Are digital writing assistants as efficient as humans?

There's been an increase in the number of digital grammar and proofreading tools, of late.

I’ve been involved with content and writing for over 30 years now, plus have an English Literature degree to boot, so I dare call myself an authority on the English language. But I will be frank, I have two digital writing aids or assistants installed on my computer. One’s for grammar, the other’s for web content optimization. The former works more like a spellcheck, the latter is too new a tool for me to get all judgemental about it (more about this one later).

To be honest, I’d rather rely on my knowledge and grasp of the English language to get by. But I leave it to my reader’s wisdom to decide whether he/she must use such writing assistants or not. (For this newsletter, I shall limit myself to writing aids for the English language.)

Here’s the thing — English is a tricky language. There’s grammar — a set of rules to go by — and then, there’s style. In between are related issues like the tone, and the voice. Except for the set of writing rules (grammar), which in the English language is also flexible and contextual, the rest are intangibles.

Straight off, today’s digital tools help you to mostly to proof-read. In fact, many that claim to be grammar check tools are nothing but glorified spell-checkers. Why? These tools have not evolved yet to “understand” the complexities of the English language — meaning grammar, or the context of your content. Inevitably, “suggestions” are incorrect, not to context or missed altogether. That’s why most of these tools have an in-built option, “Ignore in this text” (or something to that effect), leaving the final call on the writer.

Grammar cannot be treated as stand-alone, which is what many of the digital tools do. It is grammar plus usage.

I am not even getting into the area of style. Two writers can say the same thing using two different styles, and using the same set of rules that govern the English language, yet, digital grammar check tools may fault only one of them.

One more thing — if you were to use such grammar tools, chances are high that your copy may end up being sterile. And, highly sanitized. And, boring to read. Get the point?

Writing tools are still dysfunctional. Because they are unable to grasp context, they sometimes let go of obvious grammar errors. Like rightly spelled words used in the wrong context. Most of you who use such aids may have realized that they occasionally fail to point out the “difference” in the usage of “right” and “write”, for example.

To be fair, such digital writing aids have evolved over the years, and continue to do so. With the advent of machine learning, they are getting better at the game. For now, these assistants are good at treating certain types of mistakes like lexical and content errors, and not grammar. But I suspect, the intricacies of language and its nuances are such that artificial intelligence may never ever get the complete hang of it.

Where grammar, style and tone are concerned, these tools are no match for human editors. Here’s what this “dated” research paper says about such generic tools:

Generic grammar checkers have significant shortcomings. They cannot evaluate the semantics or the full syntax of the given text and as a result allow text like the following (Sanders 2000, quoted by Dodigovic 2005, 107): There are know miss steaks in my pepper be cause my word processor includes soft wear witch checks my spelling. The problem is that it doesn’t correct errors in punctuate and it will not fined words that have bean used but that are spelled write.

My Microsoft Word grammar tool could barely catch some errors in the above para, while my rather expensive writing aid failed 90%. So while Word did alert me to the fact that it should be “no” and not “know” in the above instance, and stuff like that, it did not correct “my pepper”, which should have instead been, “My paper”. If I, a human, did not know better, this piece of content would have been submitted as it is, right?

So, the question — should you use them? There’s no black and white answer. It depends on whether you are an English language expert or starter, whether you are a native speaker or non-native one, whether you are writing a letter or a thesis; things like that. I would sum it up thus — it’s a safety net just in case some things (errors) slip through the holes.

Web content optimization tools and the democratization of SEO

Writing and SEO, as I’ve often said, are two different things, but sometimes get confused for each other. If you write for the Web, your work needs to reach its intended audience. For that, your content needs to be “optimized” so that search algorithms can find it, rank it and show it off. That’s where SEO and now, web content optimization platforms come in.

The latter are similar to SEO tools but must not be confused with the average SEO on-page plugins. Today’s web content optimization tools are AI-driven, and what’s more, can be downloaded on to your computer. With the help of machine learning, they guide you on keyword research, where to use these words, help you add and rank meta tags and description, in addition to attempting to correct your grammar, even as you bang out your post.

Basically, these tools help writers to self-optimize their content for online search, not on-page, i.e. not in your website’s post but offline, too. So like Microsoft Word, you can use these tools to open a file on your PC, decide your main keyword, then write an SEO friendly headline, and the content, all in real time. You can even export your final output to your website as markdown text or as an HTML file.

More than digital writing aids, I am excited about these optimization tools. Maybe, in the coming weeks, I shall examine these tools at length for readers. For now, you could do an online search to find some of the popular brands in this line of work.

Image by Lorenzo Cafaro from Pixabay