4 Releases 6 and 7: interworking, messaging, and a stream that never stopped

4.1 What a working IMS still could not do

Rel-5 defined a system in which a call is an internet session. That leaves an obvious problem: almost nobody a subscriber wanted to call was on it.

The people at the other end were on the public telephone network, on ISDN lines, or on the circuit side of a GSM or UMTS mobile network. A multimedia session that can only reach other multimedia sessions is a demonstration, not a service.

Rel-6 is where that was fixed, and where two capabilities were added that had nothing to do with a telephone call at all. Rel-7 then opened a maintenance job that has been reopened in every release since.

The two releases are not large in the plan: 17 IMS rows are marked Rel-6 ^[5] and 13 are marked Rel-7 ^[6]. Rel-5 carried 14 ^[7].

On the plan's own measure these three releases sit at much the same size, and the releases that dwarf them all come later, in Releases 8 and 9: emergency calls, continuity and the fixed-network merge.

The rows this chapter names are these:

Acronym	Title of the row	Release
IMS-CCR-IWCS IMS-CCR-IWCS	Interworking between IMS and CS networks	Rel-6
IMS-CCR-IWIP IMS-CCR-IWIP	Interworking between IMS and IP networks	Rel-6
IMSM IMSM	IMS Messaging	Rel-6
IMSGM IMSGM	IMS Group Management	Rel-6
IMS2 IMS2	IMS Phase 2	Rel-6
IMSProtoc IMSProtoc	IMS Stage 3 IETF Protocol Alignment	Rel-7
IMSconf IMSconf	IMS Support of Conferencing	Rel-7

4.2 Reaching the networks everybody was already on

IMS-CCR-IWCS — interworking between IMS and circuit-switched networks — sits on group N3 in the plan's own code, and was opened by NP-030292 at plenary meeting NP-20 IMS-CCR-IWCS. Its objective is a single sentence, and the last four words of it are the whole ambition ^[1].

Not video. Not multimedia. Basic voice calls — the least glamorous thing IMS can carry, and the thing without which none of it is a product. Its justification says which networks had to be reached ^[2].

Four networks, named. The public telephone network, ISDN, and the circuit sides of GSM and UMTS.

The specification this row points at is TS 29.163, the rules for talking to circuit-switched networks, and it also names TS 24.228 and TS 29.061 as documents it would change TS 29.163 TS 24.228 TS 29.061.

207 documents in the store carry the acronym IMS-CCR-IWCS ^[8].

Read that number with How this subject is cut out of the work plan beside it: the store's earliest IMS upload stamp is 2015-01-23 ^[9], so the count says very little about the Rel-6 argument itself, and the register does not say what those documents are instead.

The other half of interworking got its own row. IMS-CCR-IWIP covered interworking between IMS and IP networks — the other side of the boundary — on the same group code, opened by NP-020091 IMS-CCR-IWIP. It was aimed at TS 29.162 and the Release 2000 architecture report TR 23.821 TS 29.162 TR 23.821.

4.3 One plenary paper, five rows

Rel-6 also added two things that are not calls. Both were opened by the same plenary paper, SP-020065, and both belong to SA1.

Messaging. IMSM set out to identify support of messaging services in the IMS, and its first product was to be a report rather than a specification IMSM ^[3].

Its justification says what the group thought messaging was for, and it is not "text messages over IMS" ^[4].

Group management. IMSGM set out to define and develop support of the group management capability in IMS, to make group-enabled multimedia services possible IMSGM ^[10].

Its justification is the same argument with a different subject: combine groups with presence and messaging and new services follow ^[11].

The paper that carried both is worth a paragraph of its own. The store lists five work items against SP-020065: IMSGM, IMSM, IMSM-CR, IMSM-TR and IMSM-TS SP-020065.

One plenary paper, five rows in the plan — the report, the specification and the change requests each got their own row, which is the row-versus-work-item split of How this subject is cut out of the work plan in its purest form.

The meeting it was carried at, SP-15, is one of the four in this course that cannot be dated: the meeting table holds no row for it.

4.3.1 The word both justifications reach for

Read the two justifications side by side and the same argument appears twice. Messaging is worth doing because it can be combined with presence ^[4]; group management is worth doing because it can be combined with presence and messaging ^[11].

Neither document argues that people want to send messages, or that people want groups. Both argue that the value is in the combination — that a capability added to IMS is worth more because of the capabilities already there.

A justification is a proposal written to persuade a room, and these two are worth reading as such: they promise that a shared building block will pay off, and neither describes a product anybody could buy.

What presence itself is, the register does not say. It carries no definition of the word, so this course says no more about it than these two justifications do.

