[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: [Pkg-ime-devel] Bug#644361: fcitx: Typos in package description

On Thu, Oct 6, 2011 at 08:00, Justin B Rye <jbr@edlug.org.uk> wrote:
> Jonathan Nieder wrote:
>> Some quick thoughts:
> I like this version at least as much as my own.
> [...]
>> Maybe a description like this:
>>       Description: versatile Simplified Chinese input method suite
>> Is it only for simplified Chinese or are there tables for traditional
>> Chinese and other scripts?
> Ideally we'd want the metapackage synopsis to be something that can be
> recycled in the synopses of other members of the suite, so we don't
> want to be too specific about things that aren't appropriate there.
> It seems to me that the big selling-point feature for end users is the
> fact it's configurable/themable/Windowsy, so maybe:
>        Description: friendly input method framework
> (Or maybe just keep the explain-the-name synopsis.)

Yeah, for end users is configurable, themable and works like Windows
input methods that has proved to be efficient by the test of time.

> Explaining what an IM is takes up space that could be used summarising
> its features; maybe the best compromise would be to put that
> explanation in the description for the metapackage but not elsewhere
> in the suite.  I can't see a neat way of achieving that, though.
> While trying to torment it into an appropriate shape I've ended up
> with this version instead:
>  Description: Free Chinese Input Toy of X
>  Fcitx is an Input Method framework providing a graphical interface for
>  entering Chinese characters in a variety of applications and using a
>  variety of mapping systems.
>  .

Why again "entering Chinese characters"? I'm afraid we need rephrasing
here in near future (month-ish).

>  Despite its humble name and beginnings, it offers a pleasant and
>  modern experience, with intuitive graphical configuration tools and
>  customizable skins and mapping tables. It is highly modularized and
>  extensible, with GTK and Qt front-ends, support for back-end UIs
>  based on Fbterm, pure Xlib, GTK, or KDE, and a developer-friendly API.
>  .
>  This metapackage pulls in a set of components recommended for most
>  desktop Fcitx users.
> (I'm probably garbling those features, too.  For instance, how much
> else is there that you might configure besides the skins and tables?

The configuration is very detailed. Here is an incomplete list:
1.Time to wait before actually start Fcitx
2.Enable/Disable and configure any plug-ins or modules.
 1) user defined quick phrase dictionary
 2) record all inputed characters
 3) priority of every engine/table.
 4) cloud pinyin API settings including which service provider to use,
minimal pinyin length to trigger cloud pinyin, which place to display
the result from cloud pinyin API in the candidate list.
 5) switch between XIM on-the-spot or over-the-spot on the fly.
 6) back-ends for character conversions
 7) user defined virtual keyboard and key map
 8) remote call to change status of Fcitx (useful to automatically
switch the status of Fcitx in VIM/Emacs in terminals)
3.Editing the details of the user interface, like
 1) whether to display the user's input speed
 2) the application's version number
 3) how to display the candidate list, horizontal or vectorial
 4) font size, XY positions main window,
 5) whether to use systray
 6) change any picture or color of a theme
4.Defining a number of keyboard shortcuts like changing input engine,
reload configurations, toggle full-width characters, etc.

> Would it make sense to say "intuitive graphical configuration tools
> FOR CUSTOMIZING skins and mapping tables"?)

I'm confident to say YES.

GTK2 based configuration tool has been uploaded to Sid, and the KCM
configuration module hasn't yet. I'll do it soon. They provide similar
functions and the latter is better integrated to KDE's control center.

Aron Xu

Reply to: