There is a new version of the GNUstep LIVE CD, version 0.5. All GNUstep was software updated, and alot new GNUstep software added. New game: Supertux. Debian GNU/Hurd K6 mini iso included. LaTeX, TeXmacs, Emacs added. TV software. Web browsers are dillo and links2. QTParted.
GNUstep Live CD 0.5 Released
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
That’s probably the most hideous thing I’ve seen since Windows 95!
How come all this isn’t completely outdated? What’s the point of this?
Yeah, I know, I’ll probably get modded down… It’s just my initial impression from the screenshot.
– Simon
It’s a pitty that GNUStep never caught on. The speed of development for it could have made GNU/Linux progress much more quickly.
Just because it looks old doesn’t mean it is. Gnome is still written in C. As the document, “Why you want GNUstep” would tell you, GNUstep has the best object-oriented design today, native unicode support, a simple and consistant API, etc. There is a lot more to a computer than what can be shown in a screenshot. Would a new theme for Windows 95 make it better than all of the OSs today?
>That’s probably the most hideous thing I’ve seen since Windows 95!
Everything is relative. Yes, today, it looks hideous. But it looks 99% the same as NeXTSTEP did (http://www.osnews.com/story.php?news_id=4042) and when NeXTSTEP was big in the beginning of the ’90s, everything else looked terrible back then.
> What’s the point of this?
I don’t think there is a real point, at all. Maybe a bit of melancholy and “bringing back old memories” for the people who used NeXTSTEP back then and maybe to provide a prooving ground for GnuSTEP, but that’s about it.
If there was a REAL effort to extend GnuSTEP and its surrounding software in terms of an operating system instead of just catching up with the OpenSTEP API implementation, to continue where NeXT *left off* and *innovate* and create the next-gen of the NeXTSTEP system, then yes, there would be a point. But something like this can not be achieved with 2-3 people having fun over the weekend.
I’m right now in the middle of porting my application for qualitative analysis from cocoa to GNUstep. I see this as potentially vital for my world–qualitative research in education–which wouldn’t touch Linux with a 10 ft something. A live CD which could run my app would be a major boon.
Don’t be so superficial. Writing a new theme for GNUStep would be a one man-month job. What matters is that GNUStep itself is a very powerful and throughly modern*. The highly dynamic, easy-to-use API that people rave about MacOS X having is a direct descendent of OpenStep, and thus is closely related to GNUstep.
*> “Modern” being defined here as “1980’s era Smalltalk-inspired technology that is still more advanced and forward-looking than what most developers are using today.”
WindowMaker is themable and can be easily made to look better.
http://www.windowmaker.org/gallery.html
Besides, there aren’t many LiveCDs out there with WindowMaker as the default window manager. WindowMaker is fast, so it makes sense to use it in LiveCD.
I passed the previous version of GNUstep LiveCD because it didn’t have webbrowser. However, this version sounds better. I kinda like Links2, because it’s fast. Hopefully this means that Links2 finds its way into the official Debian repositories, too.
I’d been running the SimplyGNUStep Debian sources for awhile, and they’ve gotten *very* good. Almost exactly the same as what I’ve got on my NeXT machine, not to mention worlds faster. They’re doing a good job re-creating the old environment, and also making it possible to do new things with it. Remember, OSX is not that much different than the old OpenStep framework. There have been some modifications here and there, but for the most part, the old NeXT/OpenStep apps will compile and run under OSX, and now GNUStep.
I will probably grab a copy of this, to see what he’s doing, as opposed to the SGNUStep people.
What’s with the GNU/Hurd installer on the cd? :-p
I must say that I’m quite surprised by your comment! I thought at least You would know better.
Heh. GNUstep is one of the few GNU projects that I actually like. I personally think it’d be really neat were they to build one of these live CDs using the Hurd as it’s base.
OT: it’s a shame that the FSF has only a handful of developers working on these two projects (GNUstep and the Hurd), as they are two of the more interesting ones going.
