bill-auger <[email protected]> wrote: > On Tue, 21 Jul 2020 22:38:20 +0300 Dmitry wrote: >> Since there is actually no any sane reason to hardwrap lines, > > there is a reason, actually - just last week, i watched a talk by Kevlin > Henney, in which he explained it - the reason why email clients ‹…› have the > auto-wrapping ‹…›, is because people tend to read better with short lines
Unfortunately, I do not know what talk is was and what exactly K. Henney explained there, but the above is plain wrong. The origins of hardwrapping in mail have nothing to do with userʼs comfort. It becomes pretty obvious, when one recalls that no more modern system gain such an obnoxious feature, despite human nature hardly changed over the past decades. What have changed, and dramatically have, is machine capabilities. As we know, email dates back to the very dawn of desktop computing, when slurping a whole letter (many thousands of bytes!) into memory was often unaffordable luxury, it had to be processed screamingly — by small chunks, which are sensible both to machine and to user. A line was natural choice. But how much exactly ‘a line’ should be: mail is by definition supposed to be composed by one person on one machine, but read by another person on another machine — how can one make choices for the other? Well, there was just no choice: display sizes happened to be more or less uniform back then, so line width was universal constant. > around 60-80 characters maximum Yep. So, just as I said, if you hate your correspondents, forcing them to read lines that are wider that they are comfortable with, is a perfect choice. Bonus point, if your letter will be rendered in lines of alternating width, e. g.: 50 / 30 / 50 / 30... > you will find that most news/blog websites follow that rule of thumb Do they? A few years ago they generally did not try to set max-width. Iʼve just checked: Wikipedia and LiveJournal still do not. I missed the chance to collect an amount of personal observations on everyday basis: even if there would be a limit in authorʼs CSS my userstyle would override it anyway, hence the uncertainty. Either way, itʼs not relevant: they definitely do not mix content with representation.
signature.asc
Description: PGP signature
