←back to thread

837 points turrini | 4 comments | | HN request time: 0.676s | source
Show context
titzer ◴[] No.43971962[source]
I like to point out that since ~1980, computing power has increased about 1000X.

If dynamic array bounds checking cost 5% (narrator: it is far less than that), and we turned it on everywhere, we could have computers that are just a mere 950X faster.

If you went back in time to 1980 and offered the following choice:

I'll give you a computer that runs 950X faster and doesn't have a huge class of memory safety vulnerabilities, and you can debug your programs orders of magnitude more easily, or you can have a computer that runs 1000X faster and software will be just as buggy, or worse, and debugging will be even more of a nightmare.

People would have their minds blown at 950X. You wouldn't even have to offer 1000X. But guess what we chose...

Personally I think the 1000Xers kinda ruined things for the rest of us.

replies(20): >>43971976 #>>43971990 #>>43972050 #>>43972107 #>>43972135 #>>43972158 #>>43972246 #>>43972469 #>>43972619 #>>43972675 #>>43972888 #>>43972915 #>>43973104 #>>43973584 #>>43973716 #>>43974422 #>>43976383 #>>43977351 #>>43978286 #>>43978303 #
_aavaa_ ◴[] No.43972050[source]
Except we've squandered that 1000x not on bounds checking but on countless layers of abstractions and inefficiency.
replies(6): >>43972103 #>>43972130 #>>43972215 #>>43974876 #>>43976159 #>>43983438 #
Gigachad ◴[] No.43972215[source]
Am I taking crazy pills or are programs not nearly as slow as HN comments make them out to be? Almost everything loads instantly on my 2021 MacBook and 2020 iPhone. Every program is incredibly responsive. 5 year old mobile CPUs load modern SPA web apps with no problems.

The only thing I can think of that’s slow is Autodesk Fusion starting up. Not really sure how they made that so bad but everything else seems super snappy.

replies(40): >>43972245 #>>43972248 #>>43972259 #>>43972269 #>>43972273 #>>43972292 #>>43972294 #>>43972349 #>>43972354 #>>43972450 #>>43972466 #>>43972520 #>>43972548 #>>43972605 #>>43972640 #>>43972676 #>>43972867 #>>43972937 #>>43973040 #>>43973065 #>>43973220 #>>43973431 #>>43973492 #>>43973705 #>>43973897 #>>43974192 #>>43974413 #>>43975741 #>>43975999 #>>43976270 #>>43976554 #>>43978315 #>>43978579 #>>43981119 #>>43981143 #>>43981157 #>>43981178 #>>43981196 #>>43983337 #>>43984465 #
mjburgess ◴[] No.43972245[source]
People conflat the insanity of running a network cable through every application with the poor performance of their computers.
replies(1): >>43972796 #
sgarland ◴[] No.43972796[source]
Correction: devs have made the mistake of turning everything into remote calls, without having any understanding as to the performance implications of doing so.

Sonos’ app is a perfect example of this. The old app controlled everything locally, since the speakers set up their own wireless mesh network. This worked fantastically well. Someone at Sonos got the bright idea to completely rewrite the app such that it wasn’t even backwards-compatible with older hardware, and everything is now a remote calls. Changing volume? Phone —> Router —> WAN —> Cloud —> Router —> Speakers. Just… WHY. This failed so spectacularly that the CEO responsible stepped down / was forced out, and the new one claims that fixing the app is his top priority. We’ll see.

replies(2): >>43972959 #>>43974448 #
1. mjburgess ◴[] No.43972959[source]
Presumably they wanted the telemetry. It's not clear that this was a dev-initiated switch.

Perhaps we can blame the 'statistical monetization' policies of adtech and then AI for all this -- i'm not entirely sold on developers.

What, after all, is the difference between an `/etc/hosts` set of loop'd records vs. an ISP's dns -- as far as the software goes?

replies(2): >>43973026 #>>43974584 #
2. sgarland ◴[] No.43973026[source]
You’re right, and I shouldn’t necessarily blame devs for the idea, though I do blame their CTO for not standing up to it if nothing else.

Though it’s also unclear to me in this particular case why they couldn’t collect commands being issued, and then batch-send them hourly, daily, etc. instead of having each one route through the cloud.

3. skydhash ◴[] No.43974584[source]
> Presumably they wanted the telemetry

Why not log them to a file and cron a script to upload the data? Even if the feature request is nonsensical, you can architect a solution that respect the platform's constraints. It's kinda like when people drag in React and Next.js just to have a static website.

replies(1): >>43977729 #
4. arolihas ◴[] No.43977729[source]
someone out there now has a cool resume line item about doing real time cloud microservices on the edge