summaryrefslogtreecommitdiff
path: root/doc/emacs/anti.texi
diff options
context:
space:
mode:
authorEli Zaretskii <eliz@gnu.org>2008-11-28 12:01:44 +0000
committerEli Zaretskii <eliz@gnu.org>2008-11-28 12:01:44 +0000
commite8e2bd93103909d092205b95d72bfeb8d8f6d129 (patch)
tree9aafb76b515b1b81e0083e91f3986722d916edc3 /doc/emacs/anti.texi
parent4d1243c82656339b063c3f0d027b1eb9ff90f554 (diff)
downloademacs-e8e2bd93103909d092205b95d72bfeb8d8f6d129.tar.gz
(Antinews): Add stuff about Unicode vs emacs-mule representation.
Diffstat (limited to 'doc/emacs/anti.texi')
-rw-r--r--doc/emacs/anti.texi39
1 files changed, 33 insertions, 6 deletions
diff --git a/doc/emacs/anti.texi b/doc/emacs/anti.texi
index 1c8e8c07025..83b943760f2 100644
--- a/doc/emacs/anti.texi
+++ b/doc/emacs/anti.texi
@@ -19,6 +19,34 @@ ways of specifying fonts---so-called ``Fontconfig'' and ``GTK'' font
names---are clearly redundant, and have been removed.
@item
+We have switched to a character representation specially designed for
+Emacs. Rather than forcing all the widely used scripts artificially
+into alignment, like Unicode does, Emacs treats them all equally,
+giving each one a place in the space of character codes. Thus,
+scripts do not need to fight over characters used in each one of them,
+as each has its own variant, and they all are different as far as
+Emacs is concerned. For example, there's a Latin-1 c-cedilla
+character, and there's a Latin-2 c-cedilla; searching a buffer for the
+Latin-1 variant will only find that variant, but not the others. This
+design allows us to get rid of a confusing situation in Emacs 23,
+whereby a character can simultaneously belong to any number of
+charsets.
+
+@item
+Emacs now uses an internal encoding, known as @samp{emacs-mule}, which
+is peculiar to Emacs and does not map easily into any of the existing
+character encodings, including Unicode. This was imperative to
+support several different variants of the same character, each one
+belonging to its own script: @samp{emacs-mule} marks each character
+with its script, to better discern them from one another.
+
+@item
+For simplicity, the functions @code{encode-coding-region} and
+@code{decode-coding-region} no longer accept an argument saying where
+to store the result of their conversions. The result always replaces
+the original, so there's no need to look for it elsewhere.
+
+@item
Emacs no longer performs font anti-aliasing. If your fonts look ugly,
try choosing a larger font and increasing the screen resolution.
Admittedly, this becomes difficult as you go further back in time,
@@ -34,10 +62,13 @@ and many more.
Emacs can no longer display frames on X windows and text terminals
(ttys) simultaneously. If you start Emacs as an X application, the
Emacs job can only create X frames; if you start Emacs on a tty, the
-Emacs job can only use that tty.
+Emacs job can only use that tty. No more confusion about which type
+of frame will @command{emacsclient} use in any given Emacs session!
@item
-Emacs can no longer be started as a daemon.
+Emacs can no longer be started as a daemon. We decided that having an
+Emacs sitting silently in the background with no visual manifestation
+anywhere in sight is too confusing.
@item
Transient Mark mode is now disabled by default. Furthermore, some
@@ -85,10 +116,6 @@ been removed, and so has the buffer face menu bound to
@kbd{S-down-mouse-1}.
@item
-Emacs now uses an internal encoding, known as @samp{emacs-mule}, which
-is not a superset of Unicode.
-
-@item
VC no longer supports fileset-based operations on distributed version
control systems (DVCSs) such as Arch, Bazaar, Subversion, Mercurial,
and Git. For instance, multi-file commits will be performed by