If you look at our main product page, it’s easy to believe this is a joke of some kind. However, it’s not, and in order to prove that, I will explain the technology behind it in this article.
For years we have been researching and developed a product that creates software systems based upon an existing database. It basically works by reading “meta data” from your database, for then to generate all 4 CRUD operations towards that database in a web API backend. Meta data here implies things such as what is the name of your tables, what columns do your tables have, what types of data can your columns accept, etc. CRUD here are all four axioms required to manipulate your data and it implies Create, Read, Update and Delete.
We all know that a computer is faster than a human being. If you ask a human being what 39.64 to the power of 11 is, a human will need 30 minutes to calculate the result. A computer will be able to give you the result in 0.001 second. If you ask a computer to create a software system, it can create 50,000 lines of code in one second. While the average (human) software developer according to neutral research is only capable of creating between 325 and 750 lines of code per month. There are of course exceptions, and some developers are capable of creating 10 times that amount of code, but even a developer that creates 7,500 lines of code per month is slow compared to a computer that creates 50,000 lines of code per second. Our system creates 50,000 lines of code per second using the above process, which if you extrapolate the figures becomes the productivity of a single human developer working non stop for 90 months, or 7.5 years to be accurate.
Every second our computer system generates the same amount of code a human developer needs 7.5 years to produce
This of course raises a couple of interesting philosophical questions, in addition to also questions such as “What is the quality of this code?”
Let’s start with the last question, and answer the last question first. Assuming you can correctly program a computer, the computer cannot do a mistake. It’s simply not possible, because it is a “deterministic system”, incapable of doing things you didn’t tell it to do. The computer never becomes bored, will never be lazy, and will always solve the task at hand the exact same way, every time you ask it to solve the same task. From that perspective, the generated code is actually 100% perfect! Assuming we created the program that creates your software correctly of course, which we have spent a lot of time making sure it does. However, don’t believe us, check out the code for the Sakila project below, and by all means send it to your best Angular developer and ask him if it’s good code. I guarantee you he will not be able to find a single bug in it!
Then we have the “philosophical” questions, such as where is there room for the human being in all this? Well, the advantages the computer has ipso facto illustrates where the human is needed. The computer doesn’t become bored, solves the same task the exact same way every single time, and can never do mistakes. Hence, we still need humans for the creative parts, simply because the machine cannot be creative.
So yes, we eliminate 98% of the job normally a human being needs to do, but only the 98% boring parts. The interesting parts still remains.
This is why we can deliver a fully fledged software system, 1,000 times faster than any (human) software developer can do, for 5% of the cost a human software developer would charge you, and guarantee you that it’s 1,000 times higher quality. Simply because …
This is what computers do
We just hired the computer as a “software developer” …