DNS got you down? Share ip addresses with freedom.txt | Hacker News Reader

DNS got you down? Share ip addresses with freedom.txt

github.com

23 pointsmvanveen12 comments

An open internet is more important than corporations. They have failed us because of their natural greed. They must be changed.

To be fair, evil greedy corporations also prevented SOPA

An open internet is more important than security, copyright infringement, terrorism or child pornography. We will not be fooled by the strategies of fear employed by those who wish to censor us.

This is the kind of false dichotomy the enemies of an open internet want to create. The point is that we can fight all these things and still have a free internet.

Finally, the URI to IP table is confused. Example

     http://hoop-la.ca/freedom.txt: 70.66.72.121

But http://70.66.72.121/freedom.txt fails because of the missing hostname.

This is slacktivism at its worst.

I like the concept, but why does this map http addresses (http://server.com) to ip addresses? Why not instead have a table of ips and domains (like /etc/hosts) so you can then just sync the freedom.txt file into your hosts file? That's what the hosts file is designed for after all.

This has the added benefit of working with virtual hosting configurations where the browser has to pass the Host header to the server to get a web page.

The script grabs the domain of the uri in question, so it's effectively mapping the domain to the url.

To answer your question more specifically, the freedom.txt format started in this thread: http://news.ycombinator.com/item?id=3498505 , so it links to the uri of a freedom.txt for historical reasons.

/etc/hosts integration is in the works; you're the 2nd person to recommend it so far (the first being my roommate ;-) ).

What's up now is just a quick hack to get momentum for the project in the meantime. If this takes off, the innovation will be social rather than technical, so I thought I'd spend 5 minutes to see if it's worthwhile.

Thanks for the input!

Edit: I'm working on /etc/hosts integration right now.

I know the web moves fast, but you got a good laugh out of me for calling something someone came up with in a thread yesterday "historical reasons" :)

Fair. Probably not the best wording on my part.

For context, I've been in contact with the original poster and did not want to separate our efforts. This idea was suggested in the last thread, and I thought it was worth a try.

Leveraging what's already out there seemed like a good idea.

Followed your advice and added (preliminary, beta) /etc/hosts support, thanks for the suggestion!

Two nitpicks:

1. `/robots.txt` is here to stay, but everything else should use the `/.well-known/` prefix so that they do not squat the root URI: <http://tools.ietf.org/html/rfc5785>.

From the RFC:

> To address this, this memo defines a path prefix in HTTP(S) URIs for these "well-known locations", "/.well-known/". Future specifications that need to define a resource for such site-wide metadata can register their use to avoid collisions and minimise impingement upon sites' URI space.

2. alternative DNS root systems have been tried in the past [1] and the IETF has always spoken about them on technical and political grounds. The ORSN was probably the most autoritative alternative DNS root, backed by Vixie and others, and stopped operating few years ago.

From the Wikipedia article

> The founders of ORSN are concerned that ICANN is ultimately controlled by the government of the United States. Their aim was to limit the control over the Internet that this gives, while ensuring that domain names remain unambiguous. They also expected their network to make name resolutions faster for everyone.

[1] http://en.wikipedia.org/wiki/Alternative_DNS_root [2] http://en.wikipedia.org/wiki/Open_Root_Server_Network

Much like /robots.txt and /humans.txt I think this belongs on root. I understand that we do not want to trash the root URI. Perhaps this can be merged with the humans.txt instead? It can be a separate section in it.

I also dont think this is necessarily an alternative DNS. Much like the hosts file that circulated the internet, this is is just a way to track some major IP addresses, that can be queried when the world ends.

Just start some alternative root servers. There's nothing that makes the current roots authoritative except for convention and a hardcoded list of IPs. It's not even hard to start up a server that can act as the authoritative source for blocked domains and pass through requests to the real roots for everything else.

Off topic, but can anyone tell me the history of

[SOME DATA FORMAT]

---

[SOME OTHER DATA FORMAT]

I'm sure there must be a better way to have multiple data types in a single file than just searching for '\n---\n' and splitting, no? What's the general principal when one data-block has a '\n---\n' in it's data?

I first saw it in Jekyll, I believe they call it 'Front Matter' (Though their implementation seems to be using a '---\n' to start the block) but I can't find any more information on it. Is it some sort of standard/specification/thingy I've completely missed?

Some email clients also chuck anything after a line with just two or three dashes. I believe I recall that in things as old as Eudora or as recent as Entourage.

Used to be that you'd put \n-- \n before your sig, and then replies would omit the sig.

ADDED:

"The formatting of the sig block is prescribed somewhat more firmly: it should be displayed as plain text in a fixed-width font (no HTML, images, or other rich text), and must be delimited from the body of the message by a single line consisting of exactly two hyphens, followed by a space, followed by the end of line (i.e., "-- \n").[1] This latter prescription, which goes by many names, including "sig dashes", "signature cut line", "sig-marker", and "sig separator", allows software to automatically mark or remove the sig block as the receiver desires."

1. The links in Wikipedia reference RFCs dating back to 1994.

I can tell you from direct experience this doesn't scale. That was in 1991 when the internet was a much smaller place.

What's with mapping URLs to IP addresses anyway? That's completely wrong ...

Reply on news.ycombinator.com