# Compressing Access Time Hides Friction Rather Than Removing It ## Claim Making information faster to reach does not reduce the uncertainty in it — it removes the delay that used to make the uncertainty noticeable, so the same verification is now owed and no longer prompted. <!-- atomicity: 1 idea ✓ --> ## Evidence **Source**: [[3. Resources/Readwise/Articles/The Ultimate Guide to Claude Connectors|The Ultimate Guide to Claude Connectors]], closing on Clausewitz > [!quote] > Compress the time and the friction does not disappear. You just stop noticing it. The mechanism is that friction was doing two jobs and only one of them was the cost. Waiting for a query to run, opening the source, scrolling to the passage — that delay was overhead, and it was also the interval in which a person registered *how much* they had actually looked at. Removing it removes the overhead and the signal together. The practical inversion is uncomfortable: a fast, fluent, confident answer requires **more** scrutiny than a slow one, not less, because the cues that used to trigger scrutiny were made of the delay. Nothing about the answer announces this. It reads exactly like an answer that was thoroughly derived. ## Scope & conditions - **Holds** wherever speed increases without a corresponding increase in coverage or verification — retrieval, connectors, search, summarisation, generated code. - **Sharpens** when the output is fluent. Fluency and thoroughness were correlated when producing text was slow; the correlation is what broke, and habits built on it have not. - **Breaks** where the compression came with a real verification mechanism — a test that runs, a citation that resolves, a coverage figure that is reported. Then the friction was genuinely removed rather than hidden. - **The fix is a disclosure, not a slowdown.** Requiring a tool to state what it searched and what it could not reach restores the signal without restoring the wait. - **Does not follow** that faster access is bad. It follows that the verification budget must now be spent deliberately, because nothing in the workflow will spend it for you. ## Links - [[3. Resources/MOCs/AI & the Future of Knowledge Work MOC]] — why speed gains in retrieval do not convert one-for-one into knowledge gains - [[3. Resources/MOCs/Personal Knowledge Management MOC]] — the case for keeping a deliberate verification step once the accidental one is gone - [[3. Resources/Permanent Notes/Sampling a Fraction of a Source Produces False Authority Over the Whole]] — because this is the mechanism behind that note's warning: the connector reads a sliver, and the speed is what stops you noticing that it did. - [[3. Resources/Permanent Notes/Friction Is the Feature - PKM Tools That Make Synthesis Easier Make Thinking Easier to Skip]] — because extends: the same argument in the note-taking domain, where the friction being removed was the thinking rather than the checking. - [[3. Resources/Permanent Notes/AI Erodes Team Culture by Removing the Friction That Forced Peer Collaboration]] — because extends: a third thing friction was silently carrying, this time the social contact rather than the epistemic check. - [[3. Resources/Permanent Notes/Automation Removes the Social Checks a Process Silently Depended On]] — because these are the same claim at different altitudes: automation removes a delay, and the delay was load-bearing for something nobody had written down. - [[3. Resources/Permanent Notes/Cheap Inspection Collapses the Signal-Substance Gap]] — because contrasts: that note has cheap inspection restoring truth to a signal; here cheap access removes the cue that inspection was needed at all. - [[3. Resources/Permanent Notes/The Most Reliable Evidence a Skill Has Been Automated Comes From Whoever Taught It]] — because contrasts: applied to a "the whole stack collapsed into one place" claim from an affiliate video, this note's skepticism is the right default — except for the one line in the same video that isn't a claim about the product at all, which this note's mechanism doesn't touch. ## Tests / falsifiers - Compare verification behaviour on the same question asked through a connector versus researched manually. If the checking rate is equal, the delay was not what prompted it. - Add a coverage disclosure to a fast tool and re-measure. If scrutiny does not rise, the missing signal was not the mechanism. ## Maintenance - Last tended: 2026-08-28 - Next review: 2026-11-28 ## Source [[3. Resources/Readwise/Articles/The Ultimate Guide to Claude Connectors|The Ultimate Guide to Claude Connectors]]