10 ms·

I feel like nobody really cares about optimization anymore. It's just "ship, ship, ship, move faster, ship." So there's a lot of tech-debt, a lot of unoptimized

I feel like nobody really cares about optimization anymore. It's just "ship, ship, ship, move faster, ship." So there's a lot of tech-debt, a lot of unoptimized code, and a lot of "just make it work", at the expense of writing software that utilities the hardware properly. On the flip side of that though, hardware has gotten fast enough, where this is possible.

ymerej 3y ago
Doesn't it all boil down to economics in the end? When I worked at AOL, QA was very important/valued because our SW was shipped on CDs (I know, I know), and every bug/update cost us users because we had to download an update at modem speed. Now that I work on a web development team, bugs are easily fixed by an update. So who cares if there's a bug? We can just update the site. There are other costs, of course -- if we tolerate lots of bugs and tech debt, making development slower, we will evenually have our lunch eaten -- that's economics too. Users/customers accept a certain level of quality, and it doesn't make short-term economic sense to provide more.
- ado__dev 3y ago
  Absolutely. It really is a double-edged sword and there's so many factors that play into. Do you spend a ton of time writing beautifully optimized code, but ship less features, or do you ship a ton of features that mostly work. I feel like if you take too long to ship, a competitor will eat your lunch. But if everything you ship is trash, then a competitor will also eat your lunch. All about finding that balance I guess.
- strangattractor 3y ago
  Total agree. There was a time when the deployment cycle was so long it was painful to ship poorly performing buggy software. Now people let their users do the testing.
wredue 3y ago
It’s actually worse than you think. Functional programmers have been in a massive propaganda campaign to actively paint optimization not just as “usually not necessary”, but actually as a negative thing. Ask any developer what they think of optimization, and the vast majority will respond “optimization is the root of all evil” or some other nonsense variant. Now that they’ve successfully paint optimization as bad, they’re working on painting any profiling or measurement at all as bad. They need this because the performance characteristics of pure functional programming is pretty dogshit. You also find hordes of developers that selectively care. Watching Python developers shit on Java for “being too slow” is getting pretty old.
- ado__dev 3y ago
  Yeah and personally I've been on both ends of the spectrum. I think there is something to be said for pre-mature optimization and over-optimization. But I do feel like it's a lost art these days.
  - wredue 3y ago
    The quote means to not assume something is slow, but to actually measure it. Pure Functional Programmers state that you should not even measure performance. They take this position because despite their many claims that Pure FP is performant, it turns out that if you measure it, it’s not performant. Hence “don’t measure”.
- Xeamek 3y ago
  I don't think it's just about (or even mainly) functional programmers. Look on something like java or kotlin, that are inherently slowing down EVERY.SINGLE.VARRIABLE. access by boxing them into objects. On other hand there is rust, which takes quite a bit of inspiration from fp aproach but obviously still cares about the performance.
  - rightbyte 3y ago
    Java ints etc are not boxed, right?
bgirard 3y ago
> I feel like nobody really cares about optimization anymore. You have to consider the survivorship bias. Maybe the people that care about optimization don't 'survive' to keep caring about it. > It's just "ship, ship, ship, move faster, ship." Are startup with that mentality more likely to do well? I've built my career around performance work and I've noticed that there a time and a place for everything. Sometimes shipping is more important, and sometimes performance can be a day one defining feature.
SubjectToChange 3y ago
I feel like nobody really cares about optimization anymore. Until customers care about performance/optimization, most businesses won't care. Eventually there will come a day where hardware will stagnate sufficiently to incentivize software efficiency.
- troupo 3y ago
  > Until customers care about performance/optimization They do, Businesses and programmers are incapable of listening to them
  - Gigachad 3y ago
    Businesses and programmers listen to money. When they don’t, they go out of business. Users primarily care about features first and speed second.
    - troupo 3y ago
      Users do care about speed. They complain about it all the time, you just have to listen to them. It's most of the "features" that users couldn't give a crap.
      - Gigachad 3y ago
        So why aren’t the users picking the software that’s faster instead of the ones with more features. Why does everyone use the “bloated” Gmail webui instead of some barebones desktop client that’s a few ms faster?
        troupo 3y ago
        > So why aren’t the users picking the software that’s faster This software exists? When all your alternatives are bloated enshittificated crap, what exactly can I as a user chose?
    - bufio 3y ago

