14 Reading the record yourself, and what it cannot tell you

14.1 Why the last chapter is the one about holes

Thirteen chapters of counts make a subject look settled. It is not, and the most useful thing a course like this can do at the end is draw the edge of its own evidence in ink.

Two halves, then. First how to go and look for yourself, because everything here is a pointer at documents this school does not carry. Then every question a reader would reasonably ask that this record cannot answer.

14.2 How to search this subject without drowning

The single most useful thing to know is that searching for the word "satellite" is the wrong move, and searching for a work item short name is the right one.

Everything in this course was selected by short name — 155 of them, sitting on 183 rows of the work plan ^{[2] [3]}.

Searching by word instead pulls in a different subject: 5663 documents in this record say satellite or GNSS, carry some other work item and mean satellite navigation rather than satellite access ^[4].

The short names worth typing, by what you are looking for:

• The NR radio line, one per release in this order — FS_NR_nonterr_nw FS_NR_nonterr_nw, FS_NR_NTN_solutions FS_NR_NTN_solutions, NR_NTN_solutions NR_NTN_solutions, NR_NTN_enh NR_NTN_enh, NR_NTN_Ph3 NR_NTN_Ph3.

• The internet-of-things radio line — FS_LTE_NBIOT_eMTC_NTN FS_LTE_NBIOT_eMTC_NTN, LTE_NBIOT_eMTC_NTN LTE_NBIOT_eMTC_NTN, then IoT_NTN_enh IoT_NTN_enh, IoT_NTN_Ph3 IoT_NTN_Ph3 and IoT_NTN_Ph4 IoT_NTN_Ph4.

• The system line, everything spelled 5GSAT — FS_5GSAT FS_5GSAT and 5GSAT 5GSAT for phase 1, then 5GSAT_Ph2 5GSAT_Ph2 and the phase 3 and phase 4 families.

• The trades, which are the same phase numbers with a trade on the end: _SEC for security, _CH for charging, OAM for management and _App for the application layer.

Three endings appear again and again and are worth reading as words rather than noise. -Core is the rule-writing, -Perf is the performance requirements, and -UEConTest is the device conformance testing, which always names RAN5 RAN5 and always comes last.

14.3 The four documents that say what happens next

A work item description is the shortest honest statement of what a group has promised to do. Four of them cover the front edge of this subject:

• RP-261568 RP-261568 — the Release 20 radio study on working when the satellite navigation fix goes away.

• SP-260423 SP-260423 — the phase 4 architecture work, which everything else in phase 4 is waiting on.

• CP-261281 CP-261281 — its protocol half, in CT.

• RP-253813 RP-253813 — phase 4 for the internet-of-things radio.

Each of those is a few pages and names its own targets, its own leading group and its own dependencies. Reading four of them is a shorter job than reading one specification, and it is the fastest way to know what is actually open.

14.4 A reading order for the documents themselves

This school carries the text of none of these. In the order that makes each one easier than the last:

• R1. The requirements. TS 22.261 TS 22.261 for what the system is required to do, then TR 22.887 TR 22.887 for what phase 4 is still arguing about.

• R2. The radio. TR 38.811 TR 38.811 then TR 38.821 TR 38.821 — the two studies the whole radio line rests on — then TS 38.108 TS 38.108 for what a satellite base station has to be.

• R3. The architecture. TR 23.737 TR 23.737 first, then the phase reports in order.

• R4. The internet-of-things side. TR 36.763 TR 36.763 then TS 36.108 TS 36.108.

• R5. Security. TR 33.700-29 TR 33.700-29 then TR 33.700-30 TR 33.700-30.

Start at R1 because TR 22.887 is the shortest statement of what is still open. Its scope opens on the enhancements phase 4 is chasing and names multi-orbit satellite access first:

The sentence stops there because the list under it is not in this register. What the register carries is the opening and the first item, and no more — one small example of the much longer list that fills the rest of this chapter.

The rest of the shelf, with what each document is and how much has been written about it, is The specifications this work created, and the ones it changes.

14.5 What this record cannot tell you

Everything below is a line of the register's own list of what its author looked for and did not find. It binds this course exactly as the numbers do.

14.5.1 It records no vote and no reason

No vote, ballot or objection count of any kind. 3GPP works by consensus and no column of any store behind this course holds a tally. Nothing in this subject was ever carried by a majority here, because no majority was ever counted.

No reason for any decision. The stores hold what a document's outcome word was and never why the room settled on it. So no sentence in this course says a group chose one thing because of another, and none can.

That covers most of the interesting questions. Why the phase 3 security result landed in the older core's security document. Why speech coding appears in Release 20 and in no release before it.

Why the Release 20 radio work was split into separate items instead of one. All of those have answers; none of the answers is here.

14.5.2 It records no status and no dates

No status word for a work item. The work plan holds planned days and a completion value and no field that says running, stopped or finished — which is the whole of What the shape of the paperwork says, and where the work stands's argument.

No date on which any release froze, any study ended or any specification was published. The register holds meeting codes, version numbers, the work plan's planned days and the day a file went up on a server. No sentence of the form "Release 20 completes in …" can be supported from here.

