Live data from Hacker News

IPFS is the Distributed Web

ipfs.io

241–250 of 286 posts

Re: IPFS is the Distributed Web

#241

Earlier quoted context omitted.

It works great only for a limited number of use cases. It's terrible for unpopular content, it is not immune to takedown notices nor is it very resilient in the face of a determined adversary.

Backing 'long tail' content with webseeds works well for Bittorrent. ...but that's the same as saying let's back the decentralized web with centralized resources erp.

Exactly. Which begs the question of why bother with the decentralization part at all of long tail / unpopular content?

Re: IPFS is the Distributed Web

#242

Earlier quoted context omitted.

> The site host doesn't have to participate in IPFS for this to work. Really? An IPFS user will scrape HTTP content, republish it over IPFS and update a directory of where non-IPFS content can be found over IPFS? I must have missed that part.

Yeah - there's a ton of immutable URLs on the web. All of CDNJS/JSDelivr/Google Hosted Libraries, most of raw.github.com, all Imgur & Instagram images, all YouTube video streams (excluding annotations & subtitles), and all torrent file caching sites (like itorrents.org). There are probably some large ones I'm forgetting about, but just mapping immutable URLs to IPFS could probably cover 1/3rd of Internet traffic. Che…

IPFS archives look like something entirely different from what the grandparent was talking about.

Sure, you can manually publish archives over IPFS, but that's not something that automatically creates an IPFS cache copy of whatever you are surfing.

Re: IPFS is the Distributed Web

#243
post #232

Earlier quoted context omitted.

Normal static address is assigned. Whether it comes from the external range or from the link-local range doesn't really matter. You could potentially randomise the source at the exit to prevent identification of your hosts, but you don't have to. For most practical cases just imagine you're getting a big ipv4 range. Whatever you can do with it - you can do the same with ipv6. NAT, no NAT, filtering, static or dynamic…

That's what I'm asking. IPv6 feels like a regressiin over NATv4 because it can leak which internal device made a request. Is there a standard way to randomize addresses that works with ofd-the-shelf router firmware. Also, are link-local IPv6 leaking MAC addresses?

Yes, there is a standard since about 10 years ago. It's not dependent on your router firmware, in accordance with IP's end-to-end design philosophy (keep the network dumb, and hosts smart). Here's some links to get you started:

https://slaptijack.com/networking/osx-disable-ipv6-address-p...

http://andatche.com/blog/2012/02/disabling-rfc4941-ipv6-priv...

As for link-local IPv6 addresses, those aren't even accepted by the socket API[1] in place of normal routable addresses. They're only used for low level things like neighbour discovery (IPv6 equivalent of ARP) and apps that go out of their way to use them. They aren't routable outside your L2 segment.

[1] as in:

   $ telnet fe80::something:something:42
   Trying fe80::something:something:42...
   telnet: Unable to connect to remote host: Invalid argument

Re: IPFS is the Distributed Web

#244

Earlier quoted context omitted.

Well, that is fair. I'm mainly talking about "normal" adoption though. Eg, for a long time BitTorrent was being branded as a pirate haven. The only major players i saw adopt BitTorrent were open source projects needing to distribute large ISOs. IPFS is something that they're trying to get into standards, into browsers, into operating systems, etc. If it's known for distributing kiddy porn on the dark net it will be m…

If IPFS wants get gain traction and go mainstream, they still have you address the kiddy porn and piracy issue. You can't go downloading and seeding random hashes if this exposes you to criminal charges and/or law suites.

They are addressing it, but it's a future feature. Last i saw on Github issues (it likely has evolved since then), they plan to have a central authority which allows for DMCAs/reports/etc of illegal content. The network then propagates an ignore list of sorts (or possibly watches a central list).

Any nodes on the network which abide by this will simply not download and/or unpin the data.

It doesn't prevent you from downloading some kiddy porn that a stranger sent you.. but neither does the internet. This, is purely to allow law to take place, and hopefully allow lawmakers to work with IPFS, and not block the whole damn thing.

Re: IPFS is the Distributed Web

#245

Earlier quoted context omitted.

> I worry that this is another example of throwing technology at a social and political problem. I'm so lost at this assertion. How is this a social / political problem? IPFS is not permanent hosting. It is not an attempt to take power away from others and distribute it to the masses. It is purely a technical problem, which i think is objectively true. I have a feeling that you believe IPFS is partially designed to s…

From the article: "The Internet has been one of the great equalizers in human history and a real accelerator of innovation. But the increasing consolidation of control is a threat to that. "IPFS remains true to the original vision of the open and flat web, but delivers the technology which makes that vision a reality." This presents IPFS as the technological cure for the "consolidation of control" that threatens the…

That is unfortunate, because i have yet to see how IPFS is even remotely that. Especially since it wants to comply with DMCAs and etc.

The technology and the marketing seem at odds, to me.

Re: IPFS is the Distributed Web

#246

Earlier quoted context omitted.

If IPFS wants get gain traction and go mainstream, they still have you address the kiddy porn and piracy issue. You can't go downloading and seeding random hashes if this exposes you to criminal charges and/or law suites.

They are addressing it, but it's a future feature. Last i saw on Github issues (it likely has evolved since then) , they plan to have a central authority which allows for DMCAs/reports/etc of illegal content. The network then propagates an ignore list of sorts (or possibly watches a central list) . Any nodes on the network which abide by this will simply not download and/or unpin the data. It doesn't prevent you from…

It's ironic that IPFS plans on using a central service to censor and purge it's distributed system.

Not that this central authority will solve this issue either. It just opens another can of worms.

