Example from yesterday’s news. Count ’em yourself – there’s 76 of them there.

Mass i18n efforts like this are I think an underappreciated benefit of what static site generators like Hugo can give you. Actually, especialy Hugo – it’s multi-language support is very good, like darn near everything about the platform once you get past the initial learning curve.

Another underappreciated benefit: When building HTML pages is fast, you can afford to build a lot of them. A quick hugo && cd public/ && fd html | wc -l tells us that there are about 2700 HTML files on the site, which Hugo builds in under 3000 ms on my machine. The Github Action run which built the site as of today took a glacial 35 seconds by comparison.

I have about 2 months of news articles scraped so far, so I’ll hit the “takes a minute on my local machine” mark in about 3 years at this rate. At that point I’ll likely switch to running hugo on my local machine and pushing the public/ directory to its own Git repo instead of building it via Github Actions. A little shift-left goes a long way.

EDIT; I woke up today and realized the Google Translate API costed me a cool 174 euros to run through all of these translations yesterday. Taking a look at the size of all of the Markdown documents, we get a grand total of

1
2
3
4
5
6
7

❯ fd -e md -x du -b {} | awk '{total += $1} END {print total}'

12251321 # bytes

❯ python -c 'print(round(12251321 / (1024 * 1024), 2))'

11.68 # megabytes

That’s almost 15 euros a megabyte! Luckily that all came out of my Cloud Translation free trial. And since I now know the “naive way” (A) works well and (B) is expensive, that forms a great business case to make a program that does this at a cheaper rate.