Hacker News new | past | comments | ask | show | jobs | submit login

This sounds like a proposal to make the review process giving more weight to reputation, unlike the current process which is supposed to be entirely technical[1]. This might be a good idea, but I can see how Mozilla would get a different set of complaints about reputation not being consistently evaluated.

[1] https://wiki.mozilla.org/Add-ons/Reviewers/Guide/Reviewing




That's a fair complaint, and I definitely agree that using reputation as a factor in the decision for an individual addon is a very bad idea. But why is that? (1) Because reputation does not imply trustworthiness. Someone could build up a reputation with a set of very proper addons, and then use that reputation to sneak in problematic ones. (2) Because it's unfair special treatment. The chosen person's addons would be subject to different standards than others'.

Again, this is gorhill. People are offering authors of popular addons some mind-bendingly large sums of money to sell out. (1) does not apply: gorhill is the author of the most popular addon, which implies that he has been offered if not the most money, at least a lot more than most. And the well-known history is that someone did make money off of his original version, that someone isn't him, and in response he rebirthed the addon that he didn't particularly want to maintain. Try to find someone with a more convincing backstory.

(2) is trickier, and it's why the distinction between uBlock Origin getting a free pass and gorhill being a reviewer makes sense to me, even if it seems like I'm just obscuring influence. As a reviewer, gorhill would be expected to not just automatically approve his own addons, but to apply the agreed upon evaluation criteria. This would be a farce if his integrity were in question, but see (1). It's pretty clear to see that he is the person most qualified to make that evaluation (heck, he's already doing it before releasing; he's not new to the game), so it comes down to trust.

Sure, I am not the best person to review my own code, no matter how honest I might be. But read the Technical Code Review portion of the link above[1], since it's the only part that matters here. There are some addons where those criteria might be difficult to evaluate, but we're not talking about those. If significant code changes cause those to be less clear cut, gorhill can always pass it by another reviewer. (Yes, this again requires trust. See (1).)

Plus, you don't even have to depend on (1). People can be skeptical and double-check, and news would get out very very quickly. (Even shortcomings in areas like a reproducible build would get called out.)

I don't see this being a wide open backdoor into the process. Not many people are going to come by with the #1 installed addon, together with the history of uBlock and uBlock Origin. Sure, factoring reputation into the process is fraught with problems, but I'm not suggesting that everyone above 1M installs gets grandfathered in. This slippery slope is bone dry and covered with cobblestones.

[1] https://wiki.mozilla.org/Add-ons/Reviewers/Guide/Reviewing


People should read this when they think about AI “Alignment”

Can’t even have a singular aligned person with full confidence


off of = from




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: