I don't think we need to create an entirely new alphabet. There is a created for North America Indian and Eskimo languages, noone else can read it, and there's alternative orthographic systems for those languages in Latin as well (in fact, that syllabic system was not created from scratch as it was largely inspired by Devanagari). Bearing in mind that our project exists almost entirely in WWW, that would pose really big problems concerning computer fonts. Same is true about Runes, not many browsers can handle them correctly without installing additional fonts. As for any home-made alphabet, it wouldn't be a part of Unicode, so it would be practically impossible to use it on the web.
Having said that, I like idea of runic Nynorn very much, and we've already discussed it a little bit in the past, look and . But runic Nynorn can IMHO exist only as a supplement to its Latin-based version.
In fact, no alphabet suits a language 100% correctly except the one it was created primarily for. Even Russian and other Slavic languages have issues with unstressed vowels and palatalised consonants that aren't handled unambiguously. Nynorn has a phonetic-based orthography, no rudiments like f/ph, qu/c/k etc., but still there are sounds that have no letters for them (or they do only in dialectal notation, like æ for palatal a: all vs. ællj/adl, see
http://nornlanguage.x10.mx/index.php?nynorn_dial).
We've only got 2 letters (å, ð) that could be avoided (å=o,ð=zero) and have only etymological value. But in a number of cases they are really useful, as they help us to distinguish between homonymous words having different meaning and different origin. To sum up, I don't see any reasons to change this policy, the current orthography looks very Scandinavian like a mixture of Faroese and Norwegian, it can't be better than that, can it?