And if it’s legal content we’re talking about, why on earth would a sane capitalistic company (forget Open Source for now) that pays beaucoup $$ for http web serving would be interested in you grabbing their content without first going through their heavily-commercialized website? So you can download the wrong file or call them up later complaining that your PC crashed when you downloaded and installed a sheitty old version of the file? Had you visited their website first, you’d have known that a new stable release was just made 2 weeks ago… But no, you grabbed the wrong file cuz it came to you through “other channels”, say, some guy gave you a hash (to the then-current-release) in email 3 weeks ago… Not to mentioned they just lost sales because you didn’t get the “in your face” ad for their newest game/biz-app/subscription-service/whatever. All companies that have websites WANT YOU TO BE ON THEIR WEBSITE. In other words, they’ll do everything they can (long term) so you cannot bypass their control. You won’t be able to hash their http content simply because it’ll become dynamic URLs (as many companies already do)…

In other words, I see very little practical use for the concept of global hashes. I see much more use for hashing files on all decentralized (or quasi-decentralized) networks (bittorrent, emule, etc) a la Bitzi that was mentioned, for the purpose of finding clean (commented) illegal content easily.

Bitzi is a more powerful and practical idea than global hashing. If taken seriously, then you will know problems/tips about what you are downloading right as you are downloading it (if web servers start giving hashes out before full download), or just after you finished downloading (if you have to do the hashing yourself). In my example above, just after you finished downloading the file, you’ll get a message from the devs telling you “file you just downloaded (who knows from where) is an old vers with known bugs. Trust this msg, it is digitally signed by us, the makers of the original file you’ve downloaded”.

I think commenting on URLs and unique files (hashes) is the most powerful concept yet fully realized, and not the directory of where to obtain those unique files from…

I suspect that part of the reason it is such an mess starts all the way from the bottom – the RSS feed standard – the design reflects its way all the way up the chain.

Torrents do a great job of avoiding the end UI issues, but it would be the exception not the rule and there are a fair share of bad torrent UIs from the early days – in fact, the only thing that really saved it was how easy it was for third parties to enter the market and improve on it.

This technology would have to be, at its heart, baked in just a few select pieces of software where most of the variation comes not from competition but rather different platforms. That means we are likely to see poor UI design from the masters: Microsoft, Firefox and my personal favorite for worst UI experiences if you aren’t a Unix Savant – Google/Chrome. (A notable mention goes out to Apple for making their Windows browser an alien experience to every single target user by writing copious volumns of code to make it look and act just like the mac OS)

I think this is a really cool idea. I’m a C# dev (mostly desktop app stuff) with several years of experience, email me if you need any help developing it.

-Judah

The concept strikes me as the insertion of another point of failure for reasons of rather pointless pedantry – but I’m not a mass content provider, so perhaps I’m not someone who’d see the benefits from this.

The functions you’re referring to wraps around content, and ultimately modifies the file. What I’m proposing does not change files at all. In fact every file on the internet today can be utilised.

Here’s one example of a free signed file container that could be made a standard: http://www.pkware.com/software-data-security/windows-file-encryption