פורומים טכניים » CafeTran support »
CafeTran's smart TM handling
מפרסם התגובה: CafeTran Trainer
CafeTran Trainer
CafeTran Trainer
הולנד
חבר (2006)
Oct 12, 2012

There is a new article about CafeTran's TM handling compared with some other well-known CAT tools:

http://cafetran.wordpress.com/2012/10/12/what-are-the-advantages-of-cafetrans-tm-approach-over-that-of-mainstream-cat-tools/
 
Samuel Murray
Samuel Murray  Identity Verified
הולנד
Local time: 07:10
חבר (2006)
מאנגלית לאפריקאנס
+ ...
Smart or unsmart? Oct 12, 2012

Hans Lenting wrote:
There is a new article about CafeTran's TM handling compared with some other well-known CAT tools...


Well, what does it matter what the TMX file looks like? The user will never have to see the actual TMX file. The user just translates the source text. In what way the CAT tool stores formatting information in the TMX file is irrelevant to the user as long as his formatting is retained.

Do I understand correctly from the screenshots that Cafetrans can't store formatting of the target text?

I do edit TMX files regularly, so I know what it is like, and I can tell you that I would prefer the traditional approach that uses markers to indicate inline tags, as opposed to the Cafetran method that requires me to count characters to know where a formatting change should be applied. The long markers don't bother me because I open the TMX file in MS Word and I use a very narrow font for the inline tags, so that they don't bother me because of their length.

Sure, it looks pretty scary in a text editor with full XML colouring:



But if you colour only the relevant details, and use reduced font widths for inline tags:
 
CafeTran Trainer
CafeTran Trainer
הולנד
חבר (2006)
TOPIC STARTER
Finding and replacing in TMs Oct 12, 2012

Hi Samuel,

Thanks for you kind reply!

Samuel Murray wrote:

Do I understand correctly from the screenshots that Cafetrans can't store formatting of the target text?


It does save the position of the formatting, which IMO is everything you need in a TM.

How do you perform global find and replace actions that span TMX tags?

Hans
 
Samuel Murray
Samuel Murray  Identity Verified
הולנד
Local time: 07:10
חבר (2006)
מאנגלית לאפריקאנס
+ ...
Aah, the issue is Find and Replace Oct 12, 2012

Hans Lenting wrote:
Samuel Murray wrote:
Do I understand correctly from the screenshots that Cafetrans can't store formatting of the target text?

It does save the position of the formatting, which IMO is everything you need in a TM.


From the example image in the blog post, it appears that CafeTran does not save the target text formatting in the TMX file (only the source text formatting). Please confirm whether Cafetran saves the target text's formatting also.

How do you perform global find and replace actions that span TMX tags?


Aaah, now it makes sense. I was wondering why the blog post seemed to contain screenshots of two seemingly unrelated issues (TMX handling and find/replace). Now I realise what you're trying to say: Cafetran can do find/replace across format changes because of the way it handles TMX.

Well, any CAT tool that offers find/replace should decide how to deal with find/replace that runs across format changes. Even MS Word has a way of dealing with it:



Are you telling me that if the sentence in the above screenshot was in CafeTran, and the translator did that find/replace, that CafeTran would have retained the formatting of the text better than MS Word did?
 
Rodolfo Raya
Rodolfo Raya  Identity Verified
Local time: 02:10
מאנגלית לספרדית
Stupid Oct 12, 2012

TMX is a format for exchanging translation memories. If inline information is not stored in the way prescribed in the TMX standard, interoperability is lost. That's a stupid way to store data for exchange.

The smart thing would have been to remove tags from the internal data representation, not from the exchange format.

Regards,
Rodolfo
 
CafeTran Trainer
CafeTran Trainer
הולנד
חבר (2006)
TOPIC STARTER
There you have point Oct 12, 2012

Rodolfo Raya wrote:

TMX is a format for exchanging translation memories. If inline information is not stored in the way prescribed in the TMX standard, interoperability is lost. That's a stupid way to store data for exchange.

The smart thing would have been to remove tags from the internal data representation, not from the exchange format.

Regards,
Rodolfo


Hi Rodolfo,

Glad to see you here!

Your point makes sense ...

Luckily in projects with XLIFF files the interoperability is not an issue. But yes, for exchange via TMX inline tags can be relevant.

Thanks for the info.

Hans
 
CafeTran Trainer
CafeTran Trainer
הולנד
חבר (2006)
TOPIC STARTER
I like CafeTran's approach very much... Oct 12, 2012

Samuel Murray wrote:

Well, any CAT tool that offers find/replace should decide how to deal with find/replace that runs across format changes.


Yep. And I personally like CafeTran's approach very much.

All inline formatting info is stored in the XLIFF. The TMX for internal use is lean and manageable. Perhaps the TMX created via Project | Save as ... | TMX should save a rich version of the TMX file.

2B cont.
 


To report site rules violations or get help, contact a site moderator:

מנהלי פורום זה
Natalie[Call to this topic]

You can also contact site staff by submitting a support request »

CafeTran's smart TM handling






Wordfast Pro
Translation Memory Software for Any Platform

Exclusive discount for ProZ.com users! Save over 13% when purchasing Wordfast Pro through ProZ.com. Wordfast is the world's #1 provider of platform-independent Translation Memory software. Consistently ranked the most user-friendly and highest value

Buy now! »
Anycount & Translation Office 3000
Translation Office 3000

Translation Office 3000 is an advanced accounting tool for freelance translators and small agencies. TO3000 easily and seamlessly integrates with the business life of professional freelance translators.

More info »