3 Release 5: the founding, and the ring of work around it

3.1 Where a subject begins in a table

The 3GPP work plan has no opening paragraph. A subject begins the way everything in that table begins: somebody registers an acronym, and a row appears.

For this one the acronym is bare IMS — three letters, no prefix, no suffix, no phase number — and the row is marked Rel-5 IMS. Everything else in this course hangs off that row, either as a sibling of it or as a descendant.

Its title says nothing about IMS at all. It reads "Provisioning of IP-based multimedia services", which is one of the four proofs in How this subject is cut out of the work plan that a subject cannot be found by searching titles.

3.2 What that row said it was for

The work plan points at one paper as the document that opened the row: SP-000216, carried at plenary meeting SP-8 SP-000216. Behind it sits a description, S1-000299, titled "Support of IP multimedia services", and its objective survives word for word ^[1].

Three things in that sentence are worth slowing down for.

"Release 2000" is what the release now called Rel-5 was called while it was being planned, and the sentence describes it as an evolved Release 99 plus IP multimedia services. Nothing was being thrown away; something was being added.

"multiple media components per call" is the whole ambition in four words. Not a better telephone call — a call that can carry several kinds of media at once, which is exactly what the definitions in What IMS is, in the words of its own documents were later written to pin down.

"based on existing multimedia call control standards" says the group did not intend to invent the call control. It intended to use what was already there, which is the thread that becomes the never-ending protocol alignment work in Releases 6 and 7: interworking, messaging, and a stream that never stopped.

The justification is one sentence, and it is a smaller claim than most people expect from a founding document ^[2].

That is a stage 1 job description: requirements, from a services point of view. The architecture and the protocol are somebody else's rows.

3.3 What the record cannot say about the founding paper

SP-000216 is a real document with a real address, and this course may cite it. It may not describe it, and the reasons are worth knowing because they recur through the whole first half of this subject.

• The store holds no title for it. It is one of 32 papers the work plan names as opening a work item for which the store carries the number, the meeting and the file address and nothing else SP-000216.

• Its upload stamp is not its filing day. The stamp reads 2017-10-05, which is the day the store took the file in. Well over a hundred of these old papers carry that same day.

• Its meeting cannot be dated. SP-8 is one of four meetings named in this course — SP-8, SP-9, SP-15 and NP-20 — for which the meeting table has no row at all. They may be named and never dated.

3.4 The ring of work around it

Rel-5 carries 14 IMS rows in the work plan ^[5], and the bare IMS row is one of them. The other rows are not a queue of later ideas. They are the same feature, cut up so that each group can be given its share.

Acronym	Title of the row	Groups on it
IMS IMS	Provisioning of IP-based multimedia services	NP, R5, S1, S2, S3, S5, T1, T3
IMS-ASEC IMS-ASEC	Access Security for IMS	S3, T3
IMS-LI IMS-LI	IMS Lawful Interception	S3
IMS-CCR IMS-CCR	Call control and roaming to support IMS in UMTS	R5, S2
IMS-CCR-IWMM IMS-CCR-IWMM	Impact on MM/CC/SM	N1
IMS-CCR_Test IMS-CCR_Test	Conformance Test Aspects – IMS Call Control	R5
IMS-CAMEL IMS-CAMEL	CAMEL control of IMS services	N2
IMS-OAM IMS-OAM	Charging and OAM&P for IMS	S5
IMS-ONOSA IMS-ONOSA	Support of VHE/OSA by entities and protocols of the IMS (e.g. CSCF)	N5
IMS-CODEC IMS-CODEC	Multimedia codecs and protocols for conversational PS services	S4
IMS-TEST IMS-TEST	Conformance Test Aspects - Provisioning of IMS	T1
IMS-FrWk IMS-FrWk	IMS Framework Report (work stopped)	S1
IMS-Sex IMS-Sex	Service Examples (Work stopped)	S1
IMS_CONV_IP_COM-UEConTest IMS_CONV_IP_COM-UEConTest	UE Conformance Test Aspects – IMS for Converged IP Communications	R5

Read the group column and the shape of a 3GPP feature appears. Security has a row. Lawful interception has a row of its own beside it. Call control and roaming has one, and the impact on the older mobility and call handling has another. Charging and management has one. Conformance testing has two.

The plan writes those group codes and nothing that expands them, so this course does not say what N1, N2, N5, T1 or T3 stand for.

What matters here is the pattern rather than the letters: one feature, many rows, one per group's share — which is why How this subject is cut out of the work plan insists that 345 rows is not 345 pieces of work.

