Nearly a month after Service Pack 1 for Zeta RC1 was released, yellowTAB is working on the last bits and pieces of Service Pack 2. This service pack will be the base of Zeta RC1.1, as the RC1 is completely sold out. The new SP has a new theme and a new default font (modified Vera). ZetaNews has the details and screenshots and also the videos from the NBC Giga TV show which featured YellowTAB’s Zeta recently.
Zeta SP2 Beta; New Theme Look; New Default Font; Zeta on TV
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
Why not just take a Linux kernel and a nice filesystem like XFS, JFS, or ReiserFS then re-create the BeOS GUI enviroment using your own GUI ? At the same time you keep binary compatablitly with Linux or at least allow it so major apps like openoffice or Mozilla can be ported over and maintian a app repository of OSS applications for users to access.
BlueEyedOS aims for this I tink
BeOS/Zeta isn’t just a GUI – it’s an OS.
The same goes with Linux – why not simply do any WM of choose as an explorer.exe replacement using Windows and keep compatiblity with Windows apps?
One of the great things about BeOS was that the GUI was well thought out and looked right. With zeta they’ve made changes to only part of the GUI and made it look like someone just pasted some really bad design on top of another one. Which is exactly what it is.
“The same goes with Linux – why not simply do any WM of choose as an explorer.exe replacement using Windows and keep compatiblity with Windows apps?”
Maybe because it is impossible without the source to provide the advantages of Linux to Windows in such a way.
Binnary compatability with Be, love of the “be way”, which is diffrent to the Linux way.
If you did do this all the Linux portted apps (which most, if not all of the apps would be) would not make use of the BeOS features. What is the point of having attributes, node moniting and other items of funkyness if no one uses them?
I much pref. Zeta and OpenBeOS. (but I wish BlueEyedOS all the luck
Wow, they really managed to screw the decor up. That looks like a 18 year old’s first winamp skin. But I guess it’s just in line with the rest of yT’s quality.
If you disagree, go and check out r5’s decor again and remind yourself how professional it looks in comparision.
Fads are things like Pet Rocks and Acidwash Jeans. Just because something once obscure starts to become more widely known doesn’t mean that it is less of a good thing, if indeed it is a good thing (unlike pet rocks). I assume you meant fad in a bitchy, negative sense. The “i am 16yrs old and so cool because I run linux” thing may be a fad, but its pretty safe to say that linux is here to stay…it does good things.
As for Zeta…wow, the idea of an open beos project is intriguing.
The movies don’t seem to work here.
Yes, that’s what I meant.. sorry for the confusion.
No sound in the movies here, tried both xine and mplayer.
Is it just me or is anybody else experiencing the same?
I’m having the same problem. :
@Templar (IP: —.client.comcast.net)
BeOS UI wasn’t that great. It had a ton of UI usability problems. The most annoying of all is the tab on top of each windows. It served no purpose at all. The Tracker was also annoying with the grouping tab. I found it so frustating, I only had 1 NetPositive opened and I had to get through a menu to get to it! I won’t even get started on the icons…
Okay Justin then don’t use Linux, use FreeBSD, or Darwin, etc… I don’t understand why these guys who have no access to the orginal BeOS source code are wasting their time with service packs for a dead OS. Yes I said it folks ! BeOS is deader then road kill on a Arizon backroad. Their efforts would be better spent trying to re-create the best aspects of BeOS but with source code that they can use and have access to for their own needs and maybe create something more functional that has a better chance of gaining momentuem IMHO. The reason I mentioned Linux is because of the wealth of current on-going developing applications and drivers there are for Linux compared to BeOS.
I agree that there were some usability issues. There were some things that could have been changed and there were some things that I wish other operating systems would pick up. However, I wasn’t really referring to usability, more like continuity in it’s look and feel. Zeta looks like it is putting a new GUI over top of the old one and not for usability purposes. Which makes it look like there are two designs instead of one. I could understand extending and changing the design for usability and other reasons, but zeta looks like change for the sake of change and only makes it look worse.
“The same goes with Linux – why not simply do any WM of choose as an explorer.exe replacement using Windows and keep compatiblity with Windows apps?”
Maybe because it is impossible without the source to provide the advantages of Linux to Windows in such a way.
