Home   ·   Index   ·   Search

XTM Editor


XTM is an online tool with a browser-based interface.

This guide focuses on the workflows STP linguists may encounter when working in the XTM Editor on a client's XTM server.
For detailed instructions on how to use the functionality of the XTM Editor, please refer to the official XTM manual where most of the below information is taken from (starting at page 146): http://www2.xtm-intl.com/content/XTMManual.pdf

Logging in

The PM will provide you with login credentials:

  • Link to the project
  • Client name
  • Login name
  • Password

Navigate to the link and enter the other credentials.
(Do not change the login credentials!)

Task list

After logging in, ignore the dialogue box that opens about passwords, and click straight to the “My Inbox” tab to access the files.
You need to accept any task assigned to you. There may be many files in there, assigned to different STP translators, so be careful to only select the files that the STP PM has asked you to translate.
Select Accept task before opening the file using Open Editor.

Navigating the interface

XTM displays all segments on the same page by default, but this can be changed to a number between 10 and 500 under Settings.
Navigate between pages with the buttons in the top of the screen.

  • The left hand menu bar contains the following functions:
    • Segment filter
    • Change segment status
    • Run QA
    • Go to next page with QA warnings
    • Find & Replace
    • Metrics (wordcount)
    • Preview
    • Settings
    • Help

    The right hand menu bar contains checking tools such as spellchecker and QA and navigation between flagged segments.

    • Check current segment
    • Go to previous incomplete segment
    • Go to next incomplete segment
    • Go to previous segment with a comment
    • Go to next segment with a comment
    • Go to next segment with invalid tags
    • Centralize segment
    • Restore from segment history

    The editor displays the document in four columns, which are from left to right:

    1. Segment ID
    2. The source language text
    3. The target language text (this will display the source text in case the target is empty)
    4. Segment status and character count (current/allowed)

    Clicking a segment will display matches from various sources, but never more than three. They are ordered by the type of match and then by the date created or changed.
    The order is as follows: Exact match, TM Leverage match and Repetition, DB Fuzzy matches and Internal Fuzzy matches (in score order and then in date order). MT suggestions will also appear here; this means that for MTPE projects, the MT output that you need to post edit is not present in the target segment when you receive the files. The MT content for you to post edit is in the TM; so the output that you should post edit for each segment will appear when you click on each individual segment in question.


    Keyboard shortcuts:

    Cheat sheet for XTM

    Comments

    To see all comments in the file, go to Settings and select "Yes" for "Show all comments automatically".

    Segment statuses

    The right hand column displays information about the status of the segment and origin of the target text.

    Saving

    All edits you make are automatically saved to XTM Editor.

    Concordance

    Before selecting this menu item, highlight a word in either the source or target text. There are two options:

    Default - A new window appears that displays the latest source and target language phrases in the translation memory that include the linguistic stem of the chosen word. This allows translators to see how the word was translated previously. The various segments are shown in order of relevance.
    Exact match - normally the concordance feature will search for the stem of the word and so will find all forms of the word. If you are only interested in one form of the word then use the Exact match option.

    Inline elements

    Inline elements, or tags, such as text formatting, are represented in the text for translation by numbers in curly brackets e.g. {1}. For in line elements with content, the numbers in brackets come in pairs around the text and are green. When the source document is Microsoft Word, if you hover over the inline element, a message box will display the content.

    Those inline elements without content appear alone and are blue.

    It is important that there is the same number of inline tags in the target segment as in the source segment. To insert an inline element in the target segment put the cursor in the desired location and type Alt + number on your keyboard. Alt + 0 will open a popup window where you can choose the desired element from a list of suggestions. If you are missing any inline elements, the segment will be highlighted in red and a message will appear at the bottom of the page and in the segment. If you do not correct this, the segment will be saved but the translation will not be used when the target document is generated or available in find and replace. Also if you have red segments on the page, as you move to the next page there will be a pop up message telling you that you have segments with invalid inline tags.

    QA

    The quality assurance checks happen automatically when you set a segment to translated or when you click one of the QA icons.

    Spell checking
    If a spelling mistake is detected during the check then the word is highlighted in yellow. If you click on the word then XTM displays a number of suggested spellings for the word. You then have the option to select one of the suggested words, add the suspect word to the custom default dictionary and ignore the word in this segment or in the whole project.
    The spell checking utilises the Open Office dictionaries. Separate custom dictionaries are created for each customer. This means that multiple translators working on projects for the same customer will use the same custom dictionary.

    Other QA checks
    Apart from checking the spelling, the QA module also checks for:

    • Missing or incorrectly formatted number
    • Numbers inside words
    • Missing space after punctuation marks
    • Multiple punctuation marks in a sentence
    • Multiple spaces in a sentence
    • Spaces before punctuation marks
    • Spaces around inlines
    • Spaces around double byte characters
    • Trailing tabs or spaces at the beginning or end of a sentence
    • Uneven number of opening and closing brackets
    • Check non-translatable text in curly brackets
    • Forbidden characters – as specified by the system administrator
    • Identical source and target
    • Repeated words
    • Spelling errors

    When a QA error is detected, an orange mark is inserted to highlight the problem

    Wrapping up

    The last person working on a file needs to send the file back to the client by clicking Deliver from the inbox.
    The last person will either be a reviser working in XTM or a PM (if the revision has been done outside of XTM).
    If you are translating a file you must NOT click Deliver when ready as this sends the file back without revision.