09.18
Who would have thought a recode of an OS from the early 90′s would be generating such interest? It certainly sounds quite interesting to be honest (despite the lack of wifi) and if I had more time I’d probably have a play with it.
When I heard about it I remembered this comment someone posted on an earlier post of mine in response to a discussion about not GPL’ing my stuff:
Kerberos,
Free software may not completely replace commercial software any time soon, but it will (and already has) made making money selling commercial software more difficult.
Sounds good, Until someone releases a GPL or BSD thing similar to what you make, meaning people who want to use for commercial purposes will simply skip you.
Which illustrates the source of 99% of the FOSS software that people use – clones. 
Linux is a clone of Unix. Instead of taking the time to improve some of the cruft from the 70′s it’s all faithfully reproduced – warts and all. Look in the repo’s and it’s exactly the same, hundreds of clones of classic games and the few original ones are largely gameplay copies.
I remember the announcement of Songbird on Slashdot, where it was hailed as an ‘iTunes Killer’. Check it out, screenshot on the right. Apart from a few cosmetic changes doesn’t it look exactly like iTunes.
The thing is the developers working on Linux, Songbird, Haiku and all the rest of these clones are not working on their vision for what they want to achieve, they are just faithfully copying something that exists already. Once they reach feature parity, where do they go from there? If they were truly passionate about creating software they would make their own software.
Since they are just copying someone else’s vision once they run out of things to copy they lack direction, motivation and ability – there is more to development than coding – and the projects generally just stagnate from this point on.
I am working on a custom CMS at the moment (as in the quote above) and it is nothing like anything that exists at the moment as it is based on what I want, not based on just copying an existing one – how many Mambo clones are out there now? I have enough ideas for it to keep me occupied for years as I have a vision for what the final thing should be, and am working towards that vision. Any clones will not have that, and so will largely be unable to move beyond the source of the copy.
As is the case with Haiku. Once it reaches proper parity with BeOS, where is it going to go from there? Are there any genuine innovations and improvements planned to bring it forwards in time? I somehow suspect though, since the aim isn’t making a good, modern OS but rather creating a BeOS clone, there will never be any real tangible improvements beyond what the original Be team managed. After all if these people had vision (and this goes for the FOSS community) they’d be making something original, rather than just making a carbon copy of something existing.
Linux = Unix Clone
BSD = Unix Clone
ReactOS = Windows Clone
Haiku = BeOS Clone
Take Windows 7. If Microsoft had a working copy of Windows 7 to copy when they were developing it in the first place it probably would have been available ~2003. Developing software is all about coming up with new ideas and concepts, trying them out and finding out what works. Since they didn’t have a working target to clean-room re-engineer they had to make mistakes, try new things and come up with their own ideas.
Sometime things simply don’t work in practice that worked on paper. Sometimes you make an early decision that requires a lot of work when you realise it can be improved upon. You can see the thought process by comparing Win 95 -> Win 98 -> Win 2000 -> Win XP -> Vista -> Seven. Each one brings new ideas and concepts along with new problems that the successor tries to address. Again if MS were just making clones and had a competitor’s copy of Windows 7 back in 2000 they’d have a working copy in a few years. Development is much, much harder when you have to come up with your own ideas though.
My ultimate example of this is Tetris. Any, even relatively low-skilled, programmer is generally capable of making a working copy of Tetris in under a day. Yet if you told the same programmer to make an ‘Innovative block based puzzle game’ they would be at it for weeks, if they even managed to produce anything else at all.
So the easy bit is done for Haiku – cloning BeOS. Now the question is are they up to the challenging bit of actually having their own ideas?
This is true. When you’re just copying something you already know what the final product is supposed to be thus you don’t need to waste your figuring those aspects out.
Though, to credit Linux a bit, they do come up with some rather unique Window Managers. However, Linux doesn’t seem too hip to paradigm shifts when it comes to how their OS should be put together.
You can make similar arguments for everything
MS-DOS = CP/M clone
Windows NT (kernel space) = VMS clone
Windows (user space) = Mac clone
Mac = Xerox Alto Clone
Ultimately all modern OSes stole from Multics and Xerox. There was a time where “icons” and “windows”, as well as “processes”, and “threads”, and the concept of a “kernel” where actually innovations, not common sense.
You sometimes see it on the other end. KDE 4 was, in my mind, fairly visionary in terms of what they wanted to do with their desktop shell. I would say that that particular vision has been soundly rejected by users as being too far from what they were accustomed to. The sword cuts both ways.
