5/17/2023 0 Comments Obscurity serverReduction in manual labor on repetitive tasks (not only reduces the staff's workload and boredom but also limits the amount of human errors that are bound to happen).Improved quality (in tech world businesses rarely get by with anything less than stellar without suffering from high bounce rates).Faster deployment (not only provides faster time-to-market but also generates more feedback from customers).So where do we focus our attention in order to achieve those marginal gains and improvements? Here is the starting list of areas that almost any company can gain from: You can achieve this by relying less on people and (paradoxically) technology, by automating everything (but not at the same time) and designing for technological failure (so that you can become proactive and not reactive). It's not about getting better tools, better people, or throwing more money at some combination of those - it's about working with what you have and ever so slightly improving on what already works. This is the promise of well applied DevOps philosophy to modern technological problems. After all, according to a quote attributed to Albert Einstein: “compound interest is the eighth wonder of the world”. The answer was in marginal but continuous improvement that compounded over time. They did not invent some magical bicycles that were way above the competition, nor did their competition get significantly weaker. They did not discover some undetectable doping substance. They did not replace their whole team with the world's best athletes. ![]() At the 2008 Beijing Olympics, his squad won 7 out of 10 gold medals available in track cycling, and they matched this achievement at the London Olympics four years later. They had won *drumroll* a single gold medal in its 76-year history, but that quickly changed under Sir Dave’s leadership. ![]() The answer to not only surviving but also thriving is to do what the British Cycling team did in 2002. If this sounds all too familiar to you, and you still somehow manage to survive this harsh hellscape called “running software development production in 2023”, then I want to congratulate you! Keeping it all under control by retaining your key personnel by overcompensating them will keep them happy for sure, but according to research, this will only last for few months, and then what? Adopting a security through obscurity approach will only create an illusion of safety, while toning down the functional aspects is not gonna fly with business owners. So how do we survive this? How do we make it all work without ballooning the budgets beyond feasibility and delaying launch dates beyond infinity? Employing more modern technology often only creates more complexity and steeper learning curve to your already overworked (or so they claim) staff. And we haven't even gotten into the ever-changing technological stack that keeps growing in complexity, as well as the rotation of qualified IT staff that is all-time high. Not to mention the need for maintaining regulatory GDPR compliance and industry specific standards (such as PCI-DSS for payment processing). But wait… there’s more - we want this data to be secure from ever evolving cyber threats because as common knowledge dictates - if “time is money” then “data is power”. We want to access said data in real-time, all the time, in multiple languages, across multiple database clusters on multiple availability zones with no data loss or stalling. ![]() We have access to huge amounts of data and thus a liability to use it efficiently. Nowadays we have teams of developers, sometimes located in different time zones and continents, or perhaps they work in the same office building but their work culture makes it look like they come from different planets. ![]() Alas, those times has passed irrevocably and now we deal with several orders of magnitude more complexity than we could have imagined. There is beauty in simplicity and there is nostalgia about good old times when things moved swiftly and the biggest complexity challenge was sharing code with a few peers or adapting your web project to a few hundred (!) hourly visitors. Remember when updating the code on a production server was as simple as uploading a file via FTP client and then refreshing the browser to see if the desired result was obtained? “Pepperidge farm remembers”… and so do I.
0 Comments
Leave a Reply. |