Not invented here

In the world of software a copy comes at almost no cost (yes, I know this is an oversimplification). Still engineers (and managers) go to great lengths to reinvent/recreate things that already exist instead of copying them. In fact this behaviour is so common that there is a name for it: “not invented here!” And it is so widespread that there is a comic by that name. Plus: a friend of mine asked me to write about it.

Not invented here! But why?

Here are my theories on why this phenomenon is so common:

National culture

In western Europe we learn already in school that copying is cheating. And cheating is bad!

Corporate culture

Some companies are proud of their clever ideas. It may actually be true that clever ideas were what made them successful. Or it is a case of survivors bias. Anyway. Clever and out of the box thinking is held in high regard in these companies. That’s why they apply it everywhere. A hint of such a situation might be an expression like “at this company we do things our own way!”

Dead horses

Some inventions are good in the beginning but turn sour later. It is just that nobody takes notice. Or dares to say something. Voila a dead horse! Read my earlier post for details.

The longer an own invention ist pursued, there more is at stake. Not only for the company, but also for the manager responsible. Eventually, the project becomes “too big to fail”. You could call it a case of sunk cost fallacy.

Laziness

‘What?’ you might ask. ‘Isn’t it easier to copy something than to build it yourself?’ Well, in IT to copy something you often have to delve into a different area of knowledge. God forbid you might actually have to learn something. For some people it is just easier to toil away a couple of hours than to leave their beaten paths of thought. That is especially true if the hours are paid for.

On the shoulders of giants

“If I have seen further, it is by standing on the shoulders of giants.” (Isaac Newton 1676)

If Newton stood on the shoulders of giants these people are standing on the shoulders of compiler writers. Maybe that was a vantage point in the 80s. Nowadays it is impending doom! So what can you do about it? Well, that depends on whether you are a manager or a contributor.

What to do as a manager?

Know your engineers

When you know your engineers a bit you get a feeling of how prone each one of them is to a “not invented here”. Keep a close eye on the “inventors”.

Make your engineers discuss it

It is often helpful to have your engineers discuss whether a “do it yourself” project is necessary. If you pay attention you will get a good pro and con list out of it. Plus: you make clear that you allow thinking about copying.

We are the only ones!

When your engineers claim there is nobody to copy from ask yourself: how common is the problem we are trying to solve? Can it even be that there is nobody else in a similar situation?

Ask for proof

If your engineers claim they could not find traces of others with the same problem ask them to describe their search. Try a search on your own. This makes clear that you are ready to call a bluff.

Is it a USP?

When I worked at Philips the company had an interesting policy: Philips did develop their own tools. But they were only maintained as long as no similar tools existed on the market. The thinking behind it: invest where it gives you an advantage over your competition. Copy where it doesn’t. So if presented a “not invented here” project ask yourself: will this lead to a unique selling proposition (USP)?

In case the project is already under way

All of the above assumes that a “not invented here” project is just about to start. In case it is already underway remember one thing:

It is only going to get worse!

This is not only true from Murphy’s perspective. If a project is expensive now it will not miraculously become cheap. Instead you will lose money every single day! So remember a saying of our former company lawyer:

If you have to commit an atrocity do it right away!

What to do as an engineer?

How invested is your manager?

Firstly I recommend to make an educated guess on how your manager will react. Would she dare to kill the project? Or is he too concerned about his reputation? In other words: would you face a fact oriented discussion or something else entirely. If you think your manager is open to facts make a solid case and present it to him.

If your manager is biased you could consider skipping a level and talk to your managers manager. In that case go to your manager first and if he declines go one level higher. Be aware of what you are doing, though. This move can easily backfire! And even if it works it will leave a sore spot with your boss. So you better be sure!

Convince colleagues!

If you feel that you would not stand a chance alone, you might try to rally a party. Present your thoughts to your colleagues. If you can convince them you can invite your boss together. Always make sure you have a solid case. A compromise could be to suggest to your boss a pilot project requiring limited resources.

What if?

If neither your management nor your colleagues buy in you should ask yourself: did they convince you about the necessity of inventing here? I so then everything is fine. If not it is one more bitter pill to swallow. A Kikentai Management thought: Is your walk away line reached yet?

Photo by criography