“KDE 4 was, in my mind, fairly visionary in terms of what they wanted to do with their desktop shell.”
KDE might have a good vision for where they want to take their shell, but they still have to deal with the anarchy that is the rest of the operating system – will Compiz work if we do this particular thing? Can X support it? How will any or all of the sound sub-systems react with it? Will the kernel still play ball? This is where MS have an advantage – they have clear ideas of where they want to go, they can adapt to new technologies reasonably quickly (the explosion in internet use compared to private “walled garden” networks like AOL/Compuserve notwithstanding) and most importantly from the code standpoint, they control the entire stack, top to bottom.
If there was a single Linux company, that took all the best of breed projects and made them all work as one unified whole, with one shell, one media player, one browser (but still allowing others to run of course!) and ditched all the cruft and bloat and legacy crap and went further than simply cloning UNIX and actually fixed what’s wrong with it, it’d be a sight to behold. But it won’t happen. There are great projects, some great ideas and some brilliant technical achievements in Open Source, but as they’re all separate with differing agendas and ideologies, and differing practical issues like APIs or languages, there’s no cohesiveness, no single vision. Linux as a complete operating system is always less than the sum of its parts.
What the FOSS types see as Linux’s greatest strength, I’d say is its greatest weakness.
“I would say that that particular vision has been soundly rejected by users as being too far from what they were accustomed to”
I wonder if it’s the same kind of thing as the Office Ribbon? There’s a radical change to a way of working (and of thinking), and because it’s so far from the familiar interface, it’s immediately off-putting, even though a short period of adjustment would prove it better than the original.
I’ve heard stories of people congratulating MS for new features they’d found in Office 2007 and MS saying it wasn’t new, it had been in a previous version. The user had just never been able to find it before in the mass of drop-downs and option menus.
The Ribbon is the result of a ton of usability and design studies, in my opinion, it’s incredibly well thought-out and designed (well enough that most prototypes of the new OOo interface are uncannily similar) but people still HATE it. Because it’s different.
“MS-DOS = CP/M clone
Windows NT (kernel space) = VMS clone
Windows (user space) = Mac clone
Mac = Xerox Alto Clone
Ultimately all modern OSes stole from Multics and Xerox. There was a time where “icons” and “windows”, as well as “processes”, and “threads”, and the concept of a “kernel” where actually innovations, not common sense.”
I’ll give you ‘MS-DOS = CP/M clone’ as that is fundamentally true – the rest is crap though. I nearly wrote a comment pre-emptively rebutting your other points but I didn’t think it was worthwhile. Serves me right I suppose.
Now what if Microsoft hadn’t ‘stolen’ from Xerox as you so claim? Woudldn’t they have been unable to move from a command-line based system to a windowing system without ‘stealing’. Would Windows 7 still be CLI based, or would Microsoft be a footnote in history?
Stealing ideas is fine. I just nicked the auto-load module concept from WordPress – it’s a good idea. An idea is complete, it can be described in one sentence. It doesn’t have ‘vision’ – it’s just a good idea.
The WIMP metaphor is an idea. The CLI is an idea. Window compositing is an idea. They are all good ideas and should be used – only a moron fails to take a good idea because someone else thought of it first.
Copying an entire OS verbatim is not an idea. Reproducing nearly 99% accurately a program is not an idea. I’ve read the original design concept ideas for Windows 95 and they were not saying ‘lets copy what Apple are doing’, they were discussing, conceptually, how to proceed into the unknown. Use OSX (or even OS9 if you want) and Windows XP. They are so radically different it’s untrue. Basic concepts have been borrowed back and forth, sure, but the sheer level of blatant plagarism that FOSS engages in is unparalleled pretty much anywhere else in IT. No commercial software ever rips anyone off to the degree that FOSS software does.
Gnome is a carbon copy of the Windows 98 system, right down to the inclusion of a POP3 based email client on the quicklaunch (that’s what really gave it away). I’ve covered this already.
KDE 4 sucks. It’s basically pointless developer masturbation. Rotatable widgets! Wooo! I remember thinking the same looking at Project Looking Glass – you can pin notes to the back of programs – who the hell will ever want to do that. What happens when the program closes? Personally I want user focused software development with a clear unified goal and vision. Not an avant-garde collective dedicated to useless trickery.
