shoptalk, rant

Saw this yesterday and boosted it again this morning. It's spot on about the kinds of trade offs and project methods that get employed in developing a web application -- and how those processes, born from "agile" are not the right solution for all software engineering endevours. Particularly things like the Iowa Caucus where you won't get the chance to roll back and try again if things go haywire.

It also got me to thinking about another item that's been on my mind lately. That so much web development still suffers from sticking business logic into the view layer. We have rich interactive interfaces at this point, and the hottest libraries of web development have always been front end. Go on a job board, and there will be an endless pile of positions looking for React or Vue. Javascript is everywhere.

The business side sees a need for churning out new features fast. User stories focus our attention on the UX, and draw attention away from the model -- product management types don't have the time to devote to working out a rich model that they can't see and touch. Back-end positions get advertised as Full Stack where the stack is ankle deep. Web developer's understanding of modeling business logic ends at making a bunch of tables and letting the ORM build you a bunch of objects. These objects enforce no rules, encapsulate no knowledge of the domain.

It makes some twisted sense as a trade off, as Bitlog outlines. The feature might get canned next sprint, so why put hundreds of developer hours into building anything more complex than what amounts to a prototype? If the prototype survives and the business ever has the opportunity to grow large enough to care about the model they can do a rewrite then (if).

But it can be frustrating.

Sign in to participate in the conversation

A bunch of technomancers in the fediverse. Keep it fairly clean please. This arcology is for all who wash up upon it's digital shore.