The Shortest Path from Thoughts to Words (Premium)

Some transitions happen instantly, but others require time. And in my case, switching writing tools took a long, long time.

So let’s go back in time.

? Oh, the Amiga, I miss you so

As an Amiga user in the late 1980s and early 1990s, I was painfully aware of the growing divide between the personal computing platform I loved so dearly and the real world that was dominated by the inferior MS-DOS and then Windows. This divide reared its ugly head in different ways, but the quality and diversity of productivity applications was key among them. Where the Amiga was viewed mostly as a great gaming system, the IBM PC and its clones were more focused on productivity.

There were exceptions. The Amiga experienced a brief moment in the sun thanks to what we now call the creator market with graphics apps like DeluxePaint and Photon Paint and, of course, video solutions like the Video Toaster. And an even briefer moment, like a stolen kiss, with the mainstream when WordPerfect arrived on the Amiga in 1987 … and then quietly disappeared within a few short years.

I wasn’t a writer at this time, and it’s not coincidental that my shift from whatever I was then to what I became–a software developer and then a writer–happened as I shifted from the Amiga to the PC. Or, maybe it is coincidental. By that time, Commodore, the maker of the Amiga, had filed for bankruptcy, and the future of the platform, always in doubt, was even more questionable.

But even in my pre-career days, I understood the importance of writing, and the personal computing era brought with it a sense of permanence, where the content I created, be it words or pictures, might carry forward with me and not be ephemeral. This was a big deal to me. As a child and teenager, I was an artist, but I also created comic books that ranged from the serious–adaptations of Star Wars inspired by how bad the official Marvel adaptations were–to the more comical. I also wrote stories, using just words. Most of what I created was on paper and couldn’t be duplicated, not in any meaningful way.

The writing tools I was exposed to–and could afford–were primitive, and they remained that way for a long time. In this era, electronic tools existed for two reasons: You could save your work and return to it later. And you could print it out. The end game at this time was always paper.

My first word processing application was likely PaperClip on my Commodore 64 in the early 1980s, that’s a bit fuzzy, but then the powerful MacWrite-influenced GeoWrite that strained the capabilities of that system; in both cases, the output was to an Okimate 20 color printer. When I lost my C64C in a house fire, I moved on, briefly, to an Apple IIGS and its AppleWorks GS app, and I would print the results to the Apple ImageWriter II printer I had gotten with that computer at great expense.

The II GS was a financial disaster, so I moved from that to a used Amiga 500 computer. I couldn’t afford WordPerfect on the Amiga, and so I pushed forward with a series of lackluster word processors, the best of which was either Excellence or Final Writer, though I have a vague memory of the latter app looking and working much better on higher resolution displays than my little 500 was capable of.

This was the early 1990s. I had just gotten married, and my wife was working at American Express as a writer in employee communications. She used an IBM PC–a real IBM PC, not a clone–at work. She used my Amiga occasionally, but wanted to get a PC at home for all the obvious reasons, one of which was that none of the apps she used worked on the Amiga. But IBM re-entered the home market in 1990 with its PS/1 line of PCs, which it sold at Sears. This was a low-end PC, with a 10 MHz Intel 80286 processor, 512 KB of RAM, a tiny monochrome display, and a single 1.44-inch floppy. But it could run WordPerfect for MS-DOS, which is what she used at work. And so that was my first real experience with that app.

?️ The PC era begins

By the time we moved to Phoenix in 1993, my wife had sold the PS/1, and I had moved over to a newer but slightly less useful Amiga 600. I went back to school for software development there, and it was obvious I’d need to move to a PC, that’s where everything was happening. I took a CSE100 class in the summer of 1993, and it required the use of Turbo Pascal for DOS. I found a version of Pascal that emulated Turbo Pascal on the Amiga, and it worked fine for all but the final project, which required some Turbo Pascal feature that this Amiga version didn’t support. So I got that project done using a PC at the school’s computer lab. And then started pricing out a low-end 386SX clone.

