Slashdot Mirror
BeOS For Linux
Bob Gortician writes "The BlueOS guys have posted a few screenshots of their progress in porting the BeOS interface to Linux. Note that this is an intermediary step toward a BeOS clone OS. " I actually had a Be machine for a while, and played with it - nice OS, and well thought out, just a problem of very little applications for it.
...anyone have a grammar checker?
Please go to BeNews and read the explanation for these two BlueOS screenshots. The second shot uses a "dummy" content for its windows.
Also, click here to read the Interview with the BlueOS project leader at OSNews.
Driver support. Had virtually no video or sound support, so everything was in grey and mute. I loved the interface, and it booted up as quick as can be, but there's only so much you want to do with no driver support. Why make an application when no one else has a machine it'll run nicely on?
Not that the majority of Linux users care about the Mac, but the fact is that Mac OS X represents something I believe a whole bunch of Linux users should get behind if they want their OS to succeed - It's Linux with the useability that Joe Sixpack can handle. BeOS has its uses, but aside from the glory hack of porting its interface to Linux, I'm afraid that this can only serve to fragment the already small effort behind pushing Mac OS X as Linux's true way to combat Windows, because let's face it - Neither KDE nor Gnome are going to make my mother leave Windows anytime soon...
"Mod, mod, mod...and another troll bites the dust."
If it were the other way around a lot of us would probably be running BeOs on an Alpha chip right now.
The developing of a BeOS clone via this route may yield atleast the se two main benefits:
* Linux and other *nix's will gain another easy to use, mature, comprehensive GUI.
* BeOS will gain from more exposure and may get new development.
This is a great way to continue this great product.
"UNIX was not designed to stop you from doing stupid things, because that would also stop you from doing clever things."
just a problem of very little applications for it. Why were the apps so small?
Did it help to have a bigger monitor?
When you iconify a very little application, does it disappear?
Jeez, no wonder BeOS failed.
No one would write a lot of apps until it had a larger user base, no user base would be generated until it had more apps.
It's the same set of problems Linux has faced in the past. BeOS was/is a fine OS, but it never seemed to have a good backer, nor a solid niche. Artsy types already prefer Macs, so it's hard to compete there. Ordinary desktop users have already been won over by Microsoft, so it's really hard to compete there. Linux users already had a free OS and a nice looking desktop if they wanted it (re: KDE, Gnome. You should know that by now).
I think that BeOS was a nice, stable OS that could have been a contender. It's a shame it didn't get more press or attention from major industry players. Oh well, I look forward to another nice Linux desktop all the same.
My sigs always suck.
I've been here long enough to know that this thread is going to devolve faster than you can say TinyX into the only real challenger to the vim/emacs war; the Linux on the desktop debate.
To save time, we should just symlink to the earlier desktop discussions. Here ya go.
I'm just sitting here wondering why we seek the acceptance of Joe Sixpack.
Blearf. Blearf, I say.
Lack of applications was probably the majority of the problem ... they were late in the game... very late.
But to me, personally, I think when they call software for Be "BeWare" I take it as a WARNING rather than a product.
The FAQ says its not under the GPL. Then what licence is it under? Is it free software?
Ummmm, try "just a little problem of a monopoly using illegal business practices to make sure it never reaches the market."
Hell, PC manufactures were offered the operating system for free if they would just put it on their machines! (Not even replacing Window$, but dual-booting along side it!)
To get the apps, it had to have some penetration in the market. It was penetrating the Microsoft controlled market that was the problem. Oh wait, are we talking about that anti-trust thing again!?!?!
For a time, I used (and loved) this operating system. It had enough apps to do *most* of what I needed to do.
-Derek
A lot of people, including me, are transitioning from Linux to the Macintosh. The Mac has all the deep applications people need, while including all the coolness that is Unix. (Call me when something like Final Cut Pro or After Effects runs on Linux. And when the fonts don't look like sixth-grader crayon sketches of text :-( ).
That being said, we must say that a certain amount of variety in the computing world is necessary. Some people don't have $1,299 for an iMac (assuming the low-end model comes out sometime in the next century or two). Some of them want to build computers themselves, or buy an eMachines with a crummy 15" CRT monitor picked up at a garage sale for fifty bucks.
We can't convince these people to buy Macs; Macs are always going to be a bit for the elite, a bit for those who like spending money on fine technology. They need Linux just as we need Macs. As long as they are off the dreaded Windows, we shouldn't turn against them; if they grow older and richer, like I did, they will appreciate the better things in hardware soon enough.
