This article is really shallowly and poorly-written, like one of those cheap "pop science" magazines. I wish Computer Science wasn't affected by this sort of bottom-of-the-barrel journalism.
The author embellishes, makes sweeping inaccurate statements (that users only care about price and buying the cheapest product? that's patently untrue in any field), and dismisses technical excellence (like strong static typing) as meaningless. Terrible.
In my opinion, LWN is the gold standard of good technical journalism (on fairly-advanced topics, like operating system).
I felt like I was experiencing the writer's ADHD trying to read the article. There was no real coherent narrative, just a bunch of quick cuts and asides.
I've got my own ADHD, I don't need somebody else's as well.
LWN is great. I also enjoyed this; I didn't take it very seriously, and I enjoyed the presentation of this history as a variant of the Iliad. I laughed quite a few times.
Reminds me of a line from the Common Lisp Cookbook[1]:
> Keep in mind that it was written at a time where Lisp Machines were at their peak. On these boxes Lisp was your operating system!
Common Lisp OS interfacing is not part of the standard, because there were no "operating systems" back when Common Lisp Standard started out other than Lisp itself. It, too, was a casualty of the OS wars, but it lives on in user space, where it now lurks, and bides its time.
> there were no "operating systems" back when Common Lisp Standard started out other than Lisp itself.
This isn't really true, of course.
Sure, Commonlisp was mainly the standardization of MIT Lipsm lisp (based on MACLISP), influenced by good ideas from PARC's D-Machine Interlisp-D (based on Interlisp). Both ran on the bare iron of their respective machines.
But their predecessors MACLISP and BBN Interlisp both ran on PDP-10s under Tenex, Twenex (and those latter two's official OSes) and MACLISP also on PDP-10 ITS (its original home) Multics (both hardware and an OS). Other MACLISP-ish Lisps such as VAXLISP (VMS) and franz lisp (Unix on PDP-11s and Vaxes) contributed as well. These were all user space environments.
By the way the PDP-6 (and thus its dependent the PDP-10) was explicitly designed to be a "Lisp machine" -- a machine word was a cons and many fundamental lisp operations like car and cdr were single machine instructions.
> Operating Systems were at the heart of that war. Back in 1980, they were a relatively new concept. From the 1950s to the 80s, we had been using countless computers without anything resembling an operating system.
The fact that many of the operating systems we still use to this day have origins in systems created during the 60s and 70s is one of the many problems I have with this statement.
the conclusion feels a little abrupt to me. What about the differences between web 1.0, 2.0, and 3.0? What about the failed W3C standards? What about the limitations of web apps compared to native mobile apps? What about the closed gardens that Facebook/Meta and newer social networks are trying to build? And lastly, countries are restricting web access to their own jurisdiction. The wars are just beginning.
I think he also missed the fact that Windows NT came from VMS, not unix. The article needed an editor for sure.
I think your point rings true, there are always these platform wars. But there is much more ease of interoperability now. Even reading a 5.25" disk on a different system was sometimes impossible, lots of things like that. Keyboards and monitors had different connectors.
When Apple went to USB it was like a miracle. Circa.. 1995? Maybe with the iMac.
iMac was the one with usb replacing serial, parallel, ADB and floppy with built in cdrom and usb.
This meant some issues that were quite complained about, because at its release in 1998 the only way to export data was over network (or IrDA). USB Mass Storage was supported starting next year, and pendrives debuted in 2000.
Whole article feels abrupt. The hasty pace feels like its trying to get the reader to the end as fast as possible while saying as little as possible. Not sure what the point even is other than a very brisk walk down memory lane.
Try doing some useful work after installing the Linux kernel – and just the kernel – on a device.
You can't. You need more than just a kernel, quite a bit more, and that's an operating system. "GNU plus Linux" or whatever you want to call it, or Android, or whatever. In that sense Stallman is correct that "the Linux kernel is just one piece of the operating system" (although his insistence on GNU/Linux is still rather wrongheaded IMO, but I don't feel like repeating that discussion).
There's plenty of people who do - see the https://en.wikipedia.org/wiki/Unikernel stuff and Firecracker VMs and simplified runtimes like Lambda - But agreed, a useful consumer computer requires a whole suite of programs and tools that comprise a "OS" to be useful.
the author considers an "operating system" as the ecosystem where the user live and find a convenient application "store". Same store = same OS. Different store = different OS.
So with this reasoning maybe soon Windows = Linux but Android != Linux.
I didn't bother reading the article because of that.
Big Enterprise won the kernel wars. After Balmer freaking out and paying a hit piece calling "Linux a cancer", everyone managed to work around GPL and be able to inject non-free parts in the kernel. Linux compromised with the "tainted kernel" and later just gave up. Result: Android is a linux, and is the most closed source system on earth. Thousands of people throw away perfectly good android phones just because the owner of the source code refuses to release a update.
The author embellishes, makes sweeping inaccurate statements (that users only care about price and buying the cheapest product? that's patently untrue in any field), and dismisses technical excellence (like strong static typing) as meaningless. Terrible.
In my opinion, LWN is the gold standard of good technical journalism (on fairly-advanced topics, like operating system).
Hope more folk emulate LWN.