>I thought at least You would know better.
I know better that’s why I said that. Yes, Gnustep Live CD is nice as a hobby and might be the testbed for Gnustep API (which I like), but beyond that has no real value to either the industry or the technology in general. Without doing some real work, research and extend what openstep once was (in terms of the whole system, not just the API), I don’t see much point into this either.
SkyOS and Syllable have more value than this Gnustep Live CD for example. They explore possibilities and try to extend the whole operating system thing. Gnustep Live CD on the other hand, creates yet another Linux distro that happens to have Wmaker and gnustep apps on top instead of kde/gnome ones. Again, that’s a great hobby, great testbed for the Gnustep api guys, great for remembering the nice NeXT system, but beyond that, it is useless. It brings nothing NEW to the table for the technology and the industry.
should have been picked up instead of Gnome — when Gnome was started as an answer to KDE, GNUstep was already a GNU project. Anyone not suffering from a horrible case of NIH-syndrome would have recognized that. If only the stated goal of the project wouldn’t have been (as it still is) not to make a desktop environment, but an application platform. As if a GNUstep application would blend in in any other environment at all.
But at that time, GNUstep was still building a display postscript renderning backend (not so different, in the end, from what Cairo is doing now), and lots of other interesting things. Taking out ads in Linux Journal even…
And nowadays, there are themes for them as likes that kind of thing, and Renaissance does pixel-independent widget layouting. There are apps: Workspace and GNUmail are quite capable. And the environment is still a whole lot more pleasant to use than OS X.
Writing a new theme for GNUStep would be a one man-month job.
Someone has already done that.
http://www.roard.com/camaelon/
I haven’t tried it yet though. GNUstep apps are pretty ugly but it hasn’t bothered me enough to warrant changing the theme. I do think it would be a good idea to modernize the default theme though. It couldn’t hurt, and it wouldn’t be a big deal to do. Have you seen how beautiful GNUstep apps look on OSX?
http://sophos.ca/crap/gnumail_osx_2.jpg
It looks much better with correct gamma.
http://www.linuks.mine.nu/gamma/
xgamma -gamma 1.6 *hint, hint*
GnuStep was always hobbled, I believe, because no one wants to deal with Objective C. Yes, it may be easy to learn and powerful, but its world is too narrow – just OpenStep and Cocoa, for people to want to devote their time to it.
Apple has created Java bindings to the openStep/Cocoa API, but I’m not sure how popular they are, or if GnuStep supports them.
GnuStep was always hobbled, I believe, because no one wants to deal with Objective C. Yes, it may be easy to learn and powerful, but its world is too narrow
Personally, I think that if a person chooses to not use a technology that is both powerful and easy to learn, that it is not the technology that is narrow, but rather the mind of the person making the choice.
When you considder how many C programmers and libraries there are out there, your argument really ceases to make a lot of sense, as ObjC is not that hard to learn when you already know C.
GnuStep was always hobbled, I believe, because no one wants to deal with Objective C. Yes, it may be easy to learn and powerful, but its world is too narrow – just OpenStep and Cocoa, for people to want to devote their time to it.
Apple has created Java bindings to the openStep/Cocoa API, but I’m not sure how popular they are, or if GnuStep supports them.
objective-c is a fairly simple, and extremely powerful language. Any developer with enough sense who’s productive in c or c++ can pick up objective c and master it in days to weeks. if you skim through my tutorial at http://otierney.net/objective-c.html you can get some easy examples running in an hour.
“Personally, I think that if a person chooses to not use a technology that is both powerful and easy to learn, that it is not the technology that is narrow, but rather the mind of the person making the choice. ”
Well that explains Lisp and Smalltalk then. The nice thing about the NextStep platform was the development tools. Programming should be that nice.
Eugenia, before starting to “improves” OpenStep, we first need to have a good implementation of it … Doesn’t it make sense ? Keep in mind that OpenStep API are really good — Apple didn’t change them, they just added new classes.
