We’re introducing our AutoMaintenance tool for memoQ server. Its main purpose is to help the SQL database maintain a good performance level by performing regular index defragmentation. From a database perspective, using indexes, and maintaining their good quality is crucial, as fragmentation of indexes can lead to SQL performance dropdowns.
Structure and usage behavior often result in memoQ database tables becoming fragmented, causing a dropdown in SQL performance.
The AutoMaintenance Tool’s main function is to create a regularly scheduled job in the SQL Agent to perform defragmentation. Additionally, it can also perform a one-time defragmentation and cleanup of the job and its temporary tables.
As SQL Express edition does not support SQL agent functionality – which is responsible for job scheduling, AutoMaintenance Tool is configured to schedule tasks with the help of Windows Task Scheduler.
The AutoMaintenance tool is available as a separate tool and can be accessed from the Downloads page. It can be used with memoQ regardless of version number.
Machine translation plugins are used by many memoQ users, so it’s only natural that we continuously improve them based on our users’ ideas and requests. In memoQ 9.12, we’re introducing NiuTrans and Globalese as built-in plugins. We are also introducing some updates for the Google Cloud Translation Advanced MT and the SYSTRAN plugins.
New MT plugins on board: NiuTrans and Globalese
NiuTrans is a cloud-based machine translation engine. Established in 2012, NiuTrans is used in 300+ language pairs (centered mostly around Chinese). It has recently been supported as a private plugin, but now it has become public.
To set it up, all you need to do is create a profile on the NiuTrans website to receive an API key. Once you start working on your localization project, go to MT settings in memoQ and enter the service URL, your API key, the source and target languages, and the profile name. NiuTrans is then ready to use!
Globalese is not your traditional machine translation plugin. It is not an MT engine, but a platform for building engines. So far, it has only been supported by memoQ as a private plugin, but from memoQ 9.12 on, it is also available as a public plugin (i.e., as part of the installation of memoQ).
You can directly configure Globalese engines from within memoQ. The plugin can be accessed along with all the other available MT plugins directly from memoQ’s UI.
Updates to our Google Cloud Translation Advanced MT plugin
There are two new functionalities to Google Cloud Translation Advanced. First, based on current GDPR requirements, users are now able to select any of the regional endpoints their Google Cloud account uses, including the EU regional endpoint of the Google Translate Advanced API.
Secondly, there have been changes to the glossary functionality of the plugin. To now, you had to configure a glossary on Google Cloud Platform and load it into the MT plugin to be able to access the relevant sources. To use the glossary in memoQ’s Google Advanced MT plugin, you needed some technical knowledge and to perform some background tasks.
In memoQ 9.12, we’ve made the process easier, so even those of you who don’t have an advanced technical background can access your glossaries directly from the plugin by opening the Create Glossary resource dialogue, configuring your settings, and entering your glossary data. Once the proper glossary file is uploaded to Google Cloud, the user can use the memoQ plugin to turn it into a glossary resource, make it available for Google’s MT translation, and enable it to appear in the list of glossaries in the memoQ plugin.
SYSTRAN plugin update: assign profiles to language pairs
When setting up the SYSTRAN machine translation plugin in a memoQ project, you must choose one profile (similar to a custom MT model). Profiles are usually created for specific language pairs.
In earlier memoQ versions, you only had the opportunity to choose a single profile from a dropdown menu, often not knowing the language pair associated with your profile. You had to know exactly which profile you were looking for to set it up.
In memoQ 9.12, we‘ve made this easier for you by introducing source and target language options in the plugin settings dialogue. Now all you need to do is select the source and the target languages from the lists, and memoQ will only offer those profiles that go with the selected languages—no need to remember the profile name by heart! And, more importantly, you can now associate multiple language pairs with their respective profiles, so you only need a single MT settings resource for SYSTRAN to work in any of your projects whose source and target language combination match any of the language pairs listed in the MT resource.
Other User-requested Features
The new features of memoQ 9.12 have all been created to reduce translation time and increase convenience for our users, thus making you more productive in your work. From now on you can set new term defaults in QTerm, exclude segments from auto-propagation, automatically detect whether a match is from the source or the target text when running concordance, ignore non-translatables in QA as well as see numbers in short in-line tags.
New term defaults for QTerm term bases
In the memoQ desktop client, you can choose the default settings for prefix matching and case sensitivity for all new entries in a term base from the New term defaults window. Going forward, you can do the same in QTerm, and users with sufficient permissions can change the term base’s properties (i.e., admin rights for the specific TB can set defaults for new terms).
Exclude segment from auto-propagation
Auto-propagation of repetitions can prove to be useful and a time-saver for translators. However, sometimes a segment has to be translated differently in one specific place of the text, even though all other instances of the same segment are identical. This can create occurrences of mistranslations which can be overlooked.
In memoQ 9.12, you can choose one or more instances of a specific translation to be excluded from auto-propagation. Once you set an exception, a small label will appear indicating that the segment is excluded from auto-propagation. All project participants working on the same document will see the new label, so you don’t have to worry about using the appropriate translation in that specific instance, and other project participants can remove the exception or add new exceptions for other occurrences.
Concordance from the translation editor: automatically detect source/target
When you open and run concordance from the translation grid, memoQ now automatically detects whether the specific segment appears in the source or the target text. Moreover, the Search in Target tickbox has been replaced by the Source and Target radio buttons, which also switch automatically when you open concordance from the grid, thereby clearly communicating exactly where concordance is performed.
Ignore non-translatables during spellcheck and QA
When you set non-translatables (such as acronyms or brand names) for your translation project, they can still be identified as errors during spellcheck, and you need to skip each one as you perform the quality assurance process. In memoQ 9.12, we are introducing a new feature that lets you ignore non-translatables during spellcheck, thus reducing your translation time. You can find the setting in the QA settings window or the spell-checking dialog.
Display numbers inside short inline tags
There are some industries where source files feature long inline tags, which are sometimes also nested into each other. Some of these tags also need to be translated in certain projects (such as in the gaming industry). However, it sometimes poses a problem for the translator to work with the main text because the inline tags break the text and it is hard to see the segments in context.
memoQ already had a short inline tag option, and for 9.12, we tweaked it so that the inline tags would also feature numbers. This allows the user to still see the whole text without the long inline tags disturbing the view, and also allows to see at a glance which opening and closing tags belong together in the text.
Are you using an older version of memoQ?
Check out the new features you might have missed!
BUG-9698: Changing the name or some properties of a translation memory resets the Last modified and Last used fields to N/A
BUG-10611: Documentation: The description of the IFileManagerService interface in memoQ Server's WS-API contains wrong information about a parameter
BUG-9848: Help: More detailed explanation added about how to prevent memoQ to import alternate text generated for art in Microsoft Word files
BUG-10575: Localization: (German) The term "Master" is used instead of "Haupt-" in TM and filtering duplicate context
BUG-10507: Microsoft PowerPoint (PPT, PPTX) filter: You may be unable to import documents containing the new type of comments introduced in presentations
BUG-10698: Microsoft Word (DOC, DOCX) filter: You may not be able to import documents with the Import index entries option under specific circumstances
BUG-9323: The label of the Ctrl+Shift+F shortcut mentions searching instead of filtering
BUG-10699: WorldServer Project Package (WSXZ) filter: You may receive a "WorldServer extension architecture was missing for a trans-unit" error during export under specific circumstances
Initial 9.12 release was 9.12.9