CafeTran

Perhaps we can discuss the modified handling of locked segments in Studio projects.

Is everybody happy with the current implementation?

Personally, especially for working in garbage projects, I'd like the be able to F/R in locked segments too, so that there would be needed a switch 'Deactivate locking', or so.

I now have to deal with projects that are partially pre-translated by the client and that contain lots of typos etc. -- of course in locked segments.

How to deal with them?

In the past, I could just replace them. Open the project in Studio, remove all target segments (incl. the locked once) and deliver the client a project that contained corrected locked segments--without them paying for them ;).

This is no longer possible.

Again, I can see the big advantages of CafeTran not handling locked segments. And I don't put my head under the table so I have seen the requests for it. But as our great soccer player JC (not the one with the beard, from Nazareth) always said:

Every advantage,

comes with disadvantages.

Same for memoQ xliff files. Maybe the simplest way would be to implement the locked file feature in CT. Sometimes I need to put something into a segment that should not be added to the TM (but without leaving the segment unlocked, unconfirmed or empty).

>Maybe the simplest way would be to implement the locked file feature in CT. 

Perhaps. Anyway, it'd be useful to have such a feature in CafeTran (to be able to completely lock segments).

It's extremely quiet in this 'discussion', though the topic must be relevant to more translators, having to deal with pre-translated Studio projects ...

Let me add this question: What to do with locked segments regarding the updating of the memory?

I'm beginning to think that it's wise to not commit locked segments to your clean memory, in order to avoid polluting it with badly, pre-translated but locked content.

How about you?

What's your input on this major topic?

One important observation:

CafeTran treated locked segments in SDLXLIFF and in MQXLIFF projects differently:

• Studio: Locked segments are displayed but cannot be translated, they are committed to the memory (constantly resetting the filter to hide locked segments is not a realistic option)
• memoQ: Locked segmens aren't displayed in CafeTran and thus (hopefully) not committed to the memory.
  

treated = treats

Igor, thanks for responding. But you are missing a lot of points in my argumentation here.

First. Different treatment of the same phenomenon. Locked segments are displayed/not displayed, depending on the originating CAT tool.

This is not consistent.

Secondly: how to prevent commitment of locked segments to your own memory.

I don't have any problem with PMs locking segments. But I don't want to see them (preferably) at all–so that I don't have to be annoyed by their quality. And if I cannot deactivate the display of them, then at least I don't want to have them imported into my own memory.

This is currently only possible via many work-arounds.

So yes, I'd prefer the MQXLIFF approach. If I'm not mistaken, at one time it used to be so with SDLXLIFF too. Probably some other user requested to make them visible ;).

Tough decision, but one that has to be taken. Hence my request to other users of CafeTran to participate. Before you start coding around again.

Yes, but I'm in the habit of sending all segments to the memory at the end of the project. Because I use to make modifications to prior segments as I go along and do the QAs only once I've finished the last segment. 

Is my workflow so exotic?

Nice, but indeed having a "locked segment" feature in CT would be nice. I am thinking of the following use cases:
files with single segments that shall remain in the source language, while making them invisible in Word is no option (too few segments, other file format than Word) - examples are localzations where some titles, button names etc. remain in the source language, adresses and/or departement/function names such as "Head of development".
working in a project with many repetitions, where single repeated segments need different translation and where switching off auto-propagate is not an option

What about this?

 

Two things:

I had the Action>Skip>Locked segments active. This means that segments are being searched, but not clickable in the grid to load the segment into the target editor (and if I am not mistaken, there is no visible indication for this, only after filtering at least a bit). If you are not figuring out what is going on, you might get nuts. This means I am able to edit locked segments (Action>Skip>Locked segments inactive), but segments are going to the TM, or I am unable to edit locked segments (Action>Skip>Locked segments active), while segments are kept away from the TM. Perhaps it would make sense to assign a shortcut to do a faster switch (or a button).

There might be different reasons to lock a segment. However, when the client tells me that term A should be replaced by term B and a whole bulk of segments with term A is locked, I have a problem with mqxliff files (if unable to do this in memoQ).

There currently is a strange behaviour. It used to be better some months ago.

Now: when I click on a locked segment, that is on the small grey square with the segment number, CafeTran takes me to another segment. I cannot enter the segment, not even when Skip > Locked is disabled.

There's absolutely no way to change the locked segments. In the past I was able to do this and used it a lot. I cannot change the content via direct editing nor via F/R.