4 ms·

This bizarre reasoning puzzles and infuriates me so much. The linked post shaves more than 40% of an app's size in 5 minutes, by doing nothing but being curious

This bizarre reasoning puzzles and infuriates me so much. The linked post shaves more than 40% of an app's size in 5 minutes, by doing nothing but being curious about the half-gigabyte app and knowing how to use a couple of tools. This is not a case of "I have plenty of water so let the kids splash some in the playground to have fun", this is a case of "I have plenty of water so I will load 10000 liters in a truck and drive 1000 kilometers and back just to dump them in the desert for no fucking reason". Seriously, 5 minutes? I spend more than that in the bathroom. How many 5-minutes is a "Scrum standup"? a "progress meeting" ? a <insert bullshit corporate activity here>? Do all of those things not waste the engineers' time? I'm trying to imagine the world if every profession and craft tried to justify this disgustingly irresponsible mindset : " - This combustion engine uses an inferior injection mechanism that wastes 40% more fuel compared to latest... = Don't waste my time!!! I have more important things to worry about. -.... It would take 5 minutes to fix = Customers don't notice the fuel consumption anyway, so that means robbing them out of money with inferior engineering is okay. Now hurry up, we'll miss the daily 3-hours progress meeting. " If only Donald Knuth knew what " Forget about the small efficiencies" would be used to justify.

rixed 5y ago
I share your feeling. Day by day I feel more guilty of belonging to that industry because of the laziness, sloppiness, negligence, acceptance of spying, of dark patterns, etc, and I do not disclose what I'm doing for a living more readily than I would if I worked for some mafia.
ironmagma 5y ago
There’s probably an observability bias here. It could be that the app sizes would be 3x worse, if no one ever tried. You would never know, because you only see the app sizes after people have done the work.
- Dylan16807 5y ago
  That would mean they did the 3 minute cleanup and then declared their work done. That's not a sufficient amount of trying.
  - ironmagma 5y ago
    This is in response to someone who’s clearly infuriated by a wide variety of inefficiencies they see. The point is, no one is praising the people who do optimize their <whatever> because you would never know it was even an issue that got solved.
    - Dylan16807 5y ago
      Oh, nobody is praising the other apps? That's true but these are very basic expectations. If you're a chef and you burn every tenth meal then either you or your management are making very bad decisions, even if the other 90% are fine. And app and website bloat seems to be significantly more widespread than that. The idea isn't that nobody ever tries, it's that the typical amount of trying is really bad.
      - ironmagma 5y ago
        When eating a meal, how the meal tastes (burnt) is at least #2 on my top priorities (just below whether it will lead to illness). When consuming an app though, bundle size has got to be at least #5, far below something like "Does the app do what I want?" and "Is it performant" etc. While it would be nice if everything was optimized, it's really not comparable to a meal tasting good, since that's basically the main reason to have a meal in the first place. Most people don't care about bundle size, especially when it's tens of megabytes versus gigabytes. > the typical amount of trying is really bad Again, we really have no way of measuring just how bad it could be. IMO it feels like everything could be way, way worse.
        Dylan16807 5y ago
        > When consuming an app though, bundle size has got to be at least #5, far below something like "Does the app do what I want?" and "Is it performant" etc. This much space is a pretty big cost. It represents wasted dollars, or shuffling data, or not getting it downloaded in time, or not having room for the pictures you want, etc. > Again, we really have no way of measuring just how bad it could be. If five minutes did this much, and this is within a couple orders of magnitude of typical, then we can conclude that the typical effort is really bad. How much worse it could be might be an interesting thought experiment, but doesn't affect the above conclusion. It could probably get almost infinitely bad if there was zero cleanup effort. But that's the difference between 1/10 effort and 2/10 effort. If I'm demanding 6/10 effort for a professional product, the difference between 1/10 and 2/10 doesn't matter to me.
stjohnswarts 5y ago
No doubt, there is something to be said for prioritizing concerns for your app but apparently they just gave up completely lol
saagarjha 5y ago
There’s a parable about a mechanic that spends a couple minutes finding a screw to tighten and charges a large sum once they’re done. It was a few minutes of work for them, yes, but it took a lifetime for them to get to the point where it’s a few minutes of work. Of course, the situation here is not quite that extreme, but acquiring the skills needed to know where problems are is uncommon and nontrivial.

rixed 5y ago

I share your feeling. Day by day I feel more guilty of belonging to that industry because of the laziness, sloppiness, negligence, acceptance of spying, of dark patterns, etc, and I do not disclose what I'm doing for a living more readily than I would if I worked for some mafia.

ironmagma 5y ago

There’s probably an observability bias here. It could be that the app sizes would be 3x worse, if no one ever tried. You would never know, because you only see the app sizes after people have done the work.

Dylan16807 5y ago
That would mean they did the 3 minute cleanup and then declared their work done. That's not a sufficient amount of trying.
- ironmagma 5y ago
  This is in response to someone who’s clearly infuriated by a wide variety of inefficiencies they see. The point is, no one is praising the people who do optimize their <whatever> because you would never know it was even an issue that got solved.
  - Dylan16807 5y ago
    Oh, nobody is praising the other apps? That's true but these are very basic expectations. If you're a chef and you burn every tenth meal then either you or your management are making very bad decisions, even if the other 90% are fine. And app and website bloat seems to be significantly more widespread than that. The idea isn't that nobody ever tries, it's that the typical amount of trying is really bad.
    - ironmagma 5y ago
      When eating a meal, how the meal tastes (burnt) is at least #2 on my top priorities (just below whether it will lead to illness). When consuming an app though, bundle size has got to be at least #5, far below something like "Does the app do what I want?" and "Is it performant" etc. While it would be nice if everything was optimized, it's really not comparable to a meal tasting good, since that's basically the main reason to have a meal in the first place. Most people don't care about bundle size, especially when it's tens of megabytes versus gigabytes. > the typical amount of trying is really bad Again, we really have no way of measuring just how bad it could be. IMO it feels like everything could be way, way worse.
      - Dylan16807 5y ago
        > When consuming an app though, bundle size has got to be at least #5, far below something like "Does the app do what I want?" and "Is it performant" etc. This much space is a pretty big cost. It represents wasted dollars, or shuffling data, or not getting it downloaded in time, or not having room for the pictures you want, etc. > Again, we really have no way of measuring just how bad it could be. If five minutes did this much, and this is within a couple orders of magnitude of typical, then we can conclude that the typical effort is really bad. How much worse it could be might be an interesting thought experiment, but doesn't affect the above conclusion. It could probably get almost infinitely bad if there was zero cleanup effort. But that's the difference between 1/10 effort and 2/10 effort. If I'm demanding 6/10 effort for a professional product, the difference between 1/10 and 2/10 doesn't matter to me.

stjohnswarts 5y ago

No doubt, there is something to be said for prioritizing concerns for your app but apparently they just gave up completely lol

saagarjha 5y ago

There’s a parable about a mechanic that spends a couple minutes finding a screw to tighten and charges a large sum once they’re done. It was a few minutes of work for them, yes, but it took a lifetime for them to get to the point where it’s a few minutes of work. Of course, the situation here is not quite that extreme, but acquiring the skills needed to know where problems are is uncommon and nontrivial.