Technical Writing: Threats to Future Employment



Image: [https://www.pexels.com/photo/wooden-robot-6069 CC0 License. Free for personal and commercial use. No attribution required]

Introduction
Most people take for granted that the skills they are gaining today will help them tomorrow, but, today, an increasing number professionals are becoming worried that, thanks to rapid advances on technology, this assumption might no longer hold true for them. Popular futurists, such as Martin Ford[1] and Marshall David Brain[34], predict the extinction of some of the world’s most stable jobs in the next 5-20 years, and many professionals are worried about facing the prospect of technological unemployment[7] in their semi-near future.

In the media, journalists and bloggers are worried about technological employment as well. Developments such as commercially available self-driving cars and freight trucks from large companies like Tesla, Inc. have caused many to predict the end of the truck driving and transportation professions. To put it in perspective, truck driving is the most abundant well paying job in North America, and the service sector that relies on them driving through their towns for business is worried about becoming unemployed right along with them [24].

However, it is not only one technology, like self-driving autos, that professionals wishing to hold on to their jobs are worried about; it is the convergence of several. Machine learning[7], artificial intelligence (mostly the success of machine learning and neural-networks[21]), 3D-printing, data science, robotics, and genetic engineering are popular examples. Many have also argued that such worries are unfounded and that technological progress will go slower than predicted or that roles will change rather than be gotten rid of. Nevertheless, since the worry of technological unemployment is tied up with the future of technology, the worry is not unfounded [24]. Even today, in 2017, many jobs could be automated with today's technology (One report from McKinsey estimates 51%.) [26]. The only thing that has kept this from happening has been the lack of informed management and IT professionals to implement automated systems. Therefore, with more capable and affordable and easy to use technology coming our way in the next few decades, professionals have more reason to worry[4].

What Specific Technology and Social Trends Could Reduce the Need to Employ Technical Writers?
While there are few comprehensive examples of a technology that is able to fill the full role of a technical writer, there are many technologies that can do parts of the job. Of course, technical writers play many different roles, so each one would be affected differently by different technologies[30][33]. However, there are many technologies that could potentially affect technical writers playing many different roles within science, content management, software, engineering, policy, and other areas.

Artificial Intelligence
In their most basic forms, algorithms that guide artificial intelligence are conceptually as simple as cooking directions, and the programming language they use is as simple as a recipe; they are sequences of directions carried out to create a desired result. However, the idea that computers can only think linearly, and are good for nothing more than following directions is also not correct. Using machine learning techniques, artificial intelligence is now able to go beyond simply carrying out instructions, and it can now do things more akin to forming concepts, concepts it can act upon. For example, in 2014, Google was able to train a machine learning algorithm (via a neural network) to recognize cat-videos on Youtube by feeding it millions of cat videos and pictures. The algorithm was not taught what cats look like. Instead, it learned more in the way that an animal does; it learned by observation and by making generalizations and mental-models to draw upon. It has developed a universal idea of what a cat face should be[21].

Similarly, in 2011, IBM's Watson[13], the computer that became the world's most successful Jeopardy show contestant and Alpha Go [37], a computer program that beat the world's best Go players, both used machine learning techniques to win their games. These programs rely heavily on statistical data and other cues to make guesses at answers to problems. According to futurist Ray Kurzweil[35] in his book 'How to Create a Mind'[36], as it turns out, animal brains use pretty similar techniques. What seems like intuition or innate knowledge is really often a lot of behind the scenes guesses based on probability and past experiences[3].

Technical writers could one day face algorithms similar to IBM and Google's game-winning computer programs. Similar programs could look at all the data needed to create a document (such as a business proposal or API documentation[38]) and through brute force computation find the probability of what makes a good guess at what the documentation should look like. Then, it would create that documentation from a body of text and code, known in computational linguistics as a a corpus [39]. Another thing to note is that such algorithms wouldn't have to be 100% accurate to pose a threat to technical writers; they would just need to be good enough to require light-editing, or, alternatively, a company may choose to simply warn users that the documentation might be slightly off because it was written by a computer.

Many modern-day sports articles, for example, are already written by computer programs, but that is in part because sports articles are more formulaic than many other news stories[23][30].

Artificial intelligence does not have to do the tasks of technical writers to be a threat to their job. For example, if it proved itself to be a good tool for replacing expensive web developers, there would no longer be a need to write documentation for web developers. The web development platform 'the Grid' is already doing this to a degree[19].

Drag and Drop Website Editors and 3D Printing
Drag and drop website builders are no longer limited to static web pages with very little functionality. Some website builders, like Bubble.is[15], are so powerful that they are considered as high-level programming languages in their own right. To get an idea of how capable Bubble is, one man who wasn't even a web developer built a fully functioning clone of the social media platform Twitter within a few days[22].

