Re: [lojban] Re: [Llg-members] A story of "Search %s" string in Vivaldi browser

2015-12-20 11:24 GMT+03:00 guskant <gusni.kantu@gmail.com>:

2015-12-20 7:09 GMT+00:00 Gleki Arxokuna <gleki.is.my.name@gmail.com>:
>
>
> 2015-12-18 14:37 GMT+03:00 Ilmen <ilmen.pokebip@gmail.com>:
>>
>>
>> On 18/12/2015 09:23, Gleki Arxokuna wrote:
>>>
>>> When I was translating Vivaldi browser interface to Lojban I aimed at the
>>> full translation of 100% of strings, which ended in Lojban being added to
>>> Vivaldi beta version while Esperanto and Ido got removed due to incomplete
>>> translations.
>>>
>>> However, I faced one string that I couldn't unambiguously translate. The
>>> string is
>>> "Search" and the browser uses that to form a placeholder in the browser
>>> search inputbox to form "Search Google", "Search Bing" and similar.
>>>
>>> And what I found is that Lojban presents no solutions.
>>> 1. I used {sisku sepi'o la'oi} => {sisku sepi'o la'oi Google}. The
>>> experimental cmavo {la'oi} quotes one word that must not contain spaces of
>>> pauses inside. Hence, its immediate drawback. Users can add a search engine
>>> leading to {sisku sepi'o la'oi Lojban MediaWiki} where "MediaWiki" hangs in
>>> space.
>>> 2. Neither {la'o zoi.XXX.zoi} method is fine. A search engine named
>>> ".zoi" leads to {la'o zoi..zoi.zoi} quoting an empty string and the last
>>> ".zoi" hangs in space. The browser is not able to change "zoi" separators on
>>> the fly.
>>> 3. However, the solution in English is unambiguous. It's a user interface
>>> widget, kinda {lu ... li'u} quoted text.
>>>
>>> widget (inputbox) boundaries are analogous to this {lu ... li'u}.
>>>
>>> This text can't refer to anything outside it, it's a thing in itself but
>>> this thing can be parsed and used by external mechanisms.
>>>
>>> "Search Lojban MediaWiki" means that after "Search " we have an arbitrary
>>> plain text in an unknown language until End Of Feed is reached.
>>>
>>> I imagine Lojban community could extend {la} to be able to catch
>>> non-Lojban text until the end of file/feed is reached.
>>>
>>> But in the current conditions I see it as a drawback of the language.
>>>
>>
>> Maybe this should be discussed on the BPFK mailing list rather than here.
>
>
> Missed that sentence.
> Yes and no.
>
> Technically Lojban has a solution (la'o zoi ... zoi with changing delimiters
> when needed).
> But then comes a part not dependent on the language.
> I initially posted a message about my translation
> https://groups.google.com/d/topic/lojban/da6VUtMDDLE/discussion
>
> saying "Feedback is more than welcomed."
>
> However, no such feedback has ever been provided in the mailing list.
>
> If someone disagrees with the localization or wants to suggest something
> the correct algorithm is
> TO PROVIDE FEEDBACK
>
> like Ilmen just did.
>
> The same applies to other projects where I'm asking for help. If a help or
> feedback is requested then it's needed!
> Don't hesitate to reply to such requests!
>

I'm sorry for not providing feedback there, but it's because the
problem of {la'oi} in the search box is too indifferent for me to post
a message.

{la'oi} is roughly as good as {la'o zoi ... zoi} in the conditions given.

I'm completely surprised that pure technical issues taking place beyond Lojbanistan are thought to be related to something inside.

The problem became however important after I found your
intolerance of my experimental parsers: it was a turning point that
you became my eternal enemy.

I'm intolerant to everything beyond CLL. Even my unofficial parsers are merely a playtoy, a tool to study Lojban, not to break it.

Changing Lojban is breaking its community which already happened once in 2004.

It's a pity that the current Lojban is probably irreversibly incompatible with CLL, I only have to accept that and perceive as if the language is starting from scratch. But I of course can't accept if this is going to happen on a regular basis. That's why my intolerance over even more backward incompatible changes.

It is lucky of you to be my enemy because
I will make more effort than before to report your faults.

Anyway, the current topic is out of subject of LLG meeting. I suggest
continuing it on the main list if necessary.

> By the way I have already implemented Ilmen's proposal with {ti} + {fa'o}.
> It should come to Vivaldi technical releases in a few weeks. Just feel free
> to update it regularly.
>
>>
>> If the string to be translated is of the form "... %s ..." (with %s
>> standing for the name of the search engine), then I think something like
>> {la'o cirlrbri. %s .cirlrbri} does pretty good (with some separator word
>> that'd be very unlikely to clash with the name of the browser, unless this
>> latter one is chosen intentionally to clash with the separator).
>>
>> Another option could be using something like « sisku se pi'o lo se cmene
>> be ti → fa'o %s » or « sisku se pi'o lo se cmene be lo se lidne be dei fa'o
>> %s ».
>>
>> —Ilmen.
>>

I prefer « sisku se pi'o lo se cmene be lo se lidne be dei fa'o %s »
among Ilmen's solutions. Gleki may be misunderstanding {dei} of it if
he think it contains a problem.

Oh my. I had just chosen the first option and shortened it so that the name of search engine could still be visible.

However,
« sisku fa'o %s »
is my best preference from a point of view of shortness.

Fixed to that. Given that you are interested in localization of browsers then

maybe you can join localization team and at least proofread the whole list of localization strings?

I stopped begging Lojbanists for such things long ago.

But maybe it's simply a misunderstanding that someone can control the whole process of localizing this browser?

You may join via here.

https://www.transifex.com/vivaldi-browser/

I also made a rough copy of the whole set of localization strings but it lacks context which is sometimes necessary.

So I strongly suggest using Transifex instead (it has "discussions" functionality for each localization string).

mu'o mi'e la guskant

--
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+unsubscribe@googlegroups.com.
To post to this group, send email to lojban@googlegroups.com.
Visit this group at https://groups.google.com/group/lojban.
For more options, visit https://groups.google.com/d/optout.