[TriLUG] No more Linux on WRT54G???

Greg Brown gwbrown1 at gmail.com
Tue Nov 15 10:58:18 EST 2005


Good points.  When I wrote my e-mail I was *trying* to apply the
needless layer upon layer of VLANs and Radius this, that, and backup
paths to an AP-1200 was was misbehavin'.  The AP-1200 finally did it,
but wasn't happy about all the stuff it had to do just to push
packets.  I would have never been able to do such a thing with an
OpenWRT box.

That said, I just sent off another WRT54G into the wild yesterday. 
One of my wife's employees needed a NAT/firewall/this-and-that box and
a wireless connection to her son's PS-II that only had RJ-45.  So I
got her a v4 WRT54G and a used WAP54G.  I configured both in as secure
a manner as possible (WPA vs WEP) along with firewall enabled, no
pings, etc.  All of this took about 10 minutes (if you don't count my
typo of the admin password snafu) and I put the units into grocery
bags labeled "system 1" and "system 2" wrote some directions, put it
in a box, and volia - LAN in a can (or a box).

Anyway, I'm going off the deep end there.  Here's how my OSS vs. Non
OSS network stuff breaks down:

Home user: WRT54G running Linksys code.  Easy and simple.

Wireless hot spot or hotel mesh network: OpenWRT (formerly WRT54G
testing ASUS WL-500G)

Enterprise fortune 500 anything: Cisco.

Even though mainframes are considered passe by most I was in the data
center last night looking over a sea of IBM Z Series boxes.  Granted
they aren't the 390 series from days of yore but I would still
classify them as big iron.

Anyway, certainly OSS has it's place and so does commerical hardware. 
I think that at times people just get that warm fuzzy feeling they
like when they hear "cisco", kind of like they used to get with IBM
("Nobody ever got fired for buying IBM").  Cisco's days as true top
dog can't last forever but I do like Linksys roadmap they came up with
and though I haven't seen any of the hardware yet I like the idea of
the new VoIP stuff they just released.

Oh, and that brings up another point.  Say I put in a bid at SMB
Company X and they want a VoIP system and the management stuff is not
technical.  I could propose an Asterisk system and we all know that
would work and work well but if proposed a Linksys One, powered by
Cisco, I would bet dollars to pesos that they would be more interested
in the Linksys/Cisco because of brand name recognition.  I'd certainly
propose both OSS and commercial, though.

I dig open source but appreciate commercially developed specialized
products.  They both have their place

Oh, I used to own a PC Jr. :)  And a XT (and I remember the CADET
"Can't Add Doesn't Even Try").  LOL!!  Your e-mail brought back some
good memories...

Greg

On 11/15/05, Rick DeNatale <rick.denatale at gmail.com> wrote:
> On 11/14/05, Greg Brown <gwbrown1 at gmail.com> wrote:
> > >
> > > > The real reason may well have to do with the threat that Openwrt and
> > > > the like pose to Ciscos business model, since lots of WRT54s with FOSS
> > > > software replacements seem to be displacing more expensive Cisco boxen
> > > > in commercial applications.
> > > >
> >
> > I disagree 1000%.  When you boot an AP-1200, apply layer up layer of
> > VLANs, RADIUS auth, as well as a 1000+ line tftp booted text config
> > (though it can all (or most) be done via the web config) plus TACACS
> > and every other protocol found in a true enterprise network you will
> > know that a WRT54G running OpenWRT would NEVER be able to handle the
> > load nor should it have ever been asked to do so.
> >
> > Linksys = home or SMB (look at Cisco's link roadmap and VoIP products
> > announced recently)
> > Cicso= Enterprise (or anywhere untested, unproven code should be deployed).
>
> Yes, but I think that there are some applications where the
> requirements are more than those in a home/soho environment but less
> that a full blown enterprise environment.  Internet cafe's and hotel
> applications come to mind.
>
> Without open source solutions on appliance hardware, these would
> require more expensive "industrial strength" solutions.  With them, it
> seems that lots of folks are successfully deploying those customized
> appliance boxen and nibbling away at the small end of that market.
>
> I think that there might be a parallel to what happened 20 years ago
> when personal computers were considered toys and you needed a
> mainframe to do serious stuff.
>
> I had a good opportunity to observe (and to some small degree
> participate in) IBM's odyssey through the PC revolution.  Back when
> the PC was a skunk works project, the prevailing view within IBM was
> that IBM would never come out with something to compete with the like
> of the Apple ][. Then when we did it was viewed pretty much strictly
> as a home computer, and home computers were toys. This attitude led to
> such raging successes as the PCjr <G>.
>
> The strength of IBM was seen in the huge cash cow of its market share
> in REAL computers. The 360->370 architecture and it's descendants
> ruled the roost.  There were some in IBM who thought that the PC
> needed to be warped (no not that warp which came later) to fit better
> into that  world. This led to such further huge successes <G> as the
> PC/XT370 and PC/XT3270. If anyone else remembers those, I might want
> to sign you up for my computer trivia team.
>
> Some of us, both inside and outside IBM had an inkling of where things
> might go even back in those early days of the mid-1980s.  For example,
> there was an IBM employee microcomputer hobbyist club up in
> Poughkeepsie even before the advent of the IBM PC.  We had a talented
> guy who drew the covers for the newsletter.  I vividly recall one
> cover which depicted mainframes as dinosaurs, while the more agile and
> nimble mammals (depicted in the form of Apple's and CP/M machines with
> legs and arms) were busy stealing the dinosaur's eggs (in the form of
> mainframe disk packs).
>
> The split between IBM and Microsoft came about in no small part
> because of the lack of a common vision over OS/2. IBM felt that they
> had all this mainframe expertise which needed to drive the system.
> That's why for example OS/2 and Windows used different graphics
> models. Windows was targetted towards the emerging wysiwyg/desktop
> publishing arena where the coordinate system was oriented to work well
> with text (y increases down the page), while OS/2 PM was based more on
> mainframe graphics which were primarily CAD based.
>
> Meanwhile, the old mainframe architectures HAVE been pretty much
> relegated to dinosaur status. Time and techology have moved on. The
> techology once seen as hobbyist or home oriented has been applied by
> new companies to revolutionize the IT industry.
>
> IBM has devolved from the company you went to buy enterprise hardware,
> to being primarily a successful services company. They still sell
> server hardware, but it's now mostly microprocessor based, either on
> Intel or PowerPC hardware, even the zSeries (the sole descendant of
> the old mainframes) makes a lot of use of PC technologies, and has a
> much reduced market share.
>
> Could IBM have stopped this from happening?  Probably not. The
> introduction of the IBM PC in 1981, was seen by many as legitimizing
> personal computers.  If IBM hadn't done this, it might have delayed
> the revolution, but I suspect that it wouldn't have stopped it.
>
> So it might be that Cisco is in a position today which is similar to
> IBMs in the 1980s, or maybe the 1990s.
>
> --
> Rick DeNatale
>
> Visit the Project Mercury Wiki Site
> http://www.mercuryspacecraft.com/
> --
> TriLUG mailing list        : http://www.trilug.org/mailman/listinfo/trilug
> TriLUG Organizational FAQ  : http://trilug.org/faq/
> TriLUG Member Services FAQ : http://members.trilug.org/services_faq/
>



More information about the TriLUG mailing list