ymerej 3y ago

Doesn't it all boil down to economics in the end? When I worked at AOL, QA was very important/valued because our SW was shipped on CDs (I know, I know), and every bug/update cost us users because we had to download an update at modem speed. Now that I work on a web development team, bugs are easily fixed by an update. So who cares if there's a bug? We can just update the site. There are other costs, of course -- if we tolerate lots of bugs and tech debt, making development slower, we will evenually have our lunch eaten -- that's economics too. Users/customers accept a certain level of quality, and it doesn't make short-term economic sense to provide more.

ado__dev 3y ago
Absolutely. It really is a double-edged sword and there's so many factors that play into. Do you spend a ton of time writing beautifully optimized code, but ship less features, or do you ship a ton of features that mostly work. I feel like if you take too long to ship, a competitor will eat your lunch. But if everything you ship is trash, then a competitor will also eat your lunch. All about finding that balance I guess.
strangattractor 3y ago
Total agree. There was a time when the deployment cycle was so long it was painful to ship poorly performing buggy software. Now people let their users do the testing.

wredue 3y ago

It’s actually worse than you think. Functional programmers have been in a massive propaganda campaign to actively paint optimization not just as “usually not necessary”, but actually as a negative thing. Ask any developer what they think of optimization, and the vast majority will respond “optimization is the root of all evil” or some other nonsense variant. Now that they’ve successfully paint optimization as bad, they’re working on painting any profiling or measurement at all as bad. They need this because the performance characteristics of pure functional programming is pretty dogshit. You also find hordes of developers that selectively care. Watching Python developers shit on Java for “being too slow” is getting pretty old.

ado__dev 3y ago
Yeah and personally I've been on both ends of the spectrum. I think there is something to be said for pre-mature optimization and over-optimization. But I do feel like it's a lost art these days.
- wredue 3y ago
  The quote means to not assume something is slow, but to actually measure it. Pure Functional Programmers state that you should not even measure performance. They take this position because despite their many claims that Pure FP is performant, it turns out that if you measure it, it’s not performant. Hence “don’t measure”.
Xeamek 3y ago
I don't think it's just about (or even mainly) functional programmers. Look on something like java or kotlin, that are inherently slowing down EVERY.SINGLE.VARRIABLE. access by boxing them into objects. On other hand there is rust, which takes quite a bit of inspiration from fp aproach but obviously still cares about the performance.
- rightbyte 3y ago
  Java ints etc are not boxed, right?

bgirard 3y ago

> I feel like nobody really cares about optimization anymore. You have to consider the survivorship bias. Maybe the people that care about optimization don't 'survive' to keep caring about it. > It's just "ship, ship, ship, move faster, ship." Are startup with that mentality more likely to do well? I've built my career around performance work and I've noticed that there a time and a place for everything. Sometimes shipping is more important, and sometimes performance can be a day one defining feature.

SubjectToChange 3y ago

I feel like nobody really cares about optimization anymore. Until customers care about performance/optimization, most businesses won't care. Eventually there will come a day where hardware will stagnate sufficiently to incentivize software efficiency.

troupo 3y ago
> Until customers care about performance/optimization They do, Businesses and programmers are incapable of listening to them
- Gigachad 3y ago
  Businesses and programmers listen to money. When they don’t, they go out of business. Users primarily care about features first and speed second.
  - troupo 3y ago
    Users do care about speed. They complain about it all the time, you just have to listen to them. It's most of the "features" that users couldn't give a crap.
    - Gigachad 3y ago
      So why aren’t the users picking the software that’s faster instead of the ones with more features. Why does everyone use the “bloated” Gmail webui instead of some barebones desktop client that’s a few ms faster?
      - troupo 3y ago
        > So why aren’t the users picking the software that’s faster This software exists? When all your alternatives are bloated enshittificated crap, what exactly can I as a user chose?
  - bufio 3y ago