@ned can you let us know how often your team uses the production steemit.com site each day, as in are you all eating your own dog food? When I see so few posts and comments from steemit, inc team members, I have to wonder how they can fully understand how frustrating the site has been to use these past 2+ weeks. The lack of communication from you, @steemitblog, and @steemitdev is concerning because it sends the message communication about site issues is not a priority. There's no status page explaining previous outages, and no custom error page when there is an issue to let users know you're aware of it and doing something about it.
Many in the community are frustrated (as you can see from the 150+ comments on the dog food post I linked to above). They want to be heard and have their frustrations validated along with clear communication of what the specific problems are and how they are being addressed.
Also, would it be possible for you to elaborate on this comment:
There are many things to consider for steemit that make mainstream DDoS solutions non-starters
I've tried to get a more detailed explanation from @sneak but the answers didn't make sense to me. You're already using AWS so it would seem they already have your private keys. Why not go with Cloudfront (within AWS) and have them deal with DDoS for you for around $3k a month?
I look forward to meeting you and the team in Lisbon soon so we can talk about this more in detail.
Note: my first attempt to post this comment failed.