There are plenty of open source shell replacements extant. Windows has plenty of advantages over Linux, among which are proper access control lists, a perfectly working control panel, GUI performance, widely available drivers that don’t require kernel recompilation, and a GUI that doesn’t crash when Konqueror (or Nautilus, or your pick of shoddy applications) crashes.
yup, same deal here, no sound in VLC on OS X.
Can anyone read lips in German?
Please keep the discussion on topic, regarding Zeta.
May I say Marketing for whats left of the beos community. IMNSHO Zeta is “just”, the leaked dano + drivers found on bebits + some applications developed specially for it. Take all this create some mystery marketing BS, take time to let the word spread and then release you’r stuff. Zeta has no future because they don’t touch the kernel, and they don’t touch it because they don’t have the sources. Why do I say so ? because all the improvment done in Zeta, can be achieved without kernel sources, new USB stack does not need to live in kernel. All the rest is just libs and software added. Zeta would have a future if they where able to fix issues like the 1 Gib memory limit in beos (reminds me of the AA 96 MB mem limit for beboxes), they can’t even code Java correctly and partner whit beunited so they can use beunited programming forces for “free”. All ex-be engeneers who’ve played/seen Zeta say it’s the last version they’ve used when the where at Be, two years ago. What is the roadmap for Zeat, how will they deal with 2.0, if they can’t fix issues with the kernel ? Will they rely on OpenBeOS ? if so when will they give their usb stack to openbeos (I’de like to see this happen in 2004 around september). Technically Zeta is worth nothing – marketingly they seems to go somewhere and that’s good.
I would advise people that already use beos pro , not to buy zeta but to cncenbtrate on buying 3rd party software so developers stay in the beos world. I would encourage people to show some support for OpenBeOS, so they really have a replacement OS. Let Zeta attact new commers because that’s what the beos world needs, but don’t give them your money, give it to dev. and openbeos which is the future for be related things.
Don’t get me wrong I was a strong be follower and supporter. Befoare that I was a strong Atari Fanatics. I’ve been F*** by both compagny and I’m just trying to give my experience to people so they don’t get F*** tehmselves. Zeta has no future.
—
http://homepage.mac.com/softkid
Mhmm…
I’ve worked intensively with BeOS and I loved the tab-like window titles. When working on several documents for a while, I put them all on top of each other and dragged the window titlebars so they were arranged just like tabs. Arranging was a bit of work, but you could work very smoothly from then on. Unlike with normal tabs, you could arrange documents of several apps together.
Also I liked the icons.
The worst UI problem was IMO that the popup menu in the file browser was unmarked and most people missed the only good way to go up some folders. (I filed this early as a bug, but AFAIK it was never fixed.)
Apart from that, YellowTab seems determined to ruin everything that was good about BeOS. First the fast and simple installation, now the clean look. Drop the gloss on the decor and give the buttons a better contrast. Also the font in the screenshot sucks. It looks nervous, and the spacing is terrible. Note f.e. the spacing between e and o in people.
Sorry to “toot my own horn” here, but at http://www.iscomputeron.com you can get the 4 video parts, with sound. Though they’re in mpg format, thus bigger files.
DaaT
@jeti (IP: —.pool.mediaWays.net)
Arranging was a bit of work, but…
That’s a usability problem. Working with titlebars doesn’t make any sense at all. If someone has multiple maximized apps then they all pile up in a corner of the screen. Then the user has to remember a keyboard combo to move those tabs and… IMO a full titlebar would be more pleasant and useful.
I found the icons somewhat pleasing to the eyes also but they handled like crap. Check out the behaviour of icons on BeOS then go back to a Windows desktop. You’ll see a major difference. They are hard to read and notice on a screen at an higher resolution. But then the whole BeOS GUI suffers from the “miniscule” syndrome.
Now I would not go to the point of saying YellowTab is out to “ruin” BeOS. Zeta and OBOS are the only ones left “doing” Be stuff. The other three projects (BlueEyedOS, BeFree and Cosmoe) are most likely abandoned by now.
I agree with you on the font spacing. They really need to put more work in the font dept. It’s horrible. And the fast and simple installation… I’ve seen the YellowTab installer screenshots on their website and it looks like it’s gonna be filled to the brim with useless junk that should not be included with the operating system. They need to keep it as lite as possible.
