Because this sounded outlandish, I decided to have a look on my own system. Since the author didn't give any technical details, I grepped `~/Library` and `/Library` for files containing `keystone` in either name or contents. This yielded the following interesting candidates, along with similar entries for daemons etc:
They all reference the same binary, `~/Library/Google/GoogleSoftwareUpdate/GoogleSoftwareUpdate.bundle/Contents/Resources/GoogleSoftwareUpdateAgent.app/Contents/MacOS/GoogleSoftwareUpdateAgent`, and frankly are wholly uninteresting configuration files to integrate with launchd and other stuff.
Looking through a full system trace in Instruments didn't yield anything interesting either (outside of discovering that I forgot to shut down one of my virtual machines).
All in all, this sounds like hot air, and I can't help but wonder what the motivation behind making this page was.
Thank you for digging in to it on your machine. That level of forensics is exactly what we need in this issue.
Because there is a rich history of issues that only affect certain configurations, and those configurations are typically only revealed in hindsight: your results have to be seen as a single datapoint and not a nail in the coffin.
To others who are curious about this: please contribute more datapoints
I'm not really one for conspiracies, more so I fail to see where the evidence is that Chrome was somehow related to begin with. I do agree my wording could be better there, though.
The page says "I deleted Chrome, and all the files Keystone had littered on the computer, restarted, and it was so snappy it felt like a brand new computer. Yeah, I realize this sounds like a freakin' infomercial, but it worked so well I spent $5 on a domain name and set up this website even if it makes me sound like a raving nut."
To then "wonder what the motivation is" is to suggest that you don't believe this simple positive explanation, and imply there's something less simple, and more scheming.
Yeah, there was no super master plan here, just that after dealing with one sluggish computer for days (you name it, I tried it), and another one for 5 years... the fact that it turned out to be slow because of an app that wasn't even running was pretty frustrating.
Filing a bug report that would get "works on my machine"'d and then ignored and auto-closed by a bot in 5 years didn't quite feel worthy.
Auto-updating browsers are a good idea. Keystone is bad auto-updating software. It should probably get scrapped.
Well, for one, going as far as buying a domain for a single rant is a little bit suspicious. If at least it was a website dedicated to gathering issues with chrome, but a single issue with zero evidence or research. That's not generally how people "decide to share it".
Sure looks suspicious and not the content I expect to see on HN.
"Please don't post insinuations about astroturfing, shilling, brigading, foreign agents and the like. It degrades discussion and is usually mistaken. If you're worried about abuse, email hn@ycombinator.com and we'll look at the data."
It's not suspicious, its $5 which is not much to a lot of people in 1st world countries, especially those with at least enough curiosity/knowledge to be able to set one up if he/she has be frustrated a long time by a slow computer.
Personal attacks are not ok here. Please don't respond to a comment that breaks the site guidelines by breaking them yourself. That just makes the thread even worse.
This is a serious, non-sarcastic response: honestly, sincerely, in my experience as a HN reader off and on for years, a commenter for 6 months, and occasional author of submitted articles (usually submitted by someone other than me) that have reached #1, violation of the HN guidelines is the norm on HN, and abiding by the guidelines is the exception.
I myself have been personally attacked on here, or been interpreted uncharitably, etc., countless times. Mostly with no moderators intervening. In general I have no objection to "policing" the comments, if it is done well, and consistently, but the problem with policing, not just on HN but in general, is the selective, arbitrary enforcement of rules.
Just looking at the current case, for example, why was my reply flagged, but the comment I replied to, which you also cited as violating the guidelines, was not flagged?
Moreover, IMO this whole submission seems to be a strange time to be pedantic about enforcing the guidelines. After all, the submission itself was flagged as very low quality, and many of the commenters were justifiably complaining about the low quality of the submission and wondering why it was #1. The author is deserving of some rebuke, especially for the "nefariously hides itself from Activity Monitor" bit, which is technically false, indeed impossible. The author should know better, being or having been a Mac software engineer himself.
In fairness, HN is no worse off than any other site that allows comments from the general public. HN wants to believe that its comments are higher quality than other sites, but the reality is that it's not really true, and it's hard to see how that will ever become true. Maybe it was true in the distant past when HN was less trafficked, I don't know.
> violation of the HN guidelines is the norm on HN
It happens much more than we'd prefer, but to say "the norm" is a huge exaggeration.
I understand how it can feel like the norm—it's common enough to run into frequently, and painful experiences leave a stronger imprint than pleasant ones [1]. But these false feelings of generality are dangerous because they lead to an attitude of why-bother. Current toxin levels, though bad enough to produce symptoms, at least aren't fatal. Why-bother would kill the community if it spread.
> selective, arbitrary enforcement of rules
HN moderation is inconsistent in one way but consistent in another—call them strong and weak consistency. Strong consistency would be "every post that breaks the guidelines gets moderated in the same way". That would require moderators to read all the posts, and there are far too many for that to be feasible. Weak consistency would be "when a post does get moderated, the mods do it consistently and even-handedly". That's at least possible, and what we're shooting for. We don't get everything right but we're open to hearing about and fixing our mistakes.
If you see a post that ought to have been moderated but hasn't been, the likeliest explanation is that we didn't see it [2]. (You can help by flagging it or emailing us at hn@ycombinator.com.)
> why was my reply flagged, but the comment I replied to, which you also cited as violating the guidelines, was not flagged?
Users flagged one more than the other. Your post violated the guidelines worse, so I don't think that was unfair.
> this whole submission seems to be a strange time to be pedantic about enforcing the guidelines
That's not how moderation works. When users break the guidelines and we see it, we ask them not to. Keep in mind that we don't necessarily read the threads in order—sometimes we do, but other times we're looking in via more random-access views. The guidelines don't change relative to the submission, so I don't see why moderation should.
Looking through a full system trace in Instruments didn't yield anything interesting either (outside of discovering that I forgot to shut down one of my virtual machines).
All in all, this sounds like hot air, and I can't help but wonder what the motivation behind making this page was.