top | item 24906376

How Did Things Ever Get This Good? (2009)

82 points| Tomte | 5 years ago |prog21.dadgum.com | reply

36 comments

order
[+] WalterBright|5 years ago|reply
> there must be a reason that pros can buy two-thousand dollar lenses and three-thousand dollar camera bodies.

It's pretty simple. If you make your living with a camera, the marginal improvement of high dollar equipment becomes worthwhile, as it is amortized over a lot of pictures that you make money from.

For example, in high school, I had a soldering iron and struggled to make good solder joints. They were always globby, ugly messes. I read and followed the directions, it just didn't work out.

In college I got a job soldering circuit boards. The soldering iron was a professional one, costing maybe 20x what the ones I used in high school cost. But suddenly, it was easy to get professional results soldering.

[+] malandrew|5 years ago|reply
> the marginal improvement of high dollar equipment becomes worthwhile, as it is amortized over a lot of pictures that you make money from.

I'm not even sure it is the marginal improvement, but the increased likelihood of experiencing the high cost of re-work. Rework is the last thing you want when your time is valuable.

A former partner of mine was a documentary film-maker and I observed on many occasions times where she spent way to much time trying to make up for bad or out of sync audio. I encouraged her to make changes in her equipment, but not because of the minor improvements in video or quality over her current equipment, but just because the likelihood of rework would be much lower. The video and audio quality at the end was very comparable, but the cost to get from A to B ended up being much better time-wise.

[+] taneq|5 years ago|reply
Also (at least where I live) business expenses are typically tax deductible, so the effective cost can be 20-30% lower than if you bought it privately.
[+] StavrosK|5 years ago|reply
That analogy doesn't exactly work, pro cameras do two main things differently: They allow you to take shots you couldn't before (because of some extra feature) and they allow you to do it much more easily than before (because of much better ergonomics).

It's more like if you could solder and get excellent through-hole joints, but you just couldn't solder SMD at all until you got a better iron.

[+] z3t4|5 years ago|reply
One thing that fascinated me in web development before we had Canvas or WebGL was 3d engines implemented using HTML tables, because it was the most performant way to put a triangle on the screen. Meanwhile we had very good looking 3d games built on platforms/languages with hardware level/access... That 3d game made with JavaScript and HTML tables was amazing, but it was also a "potato camera".
[+] manuisin|5 years ago|reply
That sounds very interesting. Would you happen to link to that? I tried searching for it but didn't turn up anything.
[+] muststopmyths|5 years ago|reply
>Even a language like Ruby, which tends to hang near the bottom of any performance-oriented benchmark, is thousands of times faster than BASICs that people were learning to program 8-bit home computers with in the 1980s. That's not an exaggeration, I do mean thousands.

I don't buy it. Unless we're comparing the computers BASIC ran on to computers today.

[+] pwinnski|5 years ago|reply
I think it's pretty clear that's exactly what we're doing.

BASIC on a 6502, Ruby on a VM atop an AMD Ryzen or some such.

[+] m_mueller|5 years ago|reply
What I would agree with is a factor of 1000 in programming productivity between e.g. python and BASIC for many common usecases. Having such a powerful standard lib is a huge advantage.
[+] zeroxfe|5 years ago|reply
This is a nice idea, but in the real world it really helps to align your language choice with your goals. When you're working in teams, especially large teams, then things like popularity, ecosystem, tooling, and even project fit hugely amplify overall productivity.
[+] brundolf|5 years ago|reply
> Most slams against individual languages are meaningless in the overall scheme of things. If you like Lisp, go for it.

Hah

Edit: Lots of people not noticing the scheme pun

[+] solarmist|5 years ago|reply
What is the pun? I’m not seeing it. Just that it uses scheme or is it more than just that?
[+] AnimalMuppet|5 years ago|reply
That's a pretty un-useful comment. If you actually said something specific, we might know what your point is.

I would say: If you like Lisp, all things considered, then use it.

Some things to consider: Coworkers (both existing and future). Ecosystem. Maintainability (including by people other than yourself). And, worst of all, management.

[+] jasperry|5 years ago|reply
I agree in the sense that, compared to the staggering complexity of the web/mobile as an application platform, the weaknesses of languages themselves seem pretty small (possibly excepting the lack of memory safety in C/C++.) Tooling and ecosystem are also important, but if the platform isn't overly complex there's less need for tooling. I feel like hackers who just want the joy of creating something cool should stay away from the web or mobile platforms, because it's too hard to focus on the problem itself instead of platform/dependencies/deployment issues.
[+] wollef|5 years ago|reply
I've found web to be great: you just need a single file if that's all you want. All the modern frameworks and tooling are complicated, but easy to ignore for a small project.
[+] santoshalper|5 years ago|reply
Talk about an article that felt like it ended before it began. He just makes the point and then ends it with very little support.