Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Software Disenchantment (2018)

tonsky.me

Post reply on HN
Print view View original

161–170 of 504 posts

Re: Software Disenchantment (2018)

#161

Post by TeMPOraL »

buboard wrote: post #87
rogerdb wrote: post #20

> Would you buy a car if it eats 100 liters per 100 kilometers? How about 1000 liters? I think the analogy here is backwards. The better question is "how much would you prioritize a car that used only 0.05 liters per 100km over one that used 0.5? What about one that used only 0.005L?". I'd say that at that point, other factors like comfort, performance, base price, etc. become (relatively) much more important. If bas…

I dont know, that just feels wrong. If anything, the rise of mobile means there should be more emphasis on speed. All the bloat is because of misguided aesthetics (which all look the same, as if designers move between companies every year, which they do) and fanciness. Can you point to a newish app that is clearly better that its predecessor?

I think there are two factors preventing mobile from being a force to drive performance optimizations.

One, phone OSes are being designed for single-tasked use. Outside of alarms and notifications in the background (which tend to be routed through a common service), the user can see just one app at a time, and mobile OSes actively restrict background activity of other apps. So every application can get away with the assumption that it's the sole owner of the phone's resources.

Two, given the above, the most noticeable problem is now power usage. As Moore's law has all but evaporated for single-threaded performance, hardware is now being upgraded for multicore and (important here) power performance. So apps can get away with poor engineering, because every new generation of smartphones has a more power-efficient CPU, so the lifetime on single charge doesn't degrade.

Re: Software Disenchantment (2018)

#163

Post by TeMPOraL »

buboard wrote: post #68

He has a nice follow up which gets to the reasons why https://tonsky.me/blog/good-times-weak-men/ Another take: rewrites and rehashes tend to be bad because they are not exciting for programmers. Everything you re about to write is predictable, nothing looks Clearly better and it just feels forced. First versions of anything are exciting, the possibilities are endless, and even if the choices along the path are subop…

Yet somehow, it seems to me that most software - including all the "innovative" hot companies - are mostly rewriting what came before, just in a different tech stack. So how come nobody wants to rewrite the prior art to be faster than it was before?

Re: Software Disenchantment (2018)

#164

Post by _trampeltier »

hermitcrab wrote: post #155

As a dinosaur who has been programming for >30 years, it shocks me how bloated many modern programs are. The installers for my own products are around 20MB and most of that is Qt libraries. But 100s of MB seems to be standard now. The Airbnb app on my iPhone is 210 MB. I understand that if you are shipping a 3d game with maps, textures, sound etc, but not for a mobile phone app.

The sound driver on my ZBook at work does need almost 500MB Ram. No idea what is in there .. maybe HP has these days not just keyloggers included in the drivers, maybe the have screengrabbers too.

Re: Software Disenchantment (2018)

#165

Post by gavanwoolery »

rogerdb wrote: post #20

> Would you buy a car if it eats 100 liters per 100 kilometers? How about 1000 liters? I think the analogy here is backwards. The better question is "how much would you prioritize a car that used only 0.05 liters per 100km over one that used 0.5? What about one that used only 0.005L?". I'd say that at that point, other factors like comfort, performance, base price, etc. become (relatively) much more important. If bas…

I respectfully disagree -- something that is 10 times more efficient costs 10 times less energy (theoretically). When the end user suffers a server outage due to load, when they run out of battery ten times quicker, all of these things matter. When you have to pay for ten servers to run your product instead of one, this cost gets passed on to the end user.

I was forced to use a monitor at 30 fps for a few days due to a bad display setup. It made me realize how important 60 fps is. Even worse, try using an OS running in a VM for an extended period of time...

There are plenty of things that are 'good enough', but once users get used to something better they will never go back (if they have the choice, at least).

Re: Software Disenchantment (2018)

#166

Post by arithma »

Users don't know about this performance gap, but there is a gap, and that gap is an economic opportunity.

Since the current practices are so inefficient, users are having to pay for this from pocket as hardware expenses. Buy a $1000 smartphone, to get the same experience as last-year's.

A different stack (don't need to reinvent the universe), can be branded as brutally efficient, uses slim hardware, and strict engineering practices to provide a much better experience at a fraction of the price (1/5 possible.)

