Fed up with Windows? Linux too easy? Get weird, go ALTERNATIVE It's hard to believe, looking at the modern computing world, but there is still more to life than Windows or Unix… and today, most of the alternatives run on vanilla x86 hardware and are free. Most of them need considerably lower resources than the market-leaders, too, so an old PC is ideal for trying them out. VMs are fine, …
I agree but a plug in USB key is great way to move around, with linux(xubuntu), make a system or "live CD" of my whole system , (64 gb Key) I just plug & boot from USB, and I am Home, need bigger keys, I also use 1tb Buffalo but the end is, I just copy of files downloaded to "Home Network" when checked & clean. If a bug eats a OS on KEY, ext USB HD, Laptop, a fairly quick reinstall solves all .....
Floppys have just evolved again, the interface has changed AGAIN, they are USB key now, I got over 8", 5 1/4, 3", 3.5", so a change in shape in media & storage capacity should be easy I reckon and trust me, these USB 3.0 devices are FASTER...
Oh. But no floppy drive. Hmm.
No mention of RiscOS either. Obviously not x86 but an alternative to look at for those with a Pi lying around unused (or just running squeezeplug stuck on the top of a kitchen cupboard with £18 of mains powered speakers attached as mine is - cheepest multi room music system around. My brother in law spent £1000 a room on his system. Ouch.).
This post has been deleted by its author
I have a Amstard PCW thingy with a 3" floppy, the competitor to 3.5", it also has a semi- enormous Dot matrix printer, it stills runs, even got blank discs for it, Manuals etc, it was a pig to use 25 years ago, it still is, next stop ebay for that piece of "history".....
I had those on my IBM System 34: 8" floppies and harddrives the size of attache cases made of Plywood, with motors like a landrover starter motor, that only stored 16Mb but cost $30,000 each! Those were the days!
No mention of Borland's Geoworks either, a GUI OS that ran on a IBM PC XT, killed off by Windows 95...
I know IT people do care, but does anyone believe most USERs really care ?, with now giant USB Drives that plug&play with Big screen TV, XBOX*/PS*/Audio Etc ?, the ones I meet don't give a fig, just want it to work, HERE, NOW .... So a PC OS is NOT what they want, its a embedded OS that Plays files via a remote ...
Debian since Amiga2000 (060/130mg ram), but Xubuntu is my lastest favourite, last couple of years anyway ...
Amiga OS is still a GREAT OS, one of best I ever used, but Xubuntu is free, and all my mod files etc still play ...
I believe most users do not care what is "in the box". I think there are some people who do. They have read this article, or ones similar to it.
What good does it do? I suspect that I am not the only reader here who has fired up a VM and started installing something interesting. I compare it to people who like to play with old motor bikes. The big difference is that this make less mess and probably cost less money as well.
It's been a long time since I last played. I gave my OS2 CDs away . I think the BEOS I tried is on an old HDD in a box somewhere. This article has given me a few reminders of other things. All I have done in the last while have been fairly common versions of Linux. Nostalgia...
I agree, I still play, My Parrot drone is a flying linux box, now interfaced to a laptop, I drive it with a logitech joystick, I am converting Long Range Ham Radio Control, allowing it's wifi to roam free and send data via HAM, I have had different OS's on it, TINYCORE, DSL both ultra small Linux distros, a Flying Onion/Tor Server seems a possibility.....
Why ? I dont know .... I am Playing, ordered a Pi last week, gunna see where that takes me ...
However I am surrounded here by multible systems, there are 3 MAIN computers areas in my flat, with different systems, Different OS's, VM's, set up to do different jobs/tasks/Play ....
I mean emulators are fine but when you have a Vic20,C64,Amiga's,Tandys,Ataris,segas,Nitendos laying around in boxes ...........
I see all these things (OS, Hardware) as lego ....
In answer to your post, I think Play is answer, PC industry has become too serious with itself, trying to make people believe there is only one way to do things......
The headline is misleading. When talking about alternatives to Windows and Linux I would have expected to find something a bit more contemporary, for example Solaris 11.1 or the various Illumos derivates. Or even the various *BSD variants. Not a trip down memory lane to operating systems from yesteryear and more or less dead clones of them.
QNX is great if you develop for embedded systems (i.e. in-flight entertainment systems) but its not and never has been an alternative to (non-embedded) Windows and Linux.
And ZevenOS, really? 'Get weird' with just another Ubuntu remix (as if there weren't enough already)?
>And ZevenOS, really? 'Get weird' with just another Ubuntu remix (as if there weren't enough already)?
Read the article again; the reference to ZevenOS was contained within brackets, i.e it was only mentioned as a passing remark, a footnote to the article.
"Solaris and BSD, are still based on that ancient Unix design"
The commonality with the "ancient Unix design" extends as far as the environment presented to the user - even then there are a lot of differences in the details (POSIX has helped here).
