Hi, On Thu, Aug 17, 2017 at 01:26:07PM +0100, Justin B Rye wrote: > > I would appreciate if you have any time to review > > the new debconf template for uim package. > > Okay, patch attached. Thank you for your works! Applied to git.debian.org. https://anonscm.debian.org/cgit/collab-maint/uim.git/log/?h=debian-experimental > > Template: uim/inputmethods > > Type: multiselect > > Choices: byeoru, ipa-x-sampa, latin, look, m17nlib, pinyin, viqr > > Default: byeoru, ipa-x-sampa, latin, look, m17nlib, pinyin, viqr > > _Description: Input methods to register: > > What this means isn't clear to me. Given that it's possible to > "register" multiple input methods, clearly it isn't talking about > selecting one as "active". And if it just means something like "make > it available", what reason would there ever be for UIM not > automatically registering all of them? First of all, uim has too many IMs. If all of them are registered, then uim preference tool's IM selection menu items are large. But, in practice, these "Choises" are already handpicked by me, perhaps automatically registering all "Choises" may not be so problem. I will talk about this with uim co-maintainer. -*- > But I'm carrying on in the assumption that this all makes sense to IM > users... > > > The uim package can use various input methods. The following are currently > > available: > > (The debconf standard is to use only one space after a stop, though it > hardly matters because flowed text gets converted to that format > anyway.) I see. I remember it. > > - byeoru : byeoru hangul input method > > - ipa-x-sampa: X-SAMPA IPA input method > > - latin : Latin script input method > > - look : Dictionary-based completion input method > > - m17nlib : m17nlib input method > > - pinyin : pinyin input method > > - viqr : Vietnamese Quoted-Readable input method > > If we're writing lowercase "hanyul" and "pinyin" we definitely > shouldn't capitalise "dictionary". > > Explaining "m17nlib" in a list of input methods by saying it provides > the m17nlib input method is a bit uninformative, compared to the way > byeoru (for instance) adds a mention of hangul. We could at least > hint at the explanation of the name: > > - m17nlib : m17nlib multilingual input method Thank you. Definitely. > > The following input methods require package installation. Automatically > > registered after package installation: > > That last sentence isn't clear (or a sentence). If I correctly > understand the situation, I'd suggest saying: > > Extra input methods are available for Japanese, and will be registered > automatically if you install the corresponding package: Thank you. It is suitable. > > - anthy : Install `uim-anthy` package. > > - skk : Install `uim-skk` package. > > - tcode : Install `uim-tcode` package. > > I could pedantically change a lot of trivial details - the spacing, > the capitalisation, the style of quotation marks... even the dashes > might be better as asterisks. But more significantly, it seems to be > providing the wrong information. Once I see that there's a > package named "uim-tcode", it's obvious that there's an input method > known to UIM as "tcode" that requires me to install that plugin > package. The part that isn't necessarily obvious is that "tcode" is > the input method slightly better known as "TUT-code". > > - install "uim-anthy" for the anthy input method > - install "uim-skk" for the SKK input method > - install "uim-tcode" for the T-code/TUT-code input method You are right about "TUT-code". Thank you and I apologize for the trouble. > (I'm deliberately not making it look like a natural continuation of > the previous list, because we don't want users to see "tcode" at the > bottom of the list of available IMs and immediately expect to be able > to select it from this debconf menu.) I understand. Again, Thank you for your works. -- Regards, dai GPG Fingerprint = 0B29 D88E 42E6 B765 B8D8 EA50 7839 619D D439 668E
Attachment:
signature.asc
Description: PGP signature