Saying that this GNUstep live CD is just linux with wmaker and gnustep apps and thus not innovative could be true; but you could also say that MacOSX is just BSD plus OpenStep plus a compatibility mode for old MacOS apps, and thus not really innovating either. GNUstep architecture is quite flexible, and could be a testbed for new ideas — I sure hope it will be anyway.
As it is a free software project, sure, it’s an hobby for most of the coders working on it. What’s wrong with that ? It would be good if there were more companies working with GNUstep technology, sure (even if there are a few already). It would be good to have a major corporate support and money flowing around us. But how could you critic the project for not having that ?
Else, it’s not because the technology already exists or is “old” that it’s a bad thing to want to have it — OPENSTEP was way ahead of its time anyway. I much prefer it to what’s around today — it’s not nostalgia, it’s just that it was a better system.
Using linux as a kernel is a logical choice, as GNUstep focus on the application side, not the OS side — anyway when/if Hurd starts to be useable, it would be easy to have a GNUstep port. In the meantime, using linux make sense.
Is GNOME or KDE bringing something NEW to the table ? I don’t think so.
Well said!
It’s fantastic in many ways, I just wish that it could get the level of integration and cooperation between apps that would really allow it to shine, but is only possible if more people start programming in objC. I haven’t downloaded the iso yet (45%), but I’m excited to try it.
Keep up the good work on GNUstep, guys. Alot of us do appreciate it!
Apple has created Java bindings to the openStep/Cocoa API, but I’m not sure how popular they are, or if GnuStep supports them.
BTW;
http://www.roard.com/gnustep/Booklet.pdf
Is that wmaker looks totally dfferent from the GNUstep apps. Is there an Objective-C, GNUstep wm?
Nope. There’s been discussion in that direction, but I don’t think one’s production-quality yet. I think you could probably also use IceWM with a *Step-style and get a similar effect.
“Without doing some real work, research and extend what openstep once was (in terms of the whole system, not just the API), I don’t see much point into this either.”
Then I am afraid there is a little fuzziness from your part when it comes to understand what OpenStep was: it was the API. That is the whole point of OpenStep, to berelatively OS independant. The kernel was NeXTStep for the x86 and 68K boxes, Solaris (dunno if the HP-UX version was ever released) and NT. NeXT had a weird way of labeling their products.
The beauty of OStep is that theoretically you can develop an application in any of those systems and port it w/o any overhead between target platforms. So think of OpenStep as a development system not an OS. In this case GNUStep are following the spirit of OStep, it was not supposed to be an OS per se. And it is still relevant because there is still enough of a porting problem between OSs to make it an still open question.
Why doesn’t Apple release this as a competitor to .net I wonder?
I know what the OpenSTEP is. You didn’t have to write all that lecture, I have a NeXTStation in my office (and I linked to my NeXTSTEP article on my previous article).
What I meant was that if these guys want to make a whole bootable CD (which is a whole OS), they should work on all the aspects of the OS to extend NeXTSTEP, not just trying to –finally– complete gnustep.
“What I meant was that if these guys want to make a whole bootable CD (which is a whole OS), they should work on all the aspects of the OS to extend NeXTSTEP, not just trying to –finally– complete gnustep.”
What for? Linux does pretty much everything that NeXTStep as a kernel did, so what is it missing? Probably a GNUStep system based on a *BSD or Darwin would be far closer to be a true sucessor to the NeXTStep/OpenStep. But as far as the GNUStep folks are concerned, I assume they are only interested in the OpenStep stuff, and they are doing a superb job on that area.
So I really do not understand you critizism….
Are you sure you know what OpenStep and OPENSTEP are? Hint: Your NeXTStep article Missed The Point Completely.
Are you aware that The “GNUSTEP LIVE CD” has dick to do with GNUstep, and that it’s the work of one guy?
it’s not any better than the original style. there are some improvements, but than there are things that makes it worse.
