Whether we’re talking Office or Windows, I’ve often encountered situations where open source advocates will stand, pitchfork in hand, and denounce the evil and cost of Microsoft software, and the ill intent of Microsoft.

Sure, The Firm does everything they can to keep customers on the proprietary rails of Office and Windows. But there’s more to it than that.

Organizations spend a lot on Office. That’s painfully clear to me, due to our work educating businesses on Microsoft enterprise licensing.

I also get that a lot of people get hung up on Microsoft as evil, regardless of what the company does. For example, in a recent Twitter conversation, a European gentleman (a pretty outspoken open source advocate, it seems) was complaining that his country’s higher-education entities had selected Office, and denounced the cost of it (even though the suite is technically “free” for students in many cases when faculty and staff are properly licensed for it. His point – why pay for Office, when alternatives exist, and why use Office documents, when alternatives exist?

It’s not that easy. Office may be evil, particularly in the eyes of open source advocates. (After all, while Microsoft has embraced open source in some areas, Office is not, nor should people ever expect it to be opened. )

But Office does something more important than that. Actually, it does a couple of things.

In particular, Office enables:

1. Document fidelity
2. User interface familiarity

What do I mean by these? Let’s start with document fidelity.

Document fidelity means that – I know that if I use a recent version of Office and you use a recent version of Office, I can create a document and send it to you, knowing you can read and edit it without losing the content meaning that I intended.

I can take advantage of new features in Word, and know you should be able to see them, or the document will often degrade gracefully but still let you read the basics of the document.

For most of the time that the suite has existed, Office for the Mac has been feature deficient when you compare it, “checkbox to checkbox” with Office on Windows. To be fair, there are some number of features that are exclusive to Office for the Mac. But in general, if you want the closest thing to the kitchen sink, you need to run Office on Windows – and now, that means you need to subscribe to Office 365 Enterprise E3+, on a Per User basis, and run Office 365 ProPlus (the subscriptified Office suite that gets new features chucked out to it on a regular basis). But even if the Mac version doesn’t support the entire set of capabilities, it can usually open documents from Windows, edit them, and not harm the document.

Microsoft has tried for a long time to enable fidelity when documents are shared inter-version and now inter-platform. So I know that if I’m on an iPad and you’re on Windows but we both are licensed for ProPlus, or if you’re on a Mac and I’m on Windows, but you’re running 2013, odds are that we can still exchange these documents back and forth, and not completely break the structure of the document on each other. Sure, it happens, but it’s something I believe Microsoft tries to avoid happening.

When you loop office documents into third-party editing tools, all bets are off in terms of what formatting will make it, and how the document will appear to you. Will that table still be transparent? Can you see the images I inserted? Office documents aren’t just margined words and images in doc files anymore.

Now how about user interface familiarity?

Let me start with a story… I moved to the Seattle area in 1997 to work for Microsoft. Not long after I got here, I started volunteering at Seattle Jobs Initiative. At the time, this group worked to educate low/no income individuals on Microsoft Office skills in the hopes that it would help them land a career. I saw it do just that on many occasions. You see, while the domination of Office feels like a discrete evil to some, understanding how it works and being able to make the most of it can help people land jobs and increase their worth to an employer. I get it… it’s not an easy line to draw.

The open source advocate I mentioned above, and a few others I’ve seen recently, seem to believe that using Office and using an open source Office alternative are interchangeable, or sustainable. But the reality is that even in 2019, I don’t think you can pull a “Folgers Crystals” stunt and secretly replace Office on a knowledge worker’s PC and expect them to a) seamlessly adapt, b) easily adapt, or c) not piss and moan at some key long-tail Microsoft Office feature that is widely used by a segment of knowledge workers, but which is poorly implemented by all open source Office alternatives.

What do I mean by “long-tail” features? I mean the massive set of features that exist in Office that Microsoft added over the last 30 years, most of which aren’t supported by alternative office suites, and are only used by a fraction of Office users. No

With our licensing boot camps, we often run into customers who are using (or who have “friends” who are using) Office or other Microsoft software in ways it’s not supported, and usually not licensable. Rarely does Microsoft design in code that prevents you from shooting yourself in the foot from a licensing perspective. So for a couple of scenarios, we’ve tried to think up alternatives. These are the scenarios:

1. Office server-based automation (building or parsing Word or Excel documents on the fly without an interactive user)
2. Robotic Process Automation (RPA) – similar idea, but with an interactive user that’s not really a user – just semi-intelligent code running as a user.

