Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 7 Mar 1996 13:38:22 -0700
From:      Nate Williams <nate@sri.MT.net>
To:        "Amancio Hasty Jr." <hasty@rah.star-gate.com>
Cc:        Nate Williams <nate@sri.MT.net>, chat@FreeBSD.ORG
Subject:   Re: Act Now ! 
Message-ID:  <199603072038.NAA02181@rocky.sri.MT.net>
In-Reply-To: <199603072018.MAA14436@rah.star-gate.com>
References:  <199603072003.NAA01967@rocky.sri.MT.net> <199603072018.MAA14436@rah.star-gate.com>

next in thread | previous in thread | raw e-mail | index | archive | help
>  > > Voice is just one data type. One incurs far more bandwith with
>  > > downloading images, files, etc...
>  > 
>  > That's bursty bandwidth.  Voice takes up bandwidth fairly at a fairly
>  > constant rate.
> 
> Voice does not demand constant bandwith on the internet;rather, the
> internet delivers or tries to deliver a constant rate.

To have a conversation of any kind requires that either one person or
the other is talking *most* of the time.  This means that the bandwidth
is being use *most* of the time, irregardless of the implementation.

Compare this to downloading web pages.  When you first download the
page, you will incur a BW hit, but even so if a person wants to actually
*use* the data sent to them, they won't constantly be asking for more
data every 2-3 seconds (except for those rare folks who can read and/or
understand information exceptionally fast).

These aren't issues of implmentation, but simply usage facts.  When you
talk on the phone, you tend to use more 'bandwidth' due to it's somewhat
real-time nature than when you download any static data.

> as freebsd become widely available on the internet as a matter
> of fact IUMA with their huge mpeg distribution is a good example
> of a *huge* repository for the masses.

That's irrelevant.  It's still static data that is generally downloaded
once, and then used for it's intended purpose.  Telephony data can't be
re-used for the simple fact that 10 seconds after the data has been sent
it is no longer valid as new data has superceded it.  (This is looking
at the data from a statistical point of view vs. a users point of view).
> 
>  > > By far the worst Internet polluter has got to be those zillions
>  > > of Win95 users and AT&T now providing Internet services.
>  > 
>  > No, they aren't 'polluting' the Internet any more than you are, and in
>  > actuality they are using up much less BW than you are.  However, there
>  > are lots more of them.
> 
> Okay, they are not polluters no more than voice oriented apps. My 
> point is that all of the sudden we are dumping a huge user base
> into the Internet and that we should be more concerned as to how
> to support them rather than voice or video which represents a tiny
> fraction of the user base.

Except that the voice and video users are using up a *HUGE* portion of
the total bandwidth used in comparison to the # of users.  A voice user
uses up at least an order of magnitude more BW than a non-audio user.
So, we should outlaw audio on the Internet in order to support an order
of magniture more people than we are currently.  And, if we outlaw video
users we'll get 3-4X the amount of users.

(Again, these numbers are completely bogus, but you can understand why
outlawing them is actually considered).

> So how come we are not hearing a big complain about the new rapidly
> emerging user base???

The only folks complaining about the rapidly emerging user base are
those folks who want it to be 'like it used to'.  If you've got nothing
to compare it to, then you have nothing to complain about.

>  > > One would hope that if AT&T is going to provide Internet Services that
>  > > they can also support their own infrastructure which happens to support
>  > > much of the Internet backbone infra structure.
>  > 
>  > Why should you deserve any more BW on the backbone than a new user like
>  > my father?  Just because you were there first doesn't mean he has as
>  > much right to use video/audio tools as you do.
> 
> The gating factor today is how much am I willing to pay for my 
> bandwith 8)

You are paying money to your ISP, who doesn't support the backbone
directly.  Does that mean if I get a T1, I should be able to use a T1's
portion of the backbone at all hours of the day?

That'd be like telling the U.S. government that they must build you a
bridge to your new house on the other side of the river because you
bought a GM car.  Just because GM (your ISP) gives you the ability to
be on the roads (Internet) doesn't mean the people who build the roads
(backbone) are responsible to give you the bridge (BW) you want.

(Lame example, I know, but it's the best I could come up with on short
notice).

>  > So should we expect the MBONE to go away since it's neither feasible nor
>  > useful to the general public?
> 
> The MBONE is feasible and is probably the only generally available
> mechanism to deploy audio / video.  If you are interested more on the
> MBONE please do a net search on the topic .

The MBONE is far and away *NOT* the only mechanism to do A/V.  Most of
the folks doing commercial implementation require 30kpbs (I just did
research into this topic for work) to do Audio/Video/White-Board.
That's a *heck* of alot less than MBONE.  The Mbone was designed to show
that A/V on WAN was possible, but it's certainly not the only way to do
A/V nor even the best implementation.  And, it's not generally available
to anyone but to a select few, so things like Internet-Phone are *much*
more generally available and work.  (I was suprised how well it works).



Nate



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199603072038.NAA02181>