One row of this record, assembled so it can be taken somewhere else:
what the publisher published, where this record read it, what it hashed, and
what changed between two readings it retained. Generated 2026-09-18.
It carries no assessment of what the document requires and no reading of
what any change means.
The packet, in three formats
Each is the same object.
/packet/us-21-cfr-part-211.json
carries every field;
/packet/us-21-cfr-part-211.md
is the same content as plain text; and
/packet/us-21-cfr-part-211.zip
is an archive of those two together with this row’s slice of the
published leaf file and the published root. Every byte in the archive is
served at its own address as well; it originates nothing.
Source
Identity as this record holds it for 21 CFR Part 211
Field
Value
Identifier
us-21-cfr-part-211
Title
21 CFR Part 211 — Current Good Manufacturing Practice for Finished Pharmaceuticals
The status this record holds for this document, with its dates
Field
Value
Status
in-force
Stated as of
—
In effect from
—
Last changed
—
Status read from
—
Status withdrawn
—
Decay class
live
These are the dates this record holds for the document’s own status.
They state nothing about which products or activities the document reaches;
that is a judgment, and this record does not make one.
Verification
The act behind the status above, where one is recorded
Field
Value
State
● verified
What the state states
checked against the primary source by a person
Checked by
human
Check date
—
Last verified
Terms
What this record does with the text of this document
Field
Value
Licence class
open
Text retained here
yes
A quotation appears in this record
yes
Snapshot retained here
yes
Readings with retained text
Each distinct digest of the normalised text this record
retained, addressed at the reading that first carried it
Digests on this row: 10. With retained text:
10. Without:
0. Series state:
comparable.
A digest is over the NORMALISED text of a reading, not over the bytes
served: scripts, comments, navigation and page chrome are removed and tags
are stripped before hashing. Hashing a source as served will not reproduce
these digests. The full digest for every reading is in the packet; the
column above is shortened for width.
Changes
One entry per consecutive pair of retained readings, in the order this record first saw each digest. Every entry carries a measured difference or the first condition that refused one. A refused entry was never tested against the conditions after it.
9 transition(s) between the
10 reading(s) above:
8 carry a measured difference,
1 are withheld above the word ceiling,
0 are refused by a rule,
0 differ in the digest and not in the words, and
0 could not be measured.
A faithful difference is longer than the word ceiling this site publishes under. It is withheld whole rather than cut to fit: a shortened quote is a fabricated measurement of what a publisher wrote. The figures below are counts of the difference, not the difference.
A difference here is a difference between two readings this record retained.
It is not a statement about what the document requires, about what changed
in substance, or about whether anything follows from it.
The root is over every leaf line the record holds, not over the lines below. Reducing it needs the published file, which is served whole at the address named here together with the command that reduces it. These lines are this row's own slice of that file, copied verbatim.
A count of a difference is in this packet; the words are not. They are a publisher's text, and this record checks borrowed text only on the surfaces it enumerates — a packet carrying spans under a key those checks do not read would be borrowed text nothing verifies. Each change names the comparison entry for the same pair, and says which of the two is at that address: the words, where a difference was measured and is inside the word ceiling; otherwise the same condition this entry names, with no text of the document.
this records own prose
The row's summary and its account of an unresolved address are this archive's own voice. They are leads and not evidence, however accurate, because a faithful paraphrase is indistinguishable from evidence while remaining ours. They are published on the row's page and are not here.
any field that would characterise a change
No score, no rank, no ordering by judgment, no reading of what a change means and no word for how much of it counts. A packet is evidence taken elsewhere, and a field characterising it would be an opinion with digests attached.
a recomputed root
The integrity block states the root this record published and the leaf lines for this row. It computes no root of its own: two numbers this record produced, compared with each other, check nothing.
The conditions, in the order they are evaluated
A change names the FIRST condition that fired and was never tested against
the ones after it, so a condition naming no change here is not a condition
this row passed. Not reached is not none:
Three of the conditions belong to the form of the comparison that resolves two dates. Every comparison in this packet is formed from readings addressed by the key the store gives them, so those three are never evaluated here: their absence from the entries above records that they were not tested, rather than that they were tested and did not fire.
Each condition, how many of this row’s
9 transition(s) it refused, and what it
states
Condition
Changes
What it states
licence-restricted
none
No text from this document appears on this site, so no difference between two readings of it is shown either. The digests, the dates and the addresses are the record for this row.
licence-unread
none
No reading of the terms this source states is recorded for the row. That is an absent reading rather than an absent term, and the row is handled here exactly as a restricted one until a reading is recorded.
date-not-observed
not reached
No observation of this row on that day holds retained text.
date-not-unique
not reached
That day holds more than one distinct retained digest for this row, so a date alone does not name one observation. The addressed form — the day and the sequence within it — does.
one-point
not reached
Both dates resolve to the same observation.
referent-unclassified
none
The class of one of these observations could not be read, so neither was shown to be an observation of the document itself.
referent-class
none
One of these observations is not of the document. A listing page that names the document and the document are different subjects, and a difference between them is not a difference in the document.
referent-normalisation
none
These two digests were taken over different kinds of text — one over markup, one over prose extracted from a file — so the words they yield differ for a reason that is not the publisher’s.
referent-address
none
These two observations answered at different addresses. A difference between them is at least in part a difference between two pages.
text-not-retained
none
No normalised text is held here for one of these observations, so there is nothing on this side to compare.
snapshot-unreadable
none
Normalised text is recorded for one of these observations and could not be read from this record.
span-unverified
none
A span this comparison would have shown could not be found, word for word and in that order, in the retained text it was drawn from. The comparison is withheld rather than shown: a phrase assembled out of words a publisher did not write next to each other is a fabricated measurement of what the publisher wrote.
over-ceiling
1
A faithful difference is longer than the word ceiling this site publishes under. It is withheld whole rather than cut to fit: a shortened quote is a fabricated measurement of what a publisher wrote. The figures below are counts of the difference, not the difference.
Word ceiling: 14. Where a faithful difference is longer
than that, the verdict is withheld and no words are published
for it: a shortened span is a fabricated measurement of what a publisher
wrote, so it is withheld whole rather than cut to fit.
How this record is built.