Before she landed her first big job with a publishing company in Phoenix, setting her on her own career path, my wife worked for a company called Computer Prep that published software documentation for end users. And one day, she came home, telling me that they had gotten a beta of the next version of Microsoft Office for Windows that included Microsoft Word 6.0. Did she want me to bring home a copy so I could check it out?

Why yes. Yes, I did want that. By this point, I had experienced Word 2.0 on the school’s computers, and it was head and shoulders above anything I’d seen on the Amiga. Word 6.0 was similarly impressive. This seemed like the way forward.

Or, I should say, this was the way forward. WordPerfect was already imploding by this point, having shipped a terrible first version of its word processor and correcting that mistake but never fully recovering in the market. Lotus acquired a word processor called Ami, renamed it to Ami Pro, and never really made any headway. And by the time I started working on and then writing books–not too long later, really, late 1993 and into early 1994–Microsoft Word was the standard. It was what the publisher expected, and demanded. But it was also the best product in the market, and it was on its way to becoming the dominant word processor.

And that was that.

✍️ And … suddenly, I’m a writer

During this time, I never really considered what it meant to write electronically, to think and then put those thoughts into words, typing away on a keyboard. What I dealt with at first was the bizarre needs of our publisher, which required us to install horrible (to us) templates that I guess helped them turn our digital files into printed books. I dealt with a decidedly non-technical series of edits in which the publisher would send back heavily edited stacks of paper, with each type of edit in its own color, and we would hand-edit them in return, using our own colored pens, and then ship the whole mess back for more processing. I dealt with the shift from floppies to hard drives. I bought my first laser printer, an HP that cost about $1000 and was one of the most incredible things I’d owned to that point.

I wrote. A wrote a lot. I stored my writing in .DOC files on floppy disks and then hard drives. And I backed them to more floppy disks and then an ever-changing set of backup devices that included ZIP disks, optical disks of every kind, tape drives, and probably many I cannot remember right now. The thrill of making my words, my content, persistent had shifted to a need to make sure they stayed that way. And they would stay that way, theoretically, thanks to periodic backups.

By 1995, I had upgraded my home-built 386SX to a Dell 486 mini-tower. And the career in software development I had gone back to school for had transitioned into a career in writing. That year, I had several books in various stages of development–about Windows 95, Excel 95, Office 95, general computing for the educational market, and software development titles about Visual Basic 4, the Windows 95 APIs, and more–and this is when I had to give up on the school–which had started assigning my books to fellow students–to focus solely on writing.

And so I wrote. And kept writing. Started blogging, though we didn’t call it that yet. And the words piled up. And were backed up. And I went through each and every version of Word for Windows, and several for the Mac, over a period of 15-ish years or more. I almost added “unthinking” to the end of that sentence, but the truth is, I was deeply involved in the process. And while I did, of course, test various alternatives for writing–anyone else remember AbiWord, which debuted in the early 2000s?–Microsoft Word felt unavoidable. It continued to be a requirement of the book publishers I continued to work with through 2013, a 20-year cycle, and it was a requirement of my employer from 1999 through the end of 2014, which published Windows NT Magazine and its various permutations until print publishing disappeared for the most part.

Steven Sinofsky documents most of the history of Microsoft Office in his book Hardcore Software, and to oversimplify that story, I see it unfolding in several stages: The creation of Office as a suite, the piling on of new features because that’s what customers, reviews, and competition demanded, accusations of “bloat” of both features and UI and the impossibility of creating simpler new apps that anyone would want, the shift from focusing on individuals to focus on businesses and their needs, the magical nature of the ribbon and its ability to surface commands, and the shift to the cloud (and, post-Sinofsky, to mobile). Put more simply, Office was once built for people, but it’s now built for businesses.

? Bloat

Even before Microsoft Office and Microsoft Word became enshittified, which they absolutely are, I was feeling the pull of … something. This nagging feeling, in the back of my brain, that something was off. Roughly 20 years ago, I observed that enough though I was a professional writer, I only used a tiny percentage of the features in Microsoft Word. That observation was tied to the era of “bloat” accusations noted above, and it’s a topic Sinofsky addresses in his book: This was something many customers had told Microsoft, but the paradox was that none of them were interested in paying for an Office/Word version that did less, too. Microsoft went back and forth on this many, many times. But eventually, the conclusion was clear: All users of Office/Word only used some small subset of the overall feature set, the problem was that every customer used a different subset.

