Multi-Lingual LiveCode?

[dunbarx]

I live in the United States, where we assume everyone in the solar system speaks English. Or ought to, anyway. After all, you are all reading this...

Would LiveCode grow much more rapidly if there were native German or Spanish versions? It seems to me that no core rewrite of any kind would be required, only that a one-to-one map of each native LC word would be created, and a language property set accordingly. For Spanish, say:

"put" = "poner"
"get" = "conseguir"
"add" = "anadir"

I do not know if this mapping would create syntactic oddities:

poner "foo" en mí (put "foo" into me)

Maybe this simple statement actually reads properly in Spanish (it came directly from Google Translation), but surely more complex constructions might appear stilted, like bad translations in VCR instruction booklets. I am not suggesting an effort by RunRev to make this sort of thing read correctly in idiomatic translation. That requires parser rework, I would think, in a big way.

Does anyone here know enough of both English and their native tongue to see if such a map would read decently at all? Would it actually make it more difficult for "foreign" users to write effective code? (Yes, I said "foreign" Remember that solar system thing). And anyway, is such a transliteration as straightforward a project as I think?

Craig Newman

[richmond62]

At which point you get code that is only readable by members of your speech community.

Bad idea.

[Mark]

Craig,

My mother tongue is Dutch. A few simple lines of Dutch LiveCode would look like

Code: Select all

stop 1 in veld 2
als veld 3 is veld 2 dan stop "een draad" in veld "Mijn Resultaat"
als er een veld "gebruiker" is dan
  piep
  antwoord veld "Mijn Resultaat"
einde als

The main problem here is grammar. In above script this problem isn't too apparent, but what about

Code: Select all

als er een veld "gebruiker" is piep dan

Here you see that Dutch grammar requires that the verb "piep" and the word "dan" are switched. For a Dutch version of LiveCode to really make sense, it would take a huge amount of work to make all the small changes to make LiveCode really Dutch. Moreover, Dutch programmers expect a programming language to be in English and to follow the simple English rules of grammar (as far as those rules can be applied). For Dutch LiveCode programmers, it is probably easier to programme in English than in Dutch.

Kind regards,

Mark

[dunbarx]

Mark.

I thought so. It is easier to learn a limited and narrowly defined syntactic language than transliterate to another.

I have always been amazed that people have adopted English this way. I doubt if Americans would do that.

Craig

[[-hh]]

..........

[Gurki]

Would LiveCode grow much more rapidly if there were native German or Spanish versions?

Dear Craig,

If I answer directly to your question: Yes, I think so. But not like you thought it would be in detail.

LiveCode's great-grandfather (HyperCard) had a great success in the German-speaking and French-speaking countries because of localised versions. The user interface was German or French - but not the code. That would also be very easy to implement. It was announced in 2014 (https://livecode.com/lost-in-translation), but I haven't heard a word about it since. The corresponding thread in the forum also remained a wallflower (http://forums.livecode.com/viewtopic.php?f=67&t=15918).

First Thing: Localised documentation and lessons (partly available spread over the web)
Second Thing: Localised user interface (missing)

[Lynn]

I think there is very little evidence that making the language itself "more X like" would help. While the language itself is English-like and rather forgiving, it isn't that forgiving.

HyperCard and other products have had localized UIs, but the market opportunity and situation are going to be significantly different. The industry has moved on multiple times since back then.

On the other hand, local language books and tutorials would probably help, especially with new users.

[jmburnod]

I think hypercard haden't translate code because that was a very big job and that we need a commun langage to exchange between developer from different countries.
Once you have translate code in spanish and use it, you will lost time to verify with a german developer what you mean.
As you may see, this message was written by a no native english
Best regards
Jean-Marc

[ASTU]

So what...

Let the Scriptlanguage in english... other Programming Languages are English too

Only nice to have is the GUI-Language for the Region where it´s in use.

Greetz from Germania

[Gurki]

I think hypercard haden't translate code because that was a very big job and that we need a commun langage to exchange between developer from different countries.

Hello Jean-Marc
You probabely misunderstood my posting. I never proposed to translate the code. The code is the code. The Thing that could be localised is the GUI. In Wikipedia, there is a category of the 500 most used sentence of the GUI (menues) (500 of thousands). But even that helps, because that is, what I see 90% of the time.
Gurki

[richmond62]

I would not translate the GUI for the simple reason that the vast number of
on-line lessons and examples available would suddenly become at least
partly inaccessible to learners.