There is no licensing model for Microsoft technology that can be applied to address either of these scenarios, and no technological model that can be applied to either safely. It’s always duct tape.

I’ve tried recently to see if there was an open source alternative suite that we could recommend customers try to use instead. But frankly, they’re all complete garbage. Their performance on Windows and the Mac is abysmal, and key features are missing that would probably prevent RPA or server-based automation from working as needed.

WordPerfect is effectively gone, Lotus is gone. Apple’s productivity tools are quaint, but useless for an enterprise with Office embedded in it. Google’s productivity services is interesting to consider, but you’re locking yourself into different issues instead, there’s no on-premises automation framework, and prices are rising there too – indicating pressure to make money, not just undercut Microsoft Office anymore.

So when an open source advocate says “We can give someone some open-source alternative to Office and save all that money”, it feels like a Brexit Bus to me. There’s a ton of promise, but the wheels will fall off of that idea as soon as you try to implement it. I’m serious. I often joke about “mission critical Excel” or “mission critical Access”. But the reality is that a frightening amount of systems are built on ancient Excel and Access documents and applications, and aren’t easily replaceable.

So sure, you can propose replacing the technology (Office), but unless you’ve got a plan to retrain users and deal with their nuanced scenarios each one requires, while also retooling all of the embedded solutions you’ve got that are relying on Word, Excel, and Access, you’re SOL.

There’s no way you know how much these tasks will cost ahead of time, so it’s not really a fair (or really even a possible) comparison. This is where I sometimes see people propose replacing Windows with Linux… but that’s another lengthy topic for another time.

But the tweet above wasn’t the only thing that brought this topic to mind. My own recent experiences have.

You see, above, I generically talk about Office. But the reality is that for many tasks, there’s no replacement for Office on Windows.

My colleague has been delivering Microsoft licensing boot camps for over 10 years (I joined him a few years later). For obvious reasons, we use PowerPoint. Since some of our content consists of complex diagrams, Visio has historically been brought in. When I switched to the Mac, I began developing a lot of diagrams in PowerPoint itself, using PowerPoint’s own diagramming tools. PowerPoint is not Visio. But in my book, that’s a complement, not an insult.

For many of our legacy diagrams then, this meant that my colleague had to make the changes as needed – a process my colleague is fine with. However, we recently ran into an issue. You see, you can embed Visio diagrams into a PowerPoint as an image (not surprising, right?) or a Visio diagram – even on the Mac. Why would you do that? We print our slides to a PDF and then to paper for the people who attend our boot camp. Some time ago when we took the Visio diagrams and converted them to images and embedded those into the slides, we wound up with several quality issues when we went to press. So we kept on with Visio.

I recently ran into an issue (which the Office team is now aware of) where when I’d open a particular licensing boot camp deck on my Mac – one with lots of embedded Visio diagrams, and it would fatally hang PowerPoint on my Mac. Not fun. Particularly un-fun.

Consider this feature stack: Editing of .PPT format files in a current version of PowerPoint, where that .PPT file contains embedded legacy .VSD diagrams from Visio.

This is a long-tail feature inside of a long-tail feature, inside of document formats that many people don’t use anymore – but we do, because we’ve hit print quality issues when converting to the new formats too.

By trying to edit them on a Mac, I’ve created long-tail inception. Embedded Visio support within Mac Office is the kind of well-intentioned checkbox feature that someone added to PowerPoint on the Mac in 2004 or something, but since most users don’t use it, it’s very possible that new issues crop up without being discovered during a normal automated test suite… only by nerdy power users like myself or the 31 other users on the planet who might ever hit it.

So this is an incredibly long digression… what’s my point?

Not only is Microsoft Office the standard, but the more likely your organization starts to depend on particular long-tail features in Office, the more likely you need to depend not on Office, but on Office for Windows. What do I mean by that?

People may talk big about Microsoft not being dependent on Windows anymore, ever since Office landed on iOS and Android, and with the growth of Office Online. But the reality is that long-tail features mean the following is generally true in terms of features:

Office for Windows > Office for the Mac > Office on iPads and Android tablets > Office on iPhones and Android phones > Office Online

I believe what this also means is that over the next 5-10 years, if you deeply depend on long-tail features (or on Windows-specific applications like Access, Visio, or Project, none of which have received significant features in some time), you will find yourself with a growing dependence on Office for Windows.

So for my colleague and myself, this means that unless we can kick the Visio habit completely, we’ll need to keep using legacy tools to edit legacy document formats, likely with an increasing number of weird regressions nobody else ever sees.