So don't be against this kind of project. If it can make Linux more cool, well, those who learn it are learning the same basic operating system that underlies Macintoshes. So there should be more cross-polination between the two worlds, which I feel is all for the better.
Advocate the Mac when you can, but don't consider linux the enemy. We have a common enemy, and you know what that is. All too often we get injured in internecine squbbles instead of taking care of the most important advesary.
Hope that helps.
D
That's why I prefer Windows, where all my applications are *huge.
-- @rjamestaylor on Ello
While BeOS had a nice GUI, its read strength was its highly efficient threading model, which made the OS very effecient and responsive. The OS was especially adept at efficiently utilizing multiple CPUs.
While it is certainly nice that Linux users will have the opportunity to benefit from a nice new GUI and API, the best part of the OS, alas, is being left behind...
*** Quantum Mechanics: The Dreams of Which Stuff is Made ***
I run Darwin w/o Quartz. I still have all the Quartz/Aqua stuff installed, but quit out of it at boot. I run Darwin over Debian PPC for a couple reasons:
1. Power management: Darwin works very well with the hardware (iBook). This manifests in more than one place. Longer battery life, and sleep works well (by shutting the lid).
2. The option to go into Quartz/Aqua: One of the things I hated about using Linux/x86 (what I ran as my main OS for a couple years before getting a Mac) was having to reboot into Windows to play a game or use certain useful applications for which there's no equvilent in Linux. This was true back when I used Linux a lot more than it is now, though. But with OS X, I can play games and run real, useful apps. And if I choose to run XFree86 straight out of the text console in Darwin rather in tandem with Quartz, I can always quit X11, and go back into Quartz. A lot less hassle.
Working toward a usable PDA environment in the spirit of Newton OS: Dynapad
At that time the video was MUCH better than Windows (probably still is) and I'm sure some games would have come out with BeOS versions (on the same disk as the Windows copy) and people could immediately see the better performance. That would have opened the gates to people using BeOS more and more, and applications being written for it.
BeOS may very well have failed and been wiped from the disk by every user too, but at least it could have tried!
But MicroSoft stopped this from ever happening. This is a fact and trying to pretend otherwise because "everybody here hates MicroSoft" is not going to change it.
Gotta love that stripped down minimalist look in the screen shots.
You can create the most beautiful, well-thought-out and consistant UI for an OS, but if the individual apps are written with sloppy UIs, it all falls apart.
The one problem I have with Linux is the fact that 90% of the GUI apps have simply idiotic user interfaces. I burst out laughing the first time I used Linuxconf. The dialog window that popped up the first time it ran had a "Quit" button instead of a "Close" button. That is a perfect example of the misleading, inconsistant and just difficult to use interfaces plague the platform. There needs to be some sort of effort put into implementing a consistant UI across all apps, or else all of this work will be for nothing.
On the Mac, and to a slightly lesser extend on Windows, almost every app is interacted with in the same way. A user knows what to expect when they start just about anything but a game. And while you can argue what paradigm is the best, the fact remains the consistancy is the key and Linux lacks not only that, but a core set of accepted design principles. You can argue this will somehow curtail your "freedom" or something all you want, but the fact remains it is a solution that offers much more promise than the embarassingly ameturisih one we currently have to suffer through.
Badly designed user interfaces make Linux look bad. It's simple as that. When Linux looks bad, it's adoption rate is affected. How do people expect to combat the negative stereotypes of the platform if they are unwilling to band together to overcome the easiest to fix, yet most glaring problem with the OS? This isn't as much about asthetics of Linux apps as it is about the success of Linux itself.
If you think "Oh, I just use the command line" or "Who cares, let them program it themselves" or "It's pretty, so what's the problem?" you are being ignorant of the demands and expectations of those you care attempting to bring over from Windows or wherever.
Drop the elitism, drop the selfishness, just realize what needs to be done and understand the awful truth of the computing industry, one that seems lost on most Linux developers:
Give them what they want, or they will go away.
It's not about what you want, it's about what they want, how they want to work. Never forget that. You can't force-feed them every paradigm change and excuse for every bit of laziness on your part. You have to adapt to their needs and adapt quickly. You only get one chance to make a first impression and pissing them off by acting high and mighty about changing things to make their lives easier is not the way to do it. Many a promising platfom has died because of this, don't for a second think Linux is immune to the negative effects of the choices made by its proponents.
