The Confusion On DevOps

devops in a boxI was in an interview with a candidate for a Software Engineer some time ago. We had the usual suspects at the table and me, the unusual one at the interview.

I’d read the guy’s resume. He claimed some game programming, and he also claimed to have experience with the full SDLC with both the Waterfall and DevOps methodologies. One of our people, notorious for academic questions, was pulling some of his punches related to .Net. He trotted out his usual questions on design patterns (he sure did love his MVC – let’s call him Mr. MVC), and he continued with his notorious monopolization of the interview. When it finally got to my turn, I asked a XOR question – the candidate had never used it.

So then I asked him about the strengths and weaknesses of Waterfall – and on DevOps. Mr. MVC quipped that he didn’t think we were hiring for the DevOps department, and I ignored him since I did not want to straighten out his ignorance in front of a candidate (though he had no problem being a jerk when he was wrong). The candidate answered the questions well enough, and I let it be.

But looking back, I couldn’t help but think that Mr. MVC was confused by DevOps because:

(1) We had a department by that name.

(2) He didn’t actually know about what DevOps was,

(3) He hadn’t read the resume and noted ‘using Waterfall and DevOps methodologies’.

It surprised me that Mr. MVC didn’t know about it, and it surprised me more that he’s now basically in charge of the Agile process at that company but wasn’t familiar with DevOps itself – particularly when the end goal is continuous integration. That’s sort of what DevOps is:

…Agile software development paved the way, steering away from the waterfall method of software development toward a continuous development cycle. But it didn’t include the operations side so while development could be continuous, deployment was still waterfall-oriented.

In a DevOps environment, cross functionality, shared responsibilities and trust are promoted. DevOps essentially extends the continuous development goals of the Agile movement to continuous integration and release. In order to accommodate continuous releases, DevOps encourages automation of the change, configuration and release processes…

So, by having DevOps on his resume, the candidate had opened himself up to all manner of questions on Agile, as well as continuous integration – things which, months later, would be brought up by the CTO.

And no one – not one soul – called it DevOps. Why? Likely because they got rid of the department that they used to call DevOps. This was largely because they sabotaged it by not giving it the appropriate tools for continuous integration, and the fact that the lack of documentation made the relatively fresh department more likely to fail with manual deployments.

I write all of this because I encountered it at one of my own interviews with a company. We were talking about continuous integration and the mismatch between Agile and Continuous Integration when it comes to involving Operations – they are typically not involved with Scrums and are therefore less likely to be prepared. That’s where DevOps comes in.

DevOps has been around a while. Waterfall went to V-model that went to Agile – and now DevOps is king. Or should be.

But it’s surprising how many people don’t know about it.

Suggested Reading:

The Future Is Not The Enterprise You Know

Greetings from Guyana to New YorkStory time.

When I was in Georgetown, Guyana in 20051, I snapped that picture of the television in my room. People called in to the station to relay messages to expatriates in New York so that they wouldn’t incur the cost of a phone call from the local phone company.

In the developed world, or the Global North, or the West, or whatever you want to call it, VoIP had already shaken telecommunications by 2005 and won. In the other parts of the world, state-owned or subsidized telephone companies fought to stay relevant.

In the context of Guyana, it took until 2015 for the Caribbean Court of Justice to rule against the telephone company. And I imagine that battle isn’t over. It’s certainly not the first time I’ve seen it in my travels, and it’s not the first technology either. It also won’t be the last.

Having now worked a bit directly with telephony and SIP trunks, spending late nights catching up to where the company was and then studying beyond it2, I have a good feel for what is possible. And I also know that the future is global, that infrastructure is subject to licensing across geopolitical lines, and that technology waits for nothing but ideas whose time has come. Having been involved at the World Summit for Information Society level at first directly and now vicariously, to get global the industries built around Enterprise have to change. Having been a part of a Pre-twitter clone (we copied them when we grabbed a flux capacitor?), we saw things change a year before Twitter. That change is coming regardless of how much people are in love with the present enterprise. Evolution awaits no one.

So, what’s the future?

We hear a lot about the Internet of Things (IoT). How disruptive it is. Most of that is marketing hype to get all of us to buy things that we really have no need for – and, to be fair, people usually buy it for reasons that I might write about on my other blog3. Behind all of that is an undeniable force of change that goes beyond the buzzwords.

A few data-points to draw the line for you.

Data-point 1: Telecommunications infrastructure is no longer the product it once was because of SIP (which most of you know as VoIP, but it’s bigger than that). It’s about the on demand use of the telecommunications infrastructure. You can think of it as time-sharing real estate without the need to worry about the last person leaving dishes in the sink.

Data-point 2: The laws governing telecommunication infrastructure vary across geopolitical lines and proceed at the rate of the internal geopolitical bureaucracy. That’s a nasty factor that everyone should know, but most people don’t.

Data-point 3: Oh, that little Internet of Things has spawned all manner of things, like the 10 Pine64s I have coming next month. I’ll be clustering those for my own purposes – but imagine those as part of a solution that, for less than $500. Do you think I’ll be paying Oracle or Microsoft for licensing for a database? If you think so, you’re nutty and should have your head examined. Even open source DBAs are cheaper.

Data-point 4: Bitcoin brought the block chain to light.  Think of a client as part of a Peer to Peer network where the client deals with licensing within their geopolitical sphere (see 2), thus avoiding licensing fees across geopolitical boundaries wherever possible, and otherwise diminishing them. Take a look at this post on blockchain, posted by Arvind Krishna, Senior Vice President and Director, IBM Research. Or consider how Microsoft has been rolling Windows 10 out… peer-to-peer.

Data-point 5: Open source software has come so far that the cost of the software itself for applications has diminished significantly – you don’t pay for software, you pay for the changes to it if you want the changes… or you pay for people to configure it for you.

Data-point 6: ‘Big data’, another overblown marketing phrase, is a driving force that will not be stopped – it will hopefully be curtailed for reasons of privacy, but again and again the world has shown that it abhors censorship and will – at the cost of individuals, corporations, or entire governments, if necessary – be had.

What does it all mean?

It means that a company’s infrastructure, unless it’s spread out over a large area, is pretty much going to be an antique soon. People espousing details on the how of Software Engineering will develop are likely to completely miss the what of the development; the what of development should be defining the how (Software Process 101).

The ‘data-center’ will not die. It will become less important and probably used to roll out continuous integration to a peer-to-peer network of SaaS. Data will make its way to whatever monstrosity of a database that’s out there, and if I were a betting man I’d go with Oracle more than Microsoft on that since they’re acquisition of MySQL probably wasn’t an accident. Sure, it’s not an RDBMS, but what’s the most used database on the Internet? And who now has a thermometer?

It’s happening. Now.

1: Doing some volunteer work with St. Joseph Mercy Hospital. IBM had quoted a million Guyanese dollars to do the local network for the hospital; we got a group of volunteers around the hospital to do it for the price of a pizza on a weekend. Sadly, because of internal bureaucracy, the network was not used when I left, but I do hope that changed in my absence.

2: Trying to plan for the future like any good engineer.

3: RealityFragments.com, where I focus on more creative and opinionated writing on things that aren’t technology.