MW-003 · room 3 of 12 · 1990s

An Honest Progress Bar

A bar that reports exactly what it knows, including that it does not know.

Compare the estimate with the truth underneath it. The gap between them is the whole history of progress bars.

Runs for about 26 seconds in real time. It cannot be skipped, because a wait you can skip is not a wait.

Museum label

The progress bar promises a proportion: this much done, that much left. The promise is often impossible to keep, because the remaining work cannot be measured until it is attempted.

So bars learned to lie politely. They advanced quickly through the parts that were easy to count, then slowed near the end, where the uncertain work lived. The infamous pause at ninety-nine per cent is usually not a bug. It is the moment the estimate ran out.

This exhibit shows a bar that refuses to pretend. When it does not know, it says so, and keeps moving anyway.

Research dossier · MW-003-R

The wait behind the reconstruction.

Representative period
1970s–present
Mechanism
Estimated computational latency
Source records
4
01

Inside the machine

What was actually happening?

A percent-done indicator needs two numbers: work completed and total work expected. The first may be countable; the second often does not exist until the program opens files, discovers dependencies or asks a remote service what remains.

A bar therefore combines measurement with prediction. It may weight phases, smooth sudden changes, reserve the last few percent for cleanup, or abandon percentage entirely and show indeterminate motion.

The modern HTML standard preserves the distinction. A progress element with a value is determinate. Remove that value and the same element means only that work is happening and its duration is unknown.

02

Outside the machine

The human loop.

The bar turns machine time into a social promise. Once a finish time is displayed, a correction feels like betrayal even when the new estimate is more accurate.

An unmoving interface and an unmoving process look identical from outside. Progress feedback does not make work faster; it prevents uncertainty from consuming the wait.

Catalogue measurements

Three numbers to carry out of the room.

48participants
The size of Myers' 1985 controlled comparison
86.1%preferred feedback
Participants preferring the percent-done condition
p = .0006reported result
Statistical significance reported in the original paper

Contemporary descendant

The wait did not disappear.

Today the bar appears in uploads, operating-system updates, deliveries and bureaucratic applications. It is no longer merely a control; it is a compact theatre in which an institution demonstrates that it has not forgotten you.

Open bibliography

Research notes.

The interpretation above is original. These institutional records support its technical claims and date ranges.

  1. 01
    The importance of percent-done progress indicators for computer-human interfaces

    ACM · 1985 · paper

    Brad A. Myers' 48-participant experiment: 86.1% preferred the percent-done condition, a result reported with p = .0006. The paper studied rather than invented the progress bar.
  2. 02
    The Importance of Percent-Done Progress Indicators (author copy)

    Carnegie Mellon University · 1985 · paper

    A publicly available author copy of the CHI paper, including experimental procedure, participant counts and the reported preference result.
  3. 03
    The progress element

    WHATWG HTML Standard · living standard · standard

    The modern web platform's distinction between determinate progress and a task whose duration is unknown.
  4. 04
    Response Times: The Three Important Limits

    Nielsen Norman Group · 1993–2024 · paper

    A durable summary of the rough thresholds at which interaction feels immediate, continuous or attention-breaking.
Reconstruction notice

No commercial interface, logo, screenshot or recorded device audio is reproduced in this room. The apparatus compresses historical time to remain visitable; its mechanism and uncertainty are the objects being preserved.

Open the complete reference room →

Cross-reference desk

This room continues elsewhere.

The apparatus ends. Its mechanism passes into essays, archive fragments and other rooms where the same kind of delay changes scale.