To explain more what's been happening with the old explorer: To develop faster we decided to keep the whole explorer in memory. By not needed to add a database, I was able to shave a few days off of the coding time. Well, the database kept growing (as blockchains tend to do) and it became too much for the explorer to hold in memory. So the explorer started crashing.
The new explorer is less tested, but has everything in a database. Also ,we fixed some of the heavy networking code. So, the new explorer should have a lot more stability. Once we're convinced that there are no major bugs, we'll publish it.
I will say that our primary focus is very much not the explorer at this time, and overall the quality of code on the explorer does not match the standards we typically strive for. This is purely because we are time-constrained, and have chosen to focus all of our effort on the renter and host.
You need to purchase Bitcoins first, using any method, and then you can exchange them to Siacoins on the Poloniex exchange.
You don't have a single wallet address, but a different one should be generated by the client for each transaction; you can obtain a fresh one by typing "siac wallet address".
@jthor Yes the connection speed is important as well as uptime, whether you're accepting incoming connection or not, etc.
For example on my fastest server I have going, I typically am connected to between 22-36 peers at any given point, when using the client on home connection for testing uploading of my own files, I am sitting between 7-11 peers.
If your chain stops updating entirely, you are probably uh bottlenecked and choked with orphan blocks or similar, I did once delete the entire chain data to force resync and that worked in my case but you shouldn't need to do that generally.
The 16SC doesn't come in the form of transactions, it comes in the form of completed file contracts. Right now the wallet doesn't have good support for reporting balances from completed file contracts, but it will know that it got the money and it will be able to spend the money.
As for the <nil> errors, @allejuppa already created a PR (and it was merged) to give the correct error information.
I can't be of too much help, except to suggest that most of the network is experiencing a similar number of errors in their logging. I think at this point we've addressed some of these errors, but we're continuing to work on the renter and host protocol right now.