Revised: 29.12.25
Object lesson.
These stories were not given as perfect examples of the Czech language, as noted. One was proofread by a native speaker and corrections were made. Although DeepL was considered by many among my friends and acquaintances, it is flawed. Consider what an everyday CS reader did with the Amalka story.
https://inter-linearnichesite.org/amalka.html
about a young girl "telling stories" and
https://inter-linear.nichesite.org/danny.html
about a young boy learning languages
These were created as gifts to my CZ grandchildren. CS is not easy to translate . . . I used DeepL to do it, having been recommended this engine by current and former EFL students.
It is not that DeepL doesn't make mistakes. In a story about a child and a tree, the machine translation came up with car boot or luggage compartment for trunk, as in tree trunk. How could it miss the context? If just did, so proofreading is always needed and recommended . . . until machine language translation perfection is reached, which I believe will be never, especially with literary texts.
What DeepL did that others messed up badly was correctly use CS-convention quotation marks. A minor detail, yes, but I saw that and didn't have to ask native CS speakers to proofread those translations.
BECAUSE these DeepL translations were to aid/help decode/learn/practice reading EN. Imperfect as these translations may be, they will be good enough for the intended purpose.
jkm,Prague, 12.25
No comments:
Post a Comment