I believe nobody is doing that yet since there are two main barriers:

1- Risk of no-market (I think this might be proven unfounded, given the current trend in price hikes)

2- Capital investment necessary to get started (But this also can be solved given the obvious appetite for the next-big-thing money being poured left and right, without anything catching on yet)

Re: Software Disenchantment (2018)

#167

Post by UweSchmidt »

jupp0r wrote: post #47
johnr2 wrote: post #35

One thing nobody seems to mention is the environmental cost of inefficient software. All those wasted CPU cycles consume electricity. A single laptop or phone on it's own is insignificant, but there are billions of them. Combine that with the energy wasted shovelling unnecessary crap around the internet, and it adds up to a big CO2 problem that nobody talks about.

I hear that argument very frequently and I don’t buy it. Think about all the gas that is saved because people don’t have to drive to the library, all the plane trips saved by video conferencing, all the photo film, all the sheets of paper in file cabinets, all the letters being sent as emails, all the mail order catalogues, ... you get the idea. Does anybody know of a comprehensive study on this?

If websites and business software were as lean as they could be, most computers could have amazingly weak, low-powered processors.

I'm quite disenchanted with software myself. It takes way too long to open any program, for this JIRA ticket to properly display.

One thing that has improved was boot times, I seem to remember that Windows 7 was quite a bit faster than XP. Maybe someone in upper management wanted it to be as fast as MacOS? So speed IS possible, if it is prioritized.

Re: Software Disenchantment (2018)

#168

Post by jodrellblank »

johnwatson11218 wrote: post #119
MaxBarraclough wrote: post #113

Earlier quoted context omitted.

> I'm thinking why not train a LSTM to take in http requests and generate the http response? Why? With responses generated according to what? Are you really just suggesting using neural networks in the compiler's optimiser? > Then try using a smaller network until something like your registration flow, or a simple content management system was just a bunch of floating point numbers in some matrices saved off to disk.…

I'm suggesting that you take an existing system and build up a corpus of request/response pairs. Then you use the LSTM to build a prediction model so that given a request it will tell you that the current production system will produce the following sql statement and this http response. Once the LSTM's output is indistinguishable from your current production system , for all use cases, then you replace the production…

The main idea being that for a given input the output really is 100% predictable [..] I think it would be super fast once you got it working.

I imagine it would be fast, then you realise you've made a static content caching layer out of a neural network and replace it with Varnish cache and it would be hyper fast.

Re: Software Disenchantment (2018)

#169

Post by kinkrtyavimoodh »

gavanwoolery wrote: post #165
rogerdb wrote: post #20

> Would you buy a car if it eats 100 liters per 100 kilometers? How about 1000 liters? I think the analogy here is backwards. The better question is "how much would you prioritize a car that used only 0.05 liters per 100km over one that used 0.5? What about one that used only 0.005L?". I'd say that at that point, other factors like comfort, performance, base price, etc. become (relatively) much more important. If bas…

I respectfully disagree -- something that is 10 times more efficient costs 10 times less energy (theoretically). When the end user suffers a server outage due to load, when they run out of battery ten times quicker, all of these things matter. When you have to pay for ten servers to run your product instead of one, this cost gets passed on to the end user. I was forced to use a monitor at 30 fps for a few days due to…

Yes, but it's not just relative quantities that matter, absolute values matter too, just as the post you replied to was saying.

Optimizing for microseconds when bad UI steals seconds is being penny-wise and pound foolish. Business might not understand tech but they do generally understand how it ends up on the balance sheet.

Re: Software Disenchantment (2018)

#170

Post by peter_d_sherman »

Excerpt:

"An Android system with no apps takes up almost 6 GB. Just think for a second about how obscenely HUGE that number is. What’s in there, HD movies? I guess it’s basically code: kernel, drivers. Some string and resources too, sure, but those can’t be big. So, how many drivers do you need for a phone?

Windows 95 was 30MB. Today we have web pages heavier than that!

Windows 10 is 4GB, which is 133 times as big. But is it 133 times as superior? I mean, functionally they are basically the same. Yes, we have Cortana, but I doubt it takes 3970 MB. But whatever Windows 10 is, is Android really 150% of that?"

My favorite line: "Windows 95 was 30MB. Today we have web pages heavier than that!"

If there's a new saying for 2020, it shouldn't be that "hindsight is 2020"...

Also... each web page should come with a non-closable pop-up box that says "Would you like to download a free entire OS with your web page?", and offers the following "choices":

"[Yes] [Yes] [Cancel (Yes, Do It Anyway!)]".

Post reply on HN