And don’t call it fucking 4.0 and ‘Release Candidate’ if it’s actually an Alpha unless you want to have the piss taken out of you. Seriously. Way to go to spoil any possible goodwill. That’s why people think it’s shit – once bitten, twice shy. They don’t hate KDE4 because it was revolutionary, they hate it because it was launched fundamentally broken and has taken years to even get to the point where it is usable. If MS released the ribbon and it crashed every 3rd click they would have been crucified – and rightfully so. If the KDE dev’s had held their wad and released something new, cool and polished they might have gone somewhere, but since lots of stuff from 3.x that used to work was just plain broken or fatally sucky in 4.0 with no recourse everyone hated it. But no, it’s because it was ‘different’ that everyone hated it, not because it was pushed out the door unfinished.
The amazing thing of is a google search of “kde 4.0 problems” reveals loads of people bashing it, and given the fact that critics of FOSS are generally set upon this says a _lot_.
It also looks like a bad copy of Vista.
Gnome is no way a clone of Win98, have you ever even used Win98?
Sorry Joe, my bad, I meant windows 95.
piestar.net/2009/03/20/gnome-226-partying-like-its-1995/
I did say I had already covered this
cloning software makes me horny
Okay, whatever. You are free to your opinions.
Gnome is not a Windows 98 clone. Gnome is a very good GUI, with some things I’m surprised Windows hasn’t already gained, such as programs being organized by their category. You can also access every aspect of your computer from the Gnome menu, something I know you can’t do with Windows. Gnome is much more organized and speedier than even the current Windows Explorer.
I’m a fruity weirdo
For Thomas B and his idol, Adam King:
http://www.dresan.com/images/doing-it-wrong.jpg
Haiku is interesting in that unlike Linux:
It is built for the desktop.
The people working on it do not clone design choices from the 70′s.
It isn’t Linux. A dozen articles proclaiming the year of the Haiku desktop would at least be a change of pace.
Cons:
Like Gnome it appears to be from the 90′s
Software selection is worse than Linux and will be for years.
No one really cares
Oh and I want to want to wish a very special Happy Douchebag Day to Thomas and Adam
http://www.jfplayhouse.com/2009/09/happy-douchebag-day-out.html
“But no, it’s because it was ‘different’ that everyone hated it, not because it was pushed out the door unfinished.”
First impressions seriously count in Interfaces, whether for shells, or applications. The first “It’s different!” might be enough to put people off trying it out properly. In the case of Office 2007, they might find out they like it (or not, in which case there’s always 2003). In the case of KDE4, they get over the visual changes, try to use it, _then_ find out it’s a buggy, unstable mess.
It is, too. I tried it in Kubuntu. Turn on all the desktop effects to see what all the fuss is about, then watched the desktop die a painful black with random white rectangles death. No way of getting it back found, rebooting just brings up the same borked desktop.
Microsoft do things like this right – any change to display settings has to be confirmed within a countdown period, or it’s reset back to the previous WORKING settings. No such joy in KDE.
Could this be yet more FOSS double-standards at work??
KDE4 – FOSS-developed, badly designed, buggy, unstable and different = “It’s just different!”
Ribbon UI – Microsoft developed, well-designed, usable, stable and different = “It’s shit!”
“It also looks like a bad copy of Vista.”
Very true.
I saw a blog of someone critiquing the KDE4 desktop (I’ll see if I can find it again and post a link) He took a couple of screen grabs, and then proceeded to take it to the fucking cleaners; lousy graphic design, waste of screen space, bad usage of fonts, you name it. The first thing I noticed though, was just how many elements of Vista had been ripped off; glossy black Taskbar, large analog clock, smaller Start Menu button (You can call it the “Kicker” all you want, it’s still the Start Menu).
When Microsoft copy, it’s theft. When FOSS copies, however badly, it’s innovation.
I’d still rather have kde over gnome.
At least they are copying a current OS rather than one from 1998.
I’m surprised though by how few distros use the lightweight window managers like xfce when running on old hardware is often used as a selling point.
“I’m surprised Windows hasn’t already gained, such as programs being organized by their category.”
Since Vista the start menu folder hierarchy is completely irrelevant. You don’t have to go through menus anymore, first because the 10 most/last used programs are right in your face when you pop the windows menu, secondly because if it’s not there, you don’t lose time to find it in menus and just type the 3 first letters of the program and press enter.
In Windows 7 i personally do not open the windows menu anymore because of the new taskbar.