On Tuesday 13 May 2008 20:30, Michael T?nzer wrote: > Matthew Toseland schrieb: > > > > IMHO the bigger problem is user retention. And how do we get better user > > retention? > > - More content. > > - Better performance. > > - A usable chat client. (Whether or not we choose to spend project resources > > on this, IMHO it will result in higher user retention). > > We don't need to attract nerds, they are already on Freenet (at least a > part of them). We need to attract artists, normal users, and other > people (they will bring in some new content) so we have to find out > about use cases which actually would provide them with some > functionality which they can't get elsewhere and which is a really > useful for them and then we have to build the software which makes it > happen. We have to give them reasons to use Freenet, not only give them > no reason not to use it. The question should be "Why should I use > Freenet?" not "Why shouldn't you use it?"
Anonymous web hosting is one obvious thing that you can't really get elsewhere - Tor is rubbish at it. > > Also we should explain the difference and the advantage between Freenet > and Tor better and in a well visited place (e.g. the whatis.html), there > are many users who think Freenet provides the same core functionality as > Tor and therefore use Tor (I don't mean to bring up a flamewar against > Tor but we have to clarify what the difference is and why it may be > useful to have both systems installed). It's in the FAQ, but maybe we need to allude to it or even give a very short explanation with a link to the FAQ. > > Apart from that we should have an ungeekish way to provide support to > people having questions about Freenet. IRC is definitely a bit geekish > and not everyone would subscribe to a mailinglist. But everyone who's on > the web knows how to use forums. Yes I know I don't really like them > either but ask yourself "would your mother subscribe herself to a > mailinglist?" she probably wouldn't. A functionality we could easily > provide with the new CMS. Ugh. Can we gateway them to the support list? Nextgens? IMHO this is orthogonal to to-CMS-or-not-to-CMS. > > Also we have to get rid of our filesharing image. I know you can do > filesharing over Freenet but it's not the right place to get the latest > blockbuster (one reason is because it's too slow). The current situation > is that those who think Freenet is yet another filesharing app (but with > better security) and who approve of that get disappointed and those who > disapprove it don't want to have it for that reason but if they had > tried maybe they found out it was the right thing for them. We should > emphasize this more on upcoming press releases. I disagree. Filesharing is something we do reasonably well. Yes it's slower than bittorrent for the most popular stuff, but for less common stuff IMHO it is competitive. > > For the killer application chat client talk to saces, he had a prototype > of some java app called minichat which actually could exchange messages > in almost real time (few seconds delay (<<10s), I guess with ULPRs it's > even less). The app is based on his message system Freeze (slogan > "cooler than Frost") which he started programming when the DOS on Frost > began (the first versions where published before we knew about FMS). > Freeze is a Freenet plugin (no port needed) has some features like queue > forking and so on and it should be possible to use a WoT on it (we have > talked about some nice ideas about entering the WoT), it is designed to > use GPG as encryption/signing (I think there where more cool features > but I understood only half of it). We also talked about maybe doing a > hybrid plugin which can talk Freeze and FMS, then we could migrate our > users softly over to Freeze. Well talk to him if you need to know more. I doubt very much that that will work well with WoT. ULPRs are unreliable, and it only works fast now because it polls constantly. > > regards > Neo at NHNG -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: <https://emu.freenetproject.org/pipermail/devl/attachments/20080513/30a629bf/attachment.pgp>