Where I stay in Bulgaria I can wander down the road to the bookshops and
buy books in Bulgaria about "the usual suspects" (VB, C++, C#, etc.).

If a programming language/system/IDE is to gain traction in a country
what is needed is instructional literature in at least one of the dominant
languages in that country.

No one that I know of in Scotland is banging on about interfaces in Gaelic or Scots!

[bogs]

No one that I know of in Scotland is banging on about interfaces in Gaelic or Scots!

No, but that menubar is priceless!

[teriibi]

Hi all interested,
Ok I though I could brings my 2 cents here since I did help in small translation work with other softwares.
One question is not so much When or Why ? but ---How ?
You can only succeed in translation with a good support from volunteers ...and the right tool.
(Opening a Topic to find out How many Vol in each language would be a start.)

One thing seems obvious, Translating LC is refering to anything else but its scripting code. Very few IDE offer a "translated" code for scripting (and LC Script Editor is not ready for that at all.)
So The Doc, the GUI, IDE, mesages could eventually be translated to other language with a lot of support from translators.

Down below, I ll invite you to give a look (or open your account to get a better view) of a good Translation "tool" that telegram (.org) is using to "succeed" in producing various versions.
(and they dont offer many languages indeed..)
One essencial tool used is called (www.) transifex (.com/ ).

The Telegram Co. is NOT doing all the work but heavily relying on crowd sourcing for that. I cant say How Transifex and Telegram org are related. Transifex is the page that will show you in %% How much of a Section is pending..or completed, and thats all what matters to the volunteers.
Someone at Telegram must be in charge of the final agreement to close a translated portion and make it public.
From there, anyone can update its own soft with the wished Language.

I believe LC will only be able to produce any good translated version if they too :
1) rely on Crowdsourcing translators, beside using
2) some tools sucha as Transifex or similar tool.
if the Russians made it, why not "they"...(dont know what country LC team is from )

The captures down below can give you a good idea of how any Software content can be made available to as many language as you have volunteers for.
Beside it seems the Interface is NOT what requires the most work - if compared to the Help & Tech Docs.
So actually you can make a multi language version of an english interface in quite a short time, if compared to the rest.
next.. for later - eventualy - Translated doc. you´re better with Captures done into your own interface language (not just in english), that way it also relates perfectly to your Doc.

Next, the same can be applied to the Docs, but because of the text Volume it would take much more time, translating is just one part, proofreading is the other one...more text content...longer to double check.

Here are the pix to give all and idea of the task :

I´d Hope to participate if Runrev ever go this way someday...
*beside I would say myself that the Telegram/transifex translating system with :
- context photos
- Voting system (from other translators)
- Section organization by OSes and Use
.. is one of the best I ve ever used, really excelent. !.
so lets try not to reinvent the wheel here...
(Oh, and I dont work for either Telegram or Tansifex )

More info :

https://translations.telegram.org/

https://www.transifex.com/telegram/telegram/

Enjoy the viewing.

[sphere]

If i'm correct the team is from Scotland with some foreign influences, Greece and perhaps some others.

But i disagree, let's leave it in English. I'd rather focus on quality and bug free IDE.
All programming languages are in English.
Even Python which is created by a Dutchman is in English.
Ruby is created by a Japanese guy and also in English.

English is just simply the main language in IT.

If you look at the Dutch example Mark gave, then it becomes unreadable and linguistic false.

[Klaus]

Please do not translate the scripting language! Never!

Believe it or not, in early 90s, the german version of Microsoft EXCEL came with a translated VBA, yes to GERMAN!, and although I had no idea of programming (but of english) at that time, I could see that this was not something one would use, it just looked and felt wrong.

The next year they had switched back to "normal" (english) VBA, go figure.

And since Livecode is far more "english" like than VBA ever was, a translation seems an almost impossible job to me, not only due to the highly complex german syntax and other language differences.

Example, that might cause nightmares to native english speakers, english and german articles:
Nominativ: the man -> DER Mann
Genitiv: the man -> DES Mannes
Dativ: the man -> DEM Mann
Akkusativ: the man -> DEM Mann

And:
the woman -> DIE Frau
the house -> DAS Haus
etc.

Most german folks do not even get this right...

All in all, a translation of the language can't become anything but a linguistic nightmare!