Home » Posts tagged 'miis'
Tag Archives: miis
Creating a Talent Management Office
Creating a Vendor Management Office — or “Talent Management Office” (TMO) per the growing trend — for language services is much the same as establishing any new department within a company. You need to decide who the main operators of it will be, their responsibilities, and how they are organized. As a TMO, you will also need to determine the onboarding process for the vendors — talent — that you’ll be taking on. This includes screening and contacting potential talent, doing a background review, testing their linguistic skill, and then performing the actual onboarding process.
All of this requires some specific documentation — some familiar, some that might be new. An initial questionnaire to get a good snapshot of the candidates is a good idea, as well as keeping a database of all the talent you run into. This database can include background information and a scorecard referring to the linguistic test. And of course, an Independent Contractor Agreement (ICA) is a must-have.
- Best practices
- Team Organizational Chart & Job Description
- Roles/Authority
- Talent Onboarding Process
- Contract Management
- Onboarding Processes
- Screening & Initial Contact
- Request for Information
- Due Diligence Review
- Linguistic Testing
- Onboarding
- Templates & Documentation
- New Talent Questionnaire
- Talent Database
- Independent Contractor Agreement
- Test/Scorecard
In this post I will outline a (fictional) example of a TMO established within a medical translation and interpretation language services provider (LSP). This example is the result of weeks of work by myself and my team members in a Talent Management course at the Middlebury Institute of International Studies, led by Alaina Brandt.
Our fictional LSP is called MediTrans Inc., which specializes in medical translation and interpretation. The sensitive nature of medical information informed many of our decisions going forward.
In this scenario, MediTrans is a smallish LSP, and the TMO only has six people. The following titles and descriptions are based on research we did on similar job postings around the web, customized for our purposes.
Head of Talent Office
Develops and drives our recruiting strategy and execution to support our rapid growth. Optimizes the talent acquisition function and processes.
Financial Manager
Reviews financial reports, monitors accounts, and prepares activity reports and financial forecasts.
Talent Manager
Provides analyses to the organization on vendors, costs, and ensures the terms of contracts are followed by all parties.
Recruiters (2)
Work with outside agencies to discover new talent. They also review resumes, explore job search websites for potential talent and contact those individuals personally.
Interview potential candidates, searching for talents that match the specific needs of Meditrans’ work. Introduce new hires to the company and walk them onboarding process. Facilitate all new-hire paperwork of candidates who meet the expectations of the job.
HR Coordinator to TMO
Works with MediTrans’s HR department to develop policy and direct and coordinates human resources activities, such as employment, compensation, labor relations, benefits, training, and employee services.
We used the software Process Street to help with organizing the onboarding process. Process Street is sort of a checklist software with a bunch of other bells and whistles, like the ability to make automated lists that will send out emails that auto-fill with client info. The following is a snapshot of out Process Street… process.
One thing in particular that we focused on as a medical translation/interpretation LSP was strict requirements for contractors. Our drafting process included a profile screening for staff performing the onboarding, with harsh penalties for any issues with references, reviews, and online professional presence, in addition to the obvious penalties for plagiarism and illegitimate payment methods.

We also required at least a year of experience in medial translation, as well as the ability to keep sensitive information confidential. When evaluating the test translation, we pay special attention to meaning errors, additions, or omissions, and less attention to formatting and tone. This is to ensure maximum understanding in a field where any errors or ambiguity can mean the difference between life and death.
If the candidate gets a good score and passes the background test, they’re placed in our database ready to be worked with in the future; failing candidates are also placed in the database so that we can keep track of individuals that we’ve already tested, and to make sure we don’t accidentally perform redundant work in the future.
The onboarding process for many companies can look very similar, but care should be taken to customize the process for every industry and, frequently, every project. The above can give you some ideas for your own process, but remember that there is no one-size-fits-all solution.
One piece of advice, however: don’t cut corners that sacrifice your talent’s (vendor’s; contractor’s; etc) happiness. There are only so many professionals out there, and often their online community is vocal; if you treat them right, your returns will be greater than any shady cost-cutting ever could be.
Desktop Publishing Proof-of-Concept: Localizing a Japanese Children’s Book
Before coming to MIIS, I taught English in a little tourist town in Gifu Prefecture, Japan, called Gero. Gifu is a beautiful place, with amazing nature, fresh food, and — best of all — hot springs. It also, like many things in Japan, has a cute mascot: Minamo, the Water Fairy. A friendly character that brings happiness to everyone he encounters.
So, in an effort to let more people know about Minamo and, by association, Gifu Prefecture, I decided to localize the below children’s book featuring Minamo and famous sights around Gifu from Japanese to English. You can find the original book on Minamo’s official webpage (obviously he has one).
I’ll go into more detail about the process I took to localize the book, but first, the finished product:
I’m pleased with the result! It wasn’t a too difficult project, but it presented some interesting challenges.
The Process
This project can be broken up into a few distinct phases. They are:
To-Do
- Download all assets
- Prepare pages for translation
- Localize reading order
- Translate
- Insert back into website format
I wanted the final product to be as close to the original as possible, which meant recreating the simple webpage that it is embedded in. I didn’t have access to the original files, but masking out the text wasn’t too difficult, as it’s mostly on a white background with just a few sentences overlapping the pictures. In some places the text was quite narrow because of the vertical orientation of the Japanese writing. And, being a Japanese book, the layout was right-to-left, which meant reworking the page numbers and the order that the pages appeared within the website.
The last page of the book is a map of Gifu with short descriptions of famous places. This map features a few different fonts and lots of small text, and so took a little longer to localize. Finally, the cover page had some characters separated into little bubbles — a practice that looks fine in Japanese, but wouldn’t work well in English. You can see below a comparison — Japanese on left, English on right — showing how I chose to localize these pages.