That said the Solaris, BSD and V7 (one of the more widespread "ancient" Unixen) kernels share very little in terms of design, you can look at the source and see for yourself if you don't believe me. ;)
"consoles were dumb and one typed in line by line - hence the term 'line editor'"
While people were stuck with MS-DOS on 386s the UNIX bods were playing around with X-Windows, NeWS, and NeXTStep, go figure.
"BSD and Linux variants aren't alternatives; just more of the same."
Folks tend to copy successful stuff. The also rans do stuff differently. ;)
That said there are plenty of OSes that don't work to the MULTICS/UNIX model, of the Open Source ones the most interesting I came across was EROS, which in turn became Coyotos (www.coyotos.org - looks a bit quiet now :/). If you are curious as to what the inspiration for UNIX looked like (it was *very different*) the source for MULTICS is available online now too (MIT host it).
Something to be said about being the last generation (probably) to use vi. Ridiculous to learn (unless you came from ed and the line editor world) but once you do you never have to worry about file editing in virtually any Unix ever again. Worse you find everything else to be less efficient.
Try it out to see how wrong you are. Thanks to Lisp it actually has grown into a super editor. This very design and and its modular architecture still ensures adherence to the KISS principle. I can barely remember any annoying bugs within Emacs despite some heavy usage of it for the last ... 8 years, I think.
Perhaps it's farther away from the Unix ideas than vi is, yet vi or even vim are not as extensible as GNU Emacs is.
but makes little sense from the terminal
Imho, it does, I've been running it predominantly this way until pretty recently. AMOF, on Debian, vim-runtime (a dependency for vim) takes more space (about 23mb) than emacs23-nox with its dependencies (about 13mb). It still comes with the lisp interpreter and quite a few very useful modes and things.
A lot of people run their CAS software inside of emacs because it makes a lot of sense. GNU Emacs got it's own very powerful yet simple reverse polish calculator Calc, often used as embedded while editing various stuff. This one plays so nicely with other cool modes like org-mode. Emacs Calc, unlike so many other calculators, can not only operate on dates, units, it supports various formats, does symbolic calculus and so on.
No other editor can do it.
KISS and emacs in one sentence. Must be a joke. Just not a funny one.
I pointed this out above, vim occupies more space than the emacs-nox package on Debian. This is not a joke, yet actually is pretty funny. BTW, there is no emacs anymore. There is either GNU Emacs or XEmacs and a few other clones.
This post has been deleted by its author
"the thing I always liked about RiscOs was the fact that it was iin ROM - surely a bonus in these uncertain times."
... That could be a bit of a mixed blessing in my experience...
One evening I was happily hooked up to Demon via a 56K modem when my Linux machine crashed... Checked it out, rebooted it, connected to Demon, crash. I even tried Windows. Same result.
The good old ping of death.
Eventually I managed to locate the smallest possible UNIX distro I could find - which happened to be OpenBSD 2.1, and I still run OpenBSD today as a result (I also run the other stuff too but OpenBSD remains my "trusted" platform).
I would like a USB stick/SD card that is fast and big enough to comfortably run a live image, that also has a *physical* write protect switch that makes it impossible for anything to write to that storage media. I don't think SD does that unfortunately - I think it amounts to a polite request. ;)
Anyone interested in RISC OS should look at http://www.redsquirrel.fsnet.co.uk/redsquirrel.html a native RISCOS emulator for x86/Windows. Still have a copy of my final RISC OS machine running under it - and the nice thing is that with modern hardware it actually runs faster in the emulator than it ever did on the real hardware :)
Risc OS 5 is not open source; it is merely Shared Source.
Castle's licence explicitly forbids porting to x86.
In any event, if they did, it would be of little interest. Architecturally, it is primitive, with no true memory protection, no virtual memory, no disk partition support and no preemptive multitasking in the kernel (bizarrely, the *Text Editor* does this. Yes, really.)
This post has been deleted by its author
The article fails to mention two of the most used alternatives - IBM's OS/2 and Novell's Netware are still in use worldwide, especially in enterprises, banking, and point-of-sale machines. The new(er) version of OS/2 is called eComStation, a proprietary product. A free version, osFree, has been in development by an open source community for several years.
Another community developed an open source alternative of the Netware kernel called MANOS for several years, although it doesn't appear to have been updated since 2010.
When I was in my early teens I tried to install OS/2 Warp, and gave up.
When I was in my late twenties, an ATM in South America decided to reset itself whilst my credit card was insde it... upon rebooting, I took a picture of its OS/2 splash screen. I had to stay in town an extra day to retrieve my card from the bank who operated the machine. My thanks to the Peruvian Transport Police.