There’s a lot that goes into this, but as Office transitioned from a bundle of applications that into an integrated suite, it developed functional dependencies that made it difficult to separate the pieces from the whole. And that trend only accelerated with time. Many forget this, but when Microsoft created Office 365, one of its key selling points was that this service included everything, and so developing constituent parts was easier for Microsoft, and using the service was easier/better for customers. For example, Office 365 could assume that Exchange Server and SharePoint were both always present, so it could develop features that used both. In the pre-cloud world of on-premises servers, that was impossible. Each customer had a different mix of servers and apps.

A side effect of this shift–one might call this bloat, I guess–is that Office, as a suite, and Word, as a standalone application, kind of lost their identities. Yes, these things still exist in whatever ways. But really they are just parts of a giant conglomerate offering that’s now called Microsoft 365. And this thing, Microsoft 365, has the same benefits and downsides to its predecessors. You get it all, so there’s value there. But you get all this stuff you don’t need, and that’s distracting or, worse, a net negative. I’m a writer, so I barely ever had any need for apps like PowerPoint or Excel back when Office was just a suite. But today, I’m a writer, and I have even less need for most of the apps and services that comprise Microsoft 365. And there is a ton of it.

I’m a writer. I want to write.

I want to write in the most efficient way possible. I want to write without distractions, and I want to write and store those things I write–and back them up, I guess–however I see fit. I want what I had wanted 20 years ago, a version of Word that is composed only of those features that I need. This was a version of Word that Microsoft concluded it could not build 20 years ago. And it is a version of Microsoft 365 that Microsoft cannot or will not build today. And I can admit there are good reasons for that, acceptable and logical reasons. But I can also choose not to play this game anymore. There are more choices now.

And for my needs, there are better choices than Microsoft Word. For this reason, I’ve transitioned to something else.

? Text

That transition took years. This makes some sense: I’m a writer and the tools I use are important to me. But so, too, are the documents I create. And Word, for all its issues, is ubiquitous, as is its document format. And that format is now literally open. That means that any app or service can adopt and use it. And that was the first opening. I could keep using the Word document format, now called DOCX, but use another app. One that doesn’t badger me into using Microsoft’s cloud storage service, which I like, or into a specific storage scheme, which I very much do not like. For example, LibreOffice Writer.

When I shifted from publishing books through a publisher to publishing them independently on my own, I shifted to writing those books in Word to writing them in Markdown, a markup language that is similar to HTML or XML but simpler and, to my eyes, easier to read. At some point, it became clear to me that using a single tool for writing would be ideal, but I was still using Word for other writing, such as the articles for this site and its predecessor, the SuperSite for Windows. And so I would move in and out of using Markdown for all my writing. For a while, I would land back in Word’s lap because of familiarity, self-loathing, whatever. I don’t know. But then I finally crossed over completely. Looking through my archives, it’s difficult to pinpoint the exact date this happened. But I guess it was about two years ago.

Tied to this transition is the tool I used to write. And this was part of the problem: Markdown was created in the Apple ecosystem by fans of Apple and critics of Microsoft, and it was slow to come to my platform of choice. The tool I settled on in the early days of 2013 and continued using for almost a decade is a long outdated app called MarkdownPad 2 that supports a side-by-side interface in which one side is used to write Markdown code and the other displays a preview of the output. Writing in code was vaguely OK to me, thanks in part to my developer background, and it helped me learn Markdown quite well. But this was never going to be a mainstream tool. And besides, it hasn’t been updated in many years. (I still store this app’s installer, the license code I paid for, and the out-of-date library files it needs … just in case. OCD is a powerful force.)

But I’ve found two Markdown apps that make using this open, text-based document format much more usable.

