←back to thread

388 points replyifuagree | 1 comments | | HN request time: 0.248s | source
Show context
paulsutter ◴[] No.37965627[source]
The only magic wand in software development is to simplify requirements. The requirements are always wrong: too broad, too vague, based on invalid assumptions

The real genius is to propose a simplified solution, by discarding some assumptions. This is the best and only way to shrink the schedule

replies(7): >>37965809 #>>37966142 #>>37966325 #>>37966429 #>>37966768 #>>37966963 #>>37967034 #
dsego ◴[] No.37966325[source]
Oh, but we need the full search functionality right now, yes there are only tens of entries now but in a few years there will be thousands. And the designer will design all the search flows based on our twenty page product requirements doc, and we will include engineering to write stories and estimate the works once it's all planned and prepared.
replies(1): >>37966618 #
crucialfelix ◴[] No.37966618[source]
Also some designers throw in too many features because they want to fill space in the design.
replies(1): >>37967378 #
1. dsego ◴[] No.37967378[source]
I think they want to be helpful and leave their mark, which sometimes includes trying to reinvent boring CRUD designs and borrowing some cool but more involved ux pattern they saw in a big-money product. Adopting a popular 3rd party component library sometimes helps, but for some reason it's hard to push back on tweaking the behavior of these components all the time, and usually those component libraries are such that they include everything but the kitchen sink, well... except that one thing that the designer is insisting on. And that one is really hard to do with that library and takes way more time than anticipated. The value-add here is in my experience almost always negative. Looking back at delivered projects, I can't shake the feeling that involving designers at a later stage would be beneficial to the project and timelines.