> Open VMS lives on in the architecture of Windows NT
Except for a whole bunch of differences that mattered like putting graphics drivers in kernel space which has always been one of NT's Achilles heel (but also necessary for the gamers and partially alleviated with modern WDDM). I am not sure of the numbers but I think people would be amazed how many blue screens are due to poorly written 3rd party drivers (and hardware failures themselves of course) as opposed to poorly written Microsoft code (some there too though especially in past).
"....> Open VMS lives on in the architecture of Windows NT
Except for a whole bunch of differences that mattered like putting graphics drivers in kernel space which has always been one of NT's Achilles heel (but also necessary for the gamers and partially alleviated with modern WDDM). I am not sure of the numbers but I think people would be amazed how many blue screens are due to poorly written 3rd party drivers (and hardware failures themselves of course) as opposed to poorly written Microsoft code (some there too though especially in past)...."
You are not really updated. Windows have moved out the graphics from the kernel, so latest of incarnations of Windows are lot more stable than when Windows had the graphics in the kernel. For instance, Windows7 can update it's graphics driver without rebooting nowadays.
Funnily enough, Linux has moved it's graphics into the kernel. This has made Linux even more unstable, but Linux has increased it's performance at the cost of stability. How good is an OS if it is fast but unstable?
Funnily enough, Linux has moved it's graphics into the kernel.
Didn't it just move the mode-setting to the kernel, NOT the rendering operations? That makes sense, because initialization of all other hardware devices have also always been done in the kernel. The old situation where the X11 server initialized the graphics card directly was less stable, and made the screen flicker and flash more while booting and logging in to the system.
>You are not really updated.
Didn't see the caveat about WDDM huh? My whole point is that is a modern addition and not part of Dave Cutlers work 20 years ago. He was extremely disappointed when he was overruled by management and the graphics drivers weren't required to be user space (largely the same management that thought embedding a web browser deep in a OS was a great idea). As for Linux vs Windows stability its sacrilege but today they are largely the same in my experience (Linux though still allows you to run any desktop environment you wish which is why I prefer). Microsoft has come a long ways because they have been pushed some by Mac OS and Linux and because they finally did learn their lesson about stability and security being as important as marketing and cozy OEM relationships (which they have been burning lately).
And I forgot to add...
I've yet to see any OS clustering technology that works as well, as solid, or as easily as VMS clusters. Dave Cutler may have brought VMS concepts to MS when they built Win NT, but the clustering has never lived up to the standard established by the VMS clustering.
This reminds me of the DEC Alpha I got my hands on as a teenager... not content with the Windows NT implementation and slughish x86 emulation layer running on it (though strangely Blizzard's Diablo could be played on it without any problems) I went on a quest to install Debian on it via a special pre-boot-loader. This tricked the computer's firmware (it was not really a bios) which was only designed to boot Windows NT into thinking it was booting just that.
Sadly, native code and even basic drivers seemed as hard to find under Linux as they were under alpha-Windows NT ;-(.
DEC boxes are cool
Funnily enough I was gifted several DEC AXP machines by a university and managed to install the very last Debian Linux 64 bit 5.07 on it.
There's something very elegant about this classy kit, and it ran KDE without breaking a sweat.
Another one ran Windows 2000 (a beta that didn't quite make it to production before some idiots took a sledgehammer to a so much better system than Intel ever made).
Unfortunately the beta has all sorts of issues, but USB ran, you could even watch slightly jittery DVD if you bought the right drive.
What was fun was to see what happened when it swallowed a virus or worm...it sort of got a bit gooey and sluggish to begin with then slurped out...."hey sorry mate but I'm just not gonna run this kinda x86 crap".
After a few minutes it just burped, you went looking for the piece of half digested code and that was it.
I still have the entire last Linux distro set on DVD.
It ran very well and was unhackeable as a server because hackers didn't have a clue what OS was running behind it or how to break in.
I even got the dreaded Samsung RUFFIAN board running on Linux which was supposed to be a feat.
That was a BOMB proof 64bit server but all support died in about 1999.
I'm sure it will still be running in 2059.
There was ONE problem on the last Debian DEC ports.
Massive memory leaks in the networking stack.
They even ran 64bit PCI and Gigabit LAN WAY before Intel could.
Pity!
Yeah... that win2000 beta!
I Almost got my hands on that software package!, found it on somewhere (I used to look in all the nooks and crannies of the internet when was that age)... Then the website was pulled upon almost having the download complete :-( And then MS pulled the rug underneath that built and shortly after released Win2K with only x86 support. Such a pitty. I still wonder how that beta would have worked. My Alpha didn't have USB (it was a 266Mhz one) but still I think it would have also worked until 2059 it was built like a tank and had a airflow concept over the CPU that I think you only started seeing a decade later in normal PC's. A shame, the demise of the Digital Equipment Corporation (and for what?)...
Nostalgia threads, make you get older.
This was supposed to be about alternative OS not CPUs, but what has held back all these OS into the dark ages, has been the Intel medieval CPU.
Just look what strides have been made in smartphone & embedded technology since ARM carved out the market.
Intel has been trying to play catch up ever since.
It was always this way since Gates started stealing IP from just about everyone then claiming damages from all and sundry for s/w piracy.
I still have the AXP win2k CDs somewhere, both server and workstation, so I'll bore you with some details.
There was some reg hiv to make it kill off the evaluation installation timeout.
I could do all sorts of things with it, but slowly you could see the Wintel empire killing it all off, as the lack of later browser support/ Java implementation etc slowly made it into a museum piece, and ADOBE grabbed more and more desk bloat.
WOW, it takes you back now, when you don't get the obligatory flash ads or drive by hacks/malicious crap d/l for free through MS 's gruyere cheese security model.
DEC machines were good just not to have to see all that bloat they say we need.
the Miata still plays back/records 24-96 sound files thanks to the LYNX ONE professional sound board, if you can believe it though proper studio cannon connectors not some lousy 3,5mm jacks.
It just never had a sound editor later than 16 bit Sound Forge, and the x86 emulator couldn't handle anything else, so it went the way of all useless gadgets.
FATAL again for it, as you could do nothing useful apart from using it to store stuff, with a Ethernet implementation which although fantastic reliable and fast, can't keep pace with a modern 1000baseT laptop.
Win2k even ran Ultra LVD 160, which was a novelty back then, & NT4 couldn't.
In fact that machine had the onboard ITI combined multi channel Symbios/tulip SCSI/ethernet board from LSI, which gave it really fast storage over LAN.
It made anything from Adaptec look the crap, which we had always suspected it was all along.
I even got the MS office 97 word/excel native alpha version which worked absolutely great.
Everything was just so much faster and more fluid on AXP.
If you remember Windows 2000 and hated it being a bloated SLUG on X86, you got a sense of what NT5 was touted to be for years, but failed to be on Vista yet again.
There were some funny fixes you could download from all sorts of sources/drivers inc HP and of course everything gravitated to Aaron (Alphaman) Sakovich The AlphaNT Source until about 2009, which in Alpha terms is really not that long ago.
We all remember it as a Great platform then Compaq/HP committed alphacide, sent the engineers to work on titanicium, and we all know the disaster that has become.
It kept a generation of marketing bullshitters in a job, while Wintel tout multi-core to make up for the more and more sluggish memory hogs they make today.
Tbqh a 600mhz Alpha on win2k with 500Mo RAM was/is just as quick a GUI as a "modern" POS laptop on windows 7 with 3x the clock speed and 4 x the RAM.
It's fascinating to think of what "might have been", had MS been forced to have 32 or 64bit XP to run on AXP rather than that crap from Intel, and Compaq not destroyed the EVO8.
The wikis make salutary reading:-
http://en.wikipedia.org/wiki/Alpha_21464
http://en.wikipedia.org/wiki/Itanium
I guess I would still be running XP today on AXP, rather than turning them all off with LENNY*.
Apparently at last count Mars were still making chocolate bars using VMS on DEC, because there is/was never any down time.
The funny thing is how even with ancient DEC machines with the 4Mo cache processors and 600mhz, they always seemed to be that much more fluid than the equivalent dual Xeon with twice the clock speed, and of course the board build quality on a humble workstation was designed to last a life time.
In fact Aaron reckoned that ONE AXP CPU was worth more than 2 of Intel any day.
The last dialogue I had of any sort was with Esterbrook at HP about hacking MILOs.
The sense of resignation & disappointment was palpable.
The last distro from Debian was 20/11/2010*.
It's a miracle that made it out the door, so that's already 3yrs old and still had no JAVA.
I'm still waiting for a smart phone I can roll up, has a battery that lasts a week & doesn't break in half when I sit on it in my back pocket on the tube, bit like my 10yr old Nokia phone really....
"This was supposed to be about alternative OS not CPUs, but what has held back all these OS into the dark ages, has been the Intel medieval CPU."
You're right about the first bit before the comma, but wrong about the last bit. When the 386 came out (deliveries kicked off in 1986!!!) people were running MS-DOS on them, meanwhile SunOS (I was lucky enough to play with one of those Sun 386 boxes) and Linux ran just fine on my 386DX40. It wasn't until nearly a decade after the 386 hitting the streets that Microsoft barfed NT 3.51 into the world (1995 according to Wikipedia - I didn't see it running until late '96 - and it was very rare even then).
