April birthdays by day: notable people organized by date

A day by day index looks like a list. It is really a stack of decisions: who counts, which person is which, what the date is based on, and what happens when two sources disagree. Read the decisions and you can tell within a minute whether the list is worth anything.

What to take away

• The hardest part of a calendar index is not finding dates. It is deciding that two records describe the same person.
• An entry without a scope rule behind it will drift, because there is always one more plausible name to add.
• Maintenance is the whole job. A birthday index is never finished, only current.

The index is a set of decisions, not a list

Before a single name goes in, a compiler has to answer three questions and write the answers down. Who is in scope, so that the list has an edge. What counts as a source for a date. What to do when the sources conflict. Skip those and the list still gets built, but it becomes unmaintainable within a year, because nobody can say why any given row is there.

Scope is the one most often skipped. "Notable people" is not a rule, it is a mood. A working scope names a field, a level of documentation and a reason, so a later editor can apply it without guessing at the original intent.

Deciding who is the same person

Two records that share a name may be two people. Two records with different names may be one person who changed it, uses a stage name, or is transliterated differently in two languages. Sorting that out is called authority control, and libraries have been doing it for far longer than the web has existed.

The idea is simple. Each person gets one controlled record, every variant name points at it, and identifiers rather than spellings do the matching. Projects like the international authority file exist to link the identifiers different national libraries assigned to the same person. For an index, the value is that you can say what you matched on rather than trusting a string.

The failure mode is silent merging. Two people with similar names collapse into one entry, and the merged row acquires a date from one and a career from the other. Nothing in the finished page shows that it happened. Marking a match as uncertain is uncomfortable and much better than a confident row that is two people in a coat.

What each entry has to carry

Field	Purpose	Failure it prevents
Controlled name and variants	One identity, many spellings	Two entries for one person, or one entry for two
Match basis	Says what the identification rests on	A guess that reads as a finding
Date source	Names the record, not a website that quoted it	Circular sourcing between copies
Date confidence	Marks estimated, disputed or firm	An uncertain value hardening into a fact
Checked on	When a human last looked	Endless quiet aging of the page
Inclusion reason	Why this person meets the scope rule	Scope creep that nobody can reverse

Keeping it correct after publication

• Recheck the disputed rows on a schedule, not the whole list. Rows marked firm and sourced rarely change; rows marked uncertain are where movement happens.
• Keep a correction log with the old value in it. A page that silently swaps a date teaches nobody anything, including its own editors.
• Treat a reader report as a lead. It points at a row worth reopening; it is not itself evidence.
• Never fill a gap to make a day look complete. An empty date is information. An invented entry is damage.

Compare the layouts across months to see how much of this is visible from outside. The February index has to explain a date that most years do not contain, while the March index has none of that and shows the plain structure. The month pillar for April birthdays carries the scope statement the daily rows inherit. A name list such as April's better known figures is a different product with different obligations, and a role based list like April actors narrows scope by profession, which makes its edge easier to define and its identity matching harder, since stage names multiply.

Common questions

Why not just pull dates from a large database and be done?

Because databases inherit each other. Several of the big ones were seeded from the same earlier sources, so agreement between them is weaker evidence than it looks. Naming the underlying record is what distinguishes a checked entry from a copied one.

What makes an entry worth removing?

An entry that cannot be traced to a source, or whose identity match cannot be explained. Removing it is cheaper than defending it, and the removal should be logged with a reason.

How complete can a day by day index be?

Not complete, and it should say so. It covers what its scope rule admits and what its sources support. A stated boundary is a feature; an implied claim of completeness is a promise the page cannot keep.