SDL has just released Studio 2014 Service Pack 2. The big news is that Studio and MultiTerm have finally got rid of their dependency on Java for terminology. Woohoo! Other new features include TM sorting by date, global tag verification and alphanumeric string recognition. Let’s look at all these one by one.
The end of Java and other terminology improvements
Adding and editing terminology in Studio is much faster now. And it’s more reliable. Gone are the days when a Java update would flood forums with pleas for help. When we started testing SP2 with Java-free terminology and other new features, the Beta forum came alive with comments ranging from ‘absolutely loving this’ to ‘Christmas in October!’ Beta testers don’t mince their words, so this is praise indeed.
There’s a new “Quick Add New Term” option, which automatically adds and saves the term(s) you highlight (in source and target segments) to the default termbase. The shortcut is Ctrl+Shift+F2. Use this when you just want to add a straightforward term, without filling in other fields in the termbase entry. It also works for adding a synonym to an existing term.
You can also delete a term in Studio now, without having to open MultiTerm. If your Termbase Viewer is open, browse to the term you want to delete by typing the first letter with the focus in the term list. Or, easier still, if the term appears in the Term Recognition window (see image), right click it and select “View term details”. Then click “Delete this Entry” in the termbase viewer. The shortcut is Ctrl+Shift+Alt+D. This works for all open termbases, not just the default one.
TM sorting by date
This is a new default setting. It affects TM (translation memory) look-up and concordance. In the past, the oldest concordance results were returned first, but it’s usually more relevant to see most recent matches. Studio still offers the highest percentage match first, but then sorts by date.
If you prefer the old way, just disable this new default setting. You’ll find it in Project settings (or general options) > Language Pairs > specific language pair > Translation Memory and Automated Translation > Search > Show most recent translation first.
Sorting by date has also been improved in the filter display. The drop-down list of previous searches is now sorted with the most recent at the top of the list.
Global tag verification
Tag verification warns you if any tags have been added, deleted, left as ghosts (unpaired) or placed in a different order when comparing source and target segments. This isn’t a new feature.
However, pre-SP2, these tag verification settings were defined by file type. Now, they’re set globally, so you can decide what sort of warning you want for all file types in one go. You’ll find the global setting under File > Options > Verification > Tag Verifier > Common.
Alphanumeric strings are made up of at least one number and one letter and can contain dashes, underscores or dots inside the string. They are now recognised as placeables, which means that they will be automatically transferred from the source to target segment.
Here’s an example of some segments that I used to have to translate manually, because they wouldn’t give me a decent match pre-SP2. Now, they’re autopropagated as 100% matches:
TMs created before SP2 don’t recognise alphanumeric strings as such, so you need to:
- Go to the Translation Memories window. Select a TM from the list on the left (or open a TM and close it again to make it appear in the list). Enable the alphanumeric setting under Home > Settings > Fields and Settings > Alphanumeric strings.
- Continue to Performance and Tuning > Re-index your TM.
Do this in the right order. I spent some time reindexing my TMs without changing the setting first, and couldn’t understand why the feature wasn’t working!
If you have any current projects, enable this setting by going to Project Settings> Language Pairs > (Your language pair) > Translation Memory and Automated Translation > Auto-substitution:
Text replacement penalty
You can apply a text replacement penalty to alphanumeric strings and acronyms. This is useful to check that these strings (and acronyms) are replaced correctly and show that they have been taken from a TM.
Now that we have alphanumeric string recognition in Studio, I’d like to see a verification setting in the QA checker to flag alphanumeric mismatches between source and target. At the moment I use the OpenExchange Xbench plug-in to perform this check.
Other new features and changes in Studio 2014 SP2:
- New network licensing system
- Source segment editing for almost all file types
- Different word count algorithm (see Riccardo Schiaffino’s blog post for the juicy details).
- SDLXLIFF files are now automatically added to a return package
For a complete list of new features, check out the Studio 2014 SP2 Release Notes.
Over on Twitter, even memoQ was happy to come up with a teaser for the occasion:
Congratulations on finally divorcing from Java, @sdltrados! Great move, probably worth the hundreds of days invested.
— Istvan Lengyel (@Kilgraymemoq) November 18, 2014
In the pipeline
You might stumble across a feature called ‘Translation Quality Assessment’ in batch tasks. You can’t use it at the moment, but in the future it will apparently grade translation quality. Let’s wait and see.
SP2: Easy/safe to upgrade?
It’s easy. Just download the upgrades for Studio and MultiTerm from your account on the SDL website and install them over SP1 (make sure you’ve closed Studio and MultiTerm first). I’d recommend doing this between jobs, just in case.
It’s safe. Studio 2014 SP2 works with all the OpenExchange plug-ins I use. There don’t appear to be any back-compatibility issues. Critical bugs have been sorted. But upgrade at your own risk, as always.
Read my next post for some tips on minimising risks when upgrading and a look at some installation issues.
SP2: Worth upgrading?
For me, this is a no-brainer. Anyone who has had trouble with Java error messages and terminology bugs in the past – and anyone who wants to avoid them in the future – should upgrade to SP2 today.