Given that evidence I don't think you can honestly claim Intel held up OSes (however awful the x86 arch may be!), I put the blame on the punters. There were alternatives to MS-DOS out there but they carried on running 16bit apps under MS-DOS on their 386s, 486s and Pentiums. I don't think the computer magzines really helped much either - they had a habit of pushing the Wintel uber alles line and they were complicit in dishing the FUD on potential competitors.
If you can find some copies of Byte from the late 80s/early 90s you should read some of the Chaos Manor articles. The guy always has top of the line 32bit hardware and most of the prose is about how he installed a ton of software and workarounds to make use of > 640k of memory. The irony of that was that Byte maintained some non Wintel content pretty much until the bitter end, the Chaos Manor dude could have read the magazine and spaffed some cash on a SunOS box for the same kind of money he was spending on hot-rodded 386s running MS-DOS.
Jerry Pournell, the author of the Chaos Manor column, was a big OS/2 proponent. He was not fond of Windows, generally speaking, though his published opinions on the matter range dramatically from month to month. But I'm looking right now at his Sep '95 column (the first issue I happened to grab), and he starts by saying that he likes Win95, but then goes on to detail ways OS/2 Warp Connect is better.
True, I don't remember Pournell ever singing the praises of any of the UNIX variants, Linux, etc. But he isn't a computing professional; he's an SF author. The Chaos Manor columns were one user's view on the hardware and software he happened to use. Pournell made more than a few highly dubious pronouncements (I recall in particular his often-repeated criticism of multiprocessing: "You don't want anyone stealing cycles from you, even yourself"). There's little point in complaining that he didn't share your preferences - the column was never presented as a review of the best technology available.
(And Pournell rarely spent his own money on the gear he reviewed, at least once the column became popular. He reviewed what companies sent to him.)
I missed Jerry's conversion to OS/2, although it's not surprising he latched onto OS/2. He struck me as being technically literate which was why I was disappointed that he didn't look further than MS-DOS based solutions when I was reading his articles.
Back then the popular(?) computing press tended to portray UNIX as something that propeller heads used, and was therefore unfit for stuff like word processing. Of course this neglected the facts that UNIX was originally developed for preparing technical documentation and tools like LaTeX were actually very easy and vastly more capable.
Unfortunately I think a lot of people (including Jerry) bought into the idea that UNIX can't do documents without ever actually trying it. It would have been interesting to see how things turned out if MS chose to push Xenix instead of fooling about with DOS & Windows. From a commercial perspective I can see why MS chose not to do that - I wouldn't have fancied fighting Bell Labs for the right to sell software I wrote either.
> turned out if MS chose to push Xenix instead of fooling about with DOS & Windows. From a commercial perspective I can see why MS chose not to do that - I wouldn't have fancied fighting Bell Labs for the right to sell software I wrote either.
You seem unaware of history.
Microsoft purchased a full license for Unix (edition 7) from AT&T and produced their version of actual Unix for the x86 called Microsoft Xenix. Later when they purchased 86-DOS from SCP as MS-DOS 1.x and added to this to make version 2.x. They added features form Unix/Xenix such as subdirectories, redirection, executable format*, and such and claimed that they had a family of operating systems: MS-DOS for small PCs and Xenix for multiuser systems; with some superficial similarities.
In fact Excel's predecessor: Multiplan came out on Xenix before MS-DOS.
I don't know why you think that MS would have to fight Bell Labs over anything.
Later they sold Xenix to SCO who upgraded it to OpenServer by buying licences for System III and System V.
* MS-DOS 1.x had no subdirectories and stored all files on a diskette in one flat space. It didn't even have CP/M's 'user space' which separated files into up to 8 or 16 separate namespaces. 1.x also only had .COM binary executables which were a flat format similar to CP/M 8bit .COM programs which is not surprising because 86-DOS was a 16bit translation of a decompiled CP/M (with FAT instead of CP/M filesystem). MS-DOS 2.x added .EXE which was a structured format with multi-segment support and fixups, just like Xenix programs.
"Microsoft purchased a full license for Unix (edition 7) from AT&T and produced their version of actual Unix for the x86 called Microsoft Xenix"
Yes, but keep in mind that V7 was a pretty early cut, it was missing a lot of stuff we take for granted like virtual memory, networking etc (all that came later with the BSDs). They would have been facing a choice of System III, BSD or going their own route. My guess is they would have gone BSD route & got licenses with AT&T in the end.
At the end of the day they chose to go their own route - but in a very slow, anti-competitive and tedious way.
> Yes, but keep in mind that V7 was a pretty early cut, it was missing a lot of stuff we take for granted like virtual memory, networking etc (all that came later with the BSDs). They would have been facing a choice of System III, BSD or going their own route. My guess is they would have gone BSD route & got licenses with AT&T in the end.
I don't know why you think that guessing is useful.
Microsoft bought AT&T licences to keep it up to current Unix versions:
"""in September 1983. A port to the 68000-based Apple Lisa also existed. At the time, Xenix was based on AT&T's UNIX System III."""
