The pain of legacy software doesn’t hit until it’s too late. Scroll to the end for a secret video that only you will be able to watch.
We’re kicking off our developer conferences later this month in Chicago, April 26th and 27th. It’s SpringOne Tour time! This year most of the talks have something to do with app development with/on kubernetes: not all, to be sure. There’s also finally some .Net content, and I’ll be talking about DevSecOps and Platform as a Product.
Check out the whole, two day agenda. When you register, use the code S1T_SAVE25 to get $25 off.
And, if you can’t make it to Chicago - fret not! - there are six more cities coming up and some more in the works.
Also, if you fancy yourself the “executive” type, check out the executive event I’m hosting in Chicago on April 26th.
I’m back to working on the ongoing book(let) project, The Legacy Trap. Marc has been adding in the real meat of the project, how the methodologies VMware Tanzu Labs uses for planning and doing application modernization, like Swift. Here’s a corny example story I wrote for the introduction, linking together business needs with worrying about legacy software.
Let’s look at a theoretical example of that business problem in insurance. Let’s say The Mid-Eastern Warm Smiles Insurance Company wants to grow revenue. They’ve done a great job over the past 140 years insuring houses, cars, and grew revenue in the early 2000’s by acquiring a point-of-sale warranty business.
Business has slumped now, and the share price is boring. A management consulting company compiled a large report that suggested three pillars for improving shareholder value. One of them was to enter new types of insurance. Playing off the synergies of that warranty business, the consultants suggested entering short-term insurance: coverage that would last for 24 hours or less. For example, a customer might go to the beach for the day and want to insure their new, $1,300 iPhone against damage and theft. The likelihood that anything will happen (especially if they have newer models that are water and sound resistant) is low, so collecting the $50 for that one day of insurance is almost like “free money” to the insurance company. Now, imagine if that happens thousands, hundreds of thousands of times a day, globally.
After some business and actuary work, the CIO is given a new set of applications to create. First, the application for the insurance, next the actuary back-end for approving insurance, then the account management for these policies. Seems simple enough: it’s software! You can just add a feature!
A situation like this is where the legacy trap is often sprung. Developing the actual web app for an application is often easy, but integrating with the existing backend services is often near impossible. In the case of Warm Smiles, the backend systems that maintain policies only support annual policy terms. That will need to be updated. Payments must be done through bank ACH or physical checks (here, the CIO thinks, “well, it did always seem weird that customers had to fill out a PDF to pay their policies”). These payments can take up to 10 business days to clear. Warm Smiles will need to modify their payment processing system. And what if someone actually losses their iPhone and wants to file a claim? Usually, claims processing takes 5 business days and requires an adjuster visit…but we’re just talking about an iPhone here. Our CIO friend also keeps hearing the mobile team say something about “batch jobs” and something called an “enterprise service bus.” Those need to be updated as well.
And so on, and so on. Meanwhile, Warm Smiles rival, Southeastern Friendly Pats on the Back Assurance Group, has launched their own short-term insurance business and has seen a 3% rise in share price.
This is a made up example, but this kind of situation repeats itself over and over in large organizations. Warm Smiles is squarely in the legacy trap. Their software portfolio does not support the business fitness and agility needed. At some point, the portfolio was great - the company has saturated its market and survived for 140 years. But, slowly and then all of the sudden, their portfolio became “legacy” and urgently needed to be modernized.
And then, Cagan goes on to say:
We’re putting together our State of Kubernetes 2022 survey. I’m doing four tiny promo videos for it, and here’s the first one. I don’t know, a little too much hustling in the second half. Maybe I should just take it out. I mean, I’m done with this one so I’m not going to, but in the other three, I should probably just talk about the survey with a “And, hey, at VMware we’ve got a lot going on when it comes to kubernetes. Check out the free and open source Tanzu Community Edition and also Tanzu Kubernetes for Operations.”