Re: why strict Net neutrality works best: simple beats complex
Abhijit Gadgil <[email protected]> Wed, 20 May 2015 13:14:05 +0530
| Newsgroups | gmane.org.telecom.india-gii |
|---|---|
| Message-ID | <CAOFWafhiqBQtiDbE_PynpZXEZ1a_=xKAB+ydirehMRq+5m3N5w@mail.gmail.com> |
>> All of us would agree that any use of shared resource needs some >> policing - if for nothing else - to at-least prevent it from being >> hogged completely by rogue elements to the extent it becoming >> unusable. > > > Huh? All of us certainly do not agree, doesn't the rest of your argument > collapse, then? Yes, it would indeed collapse to an extent! Policing is a 'strong word', isn't moderation a right word? (especially in view of recent events on this very site?) > What rogue elements are you talking about, pray? Well rogue elements here I am talking about misbehaving traffic flows (eg. UDP flows - which have no end-to-end flow control - if not identified, metered and policed - could throttle bandwidth in the network, which could otherwise be used for useful things - (the definition of useful changes per user agreed)). The point is - what it takes to police those flows - exactly the same it takes to provide - preferential treatment to other flows (may be little extra - allocating memory etc). But the cost argument about that is not valid - as all of this is - free (software) or commodity (hardware). So if one is fine - with absolutely un-policed network, one can indeed safely argue - that there's additional complexity in 'preferring a flow' but not otherwise. > Arun Mehta -- अभिजीत [ http://oltsm.wordpress.com ] _______________________________________________ India-gii mailing list [email protected] https://lists.india-gii.org/mailman/listinfo/india-gii