It's hard to ignore no-code movement. It's getting bigger every year.
Lubo on no-code low-code automation
What's the no-code promise? It is a simple one. Build anything without having to write a single line of code. If you can click, drag and drop than you are all set to start with no-code.
This is presented as the opposite to coding. Allegedly, writing the code is difficult. Such skill takes years to master.
One can think that suddenly, people can build production-ready software without the need of a developer. Days, when businesses had to pay (handsomely) for a programmer, are gone. So it seems.
No-code fans try hard to avoid any resemblance with programming. Although, sometimes they refer to no-code as a form of visual programming.
You see the allure, right? Visual equals to sexy. Not visual - i.e. writing the code - definitely not sexy. Nor easy.
Building a functionality using visual means - or GUI, if you will - is not a new concept. Think of FrontPage, a web builder which was popular many decades ago.
For non-programmers the no-code approach looks like great idea. But is it any good for businesses?
Although I am great fan of empowering regular users, in many cases no-code tools are wrong tools selected for the job. Let me explain.
There is a good reason why businesses who need professional results hire professionals to help them.
I checked several popular no-code tools, mainly in data automation area, and I was surprised how simple and shallow they were.
The solutions they offer are usually something like this:
Someone could even call it an ETL tool, couldn't he?
Is this the great democratization of software development? I doubt so. These are simple tasks that many people could put together using some programming language, like Python and SQL.
Instead of learning no-code using proprietary tools they have no control over, learning low-code is more useful for users and their employers alike.
Does anybody still remember Learn to code movement propagated by presidents and celebrities?
Maybe it is satisfying when someone can point and click to create a record in a database automagically. But it seems so inefficient.
Compare it with learning basic concepts of some programming language. It's not that difficult and many languages available today read like plain English.
Internet is full of code ready for copying and pasting. What a great source of wisdom (and rubbish sometimes).
What I see as a main disadvantage of no-code though is dependency on the vendors.
I am amazed when I read about businesses and individuals who have woven together some elaborate process using several different no-code apps which may or may not be around in a few months.
Wouldn't it be better to invest time in gleaning some basic, low-code programming skills? Or while we are at it, hiring a developer to write a code that your business will own?
I see no-code tools as limiting. They can't do what they have not been programmed for. If there is a functionality you need but the tool doesn't have it, you are out of luck.
Well, you can ask a vendor to develop it for you. Let me know how long it takes.
In some cases it seems that vendors sell SaaS but call it no-code for marketing reasons. This resembles other hypes we've been through already.
Note: Data21.io falls into this trap too. We call this (with tongue in cheek) a no-code tool.
The reality is, many business owners and managers won't learn the no-code tools anyway. They hire some no-code specialist to do it for them. The question is whether a hobbyist who knows how to no-code is the best person fit for a job.
But maybe I see it in a bad light. Maybe it's OK to have non-experienced people building tools for businesses, public sector and figuratively for all of us. Maybe we shouldn't expect too much. Maybe we should be grateful we have so many (although imperfect) choices.