HN user

lowq

129 karma
Posts0
Comments33
View on HN
No posts found.

I sure don't miss the footguns and raw boilerplate that is having a copy constructor, move constructor, copy assignment operator, move assignment operator, and destructor, per class.

Yes, you should avoid the manual memory management that necessitates writing them. But work with a team of developers fresh out of school and next thing you know your codebase will be brimming with this kind of busywork.

"Whoever creates this stuff has no idea what pain is whatsoever."

I think it's clear that he is specifically responding to the the overall soullessness of the technique - to animate without a human understanding of what is being animated. But as others have pointed out this is well before modern AI image gen and I have been corrected in that aspect.

"Well, we would like to build a machine that can draw pictures like humans do"

"Would you?"

"Yes"

Awkward silence

From this I don't think it's difficult to extrapolate his feelings about modern AI image gen. But you are correct in that this is not a direct assessment. Appreciate the correction, thanks.

Guilty as charged. I don't think the leap was far but there was certainly a logical leap. Thanks for pointing that out.

Not to mention, 99^6 = 941,480,149,401 ≠ 99,000,000 (which TFA also quotes). But who's to say notation didn't degrade along with the rest of society? :^)

I hear you man. The fact that a `byte` in ROS's bootleg protobuf is actually signed, and `string` is ASCII-only means that if you want to send a file or binary blob you have to use a `uint8[]`, which of course C++ will turn into a `std::vector<uint8_t>`. Good luck turning that into a proper `std::string` without an extra copy. And don't get me started on the comically scattered documentation for any basic use case that only ChatGPT can seem to make sense of. Better yet, the complete insanity that is node parameters - ever heard of reading from a file? Can't believe anyone takes this seriously.

ZeroMQ, protobuf, and a little thought would go a long way at our org.

/rant

As a fairly avid Rust user I can say I'm excited to hear that! Just playing a little devil's advocate for fun.

That said I was moderately upset by this Friday after carefully assembling a development container for a project at work. Guess I'll have to make something persistent somewhere if I don't want to twiddle my thumbs before each build.

Agreed. I find that "smart objects" are much more difficult to make cohesive with one another. Punting your "smart" logic to a higher level is easier to understand, test, and change.

If the probe was single ended (e.g. not a differential probe or handheld unit), you may have shorted something to ground without realizing it. Was a long time before I figured that one out.

That's a good one. I've been trying to find a source for: "If you try to write your own TCP, you'll end up with TCP, but slow."

Cannot for the life of me figure out where I read this.

Holy cow. I never considered this!

I guess it would also explain why the games I would write in 2011 on a netbook ran so smoothly on every other computer I tested them on, despite using the legacy OpenGL 1.1 API, etc. etc.

In our process, we've been consistently told that a weight of 8 corresponds to a two week (10d) sprint, and that 80% of our time should be spent on scrum tasks. Then, if any developer closes more than 8 points in a sprint, we are told that we aren't estimating properly. They have all but equated 1 point to 1 day, but the scrum master consistently denies that weight corresponds to time.

I'm new to this. Are we agile?