--=======50D859BE=======
Content-Type: text/plain; x-avg-checked=avg-ok-5AD01FF4; charset=us-ascii;
format=flowed
Content-Transfer-Encoding: 8bit
Hello Robert -
At 06:22 AM 9/22/2003 -0500, you wrote:
>CA> This has been said about opensource very frequently. The implications are
>CA> that Microsoft hires people who are not "very sharp". I tend to doubt
>this.
>CA> The fact that the source code is available for these 'very sharp'
>people to
>CA> develop the software also means the source code is available to be
>CA> examined for flaws by the malicious programmers who need that info
>CA> to violate the software.
>
>Oops... I did not explain myself well enough. The operative word was
>supposed to be 'a lot'... not 'sharp'. Microsoft indeed does hire
>sharp people, but they cannot possibly hire as many people for a
>project that is going to run within a specified period of time, as the
>Open Source community can get involved in a venture independently,
>over an open-ended period of time.
There are many assumptions about opensource and Linux in particular. I think
if you could find a listing of 'contributors' to any one project you would
be surprised
that there aren't as many as you might think.
>Also, in spite of MS hiring sharp people, they are all hired by people
>with the same general ethic, and method of deciding who will make a
>good employee, with 'team player' being high on the list. ( I know
>more than a few people who have both worked for and been rejected by
>MS, over the years, and have been privy to numerous renditions what of
>the interview process is like)
I've not worked for Microsoft and even if I had I would only know a certain
number of people within the company and those I did not work with I would
only be guessing as to their qualifications.
>I think that the 'like-mindedness' which the company's hiring
>practices wind up with, in their employees, might be detrimental to
>the code development and debugging process, because it would wind up
>with a more 'single-perspective work pool, than might otherwise be the
>case.
For some "like mindedness" would translate into "focus". It's the cup half
empty half full thing. The tendency is to always see the downside for
Microsoft and exaggerate the positives for opensource. ;-)
>Also, in any large corporation like MS, there is a tendency for upper
>echelon decision makers not to want to hear 'bad new'... especially as
>a release deadline nears, and for lower echelon drones, operating in
>'survival mode', not to want to buck the system, by going public with
>'bad news', if they see that the honchos are going to bury it.
You have the impression that opensource 'projects' are totally democratic
and don't suppress unpopular ideas nor ever become cliquish. Tis not the
case from my experiences.
>I think that Microsoft's past behaviour where security flaws...
>especially with their IIS and SQL Data Base are concerned, are good
>examples of this.
I don't think Microsoft, or anybody else, knew that Symantec would publish
working code for MSBLAST in their newsletter. ;-)
As one who has written sofware I don't look for 100% perfection I watch to see
how rapidly the bugs are fixed.
>CA> BSD does have a reputation for being 'secure' but as John has said
>CA> that might be only because BSD is used even less than Linux is used
>CA> making it less attractive to those who wish to draw attention to their
>CA> ability to do harm.
>
>FreeBSD Server 4.8 Stable is very widely used... I do not have
>figures, but it has been a mainstay for website hosting, and large
>ISP-type operations, and other Internet appliances, for a long time,
>as were its predecessors... the line is about 10 years old, and the
>code it is bases on, is about 30 years old.
I know that people have said that maybe by 2005 Linux will be on 15% of
the servers which tells me it isn't on 15% of the servers now. BSD is on
only a small percentage of non-Windows servers. Don't have numbers but
I imagine valid stats are out there if anyone cared to know?
I don't know the lineage of BSD but I think there was OpenBSD before there
was FreeBSD. Whoever owns the real BSD has announced they are not
going to continue development any longer just recently. Think I read that
on SlashDot? Every *nix type OS wants to claim a lineage back to the first
UNIX and I don't buy that lineage myself.
>I am pretty sure that eBay uses FreeBsd 4.8 stable.
I don't know.
>It's not a desktop appliance, and I doubt if it ever will be... Linux,
>which is already much more user friendly, is much more focused on that
>niche... but FreeBSD... especially the secure version... is the kind
>of app that banks and power supply stations choose.
I suspect this is more of a guess than something you can substantiate
with stats.
{snip}
>CA> I really don't think any recent version of Linux would be easy to get
>CA> working on an i386 and by the time you get around to this it will be
>CA> difficult to find an i386 the capacitors haven't dried out on. A 'lesser'
>CA> machine of whatever that would be at the time (early Pentium?) is
>CA> certainly a possibility but "386" is becoming only a memory in the
>CA> real world.
>
>I just tossed out '386'... however, there are plenty of Linuxes, both
>older and newer versions, that would run on a 386 processor. Peanut
>Linux would be a good choice.
I went through all this when I had an i386. To begin with the typical i386 has
only 4 meg of memory. Those with 8 meg are rare and I've never seen any
with 16 meg even though people refer to them frequently here and elsewhere.
At 4 meg of memory even a mini-install would be hard pressed to function and
there is no way I can think of that Peanut would install onto that. The
typical
i386s had 100 meg drives in them too btw.
>Here is the URL to a fellow name Doyle McClennan's website page,
>listing all the the various versions of Linux that he sells, with
>links to informational pages for each distro.
I have a webpage for mini-Linux, medium sized Linux (incl Peanut), and KNOPPIX
at my tech website.
>FreeBSD will run on a 386 processor, too.
Truth is you won't be able to find anyone who has done it or is willing to
try either.
I know, I tried for over a year to find out how to install a newer Linux
'distro' onto
an i386 and it was all smoke-n-mirrors until one Linux user actually did it
but it was
a nightmare for him to do it after he found an i386 that had 8 meg of
memory. He
installed the SlackWare ZipSlack install onto it and had 20 meg left for
'other'. He
is the only person I know of that actually did install Linux onto an
i386. It's painfully
sluggish when you telnet in btw.
Typically Linux users will claim to have installed Linux on an i386 "years
ago" then
lost it when the hard drive crashed and can't remember what distribution
they used
nor how they hacked it to fit. Sure, I believe that.
>I have about a dozen 286 and
>8086/8087 boxes that are still running fine, and get occasional light
>use, as word processors, or for playing with early versions of
>software, that cannot run correctly on later processors.
All of my XTs, i286s, and i386s have died (usually the power supply) and I
really don't
care to hunt anymore down. I have two i486s with 16 meg in them though
that I still
use mainly for testing with older browsers and I have two Linux installs on
them. One
even has the old Linux Netscape v3 browser working on it. It's so slow
it's almost
laughable and the colors are messed up but it does work. My W31 browsers make
the Linux Netscape look stupid. W31 is the best GUI OS for i486s IMO, much
better
than Linux would ever be.
>Heh! You should see early versions of Leisuresuit Larry, when they run
>on a 486 box.....zzzzoooooooommmmmmm!
I had the unpleasant experience of trying to play 'Red Baron' on
one. Ain't gonna
happen. The better written games must've been testing for CPU speed because
many will play well on faster machines. I don't like playing games on a
keyboard
and much prefer my PlayStation for gaming.
Charles.Angelich
http:/www.undercoverdesigin.com/dosghost/
--=======50D859BE=======
Content-Type: text/plain; charset=us-ascii; x-avg=cert; x-avg-checked=avg-ok-5AD01FF4
Content-Disposition: inline
---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.520 / Virus Database: 318 - Release Date: 9/18/2003
--=======50D859BE=======--
To unsubscribe from SURVPC send a message to [EMAIL PROTECTED] with
unsubscribe SURVPC in the body of the message.
Also, trim this footer from any quoted replies.
More info can be found at;
http://www.softcon.com/archives/SURVPC.html
