Le Saturday 16 July 2005 05:58, David Mézerette a écrit : | [...] though kinput2 and | canna makes a great input method for japanese, I couldn't figure out | how to make a personal dictionary file [...]
if we include kinput in [extra] someone will need to explain this in the
docs.
| , and, it is not versatile
| enough: it's a japanese-only input method, and I also wanted to type
| korean.
everybody who wants multilanguage can install scim and the desired input
engines for it. but we can have also other IM's side-by-side with scim in
[extra] (like there is already gcin)
| Though apparently most japanese distributions still use canna
| as an input-method, I am not sure it is needed in archlinux. This input
| method story can already be quite confusing to basic linux users like
| me, and many input methods would only add to the confusion.
that's why we will have docs.
| By the way, when updating my system, I noticed that the repo package
| for scim-tables is older than the one I compiled: the latest release
| (my local) seems to be 0.5.1 and the repo version is 0.5.0. I will wait
| for a working skim 1.4 package to update my input method (still using
| 1.2.1)
everything in [extra] should already be running with 1.4.0. i recompiled
all scim engines against it (what triggered them to become -2)
you are right that in the meantime there came out scim-tables 0.5.1.
nobody flagged it out-of-date and i didn't know that a new version is
out. now i know the update will follow soon.
| Another question about the way the packages are distributed in
| archlinux: do the scim engine and different input methods make a group
| (like kde for example)? I think it would make things easier for users.
groups would not essentially make it easier for the users but make it a
little bit confusing for the maintainer (scim would belong to >50 groups
and scim-tables to 30 if the grouping is with languages)
i think the easiest way is to simply describe the available packages in
the wiki and make for every language and every input method a wiki page.
and in this it would be explained what possibilities there are to use
with the corresponding package names to get.
this way, we will have 1) made the newbie read some of the docs,
understand how to use pacman -S <pkg>, 2) know what is going on if she/he
wants to use a IM and 3) do it by her/himself letting her/him the chance
to choose.
if we have kinput in the repos together with scim-anthy and scim-tables,
then obviously this 3 pkgs should belong to the "lang-jp" group if we
group them. this would mean that someone trying to install this group
would be suggested to get all 3 of them. not knowing about what they are
(not reading the docs or description) this person will install all 3
without much thinking. starting X this person will be positively
surprised with the new icon appear in the systray if she/he uses gtk2
apps. this person will then use e.g. scim-anthy but will still have the
other 2 pkgs on it's system without needing them. --- that's why i think
that groups are not the best solution for this. informing the people is.
thanx for your comments + greetings,
ダミラ
(i feel like in the kindergarten: i know how to write my name! :D )
--
Most legends have their basis in facts.
-- Kirk, "And The Children Shall Lead", stardate 5029.5
pgp3fbc1OBms3.pgp
Description: PGP signature
_______________________________________________ arch mailing list [email protected] http://www.archlinux.org/mailman/listinfo/arch
