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

Maigret: Collect a dossier on a person by username from thousands of sites

github.com

Post reply on HN
Print view View original

121–130 of 203 posts

Re: Maigret: Collect a dossier on a person by username from thousands of sites

#121

Post by dolmen »

orange_fritter wrote: post #75
eek2121 wrote: post #23

Earlier quoted context omitted.

I intentionally steal nicknames I've seen. I'm an asshole as well, so sometimes it works out. I have 13 different nicknames so far that I use/have used since 1997, though I tend to rotate between all of them regularly. I DO hope they try to use "AI" to track me. That will be fun.

What's the benefit? It sounds like you might have to explain yourself one day if someone posts something racist/horrible.

What about the case where (s)he's the one that posts the dirty things?

Re: Maigret: Collect a dossier on a person by username from thousands of sites

#122

Post by BlueTemplar »

smeagull wrote: post #25
AdmiralAsshat wrote: post #15

Seems incredibly prone to false positives. For example, I can guarantee that I'm not the AdmiralAsshat on Reddit, Gmail, or Twitter, because that username was already taken by the time I tried to sign up for them.

You always need to balance Precision and Recall. For something like this, you want to be exhaustive. I've worked on search engines, and depending on who is using them, that balance gets struck differently on the ROC curve. For legal matters, for example, they want every record that might match. For ad-hoc (google style) queries, nobody reads the second page so you care more about Precision @ 20 (or really, at 3)

Somewhat offtopic, I'm curious about this "nobody reads the 2nd page" meme...

I find myself reading pages 2-5 quite often, because page 1 just didn't give enough results, and I doubt I'm that much in a minority ?

(I'm talking about actual generalist searches, not people that use a global search engine as a replacement to bookmarks or directly searching, for instance, Wikipedia.)

Re: Maigret: Collect a dossier on a person by username from thousands of sites

#123

Post by mjevans »

BlueTemplar wrote: post #122
smeagull wrote: post #25

Earlier quoted context omitted.

You always need to balance Precision and Recall. For something like this, you want to be exhaustive. I've worked on search engines, and depending on who is using them, that balance gets struck differently on the ROC curve. For legal matters, for example, they want every record that might match. For ad-hoc (google style) queries, nobody reads the second page so you care more about Precision @ 20 (or really, at 3)

Somewhat offtopic, I'm curious about this "nobody reads the 2nd page" meme... I find myself reading pages 2-5 quite often, because page 1 just didn't give enough results, and I doubt I'm that much in a minority ? (I'm talking about actual generalist searches, not people that use a global search engine as a replacement to bookmarks or directly searching, for instance, Wikipedia.)

I imagine that each additional page presentation is exponentially less likely to be reached and considered a 'reputable' result by someone normal.

If I'm looking for a specific issue I'll sometimes try out things as deep as 10 or more pages of search results if nothing on the first 100 ish hits selects the issue, but then only if I can't think of any keyword variations to use that might get me a better result match. I don't expect the average user to go even remotely that far.

Re: Maigret: Collect a dossier on a person by username from thousands of sites

#124

Post by dolmen »

bhrgunatha wrote: post #86
P5fRxh5kUvp2th wrote: post #33

Earlier quoted context omitted.

I had that happen once. It was a recipt for a medical purchase, at first I thought I was getting scammed. What tipped me off was the email was sent to firstnamelastname@gmail.com and NOT firstname.lastname@gmail.com. That was the day I realized google would even do that. I ended up using the phone number in the email to contact the person and forwarded the email. And yes, they had my first and last name :)

I'm subscribed to so many informal church groups in Alabama originally but "I've" since moved to Texas. I tried quite hard to clear up the confusion but with virtually no action ever taken, so now I just observe. From those emails, it's nice seeing how people support each other and apparently I'm in demand giving scripture classes.

I'm french.

But my e-mail address has been used by real people to subscribe to services in Sweeden, Turkey and somewhere South America. At least language helps to sort things.

Re: Maigret: Collect a dossier on a person by username from thousands of sites

#125

Post by daanlo »

BlueTemplar wrote: post #122
smeagull wrote: post #25

Earlier quoted context omitted.

You always need to balance Precision and Recall. For something like this, you want to be exhaustive. I've worked on search engines, and depending on who is using them, that balance gets struck differently on the ROC curve. For legal matters, for example, they want every record that might match. For ad-hoc (google style) queries, nobody reads the second page so you care more about Precision @ 20 (or really, at 3)

Somewhat offtopic, I'm curious about this "nobody reads the 2nd page" meme... I find myself reading pages 2-5 quite often, because page 1 just didn't give enough results, and I doubt I'm that much in a minority ? (I'm talking about actual generalist searches, not people that use a global search engine as a replacement to bookmarks or directly searching, for instance, Wikipedia.)

Source: https://www.sistrix.com/blog/why-almost-everything-you-knew-...

TLDR: on the second results page, each result gets So it‘s not nobody, but statistically speaking not very many.

Re: Maigret: Collect a dossier on a person by username from thousands of sites

#126

Post by dolmen »

lol768 wrote: post #9

This is technologically impressive, but I was disappointed (but not particularly surprised) to see no mention of the ethics involved in using this sort of project - in either the GitHub issues or the README/docs. This seems like a great tool for stalking people; particularly the recursive functionality (for tying identifiers together). I'm not saying the world is worse-off due to the release of software like this - i…

No mention of ethics on the "Philosophy" page of the documentation.

https://maigret.readthedocs.io/en/latest/philosophy.html

Re: Maigret: Collect a dossier on a person by username from thousands of sites

#127

Post by mcqueenjordan »

I think you're supposed to do this with Stylometrics[1]. So I guess this is sort of the greedy hacky approach to try and associate users across websites.

My operating assumption of Palantir as a company is that they have a very advanced system of a similar shape that tries to accomplish the same goal of linking accounts across services, but I have no insider knowledge.

[1]: https://en.wikipedia.org/wiki/Stylometry

Re: Maigret: Collect a dossier on a person by username from thousands of sites

#128

Post by tartaglia0 »

I gave it a try feeding it with one of my community nicknames. It was rather generic so I did expect some false positives. Color me surprised when it automatically started a second search iteration using my real name!

Turns out that one of the websites I'm registered to uses Gravatar to display profile pictures. I've naively been using the same WP account to set a picture for all my emails, believing that you couldn't get information about my other accounts using a simple picture.

I guess I was wrong, and I'll probably split my Gravatar account this evening.

Post reply on HN