"""Version 2.0 of Xenix was released in 1985 and was based on UNIX System V. """
I had Debian running on a bunch of DEC Alphas. Not really any problem at all, certainly no assumption that Windows would be used. Brilliant machines used to run ProE on DEC OSF that were thrown out when the company decided that an all Wintel system was the way to go. Then they had to buy Intel hardware that was nominally twice as fast with twice as much ram to ProE to work as well as it did on the Alpha's.
I got four of them for nothing.
It was a bad day when DEC went under.
I miss Alphas too. They just seemed to have a bottomless appetite for gruelling work whatever OS they ran, and most of the boxes I worked with had a fairly long working life - they were fast enough to earn their keep long after they were obsolete. :)
They were pretty decent. I had one running OSF/1 or Digital Unix in 1995. I remember using the SunOS to OSF binary translator on a few items. It was quite nice compared to the Ultrix, SunOS, Solaris 2.3, HP-UX 9 and Irix systems we had. No Linux in that mix you'll note though I ran it at home on Pentium I systems. They got a nice mix of SysV and BSD and of course 64 bit. Sun was more popular but didn't really catch up until a few years later with the Ultra Sparcs.
I've had a server at work running Tru64 until this summer. It's finally been turned off. We turned off an HP-UX Itanium the same day. It's a shame the Alpha wasn't pursued by HP. FWIW, VMS was ported to Alpha and then Itanium. HP killed that too.
Whilst most current implementations of Forth run under a host operating system, there is no reason that Forth cannot be completely stand-alone. This used to be fairly common practise on 8/16-bit microcomputers and the versions that ran under a host tended to be for mini and mainframe systems.
"Whilst most current implementations of Forth run under a host operating system, there is no reason that Forth cannot be completely stand-alone."
True.
Forth's background is process control and fast response, low memory foot print were key drivers in the late 70s and early 80s.
Take that further and you are into full embedded apps. I think the trouble is Forth is too flexible.
Everyone rolled their own OS functions for their hardware and no one developed a specific OS (Not quite true IIRC the "Gavillan" early 80's laptop was meant to have version of a Forth based OS. No idea what happened to that).
Yes, I was thinking that Forth is ideal for people who want to / have to get things done in less than the size most other OSes have for their logo graphic. As well as the tiny footprint, it also gives you unparalleled interactivity, so remains ideal for people who have to do stuff with hardware. It's much more useful on things like the Arduino than the C variant that's the official language, for example.
A fixed Forth would be going against the language philosophy. One of the reasons people want to replace X is the baggage you need to have an X driver. It will never be needed, but without it, you're not X. Forth takes the view that if you don't want something - arrays, for example - you don't have to have it. If you do want something, you probably want a particular sort of it - an array that keeps track of the max/min/average value say - that will be important for you, but not something that should be imposed on everyone. That's not being too flexible, that's power.
When designing the software for one of the early computers Alan Turing apparently took the position that he could do things in base 36, so everyone else should have to do so to.
colorForth is Chuck's current reminder that he's a bit like that - the current 27 key keyboard he uses it with is a major expansion of the one he was using before that. I can't remember if it had three keys or five...
His hardware designs have always been incredible. When a 68000 was considered a fast CPU, he had a 4000 gate one that made it look like a keyboard controller chip. If Babbage had thought of a zero register design, he might have produced a working mechanical computer.
"Lisp Machines – some say the greatest ever programmer's environment" - indeed, that's how I remember it, (on Symbolics, anyway), though NIL on VAX / VMS was a lot more accessible :)
I remember a guy (Alexander Ogorodnikov - whatever happened to him ?) buying microVAXen on his credit card (AmEx !), putting NIL on them, and selling them on for folk who didn't want something as specialised as a LISP machine. (That or they wanted something they could easily repurpose if the AI team upped and left !)
Now, there's enterprising for you - this was back in 1985 or thereabouts .... yeah, thanks for the memory ! (more memory than NIL had after running for a whole day - oops, no garbage-collection :)
/ramble
Actually I don't think you can blame BeOS on Microsoft. They needed no help to commit suicide.
As I recall, they had an extremely proprietary and anti-hacker attitude. They were very modern Apple in that they wanted you to buy it and use it w/o wondering what was going on under the hood. If you asked how it all worked, they were "we don't need your type as a customer". Basically they alienated anyone that might have been a first-adopter. Jean-Louis Gassée had an ego the size of the Hindenburg and a my-shit-don't-stink attitude too.
I was VERY interested in the cool things they were trying to do so I was sad they didn't want to cooperate with anybody.
> Actually I don't think you can blame BeOS on Microsoft
Actually you can. It was starting to be installed as a dual boot on some Windows machines. Microsoft offered an extra $5 discount for every machine if it was not installed.