Who will review claims and reports? Who will be authorized to add hashes to the blacklist? How will counternotices be handled? How will different juridistictions and conflicting laws be handled? Who will be liable for content that makes it past the filters? Who will pay for all this?

The world is a messy place and IPFS does nothing to protect its users.

Re: IPFS is the Distributed Web

#247

Earlier quoted context omitted.

They are addressing it, but it's a future feature. Last i saw on Github issues (it likely has evolved since then) , they plan to have a central authority which allows for DMCAs/reports/etc of illegal content. The network then propagates an ignore list of sorts (or possibly watches a central list) . Any nodes on the network which abide by this will simply not download and/or unpin the data. It doesn't prevent you from…

It's ironic that IPFS plans on using a central service to censor and purge it's distributed system. Not that this central authority will solve this issue either. It just opens another can of worms. Who will review claims and reports? Who will be authorized to add hashes to the blacklist? How will counternotices be handled? How will different juridistictions and conflicting laws be handled? Who will be liable for cont…

> The world is a messy place and IPFS does nothing to protect its users.

I'm a bit confused, why is it the job of IPFS to "protect its users"? IPFS has no liability here, it's simply trying to make IPFS a law-abiding software as a whole. So it is not outlawed.

IPFS doesn't have any reason to "protect" me from things i download anymore than HTTP or TPC does. It's a technology for localizing data. What makes you feel IPFS has to protect its users?

It sounds like you think IPFS somehow spreads files onto your system without your consent. You only download what you choose to, just like with HTTP/TPC/UPD and the whole tech stack.

IPFS changes nothing for your personal security. It's not intended to. It's a technology to lower bandwidth usage from the production of mass data in the current age. And in regards to it being ironic for having a central authority, - sure - but the problem is bandwidth that IPFS is aiming to solve is, not law and/or DMCAs.

IPFS must have some seriously bad marketing (i blame the whole "good for humanity angle"), because i feel like you and a lot of people have the entirely wrong idea about IPFS. It's a technology, nothing more.. especially in it's current form. The only thing it might "save the world" from is bandwidth, lol.

Re: IPFS is the Distributed Web

#248

Earlier quoted context omitted.

It's ironic that IPFS plans on using a central service to censor and purge it's distributed system. Not that this central authority will solve this issue either. It just opens another can of worms. Who will review claims and reports? Who will be authorized to add hashes to the blacklist? How will counternotices be handled? How will different juridistictions and conflicting laws be handled? Who will be liable for cont…

> The world is a messy place and IPFS does nothing to protect its users. I'm a bit confused, why is it the job of IPFS to "protect its users"? IPFS has no liability here, it's simply trying to make IPFS a law-abiding software as a whole. So it is not outlawed. IPFS doesn't have any reason to "protect" me from things i download anymore than HTTP or TPC does. It's a technology for localizing data. What makes you feel I…

IPFS needs to protect its users, if it is to have any users.

The creators of IPFS may not have any liability, but they open up its users to severe issues of liability due to the way it operates.

IPFS will become toxic real quick, if it becomes know that IPFS can make you an unintentional distributor of kiddy porn and pirated content.

The core problem is that IPFS' default mode of operation is actively user hostile. This is a big problem, and the reason IPFS needs to protect its users. IPFS not only leaks what you retrieve, it also makes you an (unknowing) distributior of everything you retrieve. All it takes for you to commit a crime and/or copyright infringement, not only as a consumer but also a as a distributor, is to be tricked into retrieving one hash.

In conclusion, IPFS needs not only to not be outlawed, it also needs to be safe to use. As it is IPFS is not safe to use. Hence IPFS needs to do something to become safe to use, i.e. to protect its users.

Re: IPFS is the Distributed Web

#249

Earlier quoted context omitted.

By allowing your phone to be used as a server, other people will let you use their phones as a server, so you can download faster. If this sharing doesn't emerge naturally, then it can be directly incentivized with ratio tracking (like private torrent trackers), or even with money like the Karma WiFi service. Also, I have WiFi on always and my battery lasts all day, so I doubt this will be a problem.

Having wifi on all day is one thing, actually using it is totally different. Just activate your wifi hotspot on your phone and start using it from your computer. See how fast the battery depletes. No matter what incentive scheme you device , it's not going to help you with your battery.

Seeding when phone is actually in use would be net positive. And everyone is on their phones all the time nowadays anyway.

Re: IPFS is the Distributed Web

#250

Earlier quoted context omitted.

> The world is a messy place and IPFS does nothing to protect its users. I'm a bit confused, why is it the job of IPFS to "protect its users"? IPFS has no liability here, it's simply trying to make IPFS a law-abiding software as a whole. So it is not outlawed. IPFS doesn't have any reason to "protect" me from things i download anymore than HTTP or TPC does. It's a technology for localizing data. What makes you feel I…

IPFS needs to protect its users, if it is to have any users. The creators of IPFS may not have any liability, but they open up its users to severe issues of liability due to the way it operates. IPFS will become toxic real quick, if it becomes know that IPFS can make you an unintentional distributor of kiddy porn and pirated content. The core problem is that IPFS' default mode of operation is actively user hostile. T…

> IPFS will become toxic real quick, if it becomes know that IPFS can make you an unintentional distributor of kiddy porn and pirated content.

Yea, this is where you're mistaken. It cannot make you distribute anything! You only distribute what you download. Don't download kiddy porn, and you won't distribute kiddy porn.

> All it takes for you to commit a crime and/or copyright infringement, not only as a consumer but also a as a distributor, is to be tricked into retrieving one hash.

That is fair - but the same could be said for every p2p system on the planet. Has anyone solved this problem? How does BitTorrent protect me from unknowingly downloading kiddy porn?

Post reply on HN