Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Getting 10TB of GitHub logs and extracting details of all users and repositories

trickest.com

Post reply on HN
Print view View original

1–10 of 61 posts

Re: Getting 10TB of GitHub logs and extracting details of all users and repositories

#3

Post by didntcheck »

I'm a big fan of neat statistical analyses, but when I look at the archive site https://www.gharchive.org/ , the overwhelming feeling I have is "creeped out". Taking periodic snapshots of repositories and their issues and wikis sounds good, but do we really need a log of every time someone watches an issue, and every commit message being irrevocably set on public record? That level of details on individual activity seems to have little value outside of cyberstalking

I did have a look at the bottom of the page where prominent uses are listed, but nothing stands out as actually useful tbqh

Re: Getting 10TB of GitHub logs and extracting details of all users and repositories

#6

Post by zx8080 »

didntcheck wrote: post #3

I'm a big fan of neat statistical analyses, but when I look at the archive site https://www.gharchive.org/ , the overwhelming feeling I have is "creeped out". Taking periodic snapshots of repositories and their issues and wikis sounds good, but do we really need a log of every time someone watches an issue, and every commit message being irrevocably set on public record? That level of details on individual activity s…

> That level of details on individual activity seems to have little value outside of cyberstalking

A selling point for business-level is that they can do whatever OKR they want on top of that details. To make any employee dance whatever business can imagine. Any amount of Tb seems OK until it helps selling "Business" tier github.

Re: Getting 10TB of GitHub logs and extracting details of all users and repositories

#7

Post by mtmail »

atum47 wrote: post #2

[flagged]

IMHO it's in the guidelines "Please don't complain about tangential annoyances—e.g. article or website formats, name collisions, or back-button breakage. They're too common to be interesting." https://news.ycombinator.com/newsguidelines.html

Re: Getting 10TB of GitHub logs and extracting details of all users and repositories

#8

Post by ziml77 »

didntcheck wrote: post #3

I'm a big fan of neat statistical analyses, but when I look at the archive site https://www.gharchive.org/ , the overwhelming feeling I have is "creeped out". Taking periodic snapshots of repositories and their issues and wikis sounds good, but do we really need a log of every time someone watches an issue, and every commit message being irrevocably set on public record? That level of details on individual activity s…

I have a GitHub account under my real name, but recently I've started using GitHub under a couple of other names instead. There's so much stuff you do in public on GitHub that I want to avoid people doing exactly this kind of analysis on.

I wish using multiple identities was at least some level of foolproof though. I have to be careful to configure my local copies of repos to use the correct username, masked email, and PGP signing key. It would be super handy if git had a global config of multiple identities and if I could have it prompt me to select one either when I clone a repo or the first time I push to it.

Re: Getting 10TB of GitHub logs and extracting details of all users and repositories

#9

Post by jxramos »

ziml77 wrote: post #8
didntcheck wrote: post #3

I'm a big fan of neat statistical analyses, but when I look at the archive site https://www.gharchive.org/ , the overwhelming feeling I have is "creeped out". Taking periodic snapshots of repositories and their issues and wikis sounds good, but do we really need a log of every time someone watches an issue, and every commit message being irrevocably set on public record? That level of details on individual activity s…

I have a GitHub account under my real name, but recently I've started using GitHub under a couple of other names instead. There's so much stuff you do in public on GitHub that I want to avoid people doing exactly this kind of analysis on. I wish using multiple identities was at least some level of foolproof though. I have to be careful to configure my local copies of repos to use the correct username, masked email, a…

Private repos are still off limits for this sort of analysis correct?

Re: Getting 10TB of GitHub logs and extracting details of all users and repositories

#10

Post by mxmlnkn »

didntcheck wrote: post #3

I'm a big fan of neat statistical analyses, but when I look at the archive site https://www.gharchive.org/ , the overwhelming feeling I have is "creeped out". Taking periodic snapshots of repositories and their issues and wikis sounds good, but do we really need a log of every time someone watches an issue, and every commit message being irrevocably set on public record? That level of details on individual activity s…

This kind of information has become even more easily accessible for everyone with the introduction of the Activity tab. I really wish that force-pushed commits and deleted branches would be permanently lost instead of being permanently stored like on a blockchain.

https://github.blog/changelog/2023-05-31-view-repository-pus...

Post reply on HN