They had done the same with Netscape Navigator previously and later threatened removal of 'loyalty' discounts over Linux being installed on Netbooks.
When tens of thousands of machines are being built a few dollars per machine adds up to a significant threat.
Indeed it did. Hitachi licensed BeOS and bundled it with specialist audio-video workstations.
Microsoft invoked an obscure clause in the licence agreement for MS-DOS and barred Hitachi from making the BeOS partition bootable. The machines *had* to include MS-DOS, *had* to boot directly into MS-DOS and were not allowed to even offer a boot menu offering BeOS as an option. Customers had to boot BeOS from a floppy and rewrite the bootsector.
Unsurprisingly, the machines did not catch on.
Typical strong-arm restraint-of-trade tactics from the 1990s Microsoft.
Still available, not open source or freeware. WP reckons there are three different versions of what became Multiuser DOS.
For a while, it was better than DESQview, because it switched between tasks with a 60Hz clock rather than the 18.2Hz one DESQview used, and was so noticeably smoother.
...if you're interested in alternative operating systems. Won't the article doesn't mention is that it's almost completely Posix compliant; fire up a terminal window and it's bash, and your configure scripts run, and stuff Just Works. There's basic graphical acceleration which makes the GUI nippy, and it's really lightweight, much more so than Linux, making it an excellent way of turning an old laptop into a box for doing ssh from. Definitely worth a look. The installation image can be burnt to CD or booted from a USB stick, which is a cunning trick I haven't seen elsewhere.
That's not to say it's perfect: the underlying syscall architecture is in C++ and dates from the days before C++ ABI standards, so there's quite a bit of mess involved in using a compiler more recent than gcc 2.95; the wireless system is still under construction, and while you can get WPA working on it, I never have; and there are a number of semi-obscure kernel bugs that are still being looked at. (There's a reason why the website firmly calls it an alpha.)
But unlike a number of these alternative OSes, there's an active developer community and even some people who are *paid* to work on it!
Linux has been bootable from USB for a good long while now, I think Windows even manages it now.
No clue about OSX, but I wouldn't be surprised if they had that option for reinstalls.
I might try this haiku, I just got everything working on 13.10 last night so I need something to occupy me.
BeOS PE booted and ran from a CD back in 1999.
The memory management was chronic, disk access speeds were dreadful, posix sucked on BE, the IP stack was horrible, the last networking rework was a disaster which even killled printer support (!) & kernel crashes happened constantly which is basically why it died, never mind the problems with apps that never appeared until it was t..ts up and they wanted to make another suck-atronic invention at least a decade too early the "BeIA".
Today it would be beautiful on ARM with a cloned and hacked iphone.
It would make Jobs have nightmares even in the grave.
Apart from that, as eye candy it was terrific, enabling you to do perfectly useless things most of the time then crowing about how brilliant it was.
...we're still talking the same old base OSes floating around, some decades old (not meaning to be derogatory), forked from other forks from other forks, but not a single one well enough developed and mature to be a proper contender to Linux or Windows?
To be anything other than the modern version of a 19th century (computer) science curioso that'll never make the popularity big-time.
Regardless of completeness though, these all seem to me to be slightly public exhibitions of up and coming computer scientists achievements, that all lack true depth and real interest, rather than real worthwhile alternatives.
I'm going to go out on a limb here, and say it'll never happen without a serious big-name company and millions in investment, not these nigh-on bedroom coding efforts, before anything beyond MS Windows or Linux lives on greater than 20% of home (even business) computers worldwide.
To summarise:
8080, Z80, etc
1. CP/M
2. El Reg once reported someone had got a Linux running, slowly.
x86
1. DOS, Windows
2. Unix, BSD, Linux, Solaris, AIX, ...
3. QNX
4. BeOS
5. OS/2, Ecomstation
Other
1. George3, VME/B
2. RT-11, RSX11various, RSTS, VMS
3. IBM mainframe various (sorry, I know little of these)
4. Other minicomputer OSes. I remember DIRECTOR, for Ferranti process control systems.
http://www.menuetos.net/ Menuet: Still being developed, tiny full featured OS for X86 machines. Give that old 486 or Pentium machine you never got around to tossing another whirl.
http://www.contiki-os.org/ Contiki: Runs on damn near anything. You'd be surprised how many Commodore 64s are being used to surf the web with it!
I sometimes think about what would have happened if I never upgraded my home computer system away from my Commodore 64. What if I kept using it and finding ways for it to work with modern technology.... would probably be using a hacked/improved version of GEOS or Contiki... with an ethernet modem and cloud storage :)
@ Liam Proven
Thank You, Hadn't seen that, got VIC20, & C64 here, very interesting the specs on that cartridge look like a C64 programmers wet dream I had @ 18, I was just looking @ contiki OS, got 6 c64, I could start a internet cafe ? I will get some a couple of those I think, the C64 & 1084s monitors are are a pig to move to set up, these days, I may be able to just a old LCD, nice ....