I hope they up the icon resolution to at least 48×48 by default and increase the font size. And update the Tracker’s toolbar icons. They were ugly then and still are. I’m glad they put in SVG icons. I like the new window borders. It’s much better than the old yellow square design. Now they need to remodel the inside too.
This is a good example how to kill an already killed os once again
. Be had the best look and now it seems like a crappy new Linux distro with some twisted Gnome/KDE theme ;(.
On the claim Be UI had usablilty issues: It was far the easiest to use. Tabs were really cool to stack windows on top of each other to browse the net using Netpositive (the most usable browser ever made). It was really nice to navigate using keys in the entire os. And the only thing the BeBar (I cannot recall it’s name, sorry) missed was d’n’d configuration of menu items. Be has it all. I’m really sorry that it went down.
With the amount of bitching I see regarding Zeta, it has to be rather good.
But closed source is rather tedious.. We don’t want another amiga/atari/beos/whatever next system that drowns into oblivion beneath marketing BS, bad management, bloat or other things that are bound to happen.
Question is, how do you have an economy based on open source.
I have Zeta, and I do like it, yeah there are problems and the GUI could be improved still but most people that complain about the ‘new look’ GUI seem to miss a big point, if you dont like the new style then dont use it.. you can revert back to pretty much the old ones, if you like full title bars and not tabs then chose a decor that has full title bars. If you see screenshots of Zeta then that doesnt mean it is THE look and the only look, it’s customisable.
And as for the install, it it the easiest and quickest inatall of any OS I’ve personally used, Win 3.1 through to XP, numerous versions of Mandrake, Debian, Slackware, SuSE and Redhat, BeOS 5 and Max edition). Yes there is a lot of stuff there, but only if you chose the full install, if you dont want that stuff then don’t chose it and you dont see it. A basic install can be completed in 10 minutes with about 5 clicks of the mouse.
I’m not saying that Zeta is perfect, by no means, but it does have a lot of nice features, and I personally like some of the new GUI styles, ok some I hate aswell, but as I say you have choice!!!
is very harmful to the scene. It is BeOS 6, like it or not. Ok, maybe it’s 5.5. But there is no other commercial offering available, people. If you really hate it, don’t use it. Buy it and put it on the shelf. Money for Zeta does support the OpenBeOSes in the long run, as a vote of confidence.
I can’t see how you can love BeOS and hate Zeta. Give them some slack. Give them some feedback, if you’re not happy with it. If they sold enough copies, they could buy the rights to the source… luckily, there seems to be international support for yellowTab, especially in Poland, Germany and France.
And I think people misunderstand the new BeDoper, but it makes a good point about infighting in the BeOS scene. I’m surprised to see so much of it on OS News, in fact.
BeOS Journal to yellowTab: “No, YOUR mom!”
http://www.bedoper.com/bedoper
Zeta is twice as fast as BeOS 5 on the same system. True the installation takes longer but that is because it installs more software. It takes 5 times longer and installs 20 times more software. So where is the problem?
The basic Zeta (NO Additional software) system takes about 20 seconds longer to install than BeOS 5 Pro, timed on my system. And that is because it installs more drivers and a few system utilities.
The shear power of the system, taking advantage of multithreading makes it better than any Linux with a BeOS Api.
Does it need more work? Yes, but remember it is not a Final Release. It is only RC1 and they are working on the SP2 now. Wait until we see that, to have a full view of what is going on.
Linux takes 1 hour to install, (Standard Install with OpenOffice)
Windows takes about 50 minutes to 2 hours to install (no choices and different time EVERY time.),
Zeta takes 30 minutes (Full).
The UI in Zeta looks somewhat like the UI designs done for OBOS except for the extremely poor execution and changes done to make it look like ths UI wasn’t lifted from the OBOS project. The OBOS UI showed some great promise. THis is simply a steaming pile of dog poo.
[quote]The UI in Zeta looks somewhat like the UI designs done for OBOS except for the extremely poor execution and changes done to make it look like ths UI wasn’t lifted from the OBOS project. The OBOS UI showed some great promise. THis is simply a steaming pile of dog poo.[/quote]
If you point the finger, so me some proove, post the screenshots of obos design with it. You just shout something in loose air :S
I think this design has potential 2, by the way, they just have to change the shade of yellow and get the gradient out.
In terms of philosophy and technology the BSD kernel would be a much better choice if the system were to be based on something.
