On 30Nov15 15:44 +0000, Samir Benmendil wrote: > On Nov 29, 2015 at 19:42, Matthias Apitz wrote: > >El d?a Sunday, November 29, 2015 a las 05:55:32PM +0100, Bernard Massot > >escribi?: > >>I'm struggling to build mails readable on small screens, ie mails whose > >>lines wrap correctly even when there are few columns. I'm indeed > >>thinking of smart phones. > >> > >>I thought format=flowed would be the answer. However the first mailer I > >>tried – K9 mail – doesn't support f=f. I guess Android's native app is > >>no better. So f=f isn't the universal solution. > >Your mail renders fine in my Ubuntu mobile phone BQ E4.5, in Dekko and in > >mutt, see the screens: > > > >Dekko: http://www.unixarea.de/screenshot20151129_180118205.png > >mutt: http://www.unixarea.de/screenshot20151129_180302430.png
f=f seems to be the esoteric way to tackle that problem. All other users just make sure their terminal is at least 80 chars wide. text/plain is just what it is, plain text. No formating meta data in there. One more thing I did not entirely get is, do you intend to reformat all messages you receive or do you want to asure that messages you send will be reformated on the recipents side? Anyway, both approaches won't work, if you ask me. There are too many MTAs out there which all behave differently when composing, formatting and paging emails. > No it does not: http://rmz.io/ff.png That screen is approx 65 chars wide/narrow. But also I would not count that as not readable. > >>What do you suggest? It seems HTML is the only widely supported format, > >>which doesn't have line wrapping problems. So should I make Mutt > >>creating automatically an HTML part of a multipart/alternative, using an > >>empty HTML page template and wrapping paragraphs in <p> tags? Is there > >>something more clever to do? > >No, please no HTML. > Indeed, please don't use HTML. I sign that, too, but I also think that you are not wrong - given that fact that most MTAs support html paging. > You might have better luck with "quoted-printable". "f=f" is much > nicer though. To my knowledge quopri is an encoding method. I would not say that could solve the problem here. Cheers, -- Bastian
