Fixed and variable.
The contrasting terms "fixed" and "variable" occur in various computing contexts. We have, for example, fixed-length records and variable-length records in file and database design; or fixed-width fonts (otherwise called monospaced fonts) and variable-width fonts (otherwise called proportionally-spaced fonts).
The terms are normally prefixed to an adjective, and I would suggest in Irish
"comh-" and "il-", which can be prefixed to adjectives such
as "fada" or "leathan", giving the following adjectives:
In the case of records, "taifid chomhfhada" and "taifid ilfhada" for sets of records which are, in the first case, all the same length, and in the second case, of unequal lengths, are terms which reveal their meanings right away, much better than the English terms do. And in the case of fonts, the same is true of "cló comhleathan" for fonts where all symbols have the same print width, and "cló il-leathan" for fonts where the symbols may have different print widths.
The official terms are now "aonfhaid", "ilfhaid", "aonleithid", "il-leithid". For example, we have "taifid aonfhaid" (FR 231, for fixed-length records); "taifid ilfhaid" (FR 341, for variable-length records); "cló aonleithid" (FR 271, for monospaced font); "cló il-leithid" (FR 297, for proportional font). This is a great improvement over the situation as it existed when this page was originally written.
At that time, TR had no term for "monospaced", while "proportional" appeared as "spásáil comhréireach", for proportional spacing, TR 155). Worse was the situation with records: we then had "taifead buanfhaid" (TR 121, for fixed-length record) and "taifead fadathraitheach" (TR 178, for variable-length record). These terms suggested an intepretation which was quite wrong, namely, that it was change with the passage of time which was in question — and this was reinforced by giving the terms in the singular, whereas in fact, you simply cannot have "a fixed-length record" or "a variable-length record", but these terms, in their correct meaning, can only be used in the plural. The worst features of these terms have now been rectified in FR.
In TR also, the misleading rendering of "fixed/variable" by "buan-/-athraitheach" extended to other official terms, including "fixed-length field" (TR 121, réimse buanfhaid) and "variable-length field" (TR 178, réimse fadathraitheach) — whereas the terms really mean that the length of the field does or does not vary from record to record; and "fixed word-length" (TR 121, giotánrafhad buan) — which really means that all the (computer) words in the machine's store are the same length (in bits) and "variable word-length" (TR 179, giotánrafhad inathraithe). None of these concepts has a temporal aspect. These terms have since been amended to "réimse aonfhaid" (FR 231, fixed-length field), "réimse ilfhaid" (FR 341, variable-length field), and "ilfhad giotánra" (FR 341, variable word-length), while "fixed word-length" has been removed from FR.
Compare Eoghan Ó Neachtain, “Tús na Céimseatan: Leabhar a 1” (1932):
• comhmhór: equal in size or area;
• comhada: equal in length.