all times. But as I said, people seem to be obsessed, so they refuse to see our strategy when they feel that their pet item they chose from it gets neglected.

The current release is the next step in our strategic plan. I can tell you that it has allowed us to ditch all other Linux distros from our systems in the past few months, including our desktops, laptops and servers. The result is not yet suitable for publication, so we use it internally, but it is a major step towards eating our own dog food.

Questionable. Compare Haiku with modern linux distros in terms of performance and API coherence. Linux has never been made for desktop use, and you can feel that in almost all low-level components of a desktop Linux system...

Such a thing does exist, it's called Pardus Linux and I use it at home. It's a very stable Linux distro with excellent hardware support, an acceptable repository, flawless KDE4 integration, and basically almost everything a desktop Linux user could wish for, combined with some unique tools where the standard Linux tools fail to meet basic desktop user's needs. However, the storm still remains to be seen...

In the case of Syllable, by "everything" you mean that less than one percent of our own code that is in Syllable Desktop?

Again, I perfectly understand. People like to get attention, but when you're a small team, you have to make a choice before talking and actually working on stuff. It's a complicated trade-off. Writing random posts about personal ramblings to let people know that you're still alive may help, it's a strategy which I've often seen applied and try to apply myself.

"The current release is the next step in our strategic plan. I can tell you that it has allowed us to ditch all other Linux distros from our systems in the past few months, including our desktops, laptops and servers. The result is not yet suitable for publication, so we use it internally, but it is a major step towards eating our own dog food.

Just joking. I personally think that long-term evolution of desktop OSs requires completely getting rid of the UNIX legacy and making something which is really optimized for the job from the ground up. But that's a somewhat extreme approach that won't give serious results until many years. If it does give results.

In meantime, I might get tempted by switching from Linux to another development platform like Haiku or Syllabe, if they provide the right tools and that innovative bit which makes the system funnier and a pleasure to work with

Keep up the good work !

I've read that page and I don't think I'm confused. On it they say that there's Syllable Desktop and Syllable Server and no where does it say "And when the Desktop GUI has been made to work on Syllable Server we will stop building a separate desktop kernel" instead it talks about the Syllable GUI being ported "so it will be a very easy to use server system"--not so that it can be a desktop system too.

So where is my mistake? Are you building a server OS that happens be quite similar to your desktop OS and is intended to serve as a complement to it or are you planning on replacing your 'desktop' system with the 'server' one when the porting is all finished? 'Cause if it's the latter you don't seem to say so anywhere.

I've seen pardus and if you're thinking "ah, like pardus" then you missed the argument I am making by several miles. Hint: Pardus still uses KDE. Pardus just builds a relatively coherent system, otherwise it's a lot like almost every other Linux distribution. I'm proposing no KDE, no X, no dbus, no pulseaudio.

If implementing your own non-X windowing system loses you some driver support... well so be it, some is better than all which is what a brand-new system would require.

"In the case of Syllable, by "everything" you mean that less than one percent of our own code that is in Syllable Desktop?

Note however, that I'm not defending any specific interpretation, just saying that this number alone doesn't mean much..

As an aside: congratulations for having chosen a port of WebKit instead of Firefox for the web browser, is-it multi-process on Syllable Desktop like Chrome(*) is?

* I've switched recently from Firefox to Chrome, and I really enjoy Chrome's responsiveness.

I've seen pardus and if you're thinking "ah, like pardus" then you missed the argument I am making by several miles. Hint: Pardus still uses KDE. Pardus just builds a relatively coherent system, otherwise it's a lot like almost every other Linux distribution. I'm proposing no KDE, no X, no dbus, no pulseaudio.

If implementing your own non-X windowing system loses you some driver support... well so be it, some is better than all which is what a brand-new system would require.

Yes. There are two considerations for us here. One is that we've already been babbling every day for more than half a decade in our groupware system. That's among the contributors, people who make an effort for us. Other people have always bluntly refused to take part in this. I know it's a darknet, and there's a big difference between the internal and external views of the project, but we're not going to compromise the efficiency of our internal work communication for people who only want to lurk and refuse to make an effort.

The other consideration follows from this. It's much less efficient for us to publish to public places such as the web, so we need to optimise that to make it acceptable. For this, I develop my CMS as a strategic solution. Such a thing takes time, so it's another part of our development plan. Our web sites have been running on it for a few years now. It supports translated web sites and the translators can post their own articles, but it's not comfortable enough yet that they use that opportunity a lot. A year ago we added comprehensive RSS feeds, so together with all the SourceForge and Ohloh output the lurkers are already catered to quite well. We will continue to extend this. We're working on making the content of our groupware system available on the web (the web sites themselves are managed in it, but there's also chat, a calendar and check lists). This will be running on Syllable Server, so everything will come together there.

Our fundamental approach to this is REBOL.

(cut) but the fact is not that we have little own code, but that we have a lot of reused code. I'm also proud of that, because we add value to that existing code, and this is the ideal thing to do in an open source ecosystem.

However, multi-process is not about responsiveness but about stability. (cut) If you open multiple Webster windows instead of tabs (which is the fundamentally correct way, anyway), you also have multi-process.

Well, that's a choice, but I think that you're being a bit harsh about the "lurkers" curious guys who just want to learn what the project is about and how it's doing without wanting further implication at the moment. They happen to be pretty handy when your software needs testing, and, what's even more important, they form the vast majority of your future customers so they should be handled with care

A little post from times to times, written in a way that's comprehensible even for non-participants, maybe together with a screenshot which shows how great you're doing, does not cost much time and can be very well-perceived... Look at Gimp's or KDE's design blogs for examples...

"Our fundamental approach to this is REBOL.

" http://en.wikipedia.org/wiki/REBOL
+ search results in the same vein, no matter which additional keywords are being used
= Fatal error. CRC checksum failed...

So what would you suggest non-developers do in terms of helping out Syllable Desktop? File bug reports for things that just don't get fixed? :-)

http://forum.syllable.org/viewtopic.php?t=1326
http://forum.syllable.org/viewtopic.php?t=769
http://forum.syllable.org/viewtopic.php?t=1220

"[q] http://en.wikipedia.org/wiki/REBOL
+ search results in the same vein, no matter which additional keywords are being used
= Fatal error. CRC checksum failed...