Re: Feature-request: locked scrolling based on paragraph breaks



Hello meld-list readers,

since my feature-request I've researched this a bit more. One related term is XLIFF, a XML based format that allows bits of "meaning" in different languages to exist interwoven in one file. XLIFF editors allow the navigation of this format, but most of them are focused on translating "units" of a few words only, like for UI projects. Only some tackle multi-line text, which is common in law, scientific or political areas but not so much common in .po gettext() / maketext() areas. Side-by-side editors seem to be a rare beast.

The sexyness of doing translations with meld would come from the fact, that meld would then allow the translator to have texts in different languages in separate documents in plain text! No mark-up needed, no paragraph ids, no vertical interweaving (all of this is what XLIFF, TMX, etc. do). meld would counter this with simple paragraph break interlocking.

@Kai and every interested developer
If it is really not that much work to add this feature to meld - please make it happen. Let's add meld to this list: http://www.linuxfortranslators.org/align.html

-Olivier



On 28.01.2012 23:13, Kai wrote:
On 14 January 2012 03:27,<O Zucker web de>  wrote:
Hello meld-list,

with meld being my favourite gui diff editor, I'd like to nag you first with a feature request that might turn out to be contrary of what meld is supposed to do but would be a great feature:
Locked side-by-side scrolling with the lock griping on paragraph breaks instead of similar looking sections. How about that?

Why? I'll tell you: Imagine a writer working on a translation of two documents. The problem is, the actual text is supposed to differ between A and B, whereas *the meaning* per paragraph remains the same. But with meld only looking for literal diffs, meld is going to get in the way with the author progressing through the translation.

A perfect solution would be that meld ignores the changes in paragraphs, but tracks the paragraphs as such (by looking at empty lines), noting if the "structure" diverges. The secondary diff of meaning would be left to the person in front of the screen.

With this functionality, meld would be the first diff editor to assist in comparisons that require an eye on "meaning" instead of literal sameness!

How does this differ from standard translation tools? Is it just that
you're translating longer chunks of text, so normal string-by-string
tools aren't applicable?

> From your description, it sounds like some code in Meld would be
useful, but the majority of the UI just wouldn't apply, so I'm not
sure that it's a natural fit. Having said that, it sounds like it
would be a relatively small project to extract this as a new tool from
the existing Meld code base, and one that I'd be happy to help with.

cheers,
Kai




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