Ian Clarke <i...@freenetproject.org> writes: > Some good suggestions.
Thank you! These are not suggestions, though, but things I see as good paths forward. They are where I consider my own time well spent. > On the UI, this would require a rewrite I disagree. To explain that, I’m making an exception from the "not a request for input" I wrote in my email. Becoming the coolest hipster interface would require a rewrite. Becoming a joy to work with would require a rewrite. Becoming a good tool on par with or better than 90% of the websites out there would not. And there is not that much work needed for that. It’s just boring work. > If we wanted to take a fresh stab at the web UI, it would be worth at least > considering Kweb. The code looks nice — and somewhat similar to what I see from the sxml-folks in the Scheme community, i.e. http://www.nongnu.org/skribilo/#&prog-line927 If someone wants to take it up and actually finish it, why not? It could start as a plugin and gradually replace parts of fproxy, just as we tried to do it with the Winterface plugin. Though for the past 10 years our problem in Freenet hasn’t been the lack of great new approaches, but rather the lack of people actually putting in the work to get to completion. A rewrite would start out lacking a lot of features. If there were no person behind it who’d really push it forward with a lot of free time, it would either stall fproxy for even longer or it would have to constantly chase fproxy as a moving target. In general I consider rewrites a big risk. Will someone have the will, skill and perseverance to finish it once the fun part stops and what’s left are the boring 90% of the work? Those 90% are what I did when I took over the ShareLink plugin and turned it into ShareSite. But I cannot do it for a task the size of the full Freenet interface. I realized that when supporting the Winterface plugin (which is *not* the Winterfacey I named in the world domination plan: the theme is a theme to make the regular fproxy look similar to the Winterface plugin; when seeing how much it achieved I realized how much we can do with simple, non-breaking improvements to fproxy. We just have to *do* them). I’m not going to write a lot more on this topic. I want to get moving again with the *doing* for Freenet, now that I finally have a working internet connection again. Aside from getting the next release done, I want to finally implement the defense against the pitch black attack. Over the past years I could convince myself in modelling and reasoning that Oscar was right and that flaws I thought I saw are far less serious than I thought. For example follow-up swap requests after a defensive swap should minimize data-loss. Best wishes, Arne
signature.asc
Description: PGP signature
