Proving a new product or feature
What’s the best way to prove a new product or feature?
Unit tests? Integration tests? Manual testing?
They are a good start, but not what I’m getting at.
How do you really know this thing you’ve created will stand up to the abuse and dangers of the real world? You know the moment your back is turned someone will find a way through your code. Your code that has 100% unit test coverage, a slew of integration tests and has had every bug and performance problem smashed out of it through numerous rounds through the best QA team in the world.
So how do you prove it works?
One idea, use it! Get other people to use it, anyone. EVERYONE! Give away free trials. Force as much real world data and usage through it as possible. Then turn the dial up to 11. See what breaks it, fix it, repeat till you’re sorry you ever had the idea to change the world for the better with your application, web site, security patch or dancing bear emoji thing.
To this end, I recently suggested we offer our new ish cloud based smtp mail security solution to anyone and everyone to send mail to.
You may have heard of disposable email addresses such as Guerrilla Mail or Mailinator. These guys essentially offer free inboxes, sometimes with no signup and completely anonymous. They get probably thousands of mails through each day, good ones, bad ones, spam and everything in between.
You might not know this, but there is a whole bunch of stuff going on in an email that the average user doesn’t see. Any of this stuff can be accidentally or deliberately manipulated to achieve different results including stopping mail flow. For an email handling system mail flow is pretty much everything.
Also different senders connect to the mail servers in different ways using different clients which may or may not adhere to the official spec.
So to recap, I suggested we put the system out there to get exposure, not just volume but types of mail, types of attachments etc.
Sometime soon we should have an endpoint out there for anyone to route mail to. I doubt it’ll be a Mailinator replacement but it’ll sure help us figure out where the weak points are and where our attention needs to be.
