One of those things to be considered is the set of inherent characteristics or idiosyncrasies of the language being translated. For example, Biblical Hebrew often uses exactly the same form for both past and future tenses, leaving the intended tense in doubt. Sometimes this works out marvelously, the best example being in Genesis 1.3, usually translated as "And God said, 'let there be light,' and there was light." Any modern language can make clear the difference between the two statements. In Spanish (the old version of Cipriano de Valera) it is: "Y dijo Dios: sea la luz: y fué la luz." In French (Version Synodale) we have "Dieu dit : 'Que la lumière soit !' Et la lumiêre fut." In English, Spanish, French, and most other languages, the situation is quite clear. But in Biblical Hebrew, God said "y'hi or" (let there be, or literally, there will be light) and "y'hi or" (there was light). The two statements look exactly the same, which marvelously emphasizes how God's word, God's command, is law and gets carried out exactly.
Due to the above feature of Biblical Hebrew, The two translations of the Book of Psalms that Saint Jerome presented to the Pope often differ in tense, for example the version based on the Greek of the Septuagint may use a past tense in Latin, where the version based on the Hebrew text extant in Jerome's day will use the future tense. It is important to remember that the Hebrew would be the same in either case, so one must choose according to context. There will still be doubt, but it is an educated guess on the part of the contemporary translator (or the ancient translator, for that matter).
Translation isn't always a matter of simple, one-for-one equivalents, which is why machines can't really do it properly. A good example is in Psalm 17 (18), verse 36 (35), where Jerome's Hebrew-based translation says "et mansuetudo tua multiplicavit me" (and your gentleness made me strong), while his Greek-based translation has "et disciplina tua correxit me in finem" (and your discipline corrected me to the end). How are such different translations, both ultimately deriving from the same Hebrew text, possible? The simple explanation is that, in both cases (ca. 150 BCE and ca. 400 CE) the Hebrew text was doubtful, and it continues to be doubtful today. The second-century-BCE translator scratched his head and came up with the Greek word "paideia." Now "paideia" has a constellation of meanings, centering on the idea of education/training of children, and it can also be construed as "discipline." But if one looks at the Hebrew, one is not sure what word was intended (remember, the texts were often in poor condition, and they were writing without vowels). The best possibility that I have been able to come up with is a word that means both "humility" and "patience." It's probably the same word that the translator of Jerome's Hebrew-based version of the Psalms came up with. So maybe it was correct. But one does not think of God as being "humble." So perhaps the intended meaning was "patience," in the sense of "forbearance, restraint." The translator of Jerome's Hebrew-based version must have thought this was the most likely possibility (I do, too), and translated it into Latin as "mansuetudo" (mildness, gentleness). The translator of Jerome's Greek-based version opted to translate "paideia" (which I can't get from the Hebrew) as "discipline." And that is, very probably, how we got two translations with more or less opposing meanings.
In 1945 the scholars of the Pontifical Biblical Institute, no doubt aware of the problem, used the word "sollicitudo" (solicitude) instead, which I can't get out of the Hebrew at all.