Remember that Linux is a toy OS and BeOS wants to be commercial…
I think it’s great to see that OBOS chose the NewOS kernel instead of the old kernels from the 80’s.
Anyway.. back to Zeta… the best news they’ve given throughout their history as a company is probably that they’ve decided to work with BeUnited on Java…. let’s hope they actually help out this time too…
Please excuse my ignorance but I thought Be was bought out by Palm so what is Zeta and yellowTab? Did someone buy or license BeOS source from Palm? I’d appreciate it if someone can explain it to me. I followed BeOS community up until the creation of BlueEyedOS, OpenBeOS, and even Cosmoe but I guess somehow I missed Zeta. Thanks.
1st, Zeta simply has to have access to the kernel if they want professional users, or more specifically developers to pick up their platform and start developing and/or porting applications to it.
Why? Because while BeOS has been in “pause mode” for the last couple of years, with regards to its lack of development and lack of parent company support, the PC hardware people have been pushing the envelope as prices have gotten lower and lower.
Thus many of us are no longer running older hardware that’s compatible with BeOS/Zeta. True, many are still running older, or in some forms updated, yet compatible hardware setups, but a lot of people, especially the power users and developers (both of whom Zeta should be targeting if they want the platform to evolve beyond the hobby/dead end level it’s currently at) are running systems that simply will not run the BeOS, and thus Zeta.
The most obvious issue to me is Ram. 1GB is quickly becoming the norm, and as 64 bit intrudes more and more, I’d expect the norm to rise soon to 2GB or more.
And realistically, anyone who develops or runs many apps simultaneously (something the BeOS was famous for in its day!) probably is pushing their Ram to as much as they can use, and/or afford.
And Zeta won’t boot with a gig of ram. I can attest to that being the case still even with updated kits such as the online developers kit, or BeOS max.
For that matter, what about other new hardware which is quickly becoming standard PC components? SATA Raid anyone (included on many of the new Intel chipsets!)? Linux is still struggling to handle the new Canterwood chipsets 100%, not to mention SATA raid, so I can imagine how long it’ll take for Zeta to support such things, assuming of course that they somehow overcome this kernel issue.
True, I can run it on my server box, which uses older hardware, but if Zeta is immediately relegated to only the “old pc in the corner”, what chance does it have of surviving long enough to get some decent NEW apps created for it? None in my book.
I understand that many of theorized that Zeta is only “treading water” until OBOS releases their open And even if Zeta manages to get thkernel, but when is this expected? Not for awhile… And what happens between now and then? Zeta/Beos continues to fall farther and farther behind in comparison to other x86 OS’s (Linux/Windows/Whatever).
I think Zeta’s got a lot of good thinking behind it, but I think that for anyone other than nostalgic geeks like us who read this site, BeOS is not an option at this point. I’d love for them to come out and say “We have the kernel and are diligently working on updates”, but they haven’t despite many questions regarding this both on this site, Zetas own forums, and on other tech sites. In fact they’ve been very obvious in not talking about this single fact, which more or less means “We don’t have the kernel source”.
I wish I could come up with something more positive to say about Zeta at this point, but for me, they’re a lost cause. That aforementioned “old computer in the corner” that I have is already serving a purpose for my home network, and BeOS/Zeta won’t run on my main box due to its new hardware and the amount of Ram installed. 8(
Interestingly enough… Not to bring up OSNews questionable posting guidelines, but why would a posting that expresses “Why not just take a Linux kernel and a nice filesystem like XFS, JFS, or ReiserFS then re-create the BeOS GUI enviroment using your own GUI ?” be modded down, while a posting whos title is “@ Alex the moron” be left up?
Just get rid of the damn guidelines at this point people. You’re alienating more people than you’re helping with your constantly changing stance of “what’s acceptable”.
Either apply the guidelines across the board equally, and apply it to the moderators as well (Hi Eugenia!) or get rid of them entirely. It’s getting too confusing for us readers to try and figure our what we can post based on how the moderators feeling from day to day.
Is the new font Prima Sans? or something else?
http://www.identifont.com/identify?22+%20+2U+J+Z+F+8N+1U7+4A+30…
“The same goes with Linux – why not simply do any WM of choose as an explorer.exe replacement using Windows and keep compatiblity with Windows apps?”
Maybe because it is impossible without the source to provide the advantages of Linux to Windows in such a way.
