In defense of serverless

I recently read the <a href="https://world.hey.com/dhh/don-t-be-fooled-by-serverless-776cd730" rel="noopener ugc nofollow" target="_blank">“Don’t be fooled by serverless”</a> article from <a href="https://www.linkedin.com/in/david-heinemeier-hansson-374b18221/" rel="noopener ugc nofollow" target="_blank">David Heinemeier Hansson</a> and was really irritated by the serverless-bashing without any real insights. By way of background, I have operated servers since the early 2000s. First normal x86 rackservers, then blades used as Citrix servers. I racked them, and connected networking …. later we used VMware to virtualize the resources and get more reliability from our data centers with fewer maintenance needs. The trouble was always that you had to plan 3 years ahead, purchase data center space with long-running contracts and you needed many DevOps personnel to operate the servers. Then, on days like Black Friday, you always hoped that you had provisioned enough capacity (that you had planned 3 years ago). That also meant that you provisioned for Black Friday and not the average requirements throughout the year. However, when <a href="https://www.linkedin.com/in/david-heinemeier-hansson-374b18221/" rel="noopener ugc nofollow" target="_blank">David</a> writes about serverless for me it seems that he misses the point. First, the end consumer does not care if the application is provided via a server in the backend (which also is true for serverless) or via a serverless function. The customer cares about costs, response time, and availability of the service. Imagine you are a customer of a video streaming service. You want to stream a movie and you want to do it now. So you only care that the movie starts instantly and that it does not stop while you are watching. You also do not want to spend a fortune on the service. <a href="https://betterprogramming.pub/in-defense-of-serverless-668497d83896">Click Here</a>