R.I.P. Byte

agree at carltonfields.com agree at carltonfields.com
Thu Jan 21 22:44:14 UTC 1999


Actually, this is an important point.  If "Linux" is just the Torvold Kernel, 
the point is tremendously well-taken.  If Linux is the rest of the operating 
system and application software, clearly there is no unspoken or spoken 
suggestion that it cannot be modified.

The same is true of Windows.  What we NOW call an "OS" includes so much of the 
functionality of the universe of computing that the demand for customization 
is enormous, hence the "permission" to make modifications.  Windows now (if 
you believe their Antitrust defense) is so deeply integrated that the 
"application aspect," "systems aspect" and "kernel aspect" are now all a 
unity.

Is the bottom line that the Kernel should be REALLY tiny and untouchable, and 
that the System surrounding the Kernel should be really, really, really 
well-designed for safe and non-competing reuse?

Is this any surprise?

-----Original Message-----
From:	MIME :briank at hex.net 
Sent:	Thursday, January 21, 1999 5:39 PM
To:	squeak at cs.uiuc.edu; janb at pmatrix.com
Subject:	Re: R.I.P. Byte

I've always believed the reason why some OS's were unstable was because
some OS's allow application developers to replace libraries critical to
the OS (Windows is a prime example of this), whereas other more stable
OS's (i.e., Linux) discourage the modification of critical/kernel OS
code by applications.

  --Brian

Jan Bottorff wrote:
 
> I don't know, one of the last issues of BYTE published in the summer of '98
> had just an incredibly good article on why OS's were unstable. It seemed
> like the bottom line conclusion was "new untried lines of code are
> unstable, the more you have the more unstable you are". I personally agree
> with their conclusion.

----------------------
--Brian Koontz      --
--Routech, Inc.     --
--briank at routech.com--
----------------------

 << File: ENVELOPE.TXT >> 





More information about the Squeak-dev mailing list