Or maybe it’s not.. ever heard of projects such as litestep?
Check up facts before speaking next time.
>Or maybe it’s not.. ever heard of projects such as litestep?
It wasn’t about “skinning” possibility.
Initial posting was, IMHO, about of lack of sense in such monkey bussiness.
Windows with LiteStep cannot utilize *nix advantages. It is only fake for those people who are unable to estimate anything besides decors and look.
Same for BeOS on Linux. YEAWM – yet another windows manager.
Why not? But it is just linux, and has nothing besides look from those goodies wich attract people to BeOS.
I can see that YT has made lots of changes from Dano to Zeta, but were they actually improving things? The new themes and fonts are not improvements to me, what was wrong with the old ones that the new ones would do better? Don’t we still have the same major drawbacks we had with the original BeOS (RAM limit, poor scaling beyond 4 CPUs, no font-sensitive GUI)?
To me, Zeta looks like it’s just a rebranded Dano, in the same sorry state as the original. Instead of bugfixes in the kernel, I get a heap of add-ons (yes, I really needed that updated “About” panel with new CPU IDs so much more than support for 1GB+ RAM).
I’m not sure how much yT is improving Zeta but I know that I need a legal copy of Dano in order to run BeOS on my laptop. Without Dano I have no keyboard so it’s nice to have a legal copy (which would be Zeta).
(RAM limit, poor scaling beyond 4 CPUs, no font-sensitive GUI)?
Yes, there are so many quad-CPU desktop boxes. </sarcasm> As for font sensitive GUIs, this is perfectly possible if written correctly, same as on any other OS. Like it or not there are tons of not properly sensitive apps on Windows, Linux, OSX, you name it. That problem’s squarely in the app dev’s camp. I’ll give you the RAM limit issue though.
“(..) but I think that for anyone other than nostalgic geeks like us who read this site, BeOS is not an option at this point.”
Sad, but true… It hurts me to agree with this though, but it’s true…
The cool thing about BeOS was the API, not so much the UI. Zeta, et al, seek to preserve the API (AFAIK). I used to find the API extremely clean to code to, until I tried Cocoa/OPENSTEP.
I’am not sure but i think Zeta 1 aim the binary compatibility with BeOS R5 hence they can’t modify the Kernel; my guess is once they’ll get more funds they’ll will improve it.
Be break the binary compatibility several time BTW..
Anonymous:
Yes I said it folks ! BeOS is deader then road kill on a Arizon backroad.
Well gee, I’m sure glad someone finally had the balls to say it.
“Well gee, I’m sure glad someone finally had the balls to say it”
Yeah, not like anyone else has ever made such a comment over the last 3 and a half years.
“(RAM limit, poor scaling beyond 4 CPUs, no font-sensitive GUI)”
Quite frankly, other than with geeks, the RAM limit is still not a major issue as by and away the vastest majority of computer users still run well under 500 MB. That will change, of course, over the next 18 to 24 months. Multiprocessors beyond four CPUs? There’s an issue that is driving XP sales through the roof. Anyone know of some good dual P4 boards? I think I may upgrade in six or 12 months (from a 400 MHZ K6-II). Font sensitive GUI? WTF is that? Must be important though as it sounds technical enough. Sarcasm aside, I do understand what you are getting at, and I more or less agree, but the required fixes you have stated are less important than you (and a great many others) make them out to be. That said, I am less than impressed with Zeta (the company), and its apparent desire to “Linuxify” BeOS. I don’t need a lot of KEWL geek gadgets. I just need my comfounded ‘puter to “just work.” Zeta seems to have lost the “way” on that account. I am also concerned about the total lack of professionalism by the Zeta crew, especially Bernd. He is his own worse PR nightmare, and needs to shut up and hire a good ad agency and PR man. What worries me the most is that Zeta will make just enough of a splash to be noticed as it plummets in flame to meet its fate, while simultaneously consigning BeOS to the dustbin of history. If Zeta fails, and is “noticed” to fail, there is no future for any “BeOS” distro no matter what the quality and no matter how good the salesmanship. A third chance for Be will not be forthcoming.
Ronald, I am sorry to read that you do not get along well with the tabs. As for me, it is by far the most usable aspect of the GUI, and am utterly lost without them. I think I may safely say that most BeOS users feel similarly.