One day soon, with the help of artificial intelligence, website builders might become so easy to use that you need only describe and draw what web interface or software you want, and it could instantly create it for you. 3D printing could have a similar story in the future as customers could simply print custom-made products in their home instead of buying them from factories that employ engineers, etc.[40].

Virtual Reality and Augmented Reality
What happens to the role of the technical writer when the user can talk directly to an automated or non-automated avatar on demand in a highly intuitive manner or setting? For example, if you were trying to find an obscure setting on your phone, would you rather flip through a manual, or would you rather put on a VR headset to walk (virtually) into a room that was social media environment (like a subreddit on Reddit or a group on Facebook) and ask around for a few minutes until someone showed you how to find the setting? Meanwhile, you could have a fun casual conversation with the person helping you. What would you prefer[43][41]?

Augmented reality offers similar opportunities for users. Microsoft Hololense, for example, allows you to draw over a live video feed of someone else's perspective while they are also wearing an augmented reality headset. This allows people to give very clear visual instructions on how to complete tasks such as fixing a bathroom sink. If this kind of technology became the norm, it would invite more crowdsourced solutions and may reduce the need for audience-friendly documentation.

Chatbots, Virtual Assistants, and Computational Linguistics
As chatbots improve with the help of artificial intelligence and the growing science of computational linguistics[12], the more they will be capable of answering people's questions through smartphone apps or websites. Chatbots offer a quick intuitive way to find information via text or voice. Amazon's Alexa, Apple's Siri are examples that are growing in popularity as the technology gets better and better[42].

Outsourcing and Crowdsourcing
As technology and education improve worldwide, it will become more attractive for employers to engage in ‘arbitrage’[8], the opportunity to buy a commodity or service at a lower cost and use it in a market that will provide a higher return on the investment. Similarly, employers will find opportunities in virtual reality to expand pay-per-task 'microtasking' opportunities to the public as a way to get work done while avoiding the need to pay full-time employees and provide them benefits. Currently, over a million new jobs are outsourced from the united states every year[27], and crowdsourcing tasks are available in the hundreds of millions[41].

A More Educated Audience
The more users know about the technology they use, the less they need help troubleshooting. For example, how many people need to read a manual before they can use a remote control? Not many. But, when the remote control first came out, even though it only had a few buttons, people poured over their manuals to get them to work. What is considered ‘technical’ is changing.

The Role Vs. The Tasks
The idea that a technical communicator can be replaced by technology, or other means, brings into question the role of the technical writer. While it is safe to say that tasks that technical writers currently do may, in the future, be done by someone else or by technology, it is not safe to say that they can be replaced altogether[2]. A technical communicator is first and foremost a 'communicator', and the role of a communicator is to make the transfer of information between people go as smoothly as possible. Therefore, this means that their role may be flexible enough to change along with the new technology adopted [5]. For example, being a manager and designer of chatbots could be a job done by future technical writers, or if crowdsourcing and virtual reality become popular, they could be the 'hosts' in online environments for user support.

Tips for Staying Employed as a Technical Writer in the Next Few Decades
Do more and more technical work. The closer you are to being an expert yourself on a particular topic, the less help you will require from coworkers. Obtain multiple degrees and minors in your subject area. Not only does this make you attractive as a technical writer, but you can also be poised to switch careers if necessary to project manager or go become a developer or entrepreneur, or go into user testing, for example.

One last thought to ease the worry of technological unemployment is that these technologies might affect more than just one profession's employment [32]. If for example, artificial intelligence was good enough, we might start living in a technological utopia where there was an abundance of resources for everyone, and there would be no immediate need for formal positions of work to provide for oneself. Formal work would become entirely optional. However, this is only one possibility out of many of the confusing and exciting events that will make up the future.

References list from migrated article
1. Ford, Martin. Rise of the Robots Technology and the Threat of a Jobless Future. New York: Basic Books, 2015. EPUB file.

2. Carl Benedikt Freya, Michael A.Osborneb. Technological Forecasting and Social Change. Volume 114, January 2017, Pages 254-280 (The future of employment: How susceptible are jobs to computerization?)

3. Judy Wajcman. Automation: is it really different this time? ISSN: 0007-1315, 1468-4446; DOI: 10.1111/1468-4446.12239 The British journal of sociology. , 2017, Vol.68(1), p.119-127

4. Frey, C.B. and Osborne, M.A. 2015 ‘Technology at Work: The Future of Innovation and Employment’, Citi GPS: Global Perspective & Solutions, February. (https://www.oxfordmartin.ox.ac.uk/publications/view/1883)

5. Hayhoe, George F. Technical Communication, Volume 54, Number 3, August 2007, pp. 281-282(2). Society for Technical Communication link: http://www.ingentaconnect.com/contentone/stc/tc/2007/00000054/00000003/art00001