Hi. My name is Gigi Sayfan and I’ll be writing here weekly. In this first post I would like to tell you a little bit about me and what I plan to cover. I’m also very interested to know what you care about and I’ll gladly take ideas, requests and suggestions.
I’m a passionate software developer. Over the past 20 years, I worked for large corporations, small startups and everything in between. I have written production code in many programming languages and for many operating systems and devices. My current role is the director of software infrastructure at Aclima. We design and deploy large-scale distributed sensor networks for environmental quality and hope to make our planet a little healthier. I still write code every day and, in addition, I like to write technical articles about software development. I used to write for Dr. Dobbs and these days I write regularly for DevX.
I have a lovely wife and three kids that I tried to infect with the programming bug (see what I did there) with varying degrees of success. When I don’t code or read and write about software development I lift weights, play basketball and do some Brazilian Jiu Jitsu.
What am I going to write about? Well, pretty much everything. There are so many cool topics to talk about. I’m getting excited already.
Theoretically all Turing-complete programming languages are equivalent, but we all know the difference between theory and practice. There are so many new developments in programming languages, such as Modern C++, EcmaScript 6, Go, Rust, Elm, C#, and Python 3, to name a few. I have always loved programming languages and I plan do to deep dives, comparisons, reviews of new versions, and more.
Databases were considered perfect for a long time. You designed your relational schema, put your data in, got a DBA or two to optimize your queries and you were good to go. But, everything changed when the data didn’t fit into a single database anymore. NoSQL was born with its many variations and companies started to innovate significantly around data storage.
Build systems are so important for enterprise development. I created a couple of build systems from scratch and I believe that a good build system used properly is critical to the success of enterprise-scale projects.
Automation of pretty much any system is key. You’ll never be able to scale by just adding people. The nice thing about automation is that it is such a simple concept. Whatever you do, just have a program or a machine do it for you. It takes some hard work and imagination to automate certain aspects, but there is also a great deal of low hanging fruit.
Testing is yet another pillar of professional software development. Everyone knows that and these days many actually take it seriously and practice it. There is so much you can test and how you go about it pretty much dictates your speed of development. There are many dimensions and approaches to testing that I plan to explore in detail with you.
Deployments, such as with a database, used to be fairly straightforward. Now, that systems are often extremely complicated, a large deployment is not so simple anymore–with private hosting, cloud providers, private clouds, containers, virtual machines, etc. An abundance of new technologies and approaches now exist, each with their own pros, cons and maturity level.
Distributed systems are another piece of the puzzle. With big data you need to split your processing across multiple machines. We will explore lots of options and innovation in this space as well.
Development Life Cycle
The software development life cycle is another topic that never ceases to generate a lot of controversy. There are multiple methodologies and everybody seems to have their own little nuances. Agile methods dominate here. However, in certain domains such as life-/mission-critical software and heavily regulated industries, other methods are more appropriate.
Open Source keeps growing and even companies, such as Microsoft, that were once considered as distant as possible from open source are now fully onboard. The penetration of open source into the enterprise is a very interesting trend.
The web is still a boiling pot of ideas and disruption. New languages, new browsers, new runtimes–here is a lot to observe and discuss in this space.
The so-called talent wars and the difficulty in finding good software engineers are very real. It appears to just get worse and worse.
A company culture is often not very tangible, but somehow when you take a step back and look at successful companies it is evident that culture is real and can make or break you. Often big undertakings can flop with no clear reason other than culture. A prominent recent example would be Google+.
The Past and The Future
There is a lot to learn from our history. Luckily, the history of software is relatively short and well documented. The phrase “Those who don’t study history are doomed to repeat it,” is just as appropriate for software.
We live in the future. You can see the change happening in real time. Science fiction turns into science faster and faster.