—
Regards
Michael
As for font sensitive GUIs, this is perfectly possible if written correctly, same as on any other OS.
You would expect the system to ship with a UI toolkit that takes care of that for you. Good examples of such toolkits would be Qt or Swing.
Like it or not there are tons of not properly sensitive apps on Windows, Linux, OSX, you name it.
“Everyone’s doing it” is a bad excuse.
That problem’s squarely in the app dev’s camp.
Just like multithreading, right? The reason why there are so many multithreaded apps on BeOS because BeOS encouraged you to use multithreading – in contrast to, say, Win32 or MacOS Toolbox. If BeOS’ API encouraged users to write font sensitive GUIs, a lot of apps would use it.
Umm… that was all exact my point. Where are you trying to get?
I said:
“Well gee, I’m sure glad someone finally had the balls to say it”
Then Michael P. Reed said:
Yeah, not like anyone else has ever made such a comment over the last 3 and a half years.
Then the Comic Book Guy said:
“As you are apparently unfamiliar with the concept of sarcasm, I will just close the register at this point.”
Thank god Zeta/BeOS is not Linux! Common Linux and Unix systems (OS X excluded) seem to suffer from the same disease still. When you put a GUI like GNOME and KDE, which are definitely the most popular ones, on top they crash even more often than explorer. No offence but despite their steady improvement they seem to be miles away from the ‘normal’ desktop.
All the negative talk about Zeta looks like they (yT) have touched the holy cow and now some folks get pissed off because they think it’s done the wrong way. In fact it’s a rough world out there and the best things in life are not always free (see Opera
. Certainly Zeta needs LOTS of improvements from what I’ve read and can tell so far but it’s their time, money and effort – so why should they create an interim OS and hand it over to OpenBeOS WHEN it’s ready? I would like to see joined forces instead of that yT/TBJ quarrel. Don’t care who started it.
I mean, really… look at these comments. They’re all the same as the comments found in the last six Zeta news item forums. Have nothing new to say? Then be quiet.
As for me… I honestly think this new “yellow tab” window decor looks pretty nice in comparison to the ugly ones that came before it. I still prefer the standard/traditional R5 look, myself, but this one finally looks like it was done with some amount of attention to detail.
Yes, all the window decors look “tacked on” because they are. They were not part of the overal OS visual design, originally. This is clearly one of the biggest problems with Zeta, but again, everyone has said this repeatedly. If yT doesn’t get it, that’s their loss. Some people are willing to let this slide for a release or two. I am personally happy to have a USB 2.0 stack, USB mass storage support, modern motherboard support and whatever else is new in there. Yes, there are some new things not found in Dano.
But come on already. Enough with the same old ignorant and BORING comments. Say something new or just shut the hell up. Even I gave up being a repetitious complainer on this topic.
Dude. Linux didn’t even exist in the ’80s. Its only a few years older than the NT kernel, and with the massive rewrites of major subsystems in the last few years, its really freaking modern.
If you mean the overall UNIX-y design, well, you do know that NT is based on the equally old VMS design, right?
As for NewOS, in what way is it better than Linux? Can’t tell me? Probably because you know *jack shit* about kernel design, am I right?
I think a lot of people are missing the point on YellowTab. Sure some of YellowTabs antics are disappointing but I see YellowTab as a useful tool in the re-emergence of BeOS. There getting on TV, traveling to conventions getting exposure for BeOS. Because of that they are trying to encourage programmers to experiment with and work on the BeOS. This will ultimately help OpenBeOS when it comes out in 2005. We will find out if they are fools by not becoming a distro for OpenBeOS.
As for NewOS, in what way is it better than Linux? Can’t tell me? Probably because you know *jack shit* about kernel design, am I right?
NewOS from day #1 was designed to have a preemptive kernel, unlike Linux/BSD, which apply a big global lock to any kernel task. These days Linux is becoming more and more preemptive, but this is more an afterthought (hack) than a core design decision. Also, Linux is very much a monolithic kernel, for better or worse, while NewOS has lesser pretentions, which suits OBOS perfectly.
Finally, OBOS wont suffer from the variety Linux and its distributions are plagued with (for better or worse). There will only be one API, one GUI, one core distribution. Some of us are sick of recompiling and the dependancy nightmare on Linux which dwarfs Windows dll nightmares from years ago. With OBOS, we stand a chance of getting something designed with different goals in mind. Goals which suit a desktop user more than an admin.
