Hacker News new | ask | show | jobs
by berikv 707 days ago
T3e is no b4r w1y to w3e t2t w2h l2g w3s t2n to n10e t2m e8e!

> perl -C -pe 's/(\w)(\w+)(\w)/$1 . length($2) . $3/ge'

Or for the less o4e among us, this v5n will only n10e words with l4h six and up:

> perl -C -pe 's/(\w)(\w\w\w\w+)(\w)/$1 . length($2) . $3/ge'

F3l v5n:

perl -C -pe 's/(\p{L})(\p{L}*)(\p{L})/$1@{[length($2)]}$3/g'

N12g w5t i18n w3d n1t b0e c6e, t2s t2s a u1f-8 c8e v5n. I c2l i0t I16r-v1.0

새0로 오0신 모0든 분1께 인3고 싶2다.

2 comments

Translation:

There is no better way to write with long words than to numeronymise them everywhere!

Or for the less obtuse among us, this version will only numeronymise words with length six and up:

I spent some time staring at l4h, after quickly reading o4e as 'obese' on the way there. I suppose this might be a good Freudian slip generation scheme?
Numeronymizing without internationalization would not be complete, thus this [???] a utf-8 compatible version. I call it Internumeronymizer-v1.0