MenuetOS is in the article. Read more carefully. :¬)
Contiki is not included because (AFAIK) it does not run on x86 - it focusses on much lower-end hardware. It was originally designed for 1980s 8-bit home micros: it runs on the Commodore 64, Amstrad CPC series and so on.
TempleOS is wonderful. It is written by a severely mentally ill man who is in long-term care. Yes, he has recently got religion bad. Earlier versions had different names and did not have the religious imagery.
It is not OK to mock people with mental illness, even if they are a genius.
All the OSes borrow and steal and then claim it's new and improved. Patent trolls like MS and Apple prevent any attempt at true innovation. While Linux is close to a full featured OS, the lack of unity within the community prevents it from becoming mainstream.
The problem with all the geniuses (seriously bright folks) working in the alternative (Linux) environment is their egos exceed their IQ. Why can't we all just get along? While diversity is a strength, it is also the greatest weakness in the community. Just because someone is a good programmer doesn't make them a good project manager. While the community is maturing, there is a need for some consolidation, if we want more general acceptance.
Remember, the lesson from Intel and MS, technology doesn't win, marketing does. SO get ur selves together and get a fully functional desktop OS on the table, then work the improvements.
Glad you mentioned squeakNOS, I had a play (download the iso and boot off it, it boots fine in VMware). It's really bloody weird, and strangely powerful though very confusing. Do give it a try (and see if you can find the control to rotate the windows to any angle with a drag - yep, windows don't have to be at right angles in sqeakNOS).
Another to perhaps try is house <http://programatica.cs.pdx.edu/House/>. It's written in haskell. I've not tried it as it wouldn't boot in a VM.
The reason MS hacked up the BS that was "Pen Windows."
Described as the first OS designed to run with a pen and mobile computing (IE Not permanently connected to a LAN, and wha that implies for document transfer).
Test hardware was 286 based so I'm guessing it'd scream along on anything more recent.
This article is very interesting and brought back many memories of these disparate and (some) weird Operating Systems software.
However, if one wishes to "replace" Microsoft Windows XP with a fully functional, robust, easy-to-use and supported OS that can be used "at home" or in a business /corporate environment, then none in the article meet such requirements.
I recommend consideration of Zorin Linux, which is a serious and stable effort to provide a more Windows like experience for Windows XP users. Being based on world famous Ubuntu Linux (Note: Ubuntu Factory installed on some Dell and other brand computers) , with" very close" Windows 7 look and feel (at least as legal as possible), and offering commercial upgrades and support, Zorin even includes the Wine implementation that can and does run dozens of "native" Windows (XP, 7) applications, including Microsoft Office - up to 2012, with similar stability to an "all" Microsoft setup.
Microsoft and their thousands of Microsoft technology (sales) Partners are attempting to paint the most bleak picture for XP users to continue on or for moving anywhere else except onto the ridiculously expensive, convoluted and just as insecure Windows 8x platform. XP users should not forget the many thousands of dollars “not” calculated in needed modern hardware and peripherals upgrades, and mandated draconian support contract policies.
Clear thinking and technically astute XP computer users or their knowledgeable and 'client orientated' advisors/ consultants will find the Windows-like or other slightly different User interface (UI) Linux solutions are exceptional good values and better choices as against the Microsoft as usual hollow drama claims.
"It's an interesting idea, but if it ever gets close to completion, I suspect that it will get sued into oblivion with remarkable swiftness."
Surely they can't sue if ReactOS can prove (and they can) that everything is a clean-room reimplementation instead of reverse-engineered code?
I'm looking forward to the day I can dump Windows for ReactOS. The only reason I'm stuck with Windows is, like thousands of others, games. Specifically, games that can't run under Wine due to stupid anti-cheat and copy-protection measures. If ReactOS is developed correctly, then those anti-cheat and copy-protection code will most likely work as well (unless M$ plays dirty and puts in additional checks to make sure their games only run on Windows. But then again I don't give a rat's arse about whatever MS Game Studios put out. Unless the other devs are stupid enough to listen to M$ and block their software from running on other OSes as well, in which then they're digging their own grave).
Mere clean-room development and reverse engineering is not sufficient if the APIs, look and feel etc. are themselves legally-protected.
Also, to retain Windows compatibility, ReactOS is developed with and must be built with Microsoft compilers. I am not sure that the whole thing /can/ be compiled with GCC, but if even parts are, then it ceases to be able to execute Windows binaries and drivers, I believe.
Seriously, I don't think they have a snowball's chance in a supernova if they ever get close enough to be any kind of a serious option or even a minor threat.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
