,
14 tweets,
5 min read
Read on Twitter
On serverless and doctrine by @PaulDJohnston - medium.com/@PaulDJohnston…
... we should expect co-evolution of practice with a new set of emerging practices probably combining development + finance ...
... we should expect co-evolution of practice with a new set of emerging practices probably combining development + finance ...
... but whether those practices become "universally applicable" (i.e. doctrine such as "focus on user need" or "challenge assumptions") is unknown at this moment. It often happens but we can't be certain until the practices emerge.
Doctrine are those universally useful principles regardless of context ... most patterns fall into climatic (i.e. patterns you have no choice over but influence the landscape) or gameplay (context specific patterns).
You do have to be careful because many apparently "universally useful" patterns turn out to be very context specific i.e. most ecosystem plays or providing through an API or open sourcing.
Of course, if you don't have awareness of context (e.g. you don't map) then categorising patterns into universally useful or context specific is next to impossible leading to endless "secrets of success" lists containing items which only work ... sometimes.
i.e. Agile development has a context - very good in the uncharted space where reducing the cost of change is essential as change is the norm but less useful in the industrialised space (compared to six sigma) where reducing deviation is desirable.
There is instead a climatic pattern (i.e. a rule of the game, something that you have no choice over) which is "no one size fits all" ...
... which in turn leads to doctrine (i.e. a universally useful principle) of "Use appropriate methods".
So, when someone sticks "Use Agile" on their "top 10 secrets of successful companies" ... the answer is ... well, it works well in some contexts, less well in others. It's not a universal principle ...
... however, you can always select positive case examples (i.e. companies / projects using agile that succeed) and then try and hoodwink people into believing it's universal.
This works until people start to notice that failing companies / projects also "Use Agile".
This works until people start to notice that failing companies / projects also "Use Agile".
Of course, by the time that happens you will hopefully have secured healthy royalties on your book covering Agile as the new messiah and afterwards you can blame failure on "people doing it wrong" and enter full on cult mode i.e. blame the people, not the method.
So, will there be universally useful principles (doctrine) that come out of serverless and are different from what exists? Probably.
What are they? Not sure at this moment, as they have to emerge.
Will there be context specific patterns? Very likely.
What are they? Not sure at this moment, as they have to emerge.
Will there be context specific patterns? Very likely.
Other than that quibble on terminology, the post by @PaulDJohnston - medium.com/@PaulDJohnston… - is very good and points to how serverless is more than technology, it's also about the practices. Ditto Compute (as a utility) + DevOps or Compute (as a product) + ITIL etc.
@PaulDJohnston Oh, and just to be clear ... there were universally useful principles in ITIL (e.g. focus on users) which were carried over into DevOps ... but as with each flag waving faction, we dismiss other groups as some monstrous version of itself. See also the current split.
