Every question of this course

Q1.1 What are the two jobs the record of this meeting shows a plenary doing?

• Choosing which companies may join, and setting membership fees
• Drafting specification text, and running interoperability trials
• Signing off work the working groups bring, and answering letters from other standards bodies

The drafting happens in the working groups. The plenary signs their finished work off and answers the post that arrives from outside 3GPP. SP-112 draft meeting report, version 0.0.5

Q1.2 How many people registered for SA#112, and how many checked in?

• 568 registered, 365 checked in
• 568 registered, 110 checked in
• 365 registered, 110 checked in

568 registered and 365 checked in. Separately, 110 of the registrations were for attending online. 568 people registered for the meeting

Q1.3 Why does checking in at the meeting matter?

• It is what keeps a member company's voting rights alive
• It decides the order the documents are handled in
• It is how a delegate gets a document number

Registering on its own is not enough. The data ties checking in to keeping voting rights alive. 365 people checked in and so kept their voting rights

Q1.4 Where was the Release 21 timeline settled?

• In a joint session with TSG RAN
• By the Project Co-ordination Group
• In the SA plenary on its own

The draft report says SP-260595 was presented in the joint session and was agreed, so it was not SA's decision alone. 'SP-260595 was presented in the Joint session and was agreed.'

Q1.5 What does the plenary do before it handles anything else?

• Approves the previous meeting's report
• Hears the working group reports
• Approves its own agenda

The report records "The agenda was reviewed and approved." at the top of the meeting. 'The agenda was reviewed and approved.'

Q2.1 How many documents were handled at SA#112?

• 120
• 346
• 689

346 documents. 120 of them were change-request packs, and 689 is the count of individual change requests inside those packs and on their own. 346 documents at SA#112

Q2.2 Which kind of document was the most common at this meeting?

• New work item descriptions, 53 of them
• Reports, 29 of them
• Change-request packs, 120 of them

120 of the 346 documents were change-request packs, the biggest kind of document at this meeting. 120 documents of type 'CR pack'

Q2.3 Of the 62 sources that submitted documents, which five sent the most?

• Five SA working groups
• Three working groups and two other standards bodies
• Five member companies

SA WG5, SA WG3, SA WG6, SA WG2 and SA WG4 are the five busiest senders, and no single company comes close. 62 different sources submitted documents

Q2.4 This data counts the meeting's change requests two ways, and the two totals differ. Which is wrong?

• The smaller one, which misses the standalone change requests
• Neither — they count different things
• 689, which double-counts packs

689 comes from splitting the specification column of every change request and pack; the four release counts added together only look inside packs, so they come to less. Say which one you mean. 689 individual change requests

Q2.5 How many specifications did this meeting's change requests land on?

• 114
• 20
• 62

114 specifications were changed. 20 is the number of document kinds and 62 the number of senders. 114 specifications were changed

Q3.1 How many top-level agenda items did SA#112 have?

• 9
• 20
• 23

23, numbered AI 1 to AI 23, from the opening of the meeting to its close. 23 top-level agenda items

Q3.2 Which agenda item drew the most documents?

• AI 20, Rel-20 change requests, with 63
• AI 4, reporting from the working groups, with 31
• AI 6, work item and study item descriptions, with 80

AI 6 drew 80 documents. AI 20 with 63 and AI 19 with 40 are the next two. Agenda item 6: Work Item Descriptions, Study Item Descriptions, Specifications

Q3.3 How often does the draft report say "There were no contributions to this agenda item."?

• 40 times
• 3 times
• 23 times

40 times. Most sub-items of a plenary's agenda draw nothing at all at any given meeting. 40 times the report says 'There were no contributions to this agenda item.'

Q3.4 How many documents did release planning, AI 5, draw?

• 31
• 63
• 3

Three. The item that sounds like the heart of the meeting was one of its quietest. Agenda item 5: Release Planning

Q3.5 What do agenda items 10 to 16 cover, and how busy were they?

• Change requests to Rel-10 through Rel-16, with one document between them
• Liaison statements, with 22 documents
• Project management, with 13 documents

Each old release keeps its own numbered slot. At this meeting the seven slots drew one document in total. Agenda item 15: Rel-15 CRs

Q4.1 What does the status "noted" mean?

• Accepted with minor comments
• Approved but not yet published
• Read and filed, with no decision taken

Noted is the weakest outcome in the list. 68 documents ended the meeting that way, and none of them changed anything. 68 documents ended the meeting 'noted'

Q4.2 A document ends the meeting "revised". What happened to it?

• It was superseded by a new document number
• It was sent back to the working group
• It was approved after editing

Revised means a different document number replaced it. Whether that replacement was approved is a separate question with a separate answer. 60 documents ended the meeting 'revised'

Q4.3 SP-260417 was revised and SP-260696 was approved. What are they?

• Two different work items, one of which failed
• The same work item, first and second attempt
• A work item and the change request that implements it

Both are the new work item description on the ultra low bitrate speech codec. Only the second number was approved. SP-260696

Q4.4 Which status did the chair's guidance on the specification-modernisation study receive?

• noted
• approved
• endorsed

Endorsed — supported, but it is not specification text. Only four documents at the whole meeting ended that way. SP-260679

Q4.5 What does "approved" at a plenary tell you about publication?

• Nothing — approval and publication are separate, and this data covers only the first
• Publication follows automatically within a week
• The specification is on the server the same day

An approved specification still has to be produced and put on the server, and nothing in this meeting's data says when that happened. 26 specifications and reports were approved

Q5.1 How many new work items and study items were approved at SA#112?

• 11
• 26
• 30

30 new ones were approved and 11 existing ones were updated. 26 is the count of approved specifications and reports. 30 new work items and study items were approved

Q5.2 What happened to the ultra low bitrate speech codec work item?

• It was postponed to the next meeting
• The first version was revised and the revision was approved
• It was approved and then withdrawn

SP-260417 was revised, SP-260696 was approved, and the report says "This new WID was approved." 'This new WID was approved.'

Q5.3 What did the plenary do about the legal question on the AI training data?

• Settled it in the room and recorded the answer
• Pushed it up: the TSG SA Chair is to inform the Project Co-ordination Group
• Refused the work item until the question is answered

The record says the chair is to inform the PCG on 3GPP involvement and the legal requirements around the dataset. The work item was approved all the same. 'TSG SA Chair to inform PCG on this issue of 3GPP involvement and legal requirements related to the dataset needed to train the AI model specific to 3GPP work.'

Q5.4 How did agenda item 9, on modernising how 3GPP writes specifications, end?

• With both study papers noted and the chair's guidance endorsed
• With the item postponed to the next plenary
• With a decision to move to markdown and git

SP-260406 and SP-260407 were both noted; SP-260679 was endorsed. The guidance continues a trial and encourages demonstrations. SP-260679

Q5.5 Which company's objection does the draft report name?

• Nokia's, to the emergency-communications reply
• Qualcomm's, to the decision on 6G interface names
• Ericsson's, to the Forge demonstration

The report writes "Qualcomm objected to this decision." next to the task handed to SA WG2, and records nothing about anybody else's position. 'Qualcomm objected to this decision.'

Q5.6 How many of the 18 incoming liaison statements were replied to?

• One
• Eighteen
• Two

Almost all incoming letters were simply noted. Exactly one ended "replied to" — the letter from ETSI TC EMTEL. 1 document ended the meeting 'replied to'

Q6.1 Which release carried the most documents at SA#112?

• Release 21, with 11
• Release 19, with 26
• Release 20, with 140

140 documents carried Release 20, and 487 of the change requests inside packs belong to it. It is the release being written. 140 documents belong to Rel-20

Q6.2 How far had Release 21 got by this meeting?

• Its first specifications were approved
• It had a timeline and 11 documents
• It had not been mentioned

11 documents carry Rel-21, and the timeline itself was agreed in the joint session with TSG RAN rather than by SA alone. 11 documents belong to Rel-21

Q6.3 How many of the meeting's documents carry no release at all?

• 11
• 26
• 162

162 — agendas, reports, liaison statements and project management. A plenary is not only about releases. 162 documents carry no release

Q6.4 Which specification took the most change requests at this meeting?

• TS 28.552, 5G performance measurements, with 29
• TS 23.501, the 5G system architecture, with 29
• TS 23.482, the AI/ML enablement service architecture, with 46

TS 23.482 took 46, ahead of TS 23.502 with 38 and TS 23.369 with 33. TS 23.482

Q6.5 What did the plenary note about changes to frozen releases?

• That they must be approved twice
• That they are forbidden once a release is closed
• That care should be taken to agree or endorse them at working group level, given the existing commercial deployments

That is the principle the record carries, and 36 change requests still landed on Release 18 and 7 on Release 17. 'Care should be taken to agree or endorse CRs against frozen releases in the working group level, especially given the existing commercial deployments.'

Q7.1 Which file is the fastest way to see what a meeting did?

• The chair's welcome pack
• The agenda
• The document list, TdocsByAgenda

It lists every document under its agenda item with the type, the source and the result, so the outcome of the whole meeting is on one page. TdocsByAgenda, the SA#112 document list

Q7.2 How many of the 346 documents get a paragraph of their own in the draft report?

• 199
• 346
• 284

284. For the other 62 the only evidence of what happened is a row in an annex table. 284 of the 346 documents get a paragraph of their own in the report

Q7.3 Under agenda item 1.5 the draft report says the SA#111 report was approved, while the status line under the same document says "Not concluded." What should a reader do?

• Take the narrative, which is the fuller statement
• Take the status line, which is the machine-readable one
• Report both, or neither

The draft contradicts itself there. Picking the version that reads better is inventing a decision the record does not support. 'The report of meeting SA#111 was approved.'

Q7.4 What is the machine-written summary of this meeting good for?

• Nothing at all
• Finding your way around, cited as a summary
• Settling what the meeting decided

It was written by a language model, not by 3GPP. Useful as a pointer, never as the source of a claim about a decision. The machine-written summary of SA#112 on whatthespec.net

Q7.5 Where are the chair's final notes for this meeting?

• In the welcome pack
• In the draft report, as an annex
• Published after the meeting closes, and not in this data

The agenda says they are published after the meeting closes. Nothing in this data can say what they contain. SP-260361, the SA#112 agenda, version 2