The Low Code Manifesto

04.21.21 12:55 AM

Part 1 of an analysis of low code and the impact it has on the workplace. 

Low Code is coming for your job.

  

Okay, probably not. In fact, I will argue here, Low Code will make your job easier and more satisfying. That's not to say it won't be disruptive or cause some pain, like any change, but on the whole it'll make the working lives of a lot of people better. Moreover, the companies and customers they work for will be better served as a result.


First, a little about me. Don't tell my boss, but I'm a very lazy person. Maybe not the laziest you've ever met, but up there. I don't like to spend a lot of time at work. I prefer as much of my time to be leisure as possible. Shockingly, I don't see this as an unambiguous weakness. I combine my laziness with a couple other traits, so it works for me as a strength.

One of those traits is a need for continuity; I like to know where my next meal is coming from, or preferably, my next thousand meals. This counteracts my laziness because it impels me to produce output so that I can afford food now and in the future. It forces me to produce, but doesn't mandate that I have to spend all day producing, as long as I make enough in the time I spend working.

The other is a talent for efficiency; I see all the ways to make a small amount of work result in a big impact. I consistently hone my abilities and tools to maximize my leverage. I like it when I can take one small action and have it cascade into a whole bunch of output. I love it if I can build a tool that functions as a perpetual motion machine, churning out products without me ever having to lift a finger. I will spend hours building a tool that turns a daily five minute task into a daily two minute task. I will spend weeks building a tool that turns a daily five minute task into a once-a-month ten minute maintenance checkup.

 I don't like to spend a lot of time at work. I prefer as much of my time to be leisure as possible.

When I say 'tool', I’m not referring to a hammer or even a better mousetrap. I'm talking about software: code. My first tools were spreadsheets; put a chunk of data in one end, get a report out the other. When I discovered pivot tables (lookup, logic, and math formulae), it was a revelation. Take a chunk of data - in my case, lists of transactions in a retail automotive dealership - and my tools could produce a staggering variety of reports. Mostly to do with profitability and pay plans, but also touching on inventory, process efficiency, staffing, and accounting. It is hardly an exciting use-case, but it's a good example of the roots of Low Code.

The reason spreadsheets are so ubiquitous is that they're powerful and that anyone with a little time and inclination can harness that power. The principle behind them, which ultimately underpins all Low Code, is deceptively simple. A piece of data goes in a box (field), some other box contains a formula that draws on that and possibly some other data, then a logically deterministic result is produced. Other fields build on this process until you have something useful.

There are two important differences between what a spreadsheet does and what the most complicated and powerful computer programs do. First, powerful programs can do more with their results. They can produce images and sounds, send emails and texts, make phone calls, link vast networks into functional systems, fly planes, drive cars, and vacuum floors. The second important difference is accessibility: a spreadsheet can be programmed with much more ease, with significantly less training and practice than any programming language. Spreadsheets are not exactly intuitive, they can't be programmed with natural language commands, but it's only a matter of hours before you are building something useful. Any programming language takes much longer to learn before it becomes useful.

Low Code does for a large number of new use-cases what spreadsheets have historically for blocks of data and reports.

Think of it this way: a programmer, working with others, can (in theory) produce every piece of useful software you've ever used on your phone, desktop, or console. However, on the first day, while a nascent C++ programmer is getting their software to print 'Hello World!', a first-day Excel user can have a pivot table summarizing sales data for a large department. If you have five years and want to build a prosperous and interesting career, go study code. If you have a week and you want to bulk up your resume, study Excel. 

Low Code does for a large number of new use-cases what spreadsheets have historically for blocks of data and reports. It makes the basic functions of a system customizable, even automatable. Low Code has endpoints including email, document generation, database entry and editing, and many others. The really powerful thing Low Code does today is tie together bespoke tools in a variety of ways. You can see this in the internal ecosystems of tools like Salesforce, Zoho, Trackvia (my personal favorite as it's where I cut my teeth), and even in the sprawling web of tools now connected by Zapier. Even all these tools talk to each other too. The closest Low Code comes to outright software development is in relational database tools. I've already mentioned some, but there are dozens of others that provide a platform to develop nearly anything you can think of and link it to any endpoint with an API.

- To be continued in Part 2 -