3 ms·

> I am wondering, what they are even working on Without any inside knowledge, but having been part of many mega-projects like this I'm willing to take a stab a

> I am wondering, what they are even working on Without any inside knowledge, but having been part of many mega-projects like this I'm willing to take a stab at it: new features vs. QoL is a huge, huge tension that exists in projects like these. New features get new customers, and more importantly gets execs excited and people promoted. Refactors, performance improvements, etc, are almost never appropriately rewarded or prioritized, so QoL problems tend to accumulate until a crisis moment, where the organization is forced to shift priorities to address a user experience that has badly degraded to the point of significant customer revolt. I saw this with a front-row seat with Google Maps: features kept getting piled in to the app (did you want to know what the most popular dishes at a restaurant is? Now you can!) while renderer performance continually regressed, map overlay rendering bugs accumulated, and the original "slippy map" UX degraded. At this point the mobile app is deeply unpleasant to use - it hitches regularly and has trouble hitting 60FPS even on top-flight flagship phones. The product clearly hasn't reached the "crisis" moment yet, but it is on a trajectory to. Teams seems like it might be at (or near) the crisis moment. It's a shame that organizational dynamics prevent these sorts of issues from deteriorating to this point, but it's a pretty common feature among product teams to prioritize new feature dev while existing (and way more core) features degrade.

KennyBlanken 4y ago
The Google Maps site and application are both deeply unpleasant to use if only because, when you click on a search result, the fucking map doesn't zoom in from an almost state-level view to show you where the fuck the listing is.
- potatolicious 4y ago
  Yeah, that's the kind of basic QoL stuff that's really the nuts and bolts of a product, but tend to get glossed over in planning. When working out their OKRs teams generally do not have an entire Objective for "iterate, prevent regressions, and improve upon existing core features". They instead have a litany of Objectives to add completely new things. Performance regressions are just the tip of that iceberg (though it is most noticeable), but a component of this upkeep is keeping up with the competition on your core already-shipped stuff. IMO Apple Maps' zoom behavior, rendering behavior, overlay behavior, etc, are far superior to Google's. It wouldn't be rocket science to implement their version of these changes, but clearly it's not being prioritized.
ploxiln 4y ago
> refactors, performance improvements, etc, are almost never appropriately rewarded or prioritized Another variant that I've seen: actual refinements and improvements just aren't simple enough to sell to management, or to new junior devs. So they'll be all excited about switching over to the next big framework that finally makes things fast and refreshing, for the first time ever! But the migration still has some bugs that haven't been worked out and some features that haven't been ported ... and a year later things are actually worse than where we started, so repeat with the next big framework that finally makes good software possible for the first time!
- potatolicious 4y ago
  This is a huge failure of management where it occurs - and one of the chief causes for it is short leadership tenure. Places where leadership remains stable over long periods tend to do better on this front. You pay for some big architectural overhaul, but you insist that you live on the new thing long enough to actually ship features before you go into the next cycle of renewal. But in some companies leadership tenure is short - execs stick around for 2-3 years max, which is just enough time to complete an architectural migration but not enough to ship stuff on it!
ImaCake 4y ago
I think google maps is close to the crisis point. Where niche alternatives exist, people will use those instead. One of the reasons I bought an iphone was to have apple maps as an alternative to google maps.

KennyBlanken 4y ago

The Google Maps site and application are both deeply unpleasant to use if only because, when you click on a search result, the fucking map doesn't zoom in from an almost state-level view to show you where the fuck the listing is.

potatolicious 4y ago
Yeah, that's the kind of basic QoL stuff that's really the nuts and bolts of a product, but tend to get glossed over in planning. When working out their OKRs teams generally do not have an entire Objective for "iterate, prevent regressions, and improve upon existing core features". They instead have a litany of Objectives to add completely new things. Performance regressions are just the tip of that iceberg (though it is most noticeable), but a component of this upkeep is keeping up with the competition on your core already-shipped stuff. IMO Apple Maps' zoom behavior, rendering behavior, overlay behavior, etc, are far superior to Google's. It wouldn't be rocket science to implement their version of these changes, but clearly it's not being prioritized.

ploxiln 4y ago

> refactors, performance improvements, etc, are almost never appropriately rewarded or prioritized Another variant that I've seen: actual refinements and improvements just aren't simple enough to sell to management, or to new junior devs. So they'll be all excited about switching over to the next big framework that finally makes things fast and refreshing, for the first time ever! But the migration still has some bugs that haven't been worked out and some features that haven't been ported ... and a year later things are actually worse than where we started, so repeat with the next big framework that finally makes good software possible for the first time!

potatolicious 4y ago
This is a huge failure of management where it occurs - and one of the chief causes for it is short leadership tenure. Places where leadership remains stable over long periods tend to do better on this front. You pay for some big architectural overhaul, but you insist that you live on the new thing long enough to actually ship features before you go into the next cycle of renewal. But in some companies leadership tenure is short - execs stick around for 2-3 years max, which is just enough time to complete an architectural migration but not enough to ship stuff on it!

ImaCake 4y ago

I think google maps is close to the crisis point. Where niche alternatives exist, people will use those instead. One of the reasons I bought an iphone was to have apple maps as an alternative to google maps.