Streaming UX: Don't Leave Users Waiting
Tools run 30 seconds in the background — what should users see? Designing progress perception
THE QUESTION THIS PAGE ANSWERS
ANSWER FIRSTWhat is the key idea behind “Streaming UX: Don't Leave Users Waiting”?
Tools run 30 seconds in the background — what should users see? Designing progress perception
Make the claim earn its place. Use this page as a decision aid, not a definition to memorize. Connect the idea to one real task, one observable result, and one failure that would change your mind.
Write one question you could answer with evidence after trying this idea.
A conclusion that sounds complete but leaves the key assumption untested.
| Status Copy | "Searching for relevant information..." → "Found 5 results, now analyzing..." |
| Tool Display | Show the user the name and parameters of the tool being called, so they know what the AI is doing |
| Streaming Output | Display the model's output token by token, turning the wait into reading |
| Phase Markers | "Step 1 of 3" — even if timing can't be accurately estimated, phases are better than nothing |
| Intermediate Results | Give a rough result first, then refine it: e.g., provide an outline first, then fill in the details |
How “Comparison: With vs Without a Sense of Progress” changes an answer
“Tools run 30 seconds in the background — what should users see?” shows that a model does not process the “word count” we see. It processes Token pieces. Tokenization affects input length, how much context fits, and how much computation a request consumes.
Length, information, and context are different
As “Tools run 30 seconds in the background — what should users see?” grows, separate three questions: how many Tokens the text becomes, which pieces can change the current decision, and whether older material has fallen outside the context window. Removing repetition is often more useful than simply making the window larger.
Keep what can change the decision
Use “Tools run 30 seconds in the background — what should users see?” as an A/B test: keep the same question while removing repeated background, compressing format, and trimming irrelevant history. Compare answer quality, latency, and Token count.
From “Comparison: With vs Without a Sense of Progress” to “Three Principles of Progress-Sense Design”
“Comparison: With vs Without a Sense of Progress” grounds the problem in “✗ No sense of progress With a sense of progress Play comparison animation (simulating a 30-second wait)”. “Three Principles of Progress-Sense Design” then moves it toward “Let Users See the Process Don't just show a spinner — show what's happening: searching, analyzing, organizing Make Progress Perceivable "Found 3 results," "Analyzed 2/5 files" — numbers, changes, and a sense of…”. Together, they show that the lesson is not just a conclusion to remember, but a claim with conditions.
Carry the judgment into the next situation
For long text, keep what can change the conclusion before compressing format and history. A larger context is worth its cost only when the added information is useful.
- “Comparison: With vs Without a Sense of Progress”: ✗ No sense of progress With a sense of progress Play comparison animation (simulating a 30-second wait)
- “Three Principles of Progress-Sense Design”: Let Users See the Process Don't just show a spinner — show what's happening: searching, analyzing, organizing Make Progress Perceivable "Found 3 results," "Analyzed 2/5 files" — numbers, changes, and a sense of…
- “Common Progress-Sense Techniques”: Status Copy "Searching for relevant information..." → "Found 5 results, now analyzing..." Tool Display Show the user the name and parameters of the tool being called, so they know what the AI is doing Streaming…
The final “Common Progress-Sense Techniques” brings the discussion to “Status Copy "Searching for relevant information..." → "Found 5 results, now analyzing..." Tool Display Show the user the name and parameters of the tool being called, so they know what the AI is doing Streaming…”. The useful thing to carry forward is knowing which judgments must be revisited when input, scale, or risk changes.
I turned one judgment from this article into a small experiment I could run today. Knowing what to observe next is more useful than simply remembering the conclusion.
After reading this, I first looked for the conditions behind the idea instead of copying the method into a project. That order made the later trade-offs much clearer.
When this judgment reaches real work, which constraint should be added first? I am curious which step matters most between reading and the first practical attempt.
No discussion on this article yet.