[Year 12 Its] Ongoing misinformation about topology

[a.hopkins at bcc.vic.edu.au]

Frank's comments strike a chord with me.  I came into computer teaching without
formal qualifications in the area, but having done programming out of personal
interest for perhaps twenty years.  The whole networking and hardware side of
things was not on my radar.  I have found the textbooks to be lacking in
clarity and concision.  I don't mind my students knowing more about networking
than I do, because some of them are acquiring that knowledge in very practical
ways.  But other students who aren't attuned to hardware aspects, and I myself,
could surely be served better by the textbooks we use.  When some time ago I
asked my techie about some of the explanations offered, he thought they were
pretty much irrelevant.

Alex Hopkins


Quoting Frank Van Den Boom <vandenboomfj at aquinas.vic.edu.au>:

>
>
> There have been several discussions over the past couple of years
> concerning the poor understanding and confusion about network topology
> in texts, exam questions etc. I was looking forward to new editions of
> the texts in the hope that we would finally get it right.
>
> I was going through page proofs of the Thomson/Nelson book (new edition
> of Building Information Systems), and this is what it says about Star
> network topology.
>
> "The most common topology is the star network. The main type of star
> network has a central computer, usually a server computer, and all
> computers and devices are connected directly to it. This configuration
> is useful when the data to be used is required by many people and needs
> to be centralised so that its integrity and security can be easily
> managed. The access to the network is usually controlled by the network
> operating system, that is run from the central computer.This topology
> operates as a client/server network. A simple star network is a network
> or segment of a network that is controlled by a switch or hub. In this
> case the network is operating as a peer-to-peer network as there is no
> central coordinating computer. There may be various servers operating on
> the computers."
>
> I won't put in the text on Bus and Ring networks other than they do
> focus more on the protocol and cabling topology. I don't know what the
> above treatment of star network is based on and I find it all very
> confusing. There are a ton of questions I would love to ask about this
> paragraph. For starters - when was the last time any of us
> saw/installed/used or even read about a star network where all the
> devices were connected to a central computer? I won't bother with the
> rest of the questions for now...
>
> In looking at the new Janson/Dawson book, at least the focus is largely
> on the protocol.
> I have not looked at any other of the new books on this topic but I am
> not confident that this confusion will go away.
>
> So are we really clear on what we are teaching about "topology" ? I'm
> not. The way the protocol works? The way the cabling is arranged?
> Do we all realise that even 10 years ago, it was common to find a
> 24-port token ring hub, which used a star cabling topology to connect
> its devices, which could have connected to it a bunch of PC's
> functioning on a peer-to-peer basis as well as a database server which
> was being used by the same PC's for a client-server application.
>
> So often, I read about this stuff in our IPM and IS texts in a way which
> treats them all as mutually exclusive options.
>
> The classic table that all of us have seen showing a list of advantages
> and disadvantages for each of these topologies in many cases is just a
> lot of rubbish in my view. For example, bus topologies are cheaper to
> implement because they use less cable - might be true for coax but not
> an old UTP/hub installation.
>
> I just had a quick look at the VITTA Networking CD to see how it treats
> protocol - more consistent in what it considers topology to be, but
> there are still things there that are oversimplified. For example, "A
> disadvantage of a ring topology is that if any device is added to or
> removed from the ring, the ring is broken and the segment fails until it
> is reconnected." In theory that is true, but most places would have used
> a token ring hub for years, in which case it is irrelevant. But we are
> not really in a position to show our students this when we teach it, and
> it is also meaningless in my view.
> I am not a trained communications person, and only built up an
> understanding of some of these things in a large network environment
> that I worked in a few years ago. So there is a good chance that some of
> my definitions/understanding are not quite right. But I can assure you
> that the IPM and IS books that I have used for the past 5 years have
> done very little to clarify much of this.
>
> It's probably too late for the authors of the new texts to review this
> area. Is there some other way we can build a body of content that we
> agree with, understand and can teach to kids in a meaningful way on this
> subject?
>
> Does anybody else out there feel the same way about this as me???
>
> Frank
>
>