How often do you patch or upgrade your mental models?

How often do you apply patches?

Patching isn’t just a job for sys admins any more. Even people who have never performed a technical role routinely apply patches to their PCs and the apps on their phone: if they’ve turned on auto-updates, they may not even realise that they’re doing it.

Major upgrades are a different matter, though. Sys admins still approach them with caution, accompanied by a lot of planning and testing. And the rest of us approach major upgrades to our PCs and phones with some suspicion. We watch the Internet to find out how others have got on with the latest OS version, to see whether it’s going to break our apps, drain our battery or have other weird and unexpected consequences.

I think that we manage our mental models of technology in a similar way.

Anyone in a technical role carries some mental model of the technology they use, whether that model is functional map of a set of applications and data, a physical map of networks and servers, or something else entirely. That model informs the ways in which we build, design and run systems.

And, if we are paying attention, we all patch our mental models all the time. We learn new things every day, just by doing our jobs. We solve problems, we search for solutions online, we read news articles, we try out new techniques and, most importantly of all, we work with other people.

Upgrading our mental models is different, though.

When technology changes in fundamental ways, our mental models get out of step, and may mislead us badly if we don’t update them. But, just like a major OS upgrade, upgrading our mental models takes time, effort and preparation. Sometimes the world won’t quite work the same way afterwards. Upgrading our mental models can be daunting, but unless we do it, we will fall behind.

I’ve had to upgrade my mental models many times in my career, and it hasn’t always been easy. In my very first day in a real computing job, I had to shift my mental model from the microcomputers I had learnt to code on, to a big enterprise mainframe. Later, when PCs turned up, I had to shift from the mainframe model to the two tier client / server model. Then three tier web applications, then mobile applications. Along the way I had to shift from suites of batch programmes, to online transactional systems, through SOA to micro-services. And, of course, from physical machines to virtual machines to containers and Cloud, and now to serverless functions. And from network databases to relational databases to noSQL databases (although, if I’m honest, I have to admit that the noSQL upgrade hasn’t quite taken, and my mental models of databases still contain simple rows, tables, foreign keys and indexes).

There’s nothing unique about the upgrades I have made to my mental models: anyone who has been doing a technology job for thirty years or so has been through the same changes, and we will go through many more such changes in the future.

One lesson I have learnt, though, is that it’s a lot easier to make these changes when you’re doing a technical role. I am very privileged to do a job which requires me to stay current with a wide range of technology, so I don’t have the option of letting my mental models get out of date. On the occasions when I have done general management roles with less technical responsibility (not very often: such roles are definitely not my strength), I have found myself getting stale quickly, and have had to make a conscious effort to apply upgrades, just like a server that’s not been on the network for a while.

This lesson has implications for people in two quite different roles.

First, for people who have moved from technical roles into management roles, it’s important to check regularly whether your mental models are sufficiently up to date. Do you really have a clear idea of what your teams are doing? Or have you inadvertently become like the last remaining Windows XP machine that sits in the corner, off the network? If you have, you should probably find the time to make that upgrade.

Second, for architects, one of the most important things you can do is to help people make this upgrade. It’s rarely the fault of general managers that their models have got out of date: things move fast and they’re very busy. Also, despite the wealth of literature about technology, it normally takes a while for clear expressions of new mental models to emerge. You can fill that gap: architects have an opportunity to embody the Zang Jing Ge attributes of technical excellence, communication mastery and leadership power by understanding new technology, explaining how to think about it, and by leading people in its adoption.

These days, we all know that deferring systems patches and upgrades can be dangerous and can degrade performance: this is just as true of patches and upgrades to our mental models.

Debsundar Pan

Senior Technology Leader | Collection and Recoveries (Retail Banking) | HSBC | Certified SAFe 5 Program Consultant | IIMC SMP 2021-22

6 年

Can't agree more, real challenge been simplified and told with such a wonderful analogy ! We are in the time where metal patching should be made automated and made a part of our routine & habits else very difficult to catch up with the rapid technology transformation ..

Nupour Mukherjee

GENAI Competency Head GSK driving Agentic AI impact Value in Pharma | Global Leadership, P&L Mgt, Expert in Data AI ML & Banking, Pharma, SCM, ESG GreenTech, Energy| Independent Director NBFC | NASSCOM AI CoE Advisor

6 年

ZangJingGe is truly we as IT warriors need to embed into our daily bread as leaders. Interestingly there is a similar treatise in india called Arthashastra or PurnaYogamey which puts Value creation and constant checking for the value created , at the centre of communication , leadership and new skill gained. Involves not only getting better yourself but bettering the circle of influence and those around you. Could we really communicate more about this approach @Hsbc as part of healthy human system?

James Linsell-Fraser

Principal Industry Architect

6 年

Like in the world of computing sometimes applying a mental patch can also result in an abend!

One of the telling moments when you realise you need to an upgrade involves experiencing cognitive dissonance. Every time you experience two competing heuristics that both seem right, it's time to step back and think about your thought process :)? Some examples of where I have experienced cognitive dissonance recently: 1. Data should be in a central repository to be ATOMIC- data? can sit in different locations and still be ATOMIC. 2. Create the algorithm and let data run through it - Let the data be create the algorithm it subsequently operates on. 3. You need to fix the scope of your work before you start - you need to start and then fix your scope based on what you learn.?

Ramesh Vasamsetti

Technology leader; DevOps Enginering Quality and Reliability; Resilience, Observability, Chaos Engineering, Application of AI and ML in engineering quality

6 年

Excellent analogy David. As Simon mentioned the current patching level is fast getting outpaced by the industry pace. But that's how the life is and we need to gear up.

要查看或添加评论,请登录

David Knott的更多文章

社区洞察

其他会员也浏览了