From a paper sheet to a system the IT department owns
The problem
Every classroom had a paper sign-out sheet. A student wrote their name and the time when they left, and wrote the time again when they came back. Once a week an adult walked round, collected the sheets, and typed them into a spreadsheet.
Two things were wrong with that, and only one of them is obvious. The obvious one is that it depended entirely on students filling it in honestly, and on filling it in at all. The less obvious one is that even when it worked perfectly, the information arrived a week late and cost an adult an afternoon. By the time anyone could see a pattern, the week that produced it was over.
What I built, and what it was for
I built a working demonstrator. Apps Script and HTML, with a student view, a teacher view and an admin view: a student requested a pass, the teacher approved it on their own device, the student returned, the teacher confirmed, and the whole round trip was timestamped.
It was never run with students. It existed to answer one question, which was whether this was possible with the tools the school already had. I tested it myself, showed it working, and that was its entire job.
That distinction matters, because the thing I learned from it was not that it worked. It was that one part of it should not exist.
The step that had to go
The teacher approval was digital. A student requested, and the teacher had to open their email to approve it.
That is a fine idea on a whiteboard and a bad one in a classroom. It asks a teacher who is mid-sentence to go into their inbox, find a message, and click a button, in order to formalise a decision they have already made out loud. The approval was not adding a decision. It was adding an interruption to a decision that had already happened.
So the production design cuts it. The teacher says yes the way they always have. The student scans their ID at the tablet on the teacher's desk, chooses where they are going, and leaves. The record is a by-product of the student's own action, not an extra job for the adult.
The prototype's job was to end the argument about whether it was possible. Once it had done that, the right move was to stop building.
Designing the real one
The production system was designed with the Assistant Principal and the IT department in the room, because they are the people who will run it and maintain it. Four decisions did most of the work.
The tablet stays on the teacher's desk. The student walks out empty-handed. That single constraint rules out every design where a student carries a pass on their phone, and it rules them out for good reasons: a phone pass can be shown by the wrong person, screenshotted, lent, or lost, and it puts a device in the corridor that nobody wanted there. The verification problem gets solved at the door instead, not in the hall.
The teacher's yes stays verbal. The system records that it happened rather than trying to be the place it happens. This is the change that came out of the demonstrator, and it is the difference between a tool teachers tolerate and one they forget is there.
Bathroom, water and locker are short by nature. The health office and the front office are not. Giving those destinations a longer alert threshold means a student who genuinely spent twenty minutes with the nurse does not generate a follow-up, and the alerts that do fire are worth reading.
Admin get a live dashboard and searchable history. IT get a synced spreadsheet of the underlying records, so they can build their own views without asking anyone for an export. Teachers deliberately get no dashboard at all: this is a tool for spotting patterns across a school, not for watching colleagues' classrooms.
See it from both sides
The classroom tablet and the office dashboard, as specified. Switch between them.
Hall Pass
Room 207 · Period 3 · onlineWhere are you going?
The teacher has already said yes. This records it.
Office visits use a longer alert threshold, so a genuine long visit does not trigger a follow-up.
Walk out empty-handed and scan again on the way back in. Nothing goes into the hallway, which is the whole point: no phone to show, nothing to lose, nothing to lend to someone else.
Hall Pass
Room 207 · Period 3 · onlineWelcome back. An open pass was found for this ID.
The teacher never has to remember who left or when. That was the failure point of the paper sheet, and it is the one job a computer does better than a person.
Hall Pass, admin
Wed 11:23 · Admin| Student | Destination | Class | Left | Elapsed |
|---|---|---|---|---|
| Student AGrade 10 | Bathroom | P3 Biology, Rm 207 | 11:14 | 9:12 |
| Student BGrade 11 | Water | P3 Chemistry, Rm 211 | 11:19 | 4:11 |
| Student CGrade 9 | Locker | P3 Physics, Rm 215 | 11:06 | 17:24 · follow up |
| Student DGrade 10 | Bathroom | P3 World Lit, Rm 122 | 11:21 | 2:08 |
| Student EGrade 12 | Other: water bottle | P3 Calculus, Rm 208 | 11:22 | 1:15 |
5 of 5 active passes · refreshes every 15 seconds
- Bathroom28
- Water9
- Locker5
- Other5
Period 3 is the peak, every week. That is the kind of thing a paper sheet in a drawer can never tell you, and it is a timetable question, not a discipline one.
Hall Pass, history
Same filters as the live view| Student | Destination | Class | Date | Duration |
|---|---|---|---|---|
| Student AGrade 10 | Bathroom | P3 Biology | Wed 11:14 | 6:02 |
| Student FGrade 10 | Health Office | P2 Geography | Wed 09:41 | 24:17 |
| Student BGrade 11 | Water | P1 Chemistry | Wed 08:55 | 3:24 |
| Student GGrade 9 | Front Office | P5 Art | Tue 13:30 | 11:48 |
| Student CGrade 9 | Locker | P3 Physics | Tue 11:06 | 4:55 |
Same question, two timeframes. "Who is out of P3 Biology right now" and "how many passes did P3 Biology open this semester" are answered by the same set of filters.
Where it is now
The school deployed RFID student ID cards this year. That changed the cheapest part of the design, because the scan the whole flow depends on now works with a card every student already carries.
IT is building version two on that infrastructure, from this design. It is their system now: their backend, their integration with the student information system, their security review, their maintenance. I am still in the conversation with the Assistant Principal, but I am not writing it, and it is better that way. A hall pass system that only one teacher can maintain is a hall pass system with a single point of failure who also has classes to teach.
Where else this fits
The shape of this problem is common and the hall pass is just the version with the clearest paper trail. Somewhere in most schools there is a record being made by hand, collected on a cycle, typed up by someone whose time is worth more than that, and read too late to act on.
What made this one work was not the software. It was deciding that the record should be a by-product of something a person was already doing, rather than an extra task laid on top of it. Get that right and compliance stops being a behaviour problem. Get it wrong and you have built a more expensive clipboard.