Re: Talk about other OSes(was: Functional programming)

Sergio Tortosa Benedito <[email protected]>
Newsgroups gmane.linux.distributions.gobo.general
Message-ID <CAPiRi8NjgZ245aX3jAt7QMBUkbDxgGgUpLH3am1FUw-w+6yJgA@mail.gmail.com>
El 7/9/2015 2:52 p. m., "Liam Proven" <[email protected]> escribió:
>
> On 7 September 2015 at 13:41, Sergio Tortosa Benedito
> <[email protected]> wrote:
> > Oh, I love reading about OSes, the problem is that none of them tend to
be
> > useful, mainly due lack of drivers, indeed I wish someone made a thesis
> > about common/universal drivers.
>
> Well, it depends what you mean by "useful"!
>
> Different people have different definitions. I know quite a few people
> running FreeBSD as their desktop OS. I find it too hard, myself, short
> of both apps and drivers -- but it is not so for everyone.
Of course, FreeBSD is notthat bad as desktop, but I feel like it has not
much great adbantages over Linux.

> One ray of hope is that today you can do almost anything with a rich
> web browser: email, chat, newsgroups, productivity apps and much more.
>
> So if a new OS has a really good browser, it has the single most
> important tool.
Yes, however, for me the main reason against those OSes are drivers.

> > BTW, let me ask, is Plan 9 much better tjan Unix? And why do you say
modern
> > unix is not lightweight? There's any way we could make them more
> > ljghtweight?
>
> Depends what you mean by "better". I don't know a lot about Plan 9,
> but it is what the creators of Unix did next. It is Unix 2.0.
>
> In Unix, "everything is a file" -- but not all files are equal. Some
> files have special semantics for device control. And original Unix did
> not include networking, or a GUI, so when those were added on later,
> they don't appear as simple files in the filesystem. Later, the
> process map was added as a virtual filesystem, so today, Linux has
> /proc and /sys and you *can* manipulate processes via special magic
> files.
>
> Plan 9 came later. Far more things are files in Plan 9 than they are
> in Linux. So, it's conceptually cleaner, simpler, more logical.
>
> And Plan 9 integrates networking into the kernel. Not only your own
> machine's processes are visible in the filesystem, but those of all
> connected Plan 9 machines as well. Processes can move around the
> network and so on -- that is conceptually "better" but obviously has
> major security concerns.
>
> But if you're on an x86-32 machine, you cannot run a binary from your
> colleague's ARM machine, say. Different processors, different
> binaries.
>
> So Plan 9 2.0 fixed that by abstracting away the CPU instruction set
> as well. It's platform-neutral, like Java, but as part of the OS
> kernel. So the same binary can run on an x86-64 or on ARM or on
> anything else that the OS runs on. It doesn't even need its own
> computer -- it can run as a web-browser plugin on top of another OS.

While awesome, won't this hurt performance? I'm not very fond of the JVM,
and yet it's the fastest one pretty much (but is quite bad in regards to
RAM). I think it would be possible to implement this in Linux by
serializing the process memory (of course without the VM part).

> Plan 9 2.0 is called Inferno, and it is arguably "better" than Plan 9
> as much as Plan 9 is "better" than Unix.

> However, Plan 9 kept developing after Inferno forked off. They are not
> in sync any more.
>
> But neither caught on, neither was a commercial success -- one element
> of this being that neither was originally FOSS as we know it.
>
> So Unix moved on without them.
If you are willing ti answer I have two more questions: have you toyed with
microkernels? How big is their performance impact?, and is really
beneficial that view of  everythins is a file?
-- Best regards, Sergio

_______________________________________________
gobolinux-users mailing list
[email protected]
http://lists.gobolinux.org/mailman/listinfo/gobolinux-users
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.