4.4 Phase 2, about which nothing may be said

IMS2, "IMS Phase 2", is a Rel-6 row listing six groups IMS2. It is plainly the second act of the founding work item of Release 5: the founding, and the ring of work around it, and this course will say nothing whatever about what it promised.

The reason is in How this subject is cut out of the work plan: IMS2 is one of the 20 acronyms for which no work item description on this machine can be proved to belong to that row. Either the folder holds none, or it holds several and none of them settles it.

So there is no objective to quote and no justification to read. What the register holds is the row, the release, the groups and nothing else — and writing "phase 2 presumably extended…" would be exactly the invention this course exists to avoid.

4.5 The job Rel-7 opened and Rel-20 is still doing

IMSProtoc, "IMS Stage 3 IETF Protocol Alignment", is a CT1 row opened by CP-060101 at plenary CP-31 IMSProtoc CP-060101.

The work behind that title is ordinary and endless. SIP is the IETF's protocol, not 3GPP's; the IETF keeps writing new capabilities into it, and somebody has to keep deciding which of them the IM CN subsystem supports and write that down.

That is not a job with a finish. IMSProtoc is the first of a numbered chain that runs through phase 2 IMSProtoc2 and on to a Rel-20 continuation, IMSProtoc20-CT IMSProtoc20-CT. Phase 2 alone carries 248 documents ^[12], and Releases 8 and 9: emergency calls, continuity and the fixed-network merge takes it up.

Two cautions belong here. The register quotes 36 work items in their proposers' own words ^[13], and IMSProtoc itself is not among them, so nothing is said here about how its own description worded the job.

And the document the row names as its target is TR 24.930, a report of signalling flows rather than the protocol specification itself TR 24.930.

That second one repays a moment. TS 24.229 is the protocol; TR 24.930 is a report of signalling flows for session setup, and a report binds nobody TS 24.229 TR 24.930.

A work item whose named target is a report is not necessarily a small one. It may simply be booked against the document its own group owns, while the changes it causes land in somebody else's text through change requests.

The catalogue row of TS 24.229 lists a long line of the numbered protocol-alignment items as work items that declare it, among them IMSProtoc2, IMSProtoc5, IMSProtoc9 and IMSProtoc20-CT TS 24.229.

4.6 Conferencing, and where Rel-7 hands over

IMSconf, "IMS Support of Conferencing", is the other Rel-7 row worth naming: a CT1 item opened by CP-070120, aimed at TS 24.147 and at the SIP profile TS 24.229 IMSconf TS 24.147 TS 24.229.

It is a clean illustration of how the three stages line up. The word "conference" is defined in the stage 1 document — an IP multimedia session with two or more participants, each conference having a conference focus ^[14], as What IMS is, in the words of its own documents sets out.

TS 24.147 is the stage 3 answer to that definition, and its catalogue title says so: conferencing using the IM CN subsystem, stage 3 TS 24.147.

A definition in stage 1 and a protocol in stage 3 can sit several releases apart, and neither one dates the other. The register holds no date for any release at all, so the only honest statement about the gap is that the two rows carry different release labels.

Two rows started here and land in the next chapter. MIW-IMS, multimedia interworking between IMS and circuit-switched networks, is marked Rel-8 and carries two rows across CT3, CT4 and SA2 — the ambition of IMS-CCR-IWCS widened past basic voice MIW-IMS.

And the first of the TISPAN common-IMS maintenance rows, IMS_Comm_Mtce7, is marked Rel-7 while its siblings are Rel-8 IMS_Comm_Mtce7.

4.7 What these two releases did not settle

Reading the seven rows above together, one thing stands out: none of them is about IMS working when something goes wrong, and none of them is about IMS sitting on a different kind of network underneath.

Both of those arrive in the next release block, and they arrive at scale. Restoration, emergency calling, service continuity and the fixed-network documents are all Rel-8 and Rel-9 work.

Those two releases carry 64 rows ^[15] and 45 rows ^[16] against the 17 and 13 here.

What Rel-6 and Rel-7 did was narrower and, for a subscriber, more immediate: they made an IMS session able to reach a telephone, and they started the never-ending job of keeping the protocol in step with the IETF.

4.8 Where to read on

For the circuit-switched boundary, open TS 29.163 TS 29.163; for the IP one, TS 29.162 TS 29.162; for conferencing, TS 24.147 TS 24.147.

Then comes the largest block of work in this whole subject, where IMS was fitted to the packet core, to emergency calls and to the fixed networks: Releases 8 and 9: emergency calls, continuity and the fixed-network merge.