[texworks] inputenc error

Reinhard Kotucha reinhard.kotucha at web.de
Sat Jun 8 06:27:28 CEST 2019

On 2019-06-07 at 15:56:36 +0000, Taylor, P wrote:

 > [...] Are you in a position to advise which variant of \usepackage
 > [latin1] {inputenc} / \usepackage [latin9] {inputenc} / \usepackage
 > [latin$n$] {inputenc} might best address this issue ?

The short answer is "none of them" in favor of


If you are really forced to use old-fashioned 8-bit input encodings,
in most cases you would prefer latin9 (ISO 8895-15) because this makes
more frequently used characters accessible directly from Western
European keyboards.  For other keyboards one needs other input
encodings, for instance latin5 for Russian, viscii for Vietnamese,...

For ISO 8895 specifications consult


LaTeX converts all non-ASCII characters to LICR (LaTeX Internal
Character Representation) so that all non-ASCII characters are
assigned to control sequences.  Thus the choice of a paricular input
encoding is just a matter of convenience.  If you want to allow users
to use different input encodings, all strings in macro packages should
be in LICR.

The best choice today is definitely UTF-8 and I strongly recommend to
use it and to avoid national character sets due to their limitations.

Phil, I'm aware that I don't have to tell you what Unicode is good
for but you asked on a public mailing list.  Thus the more verbose


Reinhard Kotucha                            Phone: +49-511-3373112
Marschnerstr. 25
D-30167 Hannover                    mailto:reinhard.kotucha at web.de

More information about the texhax mailing list