I'm not very interested in redesigning English spelling. However, it seems feasible to automate most of the encoding. I'd think most words could be rewritten with a lookup table, except for a short list of heteronyms. For printed material, I'm not convinced teaching kids how to read "Obsolete English" would be harder than teaching them to read cursive.
Updating grammar would be a giant pain in the neck. Saying "we're gonna write 'cat' as 'kat' now sounds a lot simpler.
People don't read ancient manuscripts but recent reeditions of old works. Those can be done using new spellings, as they are in, for example, Spanish. Nobody reads the Quixote, the Mio Cid or the Amadis with the Spanish orthography of their times.
> Nobody reads the Quixote, the Mio Cid or the Amadis with the Spanish orthography of their times.
And this is actually a loss. I have a fantastic edition of the Song of Roland with facing pages in Modern English and Medieval French, and there are nuances that would be lost in a straight translation to either Modern English or Modern French.
It's also completely impractical given the wide variety of very differently pronounced English dialects and accents. In other words, it requires massive pronunciation reform too. If that's on the table, why not alter everyone's pronunciation to fit the orthography...