People need to realize that ignoring this sort of thing forever will somehow fix the problem, or that we will slowly somehow overcome it. I don't think that meshes very well with reality. It's going to take a clear and consistant vision with a lot of effort on the part of the developers and users to overcome this impasse. And believe me, it is an impasse. The platform is currently reaching critical mass and a point where it decides where it wants to go, and what it wants to be. Sure, this is going to be unpopular, but I don't care, I'd rather get modded down to oblivion than let this go unsaid. Because it needs to be said, and it needs to be appreciated, if not neccesarily liked.
I'm sorry, that's like saying Sandra Bernhard is as ugly as Madeleine Albright.
I mean, come on, they may both be ugly, but they're in completely different leagues.
Anybody remember FinalScratch?
Ironically FinalScratch now runs on Linux
-adnans
"In short: just say NO TO DRUGS, and maybe you won't end up like the Hurd people." --Linus Torvalds
While I am not involved in the BlueOS project, I think my work is complementary to theirs. Eventually, it should be possible to boot from a BeFS volume, compile and run BeOS apps, and not know that it is the linux kernel underneath it all.
Also worth a look is the OpenBeOS project, who have their own implementation of the Be Filesystem (which is actually progressing faster than mine).
Will Dyson
"We can't stop here
Not being open means that the platform is subject to the whims and fortunes of a SINGLE CORPORATION.
If you had any clue at all, that should worry you. NOTHING in your life should be so dependent.
Such situations are merely vieled communism with all the problems that implies.
A Pirate and a Puritan look the same on a balance sheet.
If you feel adventurous you can try and duplicate what this guy is doing:
t ml
:) for doing sound output and scratching. I'm currently working on a software only scratchin system, much like TerminatorX.
http://www.cs.ubc.ca/~tbeamish/digitalturntable.h
They are using AlsaPlayer (shameless plug
Incidentally, I just heard a rumour that FinalScratch is indeed using some parts of AlsaPlayer, the plugin system. Wonder if that's true.
-adnans
"In short: just say NO TO DRUGS, and maybe you won't end up like the Hurd people." --Linus Torvalds
IBM is investing a BILLION dollars in Linux and is now pushing it on their flagship server line.
If that's "dying", I would love to be so fortunate.
A Pirate and a Puritan look the same on a balance sheet.
IT seems like this crowd has a terrible 'way-back-itise'.
I realize BeOS was a great OS. It had allot of terrific ideas. Im not suggesting Be was worthless. And I also recognize that people can scratch their own itch if they damn-well-please. Im not about to tell anyone what they should spend their Coding-Karma.
BUT
What is the point of rebuilding BeOS completely and totally? Why not move on? Steal (with pride) Be's good features and ideas, assimilate them into GNU/Linux and move on. Trying to re-implement something to have a work-alike seems more like drugery and backwards-itis.
I wish the BlueOS people great luck - but I would suggest that they should implement the features that they love into GNU/Linux and move forward - the beauty of the GNU/Linux system is that the best rises to the top (because there is no impedment by irrelevant externalities(sp?)) - if BeOS's features have merit, and they are implemented in GNU/Linux, then they will have a home.
Replicating Be just doesnt make sense to me.
Part of the 'hours' were spent looking online for a player that didn't suck as much as iTunes. (I never found one for OS X, unfortunately, and I can't see running the whole classic environment for just an MP3 player.)
But the consistency problems are hardly minor. They're part of a disturbing trend with Apple -- they're moving away from usability as their primary concern and going toward flashiness. Sadly, I've had much better out-of-the-box experiences with Linux (mostly Mandrake, but Red Hat is getting better and better) than I did with OS X. OS X is frustrating to use... Linux Just Works (tm). It's all a matter of taste. But I still wouldn't set up my grandmother with a Mac.
(I pick on iTunes because it's the single most frustrating end-user app I've ever used. It won't play OGGs, ignores some directories of MP3s at random, is a pain to reorder files in (I have them sorted by filename in subdirectories for a reason, thank you. If you're going to sort by ID3 tags, at least do it by track number instead of track name!), etc. etc. It's just painful to use.)
- fader
Goodness, Windows and MacOS have been using them for /years/. Why the hell hasn't the Linux community taken a hint? Every X program I run has a different dialog, the common thread among all of them being that, for the most part, they are all crappy.
Liberty uber alles.
Implement the best features from each OS, and invent your own where applicable.
I'll list a few examples of great features from various OS' along with some features of my own, and why they should be implemented in the BlueOS or OpenBeOS projects, and any other projects which want to create a great, fast, reliable OS:
1. BeOS' "tab" window bar, which doesn't span the entire "length" of a window. Why should the window tab span the whole length of the window? That just takes up extra space. Have the close, the maximize, the minimize, and the "barrize" buttons in a tab. Have all these features automatically "display" when you move your mouse over the tab; otherwise, have the name of the app. and file displayed. Furthermore, make the "tab" movable accross the length of the window. This allows you to "semi-maximize" all your windows such that their titles display in tab form accross the screen. Should able to be pulled up by keyboard commands, and navigated by keyboard, as well as mouse.
