There are a handful of times in my career, when my eyes lit up and I felt excited about the future with a specific technology. The first time was when I worked at Lucent Technologies, my first tech job. I gained lab access to an “IP PBX” and clicked through a GUI and connected voice calls over the internet. The next moment was when I saw VMware esxi and realized I could build a full Cisco lab on a spare laptop. The latest moment I experienced awe felt the largest of them all. It was the AWS console that inspired me to pivot from a successful career in hardware to a career in software. Here I am today, working at a cloud provider, where I believe we’re just entering early stages of mass adoption. Software is incredible and I’m grateful to be a part of its evolution.

If you know me, you know I make a horrible liar. 😅 I’ve never been able to paint a rosy picture when things aren’t rosy and I’ve never been able to ‘sell’ a product that I don’t believe in. My values prevent me from ‘selling’ another human on an idea that causes harm for my gain. I work for a cloud provider because I believe it’s an important phase in our evolution and a collection of resources that has an amazing impact for my customers.

We’re all susceptible, especially when we surround ourselves with like minded people, but I try not to ‘drink the Kool Aid’. I appreciate some of this guy’s posts because they offer a different perspective which keeps me grounded with a more rounded perspective. The cloud has potential to help most businesses but it’s also not the best fit for every use case.

A few of David Heinemeier Hansson’s recent articles include the titles:

• Why we’re leaving the cloud
• We stand to save $7m over five years from our cloud exit
• Five values guiding our cloud exit
• Don’t be fooled by serverless

When we’re passionate about something, when we love it, a natural reaction is to defend against immediate threat and attack whatever disrupts equilibrium. I’ve heard other cloud believers’ reactions declaring how naive David’s take is. Others voice excitement to see the day David falls on his face and returns to the cloud. Others use a strategy of logic where they stack up all the value cloud offers. Some think he does it for shock and attention. I thank David for reminding me that it’s important to not get lucid on the Kool Aid. The openness that he demonstrates, the sharing of one’s truth, is what I want for all my relationships, both work and personal.

Growing cloud costs are a common concern so it’s not shocking. I discuss the topic regularly with my clients but as an industry we should continue to improve the experience for all customers large and small. I have customers that run parts of their business on the cloud and other parts on hardware that they own in their data centers. My goal is always to uncover the best path for my customers’ businesses to thrive. In this economic environment, cost reduction is the topic I revisit with my customers the most. If we can keep their business and profit margins are strong, we’re in a great position to grow together for the long haul. 

In the blog ‘Why we’re leaving the cloud’, the cloud is described as ‘renting computers’ and suggests that it’s generally a bad idea for mid-sized businesses that are growing. I believe the statement is too simplistic and likely driven by David’s obviously frustrating experience. Like most life and business challenges, there’s not a simple fix and it requires digging to understand the business and options. Being a part of solving business, people, and technology puzzles is what I love about my job!

“Economies of scale” are powerful when seeking the best price per piece. There’s no denying that the big 3 cloud providers can demand better pricing for hardware than I can as an individual. David insinuates being taken advantage of by a cloud provider because they bring in a 30% margin. I don’t know how I feel about that margin specifically, but it’s a reasonable target that most businesses strive for.  Being successful and making money isn’t a simple indicatio of evil or wrong doing.  Business transactions should be a win/win for both parties. For David, he doesn’t believe it’s a win for 37 Signals any more and I commend him for standing by their values.

The spirit of Moore’s law should continue to prove true in the cloud with intensifying competition from the big 3, niche cloud providers, and customers like 37 Signals who set out to build a better product that meets their needs. The presence of so many diverse perspectives and strategies demonstrates how important the market is and we’ll all benefit from the technological and pricing advantages that continue to emerge. 

In the blog “Don’t be fooled by serverless”, David perhaps suggests that ‘Serverless’ is a scam and simply a business model of buying computers in bulk and then selling them by the piece. The analogy doesn’t resonate with me when talking specifically of serverless. For me, cloud as a whole is largely economies of scale, but I believe I understand the sentiment and share a general dislike of sales tactics used by some to sell serverless as a ‘magic bullet’ that solves all. For the record I’ve never cared for the term ‘serverless’ because servers are very much still involved behind the scenes.

I have no insight into the actual profitability of services where I work but it makes sense that serverless would have a higher profit margin per compute unit than a virtual machine. The reason is simple and it’s because a serverless compute unit delivers more value in the form of  an added framework for automation. The appropriateness of a serverless approach is a business by business analysis. As alluded to, if you’re running a fixed number of instances and your CPU’s are at 100% 24×7, serverless will probably not be the most cost effective solution. It’s designed as a solution for bursty traffic that eases automation of scaling, bursting, and is even scaling to zero. If a service is only executed at the end of the month, why pay for the service to run all month? Cloud is not “one thing” to be avoided which takes us to our final quote.

This quote from the ‘leaving the cloud’ blog  is an excellent way to bring us home and with alignment. In the blog, David states:

“Anyone who thinks running a major service like HEY or Basecamp in the cloud is “simple” has clearly never tried”

Totally agree! The challenges that businesses face today aren’t simple and solutions based on cloud are not ‘simple’. The cloud is software and as we know, software isn’t simple yet we continue to demand more of it. I’m happy to talk through pros and cons of all solutions but know that cloud, when used in the right way, is an incredibly powerful collection of tools that help businesses evolve in a world that’s increasingly reliant on technology.

Thank you David for sharing your passion, experience, opinions, and what I believe are good intentions. Lots of us ‘cloud folk’ hear ya and we’re striving to improve the experience for the masses. 

Sorry to hear that your experience left a bad taste in the mount but you and 37 signals are still my ‘cloud brothers and sisters’ . In my mind, we’re all family and it’s just time to choose different adventures. I’m rooting for you and 37 Signals to succeed! What we learn on our journeys will be important contributions to our collective knowledge and community. Thanks for sharing and for braving another path!  I enjoy hearing the stories and following your learnings. 

✌️