Message-Id: <6.1.2.0.2.20050328131347.04b93aa0@mail.jefsey.com>
Date: Mon, 28 Mar 2005 13:26:33 +0200
To: ned.freed@mrochek.com, Doug Ewell <dewell@adelphia.net>
From: "JFC (Jefsey) Morfin" <jefsey@jefsey.com>
Subject: Re: [Ltru] Shelf life of data (was: Re: Question)
In-Reply-To: <01LMCEAVZ15I00005R@mauve.mrochek.com>
References: <20050325170422.OQBN4900.mta8.adelphia.net@megatron.ietf.org>
	<001301c53228$6fc9bd60$030aa8c0@DEWELL>
	<01LMCEAVZ15I00005R@mauve.mrochek.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Cc: LTRU Working Group <ltru@ietf.org>
Precedence: list
Sender: ltru-bounces@lists.ietf.org
Errors-To: ltru-bounces@lists.ietf.org

This calls for a few comments to understand what we are discussing.

At 20:32 26/03/2005, ned.freed@mrochek.com wrote:
> > > Retro compatibility is offered for a long nowhere. Try to use your
> > > 1930 telephone set, I am not sure it will work.
>
>Bad example. I have in fact done exactly this, and 1930s era phones are fully
>funcitonal today on a POTS line. (Of course they won't work on a digital
>conection, but that's to be expected.) Rotary dial phones were introduced
>around 1920, and even the phones without dials popular before that will 
>let you
>listen and talk. (Ringing is another, more complex story.)
>
>This poster illustrates this aspect of telephone history rather nicely:
>
>   http://www.sandman.com/images/telstory.jpg

In a nutshell it does not work, but under constraints rebuilding/protecting 
the 1930 environment it still works. I hope you can listen and talk, any 
one having tapped a line knows that. You certainly know how to dial without 
a rotary too I presume. You just document that the system you have is 
obsolete, i.e. out of use, of old design.

> > > Try to publish an
> > > Internet 1990 terminology. Try to use the root servers file of 1995,
> > > it will may be still work partly. Try to use DOS on your PC, it works
> > > very well, but you are missing a lot of things.
>
> > People have been quoting RFC 822, which is 23 years old, on this list.
> > Even though it has been updated by RFC 2822, obviously many of the basic
> > philosophies and terminology still apply.
>
>Exactly right. And more to the point, lots of other RFCs published at or 
>around
>the same time are nothing but historical artefacts. And at various points 
>along
>the way if you'd taken a survey you would have found a lot of people believed
>RFC 822 would join them in short order, having been replaced by X.400.

Many of us would have liked it to become osolete :-) and we will keep spam 
as long as it is not :-))))
I prefer STD 013 as an example.

> > > Objection to a decade obsolecense cycle?
>
> > It's much too general to apply blindly.  Some standards and technologies
> > are obsolete within a year or two; others are good for decades.  ASCII
> > is 42 years old (or 38, depending on what you mean by "ASCII"), and yet
> > many people still consider it the lowest common denominator necessary to
> > ensure reliable communications.
>
>And furthermore, we aren't good at predicting what will survive and what will
>be lost. Planned obsolescence is therefore a bad idea for standards, since
>things often don't go according to plan.

In our case we do not predict. We consider two things:

- the cycles of code reasignment of the ISO standard we consider and their 
interference
- the user behaviors we want to foster.

I am afraid you confuse the obsolecence of the "things" and the decided 
duration of validity of the data. The obsolecense of the validity of an 
entry has nothing to do wit the standard, the process, the application. The 
shortest the obsolescnence the greater the accuracy. In DNS this cycle is 
named TTL and can be from 0 to anything. Usually is one day to one week. If 
you want to be accurate (Dynamic DNS you can put id down to the minute or 
less).

The difference between those two notions are very important to address the 
questions risen by the charter.
jfc


_______________________________________________
Ltru mailing list
Ltru@lists.ietf.org
https://www1.ietf.org/mailman/listinfo/ltru