[apps-discuss] the use of registries

Erik Wilde <erik.wilde@dret.net> Thu, 04 February 2016 18:30 UTC

From: Erik Wilde <erik.wilde@dret.net>
To: "apps-discuss@ietf.org application-layer protocols" <apps-discuss@ietf.org>
Message-ID: <56B398CF.6080404@dret.net>
Date: Thu, 04 Feb 2016 10:30:39 -0800
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.11; rv:38.0) Gecko/20100101 Thunderbird/38.5.1
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"; format="flowed"
Content-Transfer-Encoding: 7bit
Archived-At: <http://mailarchive.ietf.org/arch/msg/apps-discuss/i_l97k1tp8vZgi_hzVOjJCK268U>
Subject: [apps-discuss] the use of registries
Precedence: list

hello.

i have started working on a draft that talks about the use of 
registries. it seems that discussions about why to use them, when to use 
them, and how to use them, are a recurring topic.

personally, i think registries are a fascinating topic because they are 
a "specification pattern" that occurs on all levels in the networking 
stack. in some cases, people like to argue that registries (at least on 
the app level) are a thing of the past and we don't need them anymore, 
and most often they give two reasons:

- the way the current registries are operated is a bit antiquated and a 
more modern approach (such as wikis, or maybe something a bit more 
structured) makes more sense.

- for the application level, URIs can always be safely used as 
identifiers, eliminating the need for controlled namespaces.

without discussing the merits of those arguments, i thought that it may 
make sense to try to summarize why, when, and how registries are used 
and can be used. this kind of document can serve as a reference for spec 
authors to decide on the use of registries, and maybe it can also serve 
as a reference for developers who want to better understand why a 
registry has been established.

i have submitted a first draft which contains "why", "when", and "how" 
sections, but definitely could be more complete, and also probably needs 
to look at other topics as well.

https://tools.ietf.org/html/draft-wilde-registries-01

it would be great to get some initial feedback from this list. the two 
main questions i have are:

- is writing up such a document a worthwhile thing to do? is it likely 
to result in a document that helps specification writers?

- judging from the current outline, are there any glaring omissions, 
either on the top level (why/when/how), or one level down in those sections?

thanks a lot and kind regards,

dret.

ps: https://github.com/dret/I-D/tree/master/registries is where the 
draft lives, please feel free to submit issues or PRs.

-- 
erik wilde | mailto:erik.wilde@dret.net |
            | http://dret.net/netdret    |
            | http://twitter.com/dret    |

[apps-discuss] the use of registries Erik Wilde