How does language service LinguaSys work?

If you’re into basketball, no doubt you’ve heard about Mark Cuban, the billionaire owner of the Dallas Mavericks (who won the NBA Finals two years ago, credit to an outstanding Dirk Nowitzki, but that’s a different story). Cuban is the main reason I learned about LinguaSys.

LinguaSys specialises in products and services that help international businesses understand different languages from across the globe. Its main product offering centres around customised ‘big data’ solutions for businesses that use multiple languages:

  1. Carabao Machine Translation – LinguaSys uses a software engine integrating statistical data and rules-based data to convert text in one language to another (i.e. English to Mandarin or Hindi to English). This automated translation process can be applied to both lengthy texts and shorter communications like text messages or emails.
  2. Text analysis – Text analysis to me sounds like a very broad and complex area. Terms like ‘discourse extraction’ and ‘entity extraction’ do dazzle me a bit but it effectively comes down to LinguaSys being able to take unstructured text in any language, convert it into a language-neutral attributes and then process this into the target language (see Fig. 1 below for a visual overview of this process).
  3. Natural Language User Interface (NLUI) – With NLUI linguistic phenomena such as verbs, phrases and clauses act as user interface controls for creating and modifying data in software applications. LinguaSys’ NLUI offering enables people to carry out tasks online or use software in their own language without language related constraints. It was interesting to see that LinguaSys have now extended this NLUI capability to speech enabled mobile apps and social media networks too.

Main learning point: it seems to me that if a business wants to act globally, the value of services like LinguaSys is almost a given. The more that a business operates in different markets and making its (digital) services available to customers around the globe, solutions around translation and text analysis quickly become unmissable tools. I have therefore no doubt in my mind that Mark Cuban’s investment in LinguaSys will eventually pay off…

Fig. 1 – Visual overview of LinguaSys’ text analysis process

Related links for further learning:

Book review: “Requirements by collaboration”

I guess I’m not alone when I say that a large number of the workshops that I’ve facilitated in the past have been based on gut feel and some sort of practical, logical approach to getting the desired outcomes. It was therefore good to read “Requirements by collaboration”. Agile practitioner Ellen Gottesdiener published this book in 2002. It provides some very useful pointers to anyone looking to facilitate a requirements gathering workshop, concentrating heavily on best ways to get the most out of your workshop participants.

The main problem that I’ve found with requirements gathering for digital products / software is to find a good balance between thinking about technical viability on the one hand and addressing customer needs on the other. I believe it’s critical to take user needs (even those needs that users might not yet be aware of) as a starting point for thinking about use cases and requirements.

However, I believe it’s equally important to keep sight of technical viability and the incremental nature of software development. This means that it’s very hard to fully predict what the final product is going to look like or how exactly it’s going to function, but one can still foresee – even on a high-level – the technical complexity or risks involved in a specific product. In short, both the user element and the technical aspect need to be taken into account when gathering requirements.

“Requirements by collaboration” by no means provides the holy grail towards resolving all the common problems with regard to gathering and prioritising requirements (nor does it promise to). However, the book does go a long way in providing useful ingredients of a successful requirements workshop:

  1. Create a shared purpose – Spending time drafting a purpose statement pior to the workshop is likely to be of great value when it comes to both planning the workshop structure and instilling a shared purpose among your workshop participants.
  2. Questions to ask as part of the workshop preparation – Thinking upfront about your workshop purpose and desired outputs means that some of the following questions are bound to creep up. Who has a stake in the product? Which user problems is the product looking to solve? Who will interact with the product or software? I like the way Gottesdiener in which positions these questions against a context of ‘ending with the beginning’. She subsequently applies the ‘plan-do-check-adjust’ approach to thinking about desired workshop outputs first.
  3. Techniques and deliverables to match your purpose – Whilst Gottesdiener sometimes gets a bit too academic for my liking, the book does provide a wide range of practical techniques and deliverables to consider in relation to the workshop objective(s) you’re trying to achieve. Techniques like “the parking lot” (to deal with topics that are relevant to the project but not to current activities) or suggestions on how to best deal with ‘difficult participants’ are all very useful in thinking about ways to get the most out of your workshop.

Main learning point: I found “Requirements by collaboration” particularly helpful in the way it focuses on creating a clear purpose statement for a workshop and practical workshop deliverables to aim for. The way in which Gottesdiener weaves workshop purpose, structure, participants and outcomes together is very useful. At the times the book tends to get a bit too ‘academic’ for my liking but it nevertheless gives readers plenty of practical techniques and suggestions to run with.

Related links for further learning: