[MLB-WIRELESS] Re: Node x is over this way -was- Applications

[Drew]

I'll drink to that.

d

Shane Chubb wrote:

> I now understand why so many people are unsubscribing from this list.
>
> an·ar·chy    Pronunciation Key 
> <http://www.dictionary.com/help/ahd4/pronkey.html>   ( n r-k )
> n.pl.an·ar·chies
>
>    1. Absence of any form of political authority.
>    2. Political disorder and confusion.
>    3. Absence of any cohesive principle, such as a common standard or
>       purpose.
>
>
> ------------------------------------------------------------------------
> [New Latin anarchia, from Greek anarkhi , from anarkhos, without a 
> ruler  : an-, without; see a-1 + arkhos, ruler; see -arch.]
>
>
> Source <http://www.dictionary.com/search?q=00-database-info&db=ahd4> : 
> The American Heritage® Dictionary of the English Language, Fourth Edition
> Copyright © 2000 by Houghton Mifflin Company.
> Published by Houghton Mifflin Company. All rights reserved.
>
>
>
>
> -----Original Message-----
> From: Ben Anderson [mailto:a_neb at optushome.com.au ]
> Sent: Wednesday, March 20, 2002 9:10 AM
> To: Shane Chubb; Drew; drew at no6.com
> Cc: Clae; melbwireless at wireless.org.au; jon at webprophets.com.au
> Subject: Re: [MLB-WIRELESS] Re: Node x is over this way -was-
> Applications
>
>
> > This isn't a new episode of Austin Powers...
>
> I knew the term 'mojo' wasn't ideal :)
>
>
> > Personally if anyone tried to implement a mojo-meter (for want of a
> > better word) and decided who was more important based on it, I would not
> > want to be part of that network.  Perhaps Ziggy would like to hear your
> > thoughts tho.
>
> It's not about deciding importance...  It's trying to make the network 
> scale
> to more than a small community.
>
>
> > And whilst we're on the matter, say a node was being restricted to the
> > their lame-mojo rating dont you think they are going to be looking for
> > other ways around the problem.
>
> Saying "restricted to mojo rating" implies that users are going to have
> access cut off if they spend more than they make.  Not true, they only get
> limited access to the resources under demand.  Anything that's not being
> used, goes up for grabs.
> And can you refrain from including so much emotional language in your 
> email,
> you make this sound like a flame :)
>
>
> > Not everybody plans to use the network to 100% capacity 100% of the
> > time.  If you think your going to get that, you should be setting up
> > your own private network.
>
> It's not about me.  It never was.  It's about getting this network to 
> scale
> to whole city+ size.
>
>
> > This was going and in my mind still is going to be a FREE network.  No
> > limitations can be imposed or it's not free.  This sounds like those
> > FREE internet services where they stream adds to you (you pay the price
> > by watching their adds).  If your concerned about congestion in your
> > neck of the woods, invest some time and effort into building the
> > community around you instead of trying to control the entire network.
> > Its like trying to get the Government to make the roads smoother instead
> > of fixing your suspension.
>
> I'm concerned about congestion *everywhere* -- I can't build a city-sized
> private network without more money.  And more money implies I need to 
> become
> more capitalistic, less altruistic.  You sound like someone who values
> freedom, and should understand why I hesitate to make this choice.
> In your "free" network you're making a tradeoff between who gets access to
> bandwidth.  Basically in your system, users who want low-latency 
> access get
> trodden all over by people who want high bandwidth access.  And this 
> problem
> becomes exponentially worse as more leaf nodes are added.
>
>
> > Mojo this, Mojo that, I really cant believe this is a serious
> > discussion.
>
> Look past the word.  Call them tiddlywinks.  Call them 'zoinks'  call them
> 'shanes' -- I don't care.  All I want is some QoS metric that gets
> distributed in some sort of fair fashion.  Leaving it open slather,
> unrestricted is decidedly not fair for applications that require a low
> latency, so I don't understand how you consider your tradeoff between
> access, bandwidth and latency is inherantly freeer than my design.
>
> > Smashing Baby Yeah.... NOT!
> >
> > PS - want to flame me in return for this - make it a private post,
> > others dont want to read about that.
>
> Hopefully I haven't used enough emotional language in this message for 
> it to
> be considered a flame.  I'm not attacking you personally.  Not discounting
> your ideas, or your right to say them.  I'm pointing out inherant design
> problems with the design you've chosen to call "free" -- I challenge 
> you to
> come up with alternative more "free" methods of protecting the network
> against DoS, large scale scalability and providing some QoS metric.
>
> Cheers,
> Ben.
>