Katie Bauer!

"Software" is about as vague a raw material for achieving business value as "data" is. Not every engineering org is humming along perfectly happily, and yet SWEs don't seem to have the same level of existential malaise as data professionals. Why is that?

My feeling is that it's a result of three big factors: org size, clearer expectations of the profession, and standardized patterns of collaboration across teams.

It's obvious how org size matters. If there's more people to do the work, it's easier for it to get done even without detailed planning. You don't have to manage your resources carefully because there's more redundancy

Clearer expectations for the profession is also pretty straightforward. SWEs are less likely to be asked to do something that's outside their skillset because the average tech employee as a fairly good idea of what the skillset of an engineer actually is.

More standardized patterns of collaboration across teams, though--this is one that is IMO underrated in the Data. How many differences in approach or opinion are smoothed over in the SWE world by doing things your own way and then jamming it through some protocol or API?

That agreed-upon form of contact, that common language--it means it's easier to away with having a bunch of unrelated and maybe even contradictory plans, so long as you plan for the right form of communication between two systems or services.