Localizing the reading order involved first using Adobe Photoshop to erase the pages numbers completely, and then using Adobe InDesign to add real page numbers to the project that could be manipulated. Photoshop and InDesign work seamlessly together, and you can edit an image from InDesign at the same time right in Photoshop without doing any file juggling by Right Clicking > Edit With > Adobe Photoshop. When you save within Photoshop, the changes will instantly be reflected in InDesign. Couldn’t be simpler.
I also needed to go into the HTML of the webpage to change how the reading panel behaved. To my surprise, the code for the webpage was very well commented — in English, no less. It was a simple matter of renaming files and changing a few attributes.
For places where the margins were very narrow and I didn’t want to cover up any of the illustrations, I used the Pen tool in InDesign to create custom borders around the illustrations where no text was allowed. This saved me from having to resort to awkward line breaks or squished text. You can find a great tutorial on how to do this here. The final product looked like this:
If I did it over, I might make the border a little bit further from the picture. But, not too bad for a proof-of-concept! While we’re on this page, if you read my English translation you may be thinking to yourself, “What on earth is a ‘milkvetch’?” Or maybe it’s just me. In any case, I went back and forth on whether to translate the flower’s name literally, or just localize it to “flowers”. In the end, I decided that since this book is meant to be showcasing specific things found in Gifu, I should use the flower’s real name. However, I did try to make it clear that it is, in fact, a flower.
And that’s pretty much it! I’m glad I was able to do this project. It was nostalgic seeing all these places in Gifu again and reading about them. I hope I’ve done Minamo proud!
Mini-Portfolio for Advanced CAT Tools
This semester at MIIS, a lot of focus was put on learning how to use different Computer Assisted Translation tools and Translation Management Systems, as well as developing skills to learn any programs we might meet in our work.
Our most ambitious project by far was training a SMT Engine using the Microsoft Translator Hub. Our group decided to create an engine that would be able to machine translate Playstation system instruction manuals. To look at our project in detail, you can check out our statement of work and our presentation on lessons learned here, but it will suffice to say that the scale of the project greatly exceeded our initial expectations. We had to use PDF versions of various Playstation manuals to align, and the English versions differed from the Japanese versions at many points. Hours were spent on just alignment.
But eventually we ended up with a decent engine! With a larger library of correctly aligned files and use of human post-editing, our group decided that an engine for Playstation manuals could be feasible with the proper up-stream preparation.
We also experimented a bit individually with ModernMT, and open-source translation engine option. This was a pretty interesting experiment that required installing Linux and a lot of troubleshooting for things I had never encountered before.
All things considered, I was pretty pleased with the results! By the end of it, I had a working engine, and the Linux command line did make me feel rather hacker-y.
Lastly, we each made an instructional video for any tool we choose that could help in a localization workflow. I chose to do mine on the “Automatic Domain Terminology Extraction System ‘Gensen Web'” developed by the University of Tokyo Library System — a tool for extracting terminology from Japanese documents, including PDF. It’s a pretty simple tool, but if you want a head start on working it, please check out my instructional video found here.
I hope that the different guides that I’ve posted this year have helped you out! I don’t plan on posting much over the summer, but if something comes up, I may. Looking forward to learning and sharing more in fall!
Mini-Portfolio for TMS
Many of my posts these past few months came about through the Translation Management Systems course at the Middlebury Institute of International Studies, taught by Adam Wooten. My most recent post comparing GlobalLink to XTM is part of that! As well as a few professional presentations, including my posts on a few TMS recommendations and the needs of a TMS for crowdsourcing. This post will be the last of that series, and so will detail our largest project: a group, test study of using the the SDL WorldServer translation management system on a repeatable job, with the goal of exploring automation possibilities and how to manage repeatable tasks.
Our group decided to use the changelog for LinkedIn’s Ads API for our project, under the assumption that it would be something that could reasonably be repeated frequently as LinkedIn came out with more updates, and would have fairly regular language that could be automated easily.
Our first step was to create a proposal, or statement of work, detailing our idea and all the things we’d need to do to accomplish it. We had a little fun with the name of our mock company; this course’s initials are TMS, and our professor’s daughter famously likes snakes. Thus, “TMS: Tiny Mewing Snakes” was born.
You can find all of the files for the project here, including the Deliverables, our presentation on the lessons learned, and our statement of work.
Translation Management Systems: How does GlobalLink measure up?
Today I had the pleasure of participating in an introductory course of TransPerfect’s new TMS: GlobalLink. Besides the swanky certificate they gave me for completing the course, the folks from TransPerfect gave us a two-hour demonstration on GlobalLink to let us know exactly what their new product is about.
Behind all the talk about all the different features they were implementing in GlobalLink, it did seem like TransPerfect was addressing all the problems that I (and my colleagues) have been complaining about for what feels like ages. Their Query system for translators to get their questions answered while never leaving the TMS; the integration with various APIs; quotes auto-generated; parsers; support for on and offline work; the live previews of target documents; and the planned support of tracked changes at a segment level.
I was impressed.
The next closest TMS that I can compare it to is XTM Cloud — a TMS that I hadn’t touched on yet, but recently was able to try out for a short project. XTM also promises automation, translation and review in context, and a cloud-based system.
I enjoyed using XTM, and their pricing packages are competitive, offering trials and different pricing. It’s easy to use, flexible, and supports mobile. A case could be made for both TMSs, and you should certainly take a look at both thoroughly before deciding.
However, there is one thing that GlobalLink offers that I thought was truly innovative and exciting: modules.
The idea of picking modules for different features is very intuitive, and promises to match well with the current agile development trend. Of course, I’ll have to wait to get my hands on it before making a verdict, but that combined with the already impressive suite of features that GlobalLink offers makes it a TMS to watch closely.