[smc-discuss] A technical doubt

Navaneeth.K.N navaneethkn at gmail.com
Thu Mar 10 20:56:41 PST 2011


You forgot to reply to the list.  :)

On Thu, Mar 10, 2011 at 4:17 PM, Jayadevan Raja <jayadevanraja at gmail.com> wrote:
> @Navaneeth
> Singularity does not natively run any Windows applications. Similarly, GNU
> User-land and its applications will not run on JNode or JXOS. This is
> because of lack of forward compatibility for the old systems.
>
> The absence of forward compatibility ensures that a big change will lead to
> loss of backward compatibility. Thus we are forced to use the systems made
> in 1960s and 1970s, with some modifications. We thus invest more time and
> money in old systems, and in turn get more dependent on the old systems.
>
> So, for our long-term interests, shouldn't we modernise our systems, even
> though it may mean breaking backward compatibility?

Yes. I agree with you. My point was, Singularity is still a research
project and not mature enough for people to start using it. Until it
gets stable builds, community may move. But the change will be very
slow.

I am an early follower of Singularity and I have great interest in
that project. But I still don't think that it can replace any of the
current operating systems anytime soon.

> Java, Object orientation, micro-kernels and Language-based systems are
> indeed far simpler than the Unix architecture, C-Language, monolithic
> kernels and procedural paradigm. So, we need systems far simpler than Unix.
>

I disagree. It's subjective, but after working with Java for years, I
know how easy is that to make simple things complex with it. It's an
OO problem than Java.

The simplicity that Unix and simple languages like C gives you is
nowhere comparable with Java or any other framework.

-- 
-n



More information about the discuss mailing list