The thing I was referring to was not block chain compression (that's not making a big difference) but rather pruning, i.e. deleting of old data from disk. There has been no progress on that front. Sipa worked on other things instead.
That is what I meant. It is to bad that it's not being worked on. Again, it was one of the things in the whitepaper which gave me some hope for the sustainability of the system as a more realistic community maintained solution. I've suspected for some time now that once the
perception of possible scalability in this respect was implanted, the goal of that text was achieved.
If your problem is you can't afford to even download 10G of data then you're better off using an SPV client instead.
It's easy to get spoiled when working on high capacity networks and neglect to consider the various use-cases. I've never argued that POTS or satellite should be supported as a baseline, but have argued that if they could it would result in a system which was much more difficult to subvert. I personally don't think it is worth the tradeoff though.
As soon as a simple SPV client implemented in a language which does not effectively require un-trusted dependencies exists I likely use it for certain things. Of course it adds no value to the Bitcoin network other than 'headcount' perhaps so I guess that the operators of the network will be extracting value from clients like this in other ways.
I'm pretty sure almost any VPS could run bitcoind - where did you find a VPS that has <20G of disk and bandwidth?
I typically try to plan my infrastructure investments (which are often more about time than money) for a reasonable life expectancy. If the resource utilization rate is predictable then this is possible. If, say, the transaction rate could change on a whim and necessitate a rapid escalation of the resources I need to deploy such a system then there is less likilyhood that I will bother in the first place. I doubt that I am alone in such a calculus.
- edit: to answer your question, I was looking for 1) a system which would allow me to compile my own OS, and 2) in a jurisdiction which was suitible miffed at the NSA spying that they may have take real steps to prevent it and have the technical expertise to do so. I found myself here:
http://nqhost.com/freebsd-vps.html. I also considered AWS micro instances which may or may not work.
If you ran one on a VPS you could use an SPV client locally that connects to it, and that'd be an equivalent security level.
If I ran bitcoind on a VPS there would be little or no need to run anything locally. At least on a VPS that I could have some confidence in from a security perspective (a big question mark in my mind at this time.)