I can safely say that although its not nrealy as polished as the R5 max edition that it replaced, it is however a nice stable OS in its own right, its also the only LEGAL way you will be able to get BONE on your desktop. Like it or not, thats more then Be ever did. I agree with the decore problem, they don’t look great, indeed some of them look utterly pathetic ati times, but hopefully with frans now getting to grips with the decore system, we will see some improvements in this particular department.
Regarding the kernel source and 1Gb memory issue.
Youre a first time company taking on an os from a legend. you have a limited amount of resources and indeed amount of apps that are still current. Do you A: get rid of backwards compatibility in one fell swoop by fixing the limit? or B: Get a product out of the door and then gradually reduce backwards compatibility until such time as you have a decent amount of new apps that will make the transision much more smoothly in the future (ALA Apples OS9 – OSX transfer).
If YT have the sources and start monkeying around with it in order to fix the limit and build a new media server which in turn may possibly break compatiblity with drivers and apps. What chance would they stand? Zeta if it is to survive in any shape or form requires a rolling start. This isn’t 1994 where Microsoft is still reletively fresh faced and can be pushed asside, this is 2004 where you’re dealing with an IT sector thats been saturated with their products for 10+years, you need a lot of momentum to break through that wall, and by loosing all the apps, you may well find yourself in a much worse posission then Be Inc were before their bankrupcy.
Anyway, great news regarding the new service pack, will be downloading it just as soon as its available.
You’re very wrong.
which apply a big global lock to any kernel task.
>>>>>>>>>
Both have a big kernel lock, but in Linux, almost nothing uses it. The work to remove things from under the kernel lock has been in progress since Linux 2.0. Linux scales to 64 CPU systems these days, so the big kernel lock is definately not a factor anymore.
These days Linux is becoming more and more preemptive
>>>>>>>>>>
You’re either preemptive or you aren’t. Linux is definately preemptive. Audio latencies, for example, are down to below 1 millisecond. The main problem today is just drivers that haven’t implemented fine-grained locking.
but this is more an afterthought (hack) than a core design decision.
>>>>>>>>>>>
Let me guess. You’re not a programmer, are you? Software evolves, it is not all planned and implemented at once. Adding a core feature during evolution is not a hack, it merely demonstrates the maintainability of the code. Besides, there is no one Linux kernel design. The design has been greatly modified several times over the evolution of the system. One of Linux’s original design decisions was to make the kernel i386-specific using inline asm and whatnot. Obviously, that decision has not affected Linux’s portability today! Preemption was not added as hack. Linux has heavily restructured during 2.0 to support fine-grained locking for SMP. The locks themselves were broken up between 2.2 and 2.6. Preemption is just a natural extension of the SMP locking mechanism.
Also, Linux is very much a monolithic kernel, for better or worse, while NewOS has lesser pretentions, which suits OBOS perfectly.
>>>>>>>>>>
Correction, Linux is a modular kernel. Most everything can be compiled as seperate modules. Also, NewOS is similar in scale. Both include the VM and page cache in kernel space. Both include filesystems and drivers in kernel space. Both have networking in kernel space. Both have a console in kernel space. Both have (or are planning, in the case of NewOS) the USB stack in kernel space. I’ve just covered the key monolithic/micro-kernel differences right there. What are *you* talking about?
Finally, OBOS wont suffer from the variety Linux and its distributions are plagued with (for better or worse).
>>>>>>>>>
The variety of Linux is entirely a function of the userspace. The Linux kernel’s involvement with userspace is limited to running the first userspace process (usually init, but can be any program) and interaction with a few configuration tools like ifconfig and iptables. Your argument has no merit. Do you think MacOS X would have fragmented if they used a BSD kernel? Oh wait!
There will only be one API, one GUI, one core distribution.
>>>>>>>>>
Entirely an organizational matter. Just release your code under a license that prevents forking. In fact, it would be easier to fork OpenBeOS (which uses a liberal license) than to fork the Linux-based BlueEyedOS (which uses a closed license)!
Some of us are sick of recompiling and the dependancy nightmare on Linux which dwarfs Windows dll nightmares from years ago.
>>>>>>>>>>
1) You’re only sick of it because you have no idea what you are doing. APT/urpmi/Yum solved the dependency/compiling problem long ago.
