Of course we’re looking for features. We know what we need – and what we want. Color, shape, size, capabilities. Our ‘specs’. But is that what we’re getting? Not really. We’re getting a lot more, and for a while it felt good. Now it’s haunting us.
Our ‘purchase criteria’ – many of them so obvious we don’t even reflect on their existence – make up the foundation of purchases and projects, from telephones via factory equipment to buildings and roads – even services. Requirements, expectations, technical details and schedules. Less evident but – as it turns out – just as important are the features and contents we don’t want, but get anyway. All those ‘included for free’ features you didn’t ask for, that seemed OK to have just in case, but will never be used. Maybe they didn’t matter before, but now they do – and they’re hurting. Do we need anti-specs? Maybe even pay a premium to get exacly what we want?
I’m not joking. The truth is – in all too many cases – that some or many of these features, functions, properties, built-in options etc. we didn’t ask for but got anyway, represent serious risk. In life, in business and particularly in technology. For now, let’s stick to the latter.
It’s ironic, but the analogy works: “If you have COVID, we don’t want you.” Not visiting, not in the office, not in the company or any of a shipload of similar contexts. A very specific ‘anti-feature’ – and one that many of us are (or was) very adamant about. For a good reason. And very different from common practice – in particular when acquiring products and services. The normal since forever has been to specify what we want, ignore what we don’t need and live with the consequences. A choice even if it wasn’t a conscious one.
Another relevant, although somewhat more complicated example. We need electricity, and being, or at least pretending to be, environmentally conscious, we want clean energy, from renewable sources only. The problem is – after leaving the power plant there is no difference, how do we know what we’re getting? Identical product, different origin – which brings to the discussion a third element: Trust. When we make demands that cannot be verified, can we trust our vendor(s)?
It’s getting complicated and we don’t need to go down that road yet. What we need is a different mindset, conscious attention that goes both ways: What we need/require, what we definitely don’t want – and possibly a ‘don’t care’ list.
We already know that complexity is bad for performance, cost, reliability, security, maintainability, scalability and more, so the conclusion is obvious: Anti-features need our attention. The weight of 40 years of featuritis, ‘digital obesity’ if you like, is heavy and costly. ‘Featuritis’ is a highly contagious disease with particular affinity to digital, software oriented environments: Many, probably most products we depend on, not only in business, but life in general, are loaded with useless and dangerous history. Old features kept for looks (impressive tick-off-list), urgency (will fix later), from old habit (don’t rock the boat unnecessarily), just in case (someone might need it), ego-trips (some developer just had to get her/his piece in) and for other reasons we normally don’t dare to mention (such as ‘don’t touch it, something might break and we don’t really know how it works’).
The consequences are obvious – and mostly ignored, even by those of us claiming to care about simplification. Featuritis and technical obesity in everything from operating systems and browsers via user interfaces and cash registers to phones and ever more ubiquitous IoT units.
Power consumption and heat emission are obvious waste-factors when 50-90% of the software (sometimes hardware too) is redundant or simply not needed for the task at hand. Far worse is the threat it represents – cybersecurity/exposure, reliability, resilience, size, scaleability. Nothing new, this is old knowledge swept (and kept) under the carpet for years, with the label ‘impossible to fix’. And maybe it was if not impossible, at least acceptable. For years Moore’s law delivered sufficient performance improvement to hide the software bloat.
No more. The metrics are new, the challenges are different and our mindset is shifting. We need simplicity, not because we cannot handle complexity, but because we’re moving faster. There is no time to deal with this humongous backpack of useless history. In all parts of life we need flexibility and headroom, the ability to adapt, understand, build, discard, evolve, change, innovate. In order to get there we need to cut ties, make hard decisions – including actively deselecting what we don’t need.
And here’s the good news. The momentum to do just that is increasing. Vendors and professional environments are increasingly getting serious about the challenge, a trend that for obvious reasons is led by younger companies with less history to get rid of. A contagious shift in attitude and mindset.
Case in point: One of my many contacts in Silicon Valley mentioned an Open Source VPN product the other day which immediately sent me on a hunt. VPNs are a big challenge for most of us – professionally and even domestically. Not the least in the countries now at war in Europe. According to a recent Economist article, the 30 day average search for VPNs increased more than 1.000% in Russia, 600% in Ukraine in March. The point this time though, isn’t the product or the war, but the attitude.
Algo VPN is hosted on Github and presents itself with features – and anti-features. At first sight it may seem like a list of lacking features, but to a security professional it’s the opposite: Less to worry about. No deprecated protocols and mechanisms, clear statements about what’s not included etc. Exactly what we’ve been looking for for quite some time.
There are many inspiring examples, but still too few – because the market (that’s you and me) hasn’t been ready (or too busy?). It’s a new mindset and a new metric: What we don’t know (about products and services) is hurting. What we don’t need is hurting.
The new generation of RFPs must be short, explicit and ‘final’ as in ‘nothing more, nothing less’. If it costs more, that should be OK too. We’re already used to the concept of paying more for less, as in streaming services and news: pay extra, get rid of the ads (noise).
Featuritis may not have been acceptable, but surely has been accepted for a long time. It’s time to change. We need ‘fit for purpose’, not ‘fits every situation’. Simplicity rules. And we are in charge.