Bauhaus Fragility: An Assessment of Architecture in Buildings and Information Systems

The basic premise of this article is that while it is generally true that ‘less is more’; this doesn’t always apply to physical systems, due to the need for redundancy.

Always Avoid Fragility

Modern building architects seem to be obsessed with Bauhaus designs. Almost all the new major buildings, from the post-war (and beautiful!) Berlin city buildings to the most recent Apple headquarters, use similar design artifacts to that of the original Bauhaus building (in Dessau-Roßlau) and its design school. The main artifacts behind this design school that are especially noticeable are: the extensive use of ‘natural material’ such as glass and wood; exposure/transparency to the natural environment; and the extensive use of geometric / euclidean shapes.

From a non-architect (like me) point of view, Bauhaus buildings look extremely well-shaped and a marvel to look at.

But from an engineering point of view, Bauhaus buildings have three levels of fragility: privacy, protection from chaos, and adaptability.

Privacy: It is obvious that Bauhaus designs remove many forms of privacy. This is due to their extensive use of glass. Privacy might not be important in many scenarios. But removing all privacy as an ideological decision is bound to cause unforeseen problems in the near future. The best way to think about this problem is to ask: what credible options are available in case privacy is needed? The mere fact that the outside walls in Bauhaus buildings have ‘transparency’ built into them makes it almost impossible to guarantee privacy for any case or circumstance in which it might be desperately required. A ‘black swan’ event (e.g. a hostage crisis) that requires total privacy for survival or necessity is (however statistically unlikely!) likely to lead to a total disaster.

Protection: We have built buildings in the past for protection from the natural environment. If we didn’t need protection: caves, trees and tents are extremely effective for ‘staying connected’ to the natural environment. Since the early 20th century, we have now decided to use materials that not only make us ‘visible’ to the outside eyes, but provide extremely thin layers of protection from the environment. The materials we now use are fragile, transparent and exposed to the many destructive forces that could render an entire building uninhabitable within a few hours. Once again, we built buildings in the past to dodge catastrophes, but we have now chosen to ignore their likelihood and their resulting consequence on our safety.

Adaptability: The extensive use of open space in most modern Bauhaus buildings could lend a good case for adaption to multiple uses, but this doesn’t truly hold when one observes the over-dependence on geometric shapes that are usually flat, round, square/rectangular or perfectly Euclidean. Perfect Geometric shapes are fundamentally cold and rigid. What this means is that if one wishes to ‘change’ the building for a use that was never before conceived, such as an engineering factory, there might very well have been a ‘perfect geometric shape’ that was an ideal shape for a ‘design’ look, but now makes the building completely impractical for a machine that just happens to be 5cm taller than the room assigned for it. As a result: ‘perfect shape introduces perfect fragility’ in terms of adaptability. By design, a ‘perfect shape’ makes sense for a narrowly defined ‘space’ problem. If the ‘space’ problem shifts, even by 5cm, suddenly the shape becomes … imperfect. The shape breaks because it is entirely fragile!

I am making commentary on this because as a computer engineer, there is a growing trend towards adopting ‘Bauhaus’ thinking in computer design. The problems that have been highlighted above will unfortunately equally apply to computing hardware. The adoption of ‘screen only’ interfaces, for example, introduces many levels of fragility (and maintenance expense!) that many people now accept as perfectly normal. It shouldn’t be!

Always Avoid Noise!

Any serious student of typography, human communication, information theory, artificial intelligence, decision theory, diagnostics and mathematical logic knows one very important thing: less is more.

But big data engineers at Google, Microsoft, Facebook and all those other ‘big data’ noise companies are obviously too nerdy to get the point. The man who is probably the Founder of modern ‘cloud computing’, Jezz Bezos, recently made this point:

“The thing I have noticed is when the anecdotes and the data disagree, the anecdotes are usually right. There’s something wrong with the way you are measuring it..” (emphasis mine)

Which follows a similar line of thinking to what I have said before: that heuristics, biases and semantics matter more than ‘big data’. Numbers, words and named objects are mind-bogglingly useless without context! Contrast this with mathematical logic and information theory, where the less statistically ‘available’, or the less probable, is the most important.

Short speeches are more valuable than longer ones (or are at least mostly remembered by a more concrete uttering, such as ‘I have a dream!’). Less cluttered logo designs are better, more memorable, and outlast time and space, such as the Apple logo. Better decisions are made via ‘satisficing’ logic which involves the use of a few objective parameters for ‘threshold’ detection. Satisficing logic reduces time-spent making a decision and optimises on the ‘feedback ’ of the results of the decision (rather than on the ‘correctness’ of the decision); and it delivers better results over the long run. Something is either true or false: more ‘data’ tends to make something seem less true (due to the bias problem in evidence presentation); and one only needs one piece of data to prove something completely false (cf. one needs only one black swan to disprove that ‘all swans are white’).

Information is, in many ways, extremely counter-intuitive.

The Public vs. Private

We build buildings for private use; and information is meant for ‘public’ consumption.

This is what makes the application of Bauhaus (‘less is more’ design) to everything under the sun extremely tricky.

For things that are ‘perishable’ and physically constructed, we need to build redundancy into them. There is a good reason why the Roman Colosseum is still around for us to marvel at.

It was built to outlast its time. It is extremely doubtful whether ‘Bauhaus’ designs will survive for the next 200 years. They are not built to last and they remove all forms of redundancy and architectural beauty (or complexity) in them.

Buildings should ‘self-protect’ at they are meant to protect and shelter humans. They are meant for private individuals, families or communities.

Information is non-perishable. Redundancy in information makes it less valuable, and less meaningful. Information outlasts its time in small bits, and becomes less ‘beautiful’ when presented with complexity.

Information is meant for the public and it is the public that decides to protect or to do away with it (when they do not find it useful… like this article?). (NB: Another reason to do away with patent laws!).

Modern architects practice the ‘less is more’ thinking in their designs. Modern software engineers practice the ‘more is more’ (‘big’ data!) in their designs.

The truth is: they should just swap roles because they both do not know what the hell they’re doing.

Download Our App on the Android or iOS Store /

Sign-UP : If you are interested in Our Diagnostic Software

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store