[standards-jig] JNG Ramblings.

Adam Theo theo at theoretic.com
Fri Aug 9 04:04:12 UTC 2002


David Waite wrote:
> A couple of points:
> 
> 1) JNG is really shorthand for 'all those things we have thrown in the 
> big basket because they break clients or are otherwise hard to think 
> about or implement'. We really need to think about what JNG means in 
> terms of an overall vision before we start prototyping implementations.

For me, JNG is about complete seperation between the routing/transport 
layer and the environment/application layer, with a set of optional core 
tool protocols in the middle to ease the building of the 
environment/application layer.

I believe that JNG should try to learn from Jabber's success of being 
very easy for humans to work with and manipulate, even if it is not 
extremely efficient for machines. As someone whose post I can no longer 
find once said "there is something to be said for the ability to open a 
telnet client and begin 'talking Jabber' to anything."

> 2) A binary application protocol greatly reduces our chances of getting 
> standardized. Exercise for the reader: Find three binary protocols on 
> standards track or which are standard @ the IETF which have been 
> authored within the last four years. And of course, there is the obvious 
> question: 'Why aren't you using BEEP?'

-- 
     /\  Adam Theo, Age 23, Tallahassee FL USA
    //\\   Email & Jabber: theo at theoretic.com
   //  \\  Pager: (850) 709 7738
=//====\\=
//  ||  \\  Theoretic Solutions: http://www.theoretic.com
     ||         "Building Ideas by Bringing them Together"
     ||      Jabber Protocol: http://www.jabber.org
     ||         "The Next Generation Communications Protocol"
     ||  "A Free-Market Socialist Patriotic American Buddhist"




More information about the Standards mailing list