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

IPFS Support in Brave

brave.com

Post reply on HN
Print view View original

211–220 of 464 posts

Re: IPFS Support in Brave

#211

Post by jabedude »

realdense wrote: post #190
waynesonfire wrote: post #177

Earlier quoted context omitted.

why would mozilla pay to add a fad technology or take on the tech debt?

Some of us believe that the client-server model has led us to the current situation, ruled by internet giants. A future where people are less reliant on giants is within our reach, the biggest hurdle is making it available to the masses. Hopefully uptake by non-mainstream browsers like Brave will increase the exposure, attract developers, increase the network and get us closer to having IPFS, Hypercore or some other…

> IPFS, Hypercore or some other protocol available to the mainstream, whether that be in Firefox or by some other means

TIL about hypercore! Looks like an interesting concept based on append-only logs. It reminds me of this work being done at VMware: https://github.com/vmware/node-replication

Re: IPFS Support in Brave

#212

Post by joshmanders »

bawolff wrote: post #159
wolverine876 wrote: post #152

Earlier quoted context omitted.

Are you sure you want to participate in a discussion badmouthing Mozilla? You're not directly saying those things, but you are taking part in the dialog. That sort of behavior makes me pretty angry about Brave. Focus on Brave. You have nothing to do with Mozilla.

I think its a little extreme to call this bad mouthing. People switch jobs. It does not reflect negatively on firefox that some of their employees now work at other companies in the same space. That is how jobs work - its not a life long commitment.

Not to mention Brave was co-founded by the former CEO of Mozilla and one of the influential people of the Netscape days, Brendan Eich. So it's only logical that a portion of Brave devs are former Mozilla/Netscape people.

Re: IPFS Support in Brave

#213

Post by joshmanders »

theptip wrote: post #121
Nican wrote: post #115

It is not the first time that I have looked at IPFS and I still have a hard time understanding how the ecosystem is going to work. From my understanding, the file hash is basically the file URL, such that any change to the file content is a change to the file url as well. For hosting something like Wikipedia, how would one create pages that link to one another? And if indexes need to be created on top of the content,…

https://docs.ipfs.io/concepts/ipns/ Quoth: IPFS uses content-based addressing; it creates an address of a file based on data contained within the file. If you were to share an IPFS address such as /ipfs/QmbezGequPwcsWo8UL4wDF6a8hYwM1hmbzYv2mnKkEWaUp with someone, you would need to give the person a new link every time you update the content. The InterPlanetary Name System (IPNS) solves this issue by creating an addre…

This all sounds like reinventing ICANN and tld's

First you had a decentralized internet, anyone with a computer could put something online, but sharing IP's everywhere was not easy to remember, and some weren't static... So they created DNS and Top Level Domains...

Re: IPFS Support in Brave

#214

Post by bawolff »

jjcon wrote: post #189
bawolff wrote: post #156

Earlier quoted context omitted.

The type of people who are into decentralizing the internet are also usually against a single company (google) controlling the rendering engine of most users. If you're talking about typical user of the average web browser, they care about IPFS as much as they care about the rendering engine - i.e. not at all.

> against a single company (google) controlling the rendering engine of most users That isn’t really applicable here though. That’s kinda like claiming we should be against Linux because it runs all the web servers. There is a difference between centralization and standardization.

There is a difference, and all the html rendering code being written by google is centralization not standardization.

I don't think the linux comparison makes any sense for this situation. Linux has very little to do with the web experience, where the rendering engine is the core of it. Linus's ability to change world wide web standards on a whim is close to non existent. Google is literally replacing the WWW network stack (a good change mind you, but nonetheless demonstrates the level of power)

Re: IPFS Support in Brave

#215

Post by breck »

joshuakelly wrote: post #29

Massive. I have a project that assumed ipfs:// would eventually exist natively within a mainstream(-ish) browser, and I'm very pleased to discover that after updating to 1.19.x it all just works. Excited for the forthcoming DNSLink support too, even if it's just a bridge to something even better. Best of luck to everyone who wants the web to stay bundled inside of the corporate state.

Agreed. I've kept Brave around because it seems to work better on Paywalled sites, but haven't had a hugely compelling reason to use it yet. Stuff like this is awesome. It's like they are saying to Chrome, Safari, Edge—"hello, we're here, and we're going to take risks. wanna play?"

Re: IPFS Support in Brave

#216

Post by Stephen304 »

autocorr wrote: post #175
rmm wrote: post #154

Earlier quoted context omitted.

