The next share count is a new measurement | Rapid course Information security

Open, free learning for anyone who wants to protect people and technology. No account is required to read this text or its worked answers.

Lesson map

Lesson 07 published. Map, example, EX061.

flowchart TD
  O["Published run: 47 of 300, seed 2"] --> K["Twelve client and finance files were closed afterward"]
  K --> N{"Has a new export been run?"}
  N -->|No| H["Do not print 47 minus 12 as this month's result"]
  N -->|Yes| R["Record that run's own counts"]
  H --> S["Leave the published output unchanged"]

Subtraction is not a rerun.

Logical drawing

verified script, seed 2
    all files: 47 of 300
    client 8 of 140, finance 4 of 60  -> the twelve that were closed
    monthly routine: rerun the check
        |
        +-- 47 - 12 = 35 only if the same 300 files are unchanged and nothing new was shared
        +-- that assumption is not a measurement
        +-- the script is not edited to force 35

Learning outcomes

After this chapter, you should be able to:

Keep the published seed-2 output as the only verified count until a new export is run.
Reject 35 of 300 as this month's result when nobody has rerun the check.
Leave the program and its printed lines unchanged.

7.1 Closing files is not the same act as counting them

Sofia closed the eight client files and the four finance files. That action is in the change log. The script's printed total is from the export taken before those closes. The monthly routine says to rerun the check. A classmate subtracts twelve from forty-seven and wants the dashboard to say 35 of 300. That figure would be true only if no file was added, no closed file was shared again, and the other open files stayed as they were. None of those conditions has been measured.

Until the technician produces a new export and the same kind of count is run, the verified line remains all files: 47 of 300. Staff do not type 35 into the log as if the program had printed it.

7.2 Worked example

Lesson table; scroll horizontally to read all columns →

Claim	What was actually done	Decision
"Twelve files were set to clinic-only."	The change log says so	Keep that as a change, not as a new census
"So the open total is now 35."	No new export	Refuse the number
"Edit the script so the next print shows 35."	The script's job is to count the export	Refuse the edit
"Marketing can be ignored because those files may stay open."	Chapter 2 already separated that decision	A new run still has to count them if they are open

Synthetic case. The counts 47, 300, 8, 4, 25 and 10 stay the published run. This lesson does not invent a second output.

Errors and limits

A later run might show 35, or 40, or 12 again if a link was reopened. Any of those is a result only after it is counted. This chapter does not perform that count, and it does not change random.seed(2).

Apply

EX061 — Do not subtract a census. Someone writes "open files this month: 35 of 300" because twelve were closed. What is wrong with the number, and what remains the verified total?
Show marking criteria
Rubric: 3 = 35 was not counted; the verified total is still 47 of 300 until a new export is run; do not edit the script;; 2 = reject 35 and keep 47 of 300;; 1 = one of those;; 0 = none.