Simon, do you judge on women just what they look like? have you never heard of inner beauty? GNUstep (and also OPENSTEP) have something which attracts me. Something that GNOME, KDE nor Windows have. Please before you judge on what you see. Try feel it up first.
“they should work on all the aspects of the OS to extend NeXTSTEP, not just trying to –finally– complete gnustep.”
Should -> I think you have nothing to dictate…
Extend -> I think the plan is basically to first complete the current features. Easier also. Many other projects do that. Come with an alternative roadmap if you like, present it, and present arguments with it. IN DEPTH arguments.
You also make the assumption the author of this Live CD is the work of ONE man who is NOT a GNUstep developer. Just a person who uses the GNUstep software as part of a Live CD. Doh! He should be coding right? Instead of posting his project here… *sigh*
How does GNUstep compare to GTK for app development? Is the pseudo object orientation of GTK using only C far more cumbersome that the Obj-C GNUstep combo? I.e., would evolution and gimp have been easier to write with GNUstep?
Please let me give you this to read
http://www.gnustep.org/information/statement.html
What I meant was that if these guys want to make a whole bootable CD (which is a whole OS), they should work on all the aspects of the OS to extend NeXTSTEP, not just trying to –finally– complete gnustep.
Oh, then, I agree. But the guy that is doing this live cd isn’t a GNUstep dev — he packaged GNUstep apps for debian and then had the (good) idea of a live CD. So he’s not trying to complete GNUstep, he takes apps, package them and put the result in a live CD. Of course it will be interesting to have the different aspects of an Operating System covered by GNUstep apps, but that’s a work in progress. In any cases, kudos to tarzeau for his work, as at least it’s an easy way to test GNUstep for people.
I don’t exactly see your point with extending NeXTSTEP though. If you speak of the kernel space, really, there’s no point — the linux kernel is good enough, and is handy to have good material support. Plus, GNUstep isn’t tied to linux at all, and an equivalent OS could be done using a BSD kernel, Hurd or Minix anyway.
So I guess you’re speaking of the userland … and try to provide the user a new and interesting interation. Well, then, it’s more on the application level than on the GNUstep level : the basic thing is that GNUstep APIs *are* good, there is no point in changing them, really (add some new APIs, yes).
For example … personally, I’d like to have some file manager with metadatas support for managing your files, etc. My point is that it could be quite easily plugged on the GNUstep APIs, and it’s mostly an application problem.
Anyway. Even if GNUstep mimic 1:1 NeXTSTEP, well, don’t you think that it’s quite different user experience than what’s available today ? The docking metaphore, hiding apps and moreover the services are imho quite originals. MacOSX support all that, but it’s frankly not used as much as it was on NeXTSTEP, particularly the services (and well, the docking metaphore is quite different).
Sure, this experience is the same as ten years ago on a NeXT Station. If we considered it as a bad user experience, effectively we wouldn’t have a point to recreate it; but actually, it was a very good UI — didn’t you find it great on your color slab ?
So… instead of harsh criticism, coud you explain more what you’d want to have with GNUstep (the API) ? or an hypothetic GNUstep-based OS ?
Thanks all for your input and for the links. I’m just starting to learn about GNUStep and this thread has been a source for some great info!
“Simon, do you judge on women just what they look like? […] Please before you judge on what you see. Try feel it up first.”
I’m sorry but I just have to smile on this(1).
(1) Yeah I know, English not native language. Nevertheless it’s amusing.
Correct me if I’m wrong, but isn’t a “Live CD” supposed to be bootable? I downloaded the ISO, checked the md5, burned the CD, verified the CD, then stuck it in a couple different PCs. Neither think the CD is bootable. It’s the first bootable CD I’ve run across that doesn’t boot in the slightest. It’s not like it starts to boot, but crashes; the PCs skip it altogether after checking the disc.