For me, somewhat embarrassingly, the one killer feature is the "Send to Device" feature from firefox on iOS. Being able to send links to browser for later reviewing is super useful. Not sure why Edge or other browsers don't implement

This is a killer feature for me too. Works great between all my devices across Linux, Android, and iOS. If I see an interesting article I just ship it off to my desktop to read later instead of keeping the tab open for days on my phone.

I've unfortunately had the opposite experience with FF send, I still use it all the time (or try to) when I find links on mobile that I want to read on a big screen, or when I have an image url on my laptop that I want to send from my phone via SMS, snapshat, etc. But the delay is just unbearably long and inconsistent. Sometimes (most often when sending desktop to phone) the delay is 15 minutes or more. I've found it a lot easier to use a bookmarklet that uses DDG to generate qr codes, but that only solved sharing in one direction.

I've wondered how send could be so slow but maybe something in my setup is just killing it. More than anything though, I would love an easy way to access my phone's sharing options from my laptop (perhaps an addon button that shows a list of app intents that can receive a url/page).

This is one thing I'm hoping to hack at when I get my Librem.

Re: IPFS Support in Brave

#217

Post by cinquemb »

simmanian wrote: post #195
joshuakelly wrote: post #29

Massive. I have a project that assumed ipfs:// would eventually exist natively within a mainstream(-ish) browser, and I'm very pleased to discover that after updating to 1.19.x it all just works. Excited for the forthcoming DNSLink support too, even if it's just a bridge to something even better. Best of luck to everyone who wants the web to stay bundled inside of the corporate state.

I wrote this somewhere below as well but I'm honestly confused. It seems like I can load ipfs:// links just fine on my firefox. Is there something I'm missing? Edit: you guys are right. it seems i have installed the add-on some time ago and forgot about it.

You might have add-on that redirects you to some gateway?

Re: IPFS Support in Brave

#218

Post by fledgexu »

simmanian wrote: post #193

Honest question: what does it mean that Brave supports IPFS in this case? I can load IPFS links on my Firefox just fine.

Yes, you can. "Native Supports" means that Brave has a native IPFS node in their Browser and you don't need download IPFS-Desktop or use public gateway to visit websites in IPFS network.

Re: IPFS Support in Brave

#219

Post by BrendanEich »

miles wrote: post #20
meremortals wrote: post #8

Earlier quoted context omitted.

Which criticisms? I've just started using it after Firefox's blog post

Perhaps this? The Brave web browser is hijacking links, and inserting affiliate codes https://davidgerard.co.uk/blockchain/2020/06/06/the-brave-we... https://news.ycombinator.com/item?id=23442027

The phrase "hijacking links" is, in a word, a lie. Its standard meaning implies that Brave changes the URL loaded when you click on a link (any or many links) in a web page. That never happened, and would be fatally scandalous if tried by any reputable browser.

What happened was a bug in a keyword and domain autocomplete tier we added. Reminder for those not aware: all browsers add reference codes identifying the browser make (not the user) to keyword queries when you type words into the address bar. This is industry standard. See https://twitter.com/BrendanEich/status/1273327455105773568?s....

The bug in Brave was a flag set wrong for two domain names, which are not keywords: binance.us and binance.com. These should not have default-completed when typed into the address bar with a Binance affiliate code, they should have been suggested completions the user would have had to pick by arrow down and or mouse equivalent. We fixed the defaulting bug and turned off the entire suggested completions feature. As noted by someone else in a comment nearby, we instructed Binance not to pay us for any referred new users who traded (Binance does not pay on the referral, they pay only after if the user trades, sharing a fixed proportion of trading fees).

So this was not "link hijacking" in any case. I'm sorry for the blunder, and we added a process step around any changes to address bar to audit harder.

Re: IPFS Support in Brave

#220

Post by rubyist5eva »

rmm wrote: post #154
mikece wrote: post #63

Earlier quoted context omitted.

Does Brave have an implementation of Multi-Account Containers? This is the ONE killer feature in Firefox that makes it impossible for me to leave for Brave completely: https://addons.mozilla.org/en-US/firefox/addon/multi-account...

For me, somewhat embarrassingly, the one killer feature is the "Send to Device" feature from firefox on iOS. Being able to send links to browser for later reviewing is super useful. Not sure why Edge or other browsers don't implement

I just send things to myself on Telegram...works fine.
Post reply on HN