Citation for your claim:

"The "far more capacity" you cite is roughly double, which is pretty close to the difference in deep discharge capability between lead-acid and Li-ion"

Citation please, considering that if this was even remotely close to being true, you're suggesting that all major telecom operators have fucked up in a major way, as main form of cell tower backup today is VRLA battery banks. If your capacity argument was anywhere near true, li-ion would have been long considered as operators are updating cell tower hardware very often and I have never heard li-ion being more than considered unless there was a specific need for it (i.e. weight restrictions on site). Reason is always the same - far too low capacity for cost.

Afaik (hearsay from a friend who is a professional electrician that among other things did electric wiring in large apartment building I used to live in) the main reason for this is costs. But that was often left to the electrician to decide on site in older buildings where building process was nowhere near as tightly degisned regulated (i.e. benefits of computerization of design process didn't exist yet).

Nowadays requirements tend to be tighter.

Which point makes no sense to you? The fact that snow comes every year? That fact huge earthquakes occur approximately once a century? That due to relative energy density, a single tank of an average household sized backup generator will hold much more energy than a battery bank like one advertised in the story?

On a last note, as a Finn with experience in actually clearing snow I would like to simply state that you have no clue on how hard it is to clear snow from the roof to the level where solar panel would do anything at all.

Because snow happens for extended amounts of time every year in relevant regions which are huge, whereas huge earthquakes happen only at faultline regions and only once a century or so.

Also, "severing pipelines" on regional level doesn't really matter in case of earthquake for single household in case we're discussing. If you get severed connections from disaster, chances are that you either leave the region if roads are still operable and logistics work until basic damage control has been done by emergency services, or they are too damaged to allow you to leave and you're stuck and limited to what you have directly on site.

In which case, you likely have far more energy in a single tank of your ICE generator than in a battery bank.

Considering the existence of residential flooded cells and VRLA batteries for decades and the fact that they absolutely destroy these li-ion batteries in all relevant factors except weight, I'd say you're barking up the wrong tree.

It's more about "ignorant people that didn't know battery backup for residential and small business existed for decades and is far more efficient than these batteries" vs "people who know battery backup for residential and small business existed for decades and is far more efficient than these batteries".

Considering the fact that instead of actual hard numbers, initial release mostly focused on how pretty said batteries look, it's pretty obvious who they are going to be marketed towards however.

Sun's thermal radiation heats roof, thermal energy goes into attic though the roof, there is no active cooling in the attic other than the fan that kicks in at 43C.

That is how thermal energy works. Just because temperature in well ventilated shady area is 23C doesn't mean that it's going to be the same in well lit enclosed space.

Which incidentally is likely to be the site of battery installation.

That is true for wet cells. VRLAs are far less picky.

Factually incorrect on all accounts. VRLAs are sealed and do not discharge significant amount of H2 outside of the individual sealed cell. Lead wear and tear is also minimal in residential use as batteries can have far greater capacity (than say li-ion for the same cost).

Finally realistic discharge rate for lithium batteries is about the same as lead. The only way to hit the numbers you suggest is to have minimal discharge on li-ion, in which case acid-lead will likely be able to match them.

Which is why warranty for typical VRLA is the same one as Tesla is willing to give it's li-ion batteries - 10 years. With exchange costs being massively in favour of lead-acid.

Citation desperately needed for all those claims, considering that opposite to your claims is easily demonstrable in the existing installations of VRLA and flooded cells applications which range from cell tower backup to residential backup and are widely in use.

There are plenty of companies on the market that sell the entire systems with either VRLA and flooded cells, including the hook up and the inverter. Which