The first is called iA Writer. It’s native to the Mac, of course it is, and it wasn’t available on Windows when I first encountered it, limiting its appeal. iA eventually released a version on Windows, but it was lackluster, and nothing like the more full-featured and better performing Mac version. Then iA announced a complete rewrite with iA Writer 2.0 for Windows, and though that was initially exciting, the final version is still a bit lackluster, sadly. It’s something I wish I could use, and it’s more of a minimalist, code-based experience, which, again, I like. For reasons. More on this in a moment.

The second is called Typora, and it is by far, the more mainstream option. Typora bridges the Markdown/word processor divide by providing a lightweight, minimalist word processing experience where you use shortcuts like Ctrl + B for bold and Ctrl + K to insert a hyperlink, so it’s a familiar stepping stone from Word. You can write in a source code mode if you want, but I find it easier to just write. I’ve grown to love Typora. It’s basically what I’ve been asking for, for about 20 years now. I’m writing this article in Typora, as I do with almost everything else I write. (But not, oddly, the books. I still use two different tools for writing, and these days I write my books using Visual Studio Code, which, go figure, offers a side-by-side Markdown writing/preview experience like that of MarkdownPad 2.) Someday.

? The formatting fetish

I follow the iA blog, and they recently published a post, Markdown and the Slow Fade of the Formatting Fetish, that triggered this post. This is a fascinating read, even if you don’t care about iA or Markdown, as it touches on the history of Microsoft Word a bit and makes the argument that Word 2.0 for Windows, from 1991, is arguably “the graphically cleanest version of Word ever.”

“Word is geared toward making office workers use their time with bad design decisions,” it claims. “As the screen became bigger [Microsoft] added more and more useless garbage around the text. Buttons, Rulers, Chatbots, and, now AI design assistants.” It then compares the underlying file that’s created when you create a document with just two words (“Hello World”). The Markdown file is plain text, so it’s just two words. The Word .docx file is an astonishing collection of files, many of them XML–DOCX files are basically ZIP file containers–that contain an even more astonishing amount of code. iA doesn’t do this, but on my PC that Markdown file is just 11 bytes on disk (11 bytes!) while the Word document is 13 KB (13,000 bytes). It’s over 1000 times bigger.

I don’t use Markdown because of disk space. Indeed, I couldn’t really use Markdown completely until I found tools that solved the problems I have with Word, which are all about UX and annoyances. But this iA blog post was a nice reminder of why Markdown is so great tied to some reasons I hadn’t previously considered. That is, Markdown isn’t just a “format,” it is in many ways a rejection of formatting.

“Today’s obsession with formatting, the omnipresent productivity theater, and bloated user interfaces are the legacy of Microsoft’s strategic control over document formats,” iA writes. “It paid off handsomely for the company–but at the cost of cluttered software and billions of distracted, poorly designed documents that no one wants to read.”

And while Office, Word, and the DOCX document format are still ubiquitous and even dominant, it’s interesting to note how Markdown has spread in an almost viral way across the Internet. It’s everywhere. It’s used by Discord, Google Chat, Slack, Trello, and even Microsoft Teams. It’s used by GitHub, Visual Studio Code, and every AI chatbot in the market. And it’s used in popular end-user productivity apps like Notion, which I love, and Craft. Even Google Docs uses Markdown keyboard shortcuts now, though not Markdown as a document format.

As I read that post, it occurred that the battle I’d been fighting all those years was really about wanting to write using a tool that didn’t get in my way. That it wasn’t the presence of all those Word features I’d never use that were problematic, it was the heavy-handed way that Word promotes not just other Microsoft products and services but also the optimal way to use them for Microsoft’s good. Bloat is one thing. But Word and Office aren’t just bloated, they’re enshittified. They’ve been undermined to serve the greater bad.

Markdown’s agenda is as clean as its code. It’s there so I can write. That’s it. Well, that’s not it, it’s also there to be in plain text that’s human and machine-readable and always will be. It’s like a colonic for my soul. It lets me focus on the thing that I want to do. And not be bogged down by those things that Microsoft, in this case, wants.

Or as iA says, “Markdown helps us focus on what we want to say instead of how it looks. It gives us more control over our writing–and over the digital tools we use.”

Exactly right.