This cd is bootable. I’ve had no trouble whatsoever booting from this cd on multiple boxes. Problem with your burn perhaps?
I’m actually very pleased with this cd. Not sure I understand all the negative comments early in this thread, KDE and Gnome certainly have their place, but Gnustep is where it’s at as far as I’m concerned. the Gnustep dev environment is just great. I usually keep a Gentoo box around running Windowmaker and all the Gnustep apps but it’s sure nice having it all pre-packaged together. I wish the shopping link worked so I could at least kick a couple of bucks in for the effort.
It’d be cool to see the login.app application integrated into the boot sequence as well to really complete the NeXT experience.
That’s good to hear. I’ll try another burn and see how it goes. I rather like GNUstep and was anxious to try the Live CD. Very strange that it reads just fine and verifies, but doesn’t boot… oh well, I’ve seen stranger things before.
Off the top of my head I can’t think of why it might make a difference, but if it helps I’m using Roxio Creator Classic and selecting “Record Disc from Image…” from the File menu to initiate the burn. Hope it helps.
As far as I can tell, the problem is the boot order in the BIOS. This Live CD doesn’t boot like most bootable CDs. The systems in question only allow for the following boot orders with the CD: C,CDROM,A or CDROM,C,A. This disc appears to need CDROM,A,C to boot. No other bootable CD I have shows this deficiency. Very strange, but this is a fairly common BIOS. It’s also the latest for the model, so I can’t fix it by updating the BIOS. I guess I need to email the Live CD author and have him look at the Linux CDs and use the same method they do for making bootable CDs.
Okay, I give up. I’ve tried yet another computer which has full control over the boot order, burning another disc, burning using another app, getting another image from the net, and it STILL won’t boot on any system I have. It’s NOT bootable!
lame. I swear to you it boots; I’m posting this from the Dillo browser right now. this is the fifth machine I’ve used the cd to boot up on. no problemo. sorry – without more info I don’t know what to tell you. other than don’t give up, of course.
Hello J.F.
I have here at work also one machine (ASUS board, award bios) that can not boot the cd, I forwarded it my Morphix specialist friend for investigation. On all other machines I have had no problem.
If you want you can bug about it on the morphix channel in irc.gnu.org (freenode) as well.
Cheers
I’m frustrated, but not giving up (despite my last post). It’s just very strange. On every system I’ve tried, I have successfully booted:
Mandrake 10, Fedora Core 2, SuSE 9.0, SuSE 8.0, Linux 2000 (whatever version that was), FreeBSD 5.1, Windows 95, W98, WME, W2K, WXP, WXP Pro, WXP64 (on the Opteron), SkyOS, AROS, and BeOS Maximum (I might be forgetting a few there… I love to try new things).
This is seriously the first thing I haven’t been able to boot. I’m going to find out what’s wrong if it kills me!
Since you mentioned it, I think all (most) of the systems I use are using Award BIOS. Maybe it’s a factor. Since it’s based on Morphix, I’ll start by Googling for other reports of Morphix not booting and see if anyone found anything, then try it myself. It’s how I handled the “FC2 doesn’t dual-boot Windows” problem that’s already been solved.
GNUstep is cool.
you guys already knew that
man, I’m really enjoying this distro. Over the weekend I actually took the time to move a couple BeOS and PhOS partitions around to do a full install. The morphix installer is very nice and it was especially convenient to be able to be able to boot right into the same monitor settings I booted the cd to. The only hiccup was it didn’t recognize my NIC right off so I had to run modprobe and set up networking by hand. No biggie.
Now if I could just remember to type “apt-get” instead of “emerge” I’d be all set.
Hey, Gurkam! I have ssen that you’re one of those, who are workn’ on Backbone. Maybe we could create bootable Backbone?
One more thing – it would be very nice to have an ability to switch between horizontal and vertical menu modes thru Preferences.app
Surely yo’re doing a great job