No publication day for any version. Only the upload day of the file, which is a different day, and 1057 of these documents carry no upload day at all ^[5].

14.5.3 It names nobody

No person is named anywhere. The work plan rows and the work item descriptions do carry rapporteur names, and the rapporteur store holds a name for every one of the specifications here. None of them was copied into the register and none may reach a chapter.

If you need to know who to ask, the work item description itself is where the name is.

14.5.4 It holds no technical content

This is the largest gap and the one most likely to catch somebody out.

14.5.5 Every count is a floor

6568 documents name satellites or NTN in their title and carry no work item short name at all ^[6]. They are in none of the numbers in this course, so no sentence here may say this subject has produced exactly so many documents in 3GPP.

The counts are counts of what one machine holds, not of what 3GPP holds. Where the difference could matter the right form of words is "the local record holds …".

Three families were deliberately taken out. Satellite navigation — 25 rows under 22 short names, AGNSS and ANSS among them — because positioning is not access.

Protection of the Earth Exploration Satellite Service, four rows, because that is coexistence with somebody else's satellite rather than access through one.

And the SIM Application Toolkit, which shares three letters with satellite and nothing else, and which took nothing out because no row had picked one up.

14.5.6 The counting units are not what they look like

A short name is not a work item. 155 short names sit on 183 work plan rows ^{[2] [3]}, because a feature, its core part, its performance part and its conformance part are separate rows. Any count of "work items" has to say which of the two it means.

No count of its own for 95 of the 155 short names. Only the 60 carrying at least 100 documents have one, and a work item entry prints as a name rather than as a number — so a sentence leaning on how many documents one of the smaller names carries has nothing to cite, and does not appear.

No objective and no justification for 114 of the 155. Only the 41 carrying at least 200 documents are quoted. For the rest this course may say what the work plan calls them and not what they promised. And for two of the 155, 5GSAT_Ph4 and ANSS, there is no work item description on this machine at all.

14.5.7 The leadership field is not what it looks like either

No column says which group leads a work item. A group entry's named_in_leadership list holds every group the work item's leadership line mentions, leading and secondary together, because that line is free text.

Which group leads is in the words of that line and only there — so a claim that a group led something has to be read off the line itself, not off the list. NR_NTN_Ph3-Core's line, for instance, names RAN2 first and then RAN1, RAN3 and RAN4 behind it NR_NTN_Ph3-Core.

No parent column in the work plan. The parent recorded for a row is simply the nearest row above it at a shallower level, which is all the plan gives. Where the plan's ordering is untidy that is not the work item a person would call the parent, so no argument in this course rests on it alone.

14.5.8 Holes in the documents and the meetings

No parsed text of 14 of the specifications here, TR 22.822, TR 23.700-27, TR 23.700-28, TR 23.737, TR 24.821 and TR 28.808 among them. TR 38.811 and TR 38.821 are in that group too TR 38.811 TR 38.821, so the two studies this whole radio line rests on can be named and counted and never quoted.

Two specifications whose own titles say satellite or NTN are not here at all — TR 38.763 and TR 38.908 — because no document and no work item of this subject points at either.

No specification number could be read out of six pieces of the work plan's own list of documents these work items change, so those documents are not in the register and are not named anywhere in this course.

No entry for 482 of the 550 meetings that carry these documents, so no days, no place and no attendance for those ^[7]. Of the 68 the register does hold, 32 have no attendance figure, so nothing here says how attendance has changed.

No explanation of any short name. The work plan row carries the name and the title exactly as they are spelled and nothing that expands either. Where this course says what a name stands for, it took it from a specification's own list of abbreviations.

14.5.9 It cannot compare

Nothing in this data counted sensing, or positioning, or ambient internet of things, or any other area of 3GPP work.

So there is no way to say from here that satellites is the largest, the fastest or the most contested subject of any release. The numbers in this course describe one subject and rank nothing.

14.6 How to quote this course without misleading anybody

Four habits, and they are the four this course tries to keep itself.

• Say which count you mean. 155 short names or 183 work plan rows ^{[2] [3]}; documents filed at a group's own meetings or documents carrying a short name wherever they were filed. A bare number with no question attached is the commonest way this data gets misused.

• Say "the local record holds". These are counts of what one machine holds, not of what 3GPP holds, and where the difference could matter the words should say so.

• Say "floor", not "total". 6568 documents about this subject are outside every number here ^[6].

• Never turn a meeting code into a date, and never turn an upload day into a publication day. Neither conversion is in this data.

14.7 What survives all of that

Given the list above it is worth being plain about what this course is good for, because it is not nothing.

• What each group set out to do, in its own words, with the document it said it in — Security, the bills, the management, the layer above and the voice for the four that arrive last.

• Which specifications exist, which were changed and how heavily — The specifications this work created, and the ones it changes.

• How far each piece has got in the only measure the plan carries — What the shape of the paperwork says, and where the work stands.

• How to find the rest yourself, which is the first half of this chapter.

That is a map with hop-off points, which is what this school is for. The text of the specifications is not carried here and never will be.

Every source this course cites, with the exact question that produced it, is on the course's own sources page.