Localhost | Hacker News Reader

Localhost

transparencyreport.google.com

272 pointsperihelions47 comments

One is reminded of the whole $cientology vs the internet shambles, at one point someone was deposed by their lawyers trying to find out who was running that "localhost" server because it had all their files on it ..... these are the same people who were trying to find out the identity of Major Domo, because he seemed to be a big wig in the whole internet cabal who were trying to take $cientology down

If anyone's curious, there's previous discussion about this here:

https://news.ycombinator.com/item?id=20791891

He really didn't handle the explanation of the joke properly...

Do I understand it correctly? It loops back to the origin machine so they were seeing their own files? And they thought they were seeing the files on an external machine?

Nothing is mentioned about anyone actually trying to access the URI. I don't think their server would have used that filepath.

I imagine the plaintiff simply assumed it was legitimate without investigation, i.e. it was a successful troll.

unrelated but i love your kerbal space program stuff!

A different "taniwha" I'm afraid - I've been using this handle on-line since the mid 80s ....

In 1999 I went for a class at a company who was trying to sell an XML database, there was a problem in the class when it turned out that someone in Germany on their corporate network had named their computer 'localhost'.

Some requests seem to be so garbage it makes me sad. For example https://transparencyreport.google.com/copyright/request/1195... has "localhost", "unknown", and "apple.com". I hope we get some laws which allow saying "your request has so many bad links, we're going to ignore all of it and future ones until quality improves".

I agree that "localhost" and "unknown" aren't right, but "apple.com" could be correct. Those reports show only the domain and don't show the exact url. You have to drill down and eventually supply an email to get the exact urls.

I agree in theory it could be correct. But there's a decent chance that if you submitted a batch of urls including invalid ones, your apple.com submission is bogus.

So like if the lords of stackoverflow did management consulting for the likes of Google's internal processes? Sounds like a fantastic idea

Your question is unclear please rephrase your question to satisfy my own unclear and opaque standards of clarity and resubmit with low probability of acceptance in 4 to 9 weeks welcome to googo have a lovely day...

No, the two are wildly different:

- in stackexchanges case someone is killing even popular questions with valid, helpful questions and answers because of their subjective opinion

- in this case it is quite obvious that anyone who asks for localhost/127.0.0.1 to be taken down either has no idea what they are doing or haven't vetted the data.

Or better make the reporter financially liable for making obviously incorrect takedown notices.

Or just do what youtube does.... three strikes.

If you, as a creator get three strikes (for doing something wrong), you lose your channel. Why not the same for copyright holders... three false accusations, and it's the end.

I don't think we should be using Youtube as a model for anything.

Because Youtube wouldn't be able to enforce that threat. If a copyright holder complains about an IP violation, Youtube has to address it, or the copyright holder will go to court.

If Youtube starts saying "you've abused our report system, so you're not allowed to make reports anymore", the copyright holder will just respond "alright, we'll sue you every time we find a copyright violation on your platform from now on".

For this to change, content platforms like Youtube would need some sort of special legal status that would say "Provided that we do some reasonable effort to weed out IP violations, we're allowed to ignore a certain percentage of IP claims if we think the claimant is a patent troll" or something similar. I doubt it will happen, though, most governments are in the pocket of media companies and default to assuming content platforms are guilty until proven innocent.

Inspired by today's [0], this is a list of legal threats delivered to Google over the domain "localhost".

[0] https://news.ycombinator.com/item?id=28106479 ("Anti-Piracy Firm Asks Google to Block 127.0.0.1")

Sorry for being off topic but what made you pick the name perihelions? I ask as someone who did some work in the field of solar physics and I don’t see a lot of opportunity to use the word perihelion :)

It originated from Kerbal Space Program!

It's a great word. Nice choice!

Awesome!

Can the title be updated to be more descriptive?

> Removed 90 (22.5%)

So 90 `localhost` results were actually in the index?

I would guess it's some quirk, like a domain that originally had a real DNS A record, then switched it to point at 127.0.0.1 some time later. But the cached index pages remained for some time as well.

Edit: Apparently not. See https://news.ycombinator.com/item?id=28112231 The entire request is all 127.0.0.1 links.

Interesting to find some domains with LOTS of removal requests.

Like this one with 16 million requests: https://transparencyreport.google.com/copyright/domains/new-...

Looks like a place with torrent links for movies.

So does that mean a good way to find piracy sites is to find the delisted domains?

Been looking at the reported domains. Yes, indeed.

> This domain had 400 URLs over 24 individual requests requested for delisting from our Search results. 22.5% were removed

How did they even have these results?

What does it even mean?

I do wonder

https://www.lumendatabase.org/notices/1315500

This looks to be the DMCA request for those 90 links. No doubt it was part of a dragnet, given they're listings of 3gp, aac, flv, mp3, mp4, and webm files, it's probably just some links found in some sort of scam page.

Good eye! Looks like that corresponds to this report on Google's page:

https://transparencyreport.google.com/copyright/request/6590...

90 entries "removed" -- all of them "localhost"

https://blackhat.guru/short/JpgkES

   > http://127.0.0.1:22342/2/life%20of%20pi%20audiobook%20CHAPTER%2065%20FULL%5Bfreemp3q.org%5D.flv

oh, it sounds like an audiobook abouth mathematics, interesting

Not so much math. Life of Pi was a 2001 novel about the son of a zoo manager and his life after being shipwrecked with a Tiger.

It was made into a 2011 movie, that had a home release in 2012, which no doubt also had a anti-piracy campaign by the studio around the same time as the movie and home release.

i think it was a joke...

Oh, it may have been. Wouldn't the first time I hear "woosh".

I wonder, if I installed Google desktop on an old PC image and stored "life of pi audiobook CHAPTER 65 FULL[freemp3q.org].flv" in "htdocs/2/", would it still be found? Are these delistings just empty promises? :-)

There is only one way to find out. Don't leave us hanging

It's also a little known fact that in a lot of operating systems the whole 127.0.0.0/8 is mapped to the local machine. I was able to fool my pretty tech-savvy friend by telling him that my IP was 127.23.85.66 and he even asked why do I have a public FTP server (I didn't) so it definitely worked :).

It's bonkers how they used up so much IPv4 address space just for your local machine.

I’m the “old days”, /8s were handed out like they were nothing. It wasn’t expected that it would become so bad.

It would not supprise me if Google index and cache sites on localhost, if you somehow manage to expose it to the internet...

No, you can’t expose anything on internet that is seen by Google (or anyone) as localhost.

Only way I could imagine this is if the text of the link had copyrighted information in it. But I have never seen Google link to spurious localhosts anyway (presumably they dont copy links into results but only show sites they were able to crawl).

It would be interesting to use Google as a search engine for your own computer, but I'm sure it's not that easy

IIRC you used to be able to do this, kind of, with the Google Search Appliance virtual edition: https://googlecode.blogspot.com/2008/11/announcing-google-se...

Not quite the same, but Google provided Spotlight-like software for Windows, and shown some local results when you were searching the big web. This was over ten years ago.

Reply on news.ycombinator.com