Ok, Lets sum up the complaints and the answers.
- Lazy People wont setup their own nodes.
Sure, but understand that most people are lazy when they first adopt a
protocol. They want it to work out of the box. (I do it all the time)
Any program that doesnt work within (2-3 min) is likely to be rm -fr 'ed.
An if we can agree that we want more people using freenet this is a
concern.
- Proxy Addresses shouldn't be kept in the URL (even if it's a fallback)
I see some parallelism between us and nntp.
NNTP is simulating an information "cloud" like we are. They're
decentralized in the way messages get passed around.
NNTP has a system for finding the default entry point (settings,
lookup "news") and we defaulting methods.
Yet NNTP has this: (from rfc)
The nntp URL scheme is an alternative method of referencing news
articles, useful for specifying news articles from NNTP servers (RFC
977).
A nntp URL take the form:
nntp://<host>:<port>/<newsgroup-name>/<article-number>
where <host> and <port> are as described in Section 3.1. If :<port>
is omitted, the port defaults to 119.
The <newsgroup-name> is the name of the group, while the <article-
number> is the numeric id of the article within that newsgroup.
Note that while nntp: URLs specify a unique location for the article
resource, most NNTP servers currently on the Internet today are
configured only to allow access from local clients, and thus nntp
URLs do not designate globally accessible resources. Thus, the news:
form of URL is preferred as a way of identifying news articles.
Again, an *alternative* scheme for getting data along with the
news:<newsgroup> format. Look familiar??
- Node addresses in web pages are a problem.
I've covered this earlier. We're going to specifically disallow this.
The enforcement mechanism is browsers will be configured to break these
links, making the links completely unusable. There's no incentive to put
a node's name on hypertext.
Putting a node name in the URL is only a suggestion made by a human
operator.
- Why is this point so important?
Again, the idea that good software starts working fast. It should work
fast for people without too much technical knowledge. This is how freenet
will be come poplar.
Example Situation:
Person 1: hey have you seen the new area 51 pictures?
Person 2: No, where are they?
Person 1: bring up netscape,
type in "freenet://<person 1's ip>/area/51/pictures"
Person 2: Hey it works. But what's "freenet"?
And there you've made a freenet user simply by proving it works.
Yes a person can be lazy and never setup a node, but the consequences are
an unreliable connection. With a little more research they can figure out
how to setup a node or configure their ~/.freenet to point to their
organization's freenet node.
We can make it easy to first enter freenet. It's better for users and for
freenet.
Thoughts?
-Larry
_______________________________________________
Freenet-dev mailing list
Freenet-dev at lists.sourceforge.net
http://lists.sourceforge.net/mailman/listinfo/freenet-dev
