The Eternal Cycle of Software Bloat

[asulyi]

speaker: Bryan Lunduke
topic: software and programmer quality
video: https://www.youtube.com/watch?v=qqs7FQ0sep8
length: 15'

I get Bryan know as "the guy that made 'Linux Sucks' talks at Linux conferences", but last (and possibly final) was 5 years ago.

He asks in this podcast, where and how 20-30 years hardware development gone. And he calls for programmers to write lean, light, fast software.

[attilammagyar]

I wasn't very optimistic about software becoming any faster in the foreseeable future, but this guy convinced me that there are good reasons for the industry to pay more attention to performance:

speaker: Casey Muratori
topic: Performance Excuses Debunked
video: https://www.youtube.com/watch?v=x2EOOJg8FkA&t=17s
length: 26'

The excuses that get debunked:

1. No need, today's hardware is so fast that we don't need to care about performance that much.
2. Result would be too small, optimizations would probably give no more than around 10% increase in performance.
3. Not worth it, if we can spend a fixed amount of time on performance improvements vs. developing new features, the latter is always more beneficial.
4. Niche: performance only matters for games, embedded sw, etc. but not for our field.
5. Hotspot: most programmers don't need to care about performance, most performance problems tend to concentrate around a few hotspots, so a few experts are enough to fix them.

Maybe we will do a double feature with these 2 videos.

[asulyi]

At interview:

This is great, your algorithm does exactly we asked you.
But, can you improve it to use O(n) time and O(1) space?