Basic aspects of computer processing of Esperanto are considered in relation to orthography and computer representation, phonetics, morphology, one-syllable and multisyllable words, lexicon, semantics, and syntax. There are 28 phonemes in Esperanto, each represented in orthography by a single letter. The PLATO system handles diacritics by using a control code in the internal representation of character strings. Lexical stress is always on the penultimate vowel, most function words are monosyllabic, and most content words are polysyllabic. These facts simplify the assignment of stress and intonation contours on the basis of text alone, even without any syntactic processing. Esperanto morphemes, all of which are invariant, fall into two classes, those requiring grammatical endings and those that do not require such endings. Rules for one-vowel and multivowel words are discussed. It is suggested that for many purposes computer processing of Esperanto text is much simpler than the processing of other languages, and machine translation from Esperanto to another language is aided by the grammatical endings. But it is a language used among humans, and when it comes to semantics, the computer is at a severe disadvantage. In using the PLATO system to translate Esperanto sentences to English, a basic scheme is to identify the grammatical categories of the words and to group the words into various constituents. A small lexicon containing English equivalents of individual roots, with special entries for irregular plurals and verbs, has been developed. Words must contain only one morpheme to avoid having to break down compound words. (SW)