> As an English speaker, I'd interpret 2k4 as 2004, not 2400.
Whereas as an English speaker, I'd interpret 2k4 as 2400 because, well, that's just how it's always been in my orbits (cf resistor labelling, for example.)
Lots of users of engineering notation do this.. 1M5 = 1,500,000, 22k1 = 22,100, etc. The unit takes the place of the decimal point. A missing dot doesn't change the meaning.
Exactly, putting (meaningful) characters instead of dots got really going when copy machines made some dots disappear after more than one "copy of a copy" was made.
Then when fax machines came along sometimes dots would disappear or be unclear on a single transmission.
Either way, occasionally sometimes the page would also be scattered with random dots too because of degraded photosensitive operation, or audio-frequency noise on the telephone line at the time.
And it got even more uncertain when text that is not fixed-width got within mainstream reach :\
> Can probably trade the 's' in "reqs" for the decimal point.
I was iffy about dropping the 's' from 'reqs' to get it to 79 (makes it sound like they've only violated one requirement, not multiple) but then I suppose the whole thing is that truncated/abbreviated by that point, it doesn't matter...
Occurs to me that I could have done ">2k VSPs" and avoided the "2k4" controversy but I expect it'd just cause more arguments about whether ">2k" is valid for ">2400"...
(Submitted title was "FCC Could Block Over 2,400 Providers from Robocall Mitigation Database".)