2) It’s all organizational! Nothing about using the Linux kernel forces you to manage your project in the same way as Linux! At the same time, nothing about using the NewOS kernel prevents people from fragmenting OpenBeOS! Do you even understand where the seperation between userland and kernel space is???
Vassilis Perantzakis,
I’m not sure what you meant by, “(no choices and different time EVERY time.)” however, in defense of Windows XP, you can create an answer file using SysPrep and have XP installed on an a Dell GX260 in less than 40 minutes.
From a 10/28/2003 MacSlash article:
“The process that Pixar developed and uses for migrating OS 9 users to OS X is simply incredible. Using a series of AppleScript and Unix shell tools, they can turn a stock Powerbook into a Pixar ready machine in 15 minutes. They use scripting to set user information to validate to an LDAP server, install various applications and then also handling all of the user creation and customization.”
To keep this post about Zeta though, I like the work they’ve done so far. I understand it’s only a release candidate, so I’m not too picky about the interface quirks. In fact, I like the direction they’re going with it. I appreciate the icon overhaul while still keeping the spirit of the originals.
I have always been impressed with the process handling in BeOS and Zeta doesn’t seem to disappoint in that area. I would rather contribute to a project like this than pay the MS premium all the time. Perhaps we’ll all get something out of it later when MS steals whatever it is and makes Windows better.
Do you A: get rid of backwards compatibility in one fell swoop by fixing the limit?
How would fixing the limit break backwards compatibility?
Ask Eugenia about that one. I believe she was the first to bring it up, but I’ve read from several people ‘in the know’ who all say that fixing the 1BG limit will break binary compatability.
There’s always the possibility that another layer could be added for handling legacy software but you’re talking about a small OS with a semi-small software base. I think it would be suicide to take the software repository on BeBits and throw it to the wind by breaking compatability, yet they shouldn’t take resources to add another layer to the OS. Let them get the OS working on common hardware available today and they can worry about the other stuff tomorrow.
“Is the new font Prima Sans? or something else?”
It is Bitstream Vera with a lot of new characters (and sets) added.
As for NewOS, in what way is it better than Linux? Can’t tell me? Probably because you know *jack shit* about kernel design, am I right?
Well for one thing, NewOS is very clean and neat (code-wise, I mean). If you’re developing your own OS as a hobby project (and I seem to remember that you are, too), sometimes you may take a look at Linux sources to see how things were done, and you quickly think “Forget about this, it’ll take more time to understand this than to reimplement it”. If you look at the NewOS code, it’s hard not to simply copy it lock stock and barrel, because it’s exactly how you would have done it yourself.
This is not a value judgement, by the way. The Linux code is the way it is because it has been revised and improved so many times, and scales well, works on more machines, etc.
If I were to write a desktop OS and didn’t feel like writing the kernel, I would pick NewOS as well.
but I’ve read from several people ‘in the know’ who all say that fixing the 1BG limit will break binary compatability.
So…we won’t break it today. Will it be broken tomorrow then? Or will it never be fixed?
And I’d still love to get more details how that will break things – from my software development experience in BeOS, I don’t see where I’d access memory on such a low level that changes in the VM would affect my software.
Take a look at Eugenia’s post:
http://www.osnews.com/comment.php?news_id=4156&limit=no#128253
You’d have to ask her or JBQ for further details, though.
Thanks.
But still, I think now would be the best time to do things that could hurt backwards compatibility. If YT doesn’t fix it now, they have to fix it tomorrow – they can’t have the 1GB limit forever. As stupid as it would be for Zeta 1 to not be compatible with BeOS 5, it’d be even more stupid if Zeta 1.5 or Zeta 2 was incompatible with Zeta 1. That would scare all the new customers, where breaking compatibility with BeOS would only affect owners of previous BeOS versions, and those should be used to this happening from time to time.
At least for me, the 1GB barrier will be a “to buy or not to buy” factor. I plan to upgrade my computer to beyond 1GB RAM for several reasons.
Ì do not have my copy of Zeta yet. But I will get it to play with it and find out about it. Why do so much here talk badly bout Zeta without having it. I think YT does nothing bad to the community. They sale an OS. But why does the formally great BeOS Community hurt itself? Why going down on a step with some lowlevel arguing Linux users?