Nope. Note in TFS the key phrase: 'In its default configuration'. The university that I used to work for bought Office 365. This was even before the GDPR, but the university deals with a lot of confidential commercial data from industrial partners and with health records in life sciences departments. Google's stock T&Cs were completely incompatible with this and they refused to negotiate. Microsoft's stock T&Cs were also incompatible (which is why this ruling is completely unsurprising), but Microsoft was happy to negotiate a contract that gave much stricter controls over data.
For Germany in particular, the German Azure data centres are actually owned by a joint venture between Deutsche Telekom and Microsoft and so out of US jurisdiction. Companies in Germany (and the rest of the EU) can buy an Office 365 subscription that guarantees that their data doesn't ever leave Germany.
It gets more complicated when you factor in partial charges. LiIon batteries are most efficient if you never fully charge or discharge them. If you use around 40% of their total charge cycle each time then they last a lot longer, but then you have to increase your up-front costs in exchange for the lower TCO.
Solar cells are now very cheap. They are a negligible part of the cost for a small-scale installation. The cost of deploying rooftop solar is dominated by the installation cost (putting up scaffolding and having competent people climbing safely around on the roof is not cheap). The second largest cost is the storage and the alternator system to drive AC mains. Both of these costs are amortised significantly in larger installations. Most large installations are at ground level, so require a fraction of the manpower to install each panel. They use much larger alternator installations, which also come with higher efficiency.
TL;DR: Solar power is not immune to economies of scale.
That's technically true, but not practically true. Most software gains very little (more entropy in ASLR) from moving to using 64-bit pointers. Most software gains a lot from moving from x86-32 to x86-64 as the ISA. In particular, it gains a lot more registers (and, importantly, loses the restrictions on which instructions it can use). It gains cheap PC-relative addressing (i.e. shared libraries are a lot faster). It gains SSE2 as a baseline, so floating point operations and calling conventions are cheaper.
There's a good argument to be made for using the x32 ABI, but generally speaking x86-64 code with the 64-bit ABI will still be faster than x86-32 code using any ABI.
You've inadvertently flagged the real problem with XMPP: It doesn't store messages server side, or it does store messages server side, depending on which protocol extensions a given implementation happens to have. For anything that you might want to do with XMPP, there are 2-10 different XEPs with varying levels of support, that describe how to do it.
XMPP badly needed a high-quality reference implementation of a server and a library for implementing clients. Instead it got two crappy reference implementations (that were so bad even the standards editor didn't recommend using them) and a load of partial implementations of the client part of the spec.
Statistics are no substitute for judgement. -- Henry Clay