From my own perspective, however, the issue is about impact and responsibility.
The excess traffic is their impact as it is caused by their design and is entirely of their choosing. Thus it becomes their responsibility.
The workaround they offer (as with so much in tech) means their stuff becomes opt-out and not opt-in. By which I mean their own decisions/designs cause an impact where they require everyone else suffering from it to specifically and individually opt-out of being excessively polled.
This is an outrageous requirement they can only impose due to the power imbalance.
I think its primarily about the Go team having a different working philosophy than the standard HN soft. dev philosophy, and an intolerance of difference.
Go is my second fave after C# and I've used it almost daily for many years. No animosity here.
I'm not ignoring that Drew complained and they offered to dial it down. That's a reasonable response if it were an isolated case. The problem is that it isn't an isolated case. This is how the module cache works, which means it is doing that kind of load against other module sources too.
If you're going to impose these resource demands on others then they should be managed for everyone affected not just the one who complains about it.
- Repeatedly pulling an entire repo over and over again is an antisocial act that is unfairly imposing bandwidth/CPU costs on third parties
- They should not expect every impacted party to obtain an opt-out individually because of unasked-for extra resource demands
- The system that excessively polls is broken, not its victims
This lets Google collect very granular data on who uses which modules, and also makes it opaque to webmasters who's using their services (it's all hidden behind google). It's honestly pretty nasty how Google has hidden this kind of "analytics" within a pretty dark grey area.
At the same time, the bot that handles the proxy seems to crawl everything, very often, and excessively, effectively DDOS'ing anyone hosting go modules.
It's OK to just not know, but if you don't know, it's weird to have strong opinions about it.
I feel like what I'm sticking up for here is the practice of software development. Building an automated system that generates unexpectedly unwelcome load on someone else's service is... not exactly front-page news? It happens basically all the time? The idea that because the Go team is sponsored by Google, nothing like this should ever happen seems deeply unrealistic. "We can push a button to make this stop happening; the tradeoff is that other hosting services that don't care about this load might currently have fresher cache entries [whatever that means]" seems like a perfectly cromulent response.
He should just tell them to push the button. If you think he shouldn't, you should be able to say why.
I do have a mild rooting interest here: I think the Go module proxy is pretty neat, and does something interesting for the security of the ecosystem. DeVault disagrees. That's fine, disagreement keeps things interesting.
Accepting that Drew should get them to just push the button is accepting that it is the responsibility of each victim individually to cope with the abusive load being sent their way by engaging directly with their abuser. Rather than the sender, who is truly the one responsible, fixing it for every target by reducing their polling at source.
If the behavior is unimportant enough that drew should accept eliminating it, then its value does not justify the load.
If the behavior is important enough to justify the load, then eliminating it is no solution.
It feels like some comments here are trying to reconstruct what happened axiomatically purely from comments on the thread, without any empirical input from, like, how the Go module proxy actually works, and ending up in weird places as a result.
The abuser of resources should stop abusing. It isn't the fault of the victims and they shouldn't each individually need to address it. Stop the issue at source for everyone by stopping being anti-social/parasitical on the use of resources through excessive polling. I think that's the extent of my own argument here.
If they knew DeVault's host wasn't going to be able to handle repeated module clones, a priori, then doing it anyways would be a problem. It doesn't look like anyone expected this to be a problem. It turned out it was, and there's a fix.
I think you don't always get the latest stuff due to caching or so.