2. Apple's "universal menu". Why have a menu in every window? That just wastes space. The argument could be made that its inefficient to have to move your mouse WAY to the top of the screen for a small window, but you can always make it so users can "retrieve" the universal menu to their specific window, or send it back to the universal position, for each program. Furthermore, the universal menu should have the option of "auto-hiding" away, like Apple's "warf" or Win9x's "task bar". Should able to be pulled up by keyboard commands, and navigated by keyboard, as well as mouse.
3. The desktop. This is a BIG DUH. Though task-bars and warfs are nice, having icons on the desktop is still a must; it should at least be an option. But the desktop shouldn't just be complacently left alone, it should be improved. People should be able to make desktop regions, so when they "auto-organize" icons on the desktop, some will stay on the bottom, or top, and others on the left/right side, rather than all being automatically placed on one side. Should able to be put focus on it by keyboard commands, and navigated by keyboard, as well as mouse.
4. The warf. Another duh. Apple's or OpenStep's version is a great implementation. It should be scrollable, and should "hide away". Having icons or icon names "enlarge" or change color as you move over them should be an option. Should able to be pulled up by keyboard commands, and navigated by keyboard, as well as mouse.
5. The taskbar. Not that its completely original, but it is a nice feature in Windows. Having all the application titles appear in boxes, and having a customizable start menu with lots of neat features is nice. Also, having an address bar in the task bar is nice. Of course, the management of displaying window names should be improved, and task bar should allow you to scroll left/right or up/down rather than "shrinking" down the boxes when many windows are open. Should be hide-away.
6. Apple's new "file browser", Cocoa or whatever its called. Of course, its not new, but just a pretty skin of NeXT or OpenStep's file-browser. But the new folders displaying to the left of the old one's and scrolling right is nice.
7. A throwback. F1-F12 as FILE MENU KEYS. Alt-F for "File" as is typical in Windows and Linux, or no key-control for file menu's as is typical in OSX is CRAP. The KB is quicker for accessing file menu's than the mouse, but why should we have to press TWO buttons to access the file menu's? Also, it means we always have to look at the "underlined" letter to see which letter we have to press in combo with Alt. It would be much easeir to just ALWAYWS have F1 representing the first menu. This standardizes it unilaterally.
8. Right-clicking (Win9x) and "hold-clicking" (OSX) to get the "options menu". Great features. Should be combined. On a two-mouse button with a "scroll button", there should be a function for a left click, one for a right click, one for a double left click, for a double right click, for a hold on a left click, and for a hold on a right click.
9. Space-saving by dissapearing buttons. One great idea which might actually belong to Windows, though probably not (just I first noticed it in a MS Windows program; note, I said !might!, so don't jump all over me). Anyways, the feature is in the Windows DVD player that is part of Windows Media Player (just go to Xteq, and click on "enable DVD functionality" under Windows media player, fyi). When you watch a DVD, the buttons for play/forward and all the others are initially visible, and look normal (though small). Then, after a period of inactivity by your mouse, they dissapear. A nice feature! When u move the mouse again, they reappaar. This saves space on your screen while still having all the functionality, and gives you more room for your actual work. I think people should explore implementing this strategy accross many different applications, from browsers to word processors to image editors.
10. The "between space". Most of you are probably in front of a graphical web browser now. It probably has buttons at the top of it, with forward, back, stop, home, search, favorite, and history functions. These buttons probably have a "grey space" between them which serves no purpose. Why have useless space between them? Why not make it transparent to the underlying content of the window, with the buttons as opaque layers on top of the content of the window? This can be combined with #9.
11. The UNIX power. Ok, this is broad. But what I mean by this is the vast vast vast vast array of command-line commands you see in UNIX-like OS' such as IRIX, *BSD, and *Linux. This is a feature all OS' should have.
12. The UNIX-stability/security. Again, obvious. But should be pretty self-explanatory. Unices have a reputation for being stable and secure.
13. The hardware/software support of Windows9x. This is something easier said than done. It basically happens over time. Linux is getting there, so is Apple. This would be a factor totally based off the quality of the OS, were not MS a monopoly. But, as it is, no good deed by companies competing with MS goes unpunished; no vile deed by MS unrewarded.
