undefined | Better HN

0 pointstyingq9y ago0 comments

Typically, 'U' is omitted to avoid accidentally putting the word FUCK into the generated output. No, not kidding.

See http://www.crockford.com/wrmg/base32.html, Ctrl-F, search for obscenity.

Edit: Interestingly, since it already omits the letter I, adding U covers all of George Carlins' 7 Dirty Words: https://en.wikipedia.org/wiki/Seven_dirty_words

0 comments

3 comments · 1 top-level

joshuak9y ago· 2 in thread

This is so bazar that I don't even know where to start to try an understand it.

Is it a joke? Tongue-n-cheek? It must be. I can't understand how it could possibly be a ligament concern. What about other languages? As adults who doesn't use 'fuck' conversationally? If it were a real concern then surly drawing attention to it in the design of the ID is a more obvious then the very rare occasion of 'dirty words' (by some definition of dirty) appearing in otherwise random strings.

So I have to conclude it's a joke, but then what the fuck?

This type of thing puts noise into engineering details that just ends adding to confusion and ambiguity. If there's no (important) reason for 'U' to be in the omission set, then just say it's an arbitrary choice, if there is an OCR, or legibility reason then say that.

Half of my brain says it's funny, and half my brain says it's fucked up.

tyingqOP9y ago

I see where you're coming from. On the other hand, since they were after base32, they were able to choose an additional letter to omit. By omitting the U, they happen to exclude the most popular english language expletives.

These things can and do end up in, for example, urls. Think something like an ecommerce store order id. While not likely to happen, the below could happen in an email, and might invite unneeded controversy:

"Dear Mr Customer, Please Click Here to see your order status: http://example.com/status/FUCKUP6433432334234 "

joshuak9y ago

Ah yes ok I can understand that. I can see from a marketing point of view that you might want to simply and quietly avoid the (roughly) 1 in 1,000,000 ids which start with 4 specific characters.

Still, it's a bit culturally specific. It gives the impression of solving a problem that is doesn't. Depending on your market/language a different character might be a better choice. Plus there are quite a few 'bad' words that can still appear.

Filtering and rejecting proposed IDs based on a list of objectionable words seems like a far more realistic solution, then baking a very specific exclusion into the plumbing.

1 more reply

j / k navigate · click thread line to collapse