Saturday, May 11, 2013

Saying No is cheap

Although it sounds logical, there is a danger in standardizing in software. Since I started worked on software automation, modelling and software factories I have had very different reactions.
The positive ones are like this:

  • Wow, this cuts down development time with at least 75%
  • So you are saying we don't have to redevelop when technology advances?
  • Great, so we can specialize development roles and make everybody way more productive in their own tasks
But the negative ones more or less say:
  • This is not standard software you are using (or it is not Microsoft, Red Hat, IBM, Google, Apple, ....)
  • You have great tools and we would like to use it, but you are not on our preferred suppliers list
  • Great, but I am not sure if we have the technical knowledge to be able to understand the open source software stack you are using so we better stick to something we can ask any development company to do for us
Prove it
We have been hearing this for almost 20 years now, and in the beginning there was also some talk about the bigger companies coming up with something like that soon. That never really happened, but the other arguments remained. 

Even in companies were we implemented our process and tools and saved millions of euros every year and got 100% success rate on every project year after year, the discussion remained the same. 

Yes or No
The point I would like to make is the underlying philosophy when choosing 'standard' software and 'proven' tools over advanced technology: "it is better to do development like most of the industry so we are not depending on 'smart people'". I think here are three flaws with this line of reasoning:
  1. You underestimate the potential of your developers when you enable them
  2. The consequence of this argument is you prefer doing things 'not so clever' to enable everybody to do it for you....
  3. Ultimately you are waiting for your competition to be the first to do it smarter and advance the standard

Saying NO to an idea is cheap, but in the end there might be an implicit YES to something connected to that NO, that is a lot worse. Just like in politics it is easier to get a lot of people say NO to an idea, than it is to make them say YES. When you are able to get a true YES, it can change the world so we keep telling our story. Like the people on the Software Development Automation conference in a couple of weeks!