14. The ease of use of BeOS, Amiga, and Apple-OS. This is another general feature. But these OS' are widely reputed as being easy to use. I believe its because of the KISS (keep it simple stupid) philosophy. Of course, Apple has standards almost set in stone for GUI's. But as some simple guidelines, always consider what the function of your program is, and if extra features aid in that function? Does that neat-looking (self-promoting) logo in the corner really serve any useful function? Or is it just eye-candy, something to be shown off in screen-shots? Make sure every graphical feature, button, whatever, in your GUI/apps has a function, and aids in ease of use as much as possible. In short, critically evaluate everything.
15. Condensing functions. Condense the functions of several related buttons into one button. I.e., I created a nice, efficient, easy-to-use system for media-player buttons. Have a play/fast-forward/next-track/next-CD button, a reverse/rewind/previous-track/previous-CD button, and a pause/stop/eject/open button. In each case, the first function listed would be done by single left-clicking; the second function, by holding a single left-click; the third, by double left-clicking; and the fourth by right clicking. This allows you to compress what would be 9 buttons into 3 buttons. More efficient, easier to use (as less hand-motion, and more intuitively like an MP3-player), and less wasteful of space. This would be, of course, combined with #10 and #9.
16. Load-time, run-time, RAM, and hard-drive requirements. These are all the performance-related issues. I believe a few distinguished OS' represent excellence in these fields: IRIX, BeOS, Amiga-Classic, Amiga (the new Amiga), QNX, *BSD, and some Linux' (i.e., Slackware, Debian). The reason for such excellent performance offered by these OS' is a combination of factors: efficiency, minimalist philosophy, innovative architectures/ideas, etc. I won't go into details, but sufficed to say, developers should be considering factors such as HD-size, load-time (ESP LOAD TIME for most apps, nothing worse than waiting for ever), RAM (another biggie, don't want the OS taking up half of my RAM), and run-time (a biggie for apps which do any serious crunching, such as phylogeny apps, or DNA alignment apps). A comparison of various OS' in terms of tech-stats can be found here: http://maxlinux.hypermart.net/comp_chart.htm. It can be observed that IRIX wipes the floor with everything else in every category. 9 million terabytes as max file size?
17. Transparent features. This one's a bit touchy. It shouldn't be over-used. You need to be *very* selctive when using this feature...but it can be great for certain apps, like terminals or word-processors (Office products, Vi, Emacs), and for certain parts of apps (like configuration boxes, occasionally). Obviously, its idiotic to make the material of a web-browser or image-editor (semi)-transparent.
18. Aqua/glassy/smooth/gradiated stuff. Obviously, MacOSX has a great-looking GUI. It isn't just eye candy -- it really helps you easily distinguish features from one-another. Lets not give Apple too much credit here. They just "heard" what the consumers wanted. Everyone likes gradiated stuff, which is smooth. There was gradiation and shinyness long before OSX's Aqua theme.
19. Now, an annoying, but *sometimes* useful feature. "Animation" effects on menus or windows. I.e., a menu "scrolling" into place, or window fading away when minimized, rather than doing so instantaneously. All such effects should be quick, should be such as to indicate what's happening, and should be configurable and deactivatable.
20. Plug & play, automatic hardware recognition. Another DUH. Windows and Apple have accomplished this to near perfection by sheer brute force. QNX has a more clever method, which involved some kind of "detection algorithm" to detect hardware and optimize the OS to it. I think this is the way to go. I.e., have the OS "search" for say a graphics, CD, CD-RW, CD-DVD, DVD-R, printer, speaker, sound-card, networking, etc hardware. Then when it finds the (say) graphics card, let it explore various values of that card in a conservative way (starting from very low values that won't mess up any hardware) and gradually working up, using some benchmarking and stability tests to find the optimal settings.
21. Cross-platform compatability. Amiga has achieved this by using VP Assembly; thus, their OS can run on virtually any hardware. You need to bite the bullet on this one. The initial "performance" decrease may be compensated for by less overhang because the stuff your loading from the HD is smaller...furthermore, just improving ONE part -- the virtual machine -- increases performance of everything everywhere. Also, the "performance" u might initially lose is moret han made up for by the additional efficiency you can put into it by having more time to work on better algorithms/smoother interfaces, b/c u don't have to port.
social sciences can never use experience to verify their statemen
the window in screenshot #2 has the exact same selections as the screenshot from three months ago. I certainly hope it is coincidence. Note this isn't just preserving settings; the older screenshot was proviced by an outside source.
Use my userscript to add story images to Slashdot. There's no going back.