Two of the fourteen carry their fate in their own titles. IMS-FrWk, the framework report, and IMS-Sex, the service examples, both end with the words "work stopped" typed into the name IMS-FrWk IMS-Sex.

That hand-typed note is the only thing in the plan resembling a status, and it sits in the title rather than in a field.

3.4.1 The founding row is the only one with eight groups on it

Look again at the first line of that table. Eight group codes on one row, where every other Rel-5 row has one or two IMS.

That is what an umbrella row looks like. IMS is not a job somebody did; it is the heading under which SA1's requirements, SA2's architecture, SA3's security, SA5's charging and the testing groups' work were all hung. The rows beneath it are where the actual writing was booked.

It matters for reading any later number. A count of rows treats the umbrella and the real work as one thing each, which is why How this subject is cut out of the work plan insists that no field in this plan holds "how many pieces of work there were".

3.4.2 Three titles that tell you what was already there

Three of the sibling rows are about the effect of IMS on things that existed before it, and their titles say so.

IMS-CCR-IWMM is titled "Impact on MM/CC/SM" IMS-CCR-IWMM. IMS-CAMEL is "CAMEL control of IMS services" IMS-CAMEL. IMS-ONOSA is "Support of VHE/OSA by entities and protocols of the IMS (e.g. CSCF)" IMS-ONOSA.

The register carries no expansion of MM, CC, SM, CAMEL, VHE, OSA or CSCF, so this course leaves all seven as the plan writes them.

What the three titles show without any expansion is that a new subsystem does not arrive on an empty table: three of the fourteen founding rows exist only because something was already running that IMS had to fit beside.

3.5 The codecs came with it

IMS-CODEC is the one Rel-5 sibling this course can quote, and it is a clean example of a group being handed a slice. It belonged to SA4, the media group, and its paper is SP-000398, carried at meeting SP-9 — another of the four meetings that cannot be dated IMS-CODEC SP-000398.

Two details in that quotation are worth having. The first is the word default, which is doing the load-bearing work: what was to be standardised was the set of codecs every product would have, not every codec anybody might use.

The second is the parenthesis "(likely to become Release 4)", which turned out to be wrong — the row sits under Rel-5 — and is a small, honest reminder that a work item description is a proposal written before the fact.

Its justification is a single sentence about the market it expected ^[4].

3.6 What the Rel-5 numbers mean, and what they do not

The founding row has one document against its name and IMS-CODEC has four IMS IMS-CODEC. Neither number says anything at all about how much work was done.

The reason is in How this subject is cut out of the work plan: no IMS document earlier than 2015-01-23 carries a work item acronym in this store ^[6]. Rel-5 was written and argued long before that line, so its paperwork is simply not labelled in a way any count here can see.

The planned dates are no better as history. The IMS row carries a planned start of 2000-01-03 and a planned finish of 2006-12-01, and IMS-CCR carries exactly the same pair IMS IMS-CCR. They are the plan's own bookkeeping, quotable as planned days and never as what happened.

And the completion figure is not a status. 339 of the 345 IMS rows stand at 100% ^[7] — a percentage the work plan keeps by hand. No field anywhere in this data says whether a work item is running, stopped or finished.

3.7 What Rel-5 left on the shelf

The documents themselves are the one place where the founding release is still visible, and both of the central ones carry a stored Rel-5 version.

TS 23.228, the stage 2 description, has a Rel-5 version stored at 5.15.0 TS 23.228. TS 24.229, the SIP profile, has one at 5.26.0 TS 24.229. Both go on from there through every release to Rel-20.

Neither of those is a first draft. The stored Rel-5 text of TS 23.228 is version 5.15.0 and the stored Rel-5 text of TS 24.229 is version 5.26.0, and a release's text plainly went on being reissued after the release had a number.

The only date the register holds beside either is the day that version's file went up on the 3GPP server — 2006-06-08 for one and 2012-03-13 for the other TS 23.228 TS 24.229.

That is not the day the version was approved and not the day it was published, and no chapter here may call it either.

That is the shape of the rest of this course. A release opens work, the work lands in a document, and the document goes on being edited under later releases by items with new acronyms. Following the acronyms, as How this subject is cut out of the work plan does, is the only way to see it.

3.8 Where to read on

The specification this founding row eventually produced is TS 23.228, whose stored releases start at 5.15.0 TS 23.228. The protocol beside it, TS 24.229, also runs from Rel-5 TS 24.229.

Rel-5 built a thing. Rel-6 and Rel-7 made it reach the networks people were actually on, and started a maintenance job that Rel-20 is still carrying: Releases 6 and 7: interworking, messaging, and a stream that never stopped.