To speak with people, you might look at the learning route advocated by Dr. Pimsleur, who also started a company which publishes language learning in audiobooks only, with the argument that most languages started out without a written form, and when they did develop writing it was inaccessible to a majority of speakers for most of history. Literacy is a recent phenomenon.
But learning a language is a big deal, and I don't think that either one is really worth it for anyone who isn't a student. Though if I lived in Greece, that's a different matter.
For almost all of history most languages were unwritten. While it's true that today, owing to the efforts of indigenous leaders[1], missionaries[2], and linguists[3], many languages have gained writing systems, they often are of marginal, if any, relevance in the average speaker's daily life.
E.g., Wu Chinese, at 80M speakers, lacks a robust writing system. Many Indo-Aryan languages (i.e. those related to Hindi) in Northern India lack a standard written form. Maay Maay, a language spoken in Somalia with 2M speakers, does not have a writing system.
[1]: https://en.wikipedia.org/wiki/Cherokee_syllabary
[2]: https://en.wikipedia.org/wiki/Vietnamese_alphabet
[3]: https://en.wikipedia.org/wiki/Coorgi%E2%80%93Cox_alphabet
Flippancy aside: you learn languages to have conversations with others; in the case of dead languages it's to have one-sided conversations with long ago people, no different from reading a contemporary novel. Translation is never perfect (whatever that would even mean) so some stuff is clearly worth reading in the original, and when there's a significant corpus of stuff to read the effort to learn the language can be worth it.
The article actually makes two points, neither of which I agree with. One is that interlinear translation is better than, say, paragraph by paragraph translation. The problem is linguistic structure, and the article even says he had to move words around to make interlinear translation work. That can actually "work" in languages with case like Classical Greek and Latin, but would result in a word salad with French and English. And even in a language with case, word order is not irrelevant.
My second problem is the claimed emphasis on grammar. Of course you do need to know some grammar, but in the languages I learned in HS (French, Latin, Greek) the whole point was to be able to read classical texts (some of them salacious to keep the interest of HS kids), not rote learning of grammar and vocabulary for some abstract benefit.