Transaction

j"1LAnZuoQdcKCkpDBKQMCgziGMoPC4VQUckMœ<div class="post">I'm wondering if anybody is doing anything about this idea, as I have been researching the possibility of implementing some Internet browser functionality, as per <a href="http://'http://bitcointalk.org/index.php?topic=123.0'">this thread</a> and the wonderful developments <a href="http://'http://bitcointalk.org/index.php?topic=63.0'">described here.</a> It seems that it may be easier first steps, (I'm learning as I go) to try this out and include it with whatever other features the Firefox add-on evolves. <br/><br/>Before this can happen there needs to be a consensus about what form the URI would take. I have done some quick research on the URI schemes and most approaches are frowned upon without W3C approval, there is one category whereby the W3C approves a tentative URI scheme and holds it in reserve, with no guarantees (like a provisional patent), but by far the most suitable scheme, seems to be the magnet URI, which was designed to find a resource by the hash being a product of its content uniqueness, rather than by specific address. Magnet was intended for use by applications on peer to peer networks and has reserved parameters for reference to application specific data. Basically, you simply preface the application specific parameter with an x as in xbitcoin1:?<br/><br/>Maybe it is good, not to try implementing a new top level URI scheme, as it is regarded as an ad lib maneuverer, if it is not recognized universally. Adopting the magnet system would avoid this, but in itself magnet is an informal, unregistered scheme. The beauty of magnet is that it can be shared and is open in that sense. <br/><br/>Now, that might all seem very good, because bitcoin is after all a P2P application and it's network does operate on similar principals, but what if the person clicking on the link is not a bitcoin user? What if you want to incorporate some information that is not carried in the existing network, such as the article being sold, shipping information and various other issues. My understanding is that bitcoin only handles the transaction itself, so the user is left to make all other arrangements, contracts and communications independent of bitcoin. Here's where the anonymity is compromised, because here, we are inclined to resort to conventional 'hierarchical&nbsp; hardwired, or non-anonymous technologies. That's not a criticism of bitcoin by any stretch, because bitcoin is a fantastic way to transact, no question about it. But the transaction is only one side of the coin, if you'll pardon that cheesy pun. <img alt="Grin" border="0" src="/static/img/emoticons/grin.gif"/> <br/><br/>In the process of investigating URI's I was reminded of freenet and the magnet like system they use, and I ended up back there checking it out. Given that freenet is a user/identity-space intended for information storage, couldn't it also be used as a repository of human readable data files, particular and unique to each user, and incorporating any format of data you wish to use, for embellishing your transaction? Couldn't this be used for hooking the URI into a rich database of application specific data, that has been designed for the publishers own website functionality or business system? While your bitcoin node takes care of the actual transaction, the freenet node takes care of the individual data presented. <br/><br/>This should keep everything under the same kind of cryptographic P2P infrastructure, and allow infinite diversity of application specific context. I suppose a unique URI hash key, could be borne out of the combination of your freenet+bitcoin hashes and the file encrypted with your freenet key found by decrypting with your bitcoin key, while your bitcoin node can be addressed the other way around. Does that sound right? Of course, if you use your real identity on one, then I guess then, you real identity would be divulged on the other. <br/><br/>Anyhow, I'm sure there's some way to do it. Again, freenet also uses an unregistered URI scheme, but if a variant of magnet is adopted, that will not be a problem. The question remains, if people have hopes of bitcoin becoming a formal recognized currency and a very legitimate deal, then the strict adherence to all other internet protocols might be advisable. I have noted so far nothing in particular that would prevent the registration of a magnet/freenet like URI scheme. I haven't searched them all but one may even exist. Otherwise why not register one?<br/><br/>So the URI scheme needs to be agreed upon as to the classification (i.e. what kind magnet etc, as well do we go legit/formal or not) and format, such as how it is to be laid out: <br/><br/><div align="center"><b><span style="font-size: 10pt !important; line-height: 1.3em;">Components Of A URI</span></b><br/><img alt="" border="0" class="userimg" src="https://ip.bitcointalk.org/?u=https%3A%2F%2Fdeveloper.mozilla.org%2F%40api%2Fdeki%2Ffiles%2F3267%2F%3Duri-diagram.png&amp;t=589&amp;c=IB2e7gz117wstg"/><br/>This speaks to the conventions that Mozilla will account for in parsing a URI.<br/></div><br/>A typical magnet URI, looks like this<br/><br/>magnet:?xt=urn:sha1:YNCKHTQCWBTRNJIV4WNAE52SJUQCZO5C<br/><br/>A freenet URI is like this:<br/><br/><a href="http://127.0.0.1:8888/USK@oshw3DxmJUt7q4ThF4dCez5IXbc9hCGcv0VuwLRCmeQ,ckeXv20F1gBzkqssB4RXHZ2nB1YRT8Pb8KYZk8wj-bs,AQACAAE/occamsrazor/6/f.pdf">http://127.0.0.1:8888/USK@oshw3DxmJUt7q4ThF4dCez5IXbc9hCGcv0VuwLRCmeQ,ckeXv20F1gBzkqssB4RXHZ2nB1YRT8Pb8KYZk8wj-bs,AQACAAE/occamsrazor/6/f.pdf</a><br/><br/>To be recognized at all, the freenet program needs to be installed and this includes a Java powered web server that recognizes the resource as an item in identity space of the localhost. i.e. my browser sees the whole of freenet as being on my machine. I don't know if this identity space could be addressable without the user having to run a web-server, but I suspect the server is only for the ability to serve web pages as well as recognition of the localhost as the requisite top level domain. The 'http:' indicates that the freenet URI, is nothing special in itself, but the IP address (localhost) certainly is. I suspect a program can address the freenet user space, without incorporating a web server, and the universal links can be made as magnet URIs. As for those who don't have bitcoin installed, I don't know how the URI could be made to have a default fall back, to redirect visitors of that link. to the Sourceforge download for bitcoin. That may need to implemented in code and cookie or environ variable set to record the presence of bitcoin. <br/><br/>Other relevant sites I have visited in the course of this research include:<br/><br/><ul style="margin-top: 0; margin-bottom: 0;"><li><a href="http://infomesh.net/2001/09/urischemes">http://infomesh.net/2001/09/urischemes</a></li><li><a href="http://www.search.com/reference/Magnet:_URI_scheme">http://www.search.com/reference/Magnet:_URI_scheme</a></li><li><a href="http://magnet-uri.sourceforge.net/">http://magnet-uri.sourceforge.net/</a></li><li><a href="http://magnet-uri.sourceforge.net/magnet-draft-overview.txt">http://magnet-uri.sourceforge.net/magnet-draft-overview.txt</a></li><li><a href="http://www.w3.org/Addressing/schemes">http://www.w3.org/Addressing/schemes</a></li><li><a href="http://www.w3.org/Addressing/schemes#Registration">http://www.w3.org/Addressing/schemes#Registration</a></li></ul><br/>Any ideas/comments/feedback is/are welcome.&nbsp; <img alt="Wink" border="0" src="/static/img/emoticons/wink.gif"/><br/>Steve<br/><br/></div> text/html