Re: Question about the recent changes to hi-itrans.mim

[Mike FABIAN]

"विश्वासो वासुकिजः (Vishvas Vasuki)" <vishvas.vasuki@gmail.com> さんはかきました:

> On Tue, 8 Aug 2023 at 00:59, Mike FABIAN <mfabian@redhat.com> wrote:
>>
>> I got some feedback off mailing list to the recent changes in
>> hi-itrans.mim. A mapping for “L” was already there, the mapping “ld” was
>> removed, and mapping for “LH” and “LLL” were added:
>>
>>    ("L" "ळ्")
>> -  ("ld" "ळ्")
>> +  ("LH" "ऴ्")
>>
>> [...]
>>
>> +  ("LLL" "ळ्ळ्")
>>
>> Apparently the characters produced by typing “L”, “LH”, “LLL” are not used
>> in Hindi but only in Sanskrit.
>>
>> Therefore, these characters are not necessary for “normal” users, only
>> for scholars and experts. “normal” users might be confused when
>> accidentally typing “L” when some weird character appears.
>>
>> I don’t really understand why it should be better to not map anything
>> to “L”.  Even if “L” isn’t used in Hindi, is that really a reason not
>> to put anything useful on that key?
>>
>
> Namaste,
>
> Normal vs expert boundary is very elastic and subjective.
> How does a regular hindI person properly refer to the neighboring पटाळम्म
> temple in his neighborhood in बॆङ्गळूरु (aka Bangalore going by the British
> colonial name)?
>
> How about a holiday pertaining to तिरुमङ्गै आऴ्वार्?
>
> For that matter even the vowels ऎ and ऒ and their corresponding mAtra-s
> wouldn't be used by "most hindI speakers".

>> Personally I use t-latn-post.mim to type German a lot and it doesn’t
>> bother me at all that it is possible to type many accented characters
>> which are not used in German.  I might sometimes get one of these by
>> accident if I mistype, but that doesn’t make the typo any worse, it
>> would have been a typo on any case. And from time to time I want to
>> type other accented characters so I like it that they are there.
>>
>> So why should it be bad to map “L”, “LH”, “LLL” to something useful
>> instead of leaving it empty? Even if it is only useful to scholars and
>> experts, why not add it? It cannot do harm to normal users, can it?
>
> It can inconvenience total novices slightly - by making typos less obvious.
>
> Perhaps, it may be a good idea to have some option to turn them off for
> those who are so bothered by it (should be "on" by default)?

That is possible, but I think it is not worth it to have an option for
that.  I also think that making “L”, “LH”, “LLL” do something useful
is better than nothing.

Somebody told me that using mr-itrans.mim one can write “senTa” to get
“सेन्ट” but with hi-itrans.mim that is difficult. In hi-itrans, “senTa”
produces “सेण्ट”. That is because of

("nT" "ण्ट्")

in hi-itrans.mim. While trying to find out whether it is possible to
write “सेन्ट ” using hi-itrans.mim, I found that it is possible with
ibus-m17n by typing “senS-C-ReturnTa ”. After typing an “n”, the
commit-key “S-C-Return” commits “न्” and that avoids that “nT” are
transliterated together as “ण्ट्”.

Trying that with hi-itrans in ibus-typing-booster, I found that this
didn’t work, the “S-C-Return” did always commit the current preedit and
send a Shift+Control+Return to the application, so in
ibus-typing-booster, “nS-C-Return” produced “न” plus a new line.

https://github.com/mike-fabian/ibus-typing-booster/issues/457

I have just fixed this in ibus-typing-booster, this release has the fix:


https://github.com/mike-fabian/ibus-typing-booster/releases/tag/2.23.4

With that fix ibus-typing-booster and ibus-m17n do the same when typing
“S-C-Return” while using hi-itrans.mim now.

-- 
Mike FABIAN <mfabian@redhat.com>
睡眠不足はいい仕事の敵だ。