14 The two cultures, side by side and on the evidence

14.1 Comparing without folklore

Everyone who has worked in one of these bodies has an opinion about the other, and most of those opinions are unfalsifiable. This chapter compares only what the documents and the counts on this machine actually say.

Two ground rules, stated first so the rest can be read against them:

• Nothing here compares quality, speed or openness. No file read for this course supports a sentence saying one body is faster, better or more open than the other. The counts measure how much text exists, and nothing else.

• Nothing here describes culture. There are no meeting records, no attendance figures, no membership costs and no names in the material. What follows is procedure.

14.2 The comparison table

	3GPP	IETF
Who takes part	member companies, through a partnership of organisations §3.2	individuals; no formal membership [1]
Structure	plenary, working group, sub-group §2	working groups gathered into areas RFC 2418
Who approves	the responsible plenary §4.6.4	the IESG [2]
How a decision is expressed	one status value against a document number §9.2	rough consensus, judged by the chair [3]
Between meetings	email; silence grants agreement §8.6	email; a Last Call open to anyone RFC 2026
Unit of change	the Change Request §4.6.1	a new document
After publication	new versions of the same document §4.6.5	never edited; superseded instead RFC 2026
Requirement words	shall, should [4]	MUST, SHOULD, MAY, in capitals [5]
Grouping	Releases §4.0B	nothing corresponding appears in the material read here
Implementation evidence	not part of approval §4.1.1	part of the definition of an Internet Standard [6]

14.3 Where they are doing the same thing

Four mechanisms look different and are not.

Authorisation before writing. 3GPP will not let substantial work start in a working group before the plenary has decided §6.1, and requires a scope, a schedule, a rapporteur and four supporting Member Organizations §6.0.2. The IETF charters a working group before it works RFC 2418. Both refuse to let a document exist without somebody having agreed it should.

A named person who is accountable. 3GPP has the rapporteur, who is a delegate from a member company and is answerable for the technical quality of a document §4.2, §4.1.2. The IETF has the working group chair, who is the person who determines whether rough consensus exists ^[7]. Both put a single human being where the buck stops.

Deciding by email. Both do it, both bound the period, and both write the rules down §8.0, RFC 2026. The difference is which way the default falls: in 3GPP, no comment means agreed §8.6; in the IETF, the process announces its intention and asks the whole organisation to object.

Handing numbers to somebody else. 3GPP's Support Team registers code points with outside bodies such as IANA, in the name of an Organizational Partner §3.2. IETF documents hand registries to IANA in a section written to a prescribed shape RFC 8126. Same problem, same third party.

14.4 Where they genuinely differ

Permanence. This is the one that changes how you read everything else. A 3GPP specification is reissued whenever an approved change request lands §4.6.5, so the document you are holding has a version number and a change history. An RFC is never edited after publication, and a new version of a standard goes through the whole process again as a new document RFC 2026.

Neither approach is free. 3GPP's living document means you must always check which version and which Release you are reading Reading a 3GPP number — the series, the version, the file name. The IETF's fixed document means you must always check what has obsoleted or updated it, from its own header block RFC 7841.

Whether anybody built it. 3GPP approval is a decision, taken by a plenary on a document §4.1.1. Nothing in the working methods asks whether an implementation exists. The IETF's top maturity level cannot be reached without multiple independent interoperable implementations with substantial operational experience ^[6].

What "consensus" is for. In the IETF it is the ordinary decision mechanism and there is no other ^[8]. In 3GPP the word appears once in the working methods, and it appears as an exception: a plenary may approve an otherwise forbidden change to a frozen specification if it is the consensus of the meeting that the exceptional action is justified §4.6.2.

Who may speak. The IETF's process is open to all, by rule ^[1]. How a 3GPP decision is actually carried, and with what weight per member, is not in the working methods at all — it is in the 3GPP Working Procedures ^[9], which was not read for this course. So this chapter can state the IETF's rule and cannot state 3GPP's.

14.5 The one place the rulebooks touch

For all the distance between them, the two sets of rules meet at exactly one point in TR 21.900, and it is a small, practical one. When a 3GPP specification carries a service-interface file, that file is extracted and published on its own — in a character encoding defined by the IETF.

RFC 3629 is the seventh entry in TR 21.900's reference list §1A, RFC 3629. A body that writes its own rules for everything else reaches out for a character encoding, because rewriting that would be absurd. It is a small illustration of the IETF principle of protocol ownership: the body that owns a specification maintains it, and everybody else cites it RFC 3935.

14.6 The volumes, and what they do not mean

	3GPP	IETF
Documents	3,928 specifications [10]	9,735 RFCs [11]
Meetings recorded	4,371 [12]	none in this archive
Contributions recorded	1,629,581 [13]	none in this archive

The right-hand column is mostly empty, and that is a fact about the archive on this machine rather than about the IETF. The archive holds RFCs and nothing else: no charters, no minutes, no mailing list archives, no Internet-Drafts.

14.7 Where this leads

Reading either body's paper trail from a cold start turns all of this into a practical exercise: starting from a sentence in a live document, in either body, and working back to the decision that put it there. What this record cannot tell you collects every gap this course has admitted to, in one place.