As for the blog it was having a brief issue with psycopg2, apache, and python eggs... mutter. I'm still learning the sys admin side of things (but it is fixed now). In general it has been stable... (Apache2, Lighttpd for serving media, memcached for caching...).
I think the way to look for the future is like this: find something that everyone uses, and think of what unexamined assumptions we are making about it.
For an example, look at how comments work in blogs, maybe we could make a more localized comment system (you can make your comment close to the context you want to commenta bout), or at least inserting comment footnotes.
Essentially, look for places where the "proper" shape and form for an idea have crystalized, and then find a way to do it differently. At this point I don't think that web innovation is bounded by technology (horizontal scaling to build massive and efficient systems is a reality, albeit a n occasionally unpleasant reality for the implementors), but more by ideas.
Development frameworks (Rails, Django, TurboGears, CakePHP, et al) continue to trivialize internet development, so I think we'll continue moving towards an era where (for internet applications) ideas and design are the limiting factors rather than technology.
I think the way to look for the future is like this: find something that everyone uses, and think of what unexamined assumptions we are making about it.
For an example, look at how comments work in blogs, maybe we could make a more localized comment system (you can make your comment close to the context you want to commenta bout), or at least inserting comment footnotes.
Essentially, look for places where the "proper" shape and form for an idea have crystalized, and then find a way to do it differently. At this point I don't think that web innovation is bounded by technology (horizontal scaling to build massive and efficient systems is a reality, albeit a n occasionally unpleasant reality for the implementors), but more by ideas.
Development frameworks (Rails, Django, TurboGears, CakePHP, et al) continue to trivialize internet development, so I think we'll continue moving towards an era where (for internet applications) ideas and design are the limiting factors rather than technology.