[ale] open source wars at DOD

Chris Fowler cfowler at outpostsentinel.com
Fri May 24 17:49:22 EDT 2002


What Boa said sums up why I believe future System Administrators will loose
the RTFM words
and gain the RTFS word along with "Use the source Luke".



-----Original Message-----
From: Bao C. Ha [mailto:baoha at sensoria.com]
To: ale at ale.org
Sent: Friday, May 24, 2002 4:24 PM
To: 'Thompson Freeman'; 'Pete Hardie'
Cc: 'Joseph A Knapka'; ale at ale.org
Subject: RE: [ale] open source wars at DOD



I have two personal examples:

1)  When Nimhda hit us last summer, we had a problem with our Win2K
servers communicating with each other.  It seemed like there was a
loss of the trust relationship.  Since we did not have a technical
support contract, we paid $$$ to Microsoft for assistances.  After
a week wasting our time, the best solution was "to reinstall
everything".  Heaven forbids!  We did not do what they told us.  It
was an unfortunate combination of the Microsoft's implementation of
Active Directory/LDAP, Win2K SP1, and Nimhda just confused the whole
issue.  And we fixed it without Microsoft help.

2)  Also last summer, about kernel 2.4.8, embedded JVM (Java Virtual
Machine), from both Sun and IBM, failed on the SH4 architecture.  We
reported and waited and waited while IBM/Sun were pointing fingers to
the tool chains, glibc, hardware, software, ...  I broke down and
traced it back to Linuxthreads.  Within a couple of days, it got
fixed.  Niibe, one of the kernel developers on the SH4 architecture,
is so efficient that he optimized and improved the kernel to more
than twice the performance.  He also unwittingly broke Linuxthreads!
You need something as bloated as these JVMs to make it failed.  But
it was fixed.  The optimization and performance are still there.  And
I still blame Linuxthreads to depend on some obscure, non-standard
behavior of Linux's mmap system call.

It is just an illusion if you think that a commercial outfit will
provide better support than open-source developers.  If it can be
proven that there is a bug in his code, a developer will fix it.  In
an open-source project, developers are more accessible.  In a
commercial environement, you will be talking to droids, 1st, 2nd, and
3rd level of support before they let you contact developers to explain
the problem.  And they are quite happy to be fed with your money while
you twisting your fingers waiting for an answer.

But then I don't need hand-holding either and would not care less!

Bao

> -----Original Message-----
> From: Thompson Freeman [mailto:tfreeman at intel.digichem.net]
> Sent: Friday, May 24, 2002 12:22 PM
> To: Pete Hardie
> Cc: Joseph A Knapka; ale at ale.org
> Subject: Re: [ale] open source wars at DOD
>
>
> On Fri, 24 May 2002, Pete Hardie wrote:
>
> > Joseph A Knapka wrote:
> > > All this recent hoopla about security in open source, integrity
> > > of public data, etc. has got me asking myself the question: Is
> > > there *ever* any advantage to closed-source software, for anyone
> > > besided the vendor? I can't see one, myself. It is self-evidently
> > > in the buyer's best interest to insist on open-source solutions;
> > > only the gigantic dead weight of M$'s marketing machine obscures
> > > this fact in practice. Eventually open-source will prevail in
> > > a free marketplace.
> >
> > I'd say that closed-source from a typical vendor is faster
> to respond
> > to specific customer requests than an open-source project, since the
> > money paid for the program can fund a dedicated customer
> support group
> > that is acutely aware that failing to respond to customers
> will result
> > in the loss of their jobs; typical open-source projects are not as
> > responsive because there is no immediate negative feedback for slow
> > fixes.
>
> Huh??
>
> I'm tempted to say "Prove it", but I'm time strapped to do my
> own research
> to support the contrary. I do seem to recall seeing that open source
> companies were among the faster to patch security holes. But
> then again,
> my memory is a security hole being exploited 8-(.
>
> >
> >
> >
> >
>
> --
> ===========================================
> The harder I work, the luckier I get.
>                     Lee Iacocca
> ===========================================
> Thompson Freeman          tfreeman at intel.digichem.net
>
>
> ---
> This message has been sent through the ALE general discussion list.
> See http://www.ale.org/mailing-lists.shtml for more info.
> Problems should be
> sent to listmaster at ale dot org.
>
>

---
This message has been sent through the ALE general discussion list.
See http://www.ale.org/mailing-lists.shtml for more info. Problems should be
sent to listmaster at ale dot org.




---
This message has been sent through the ALE general discussion list.
See http://www.ale.org/mailing-lists.shtml for more info. Problems should be 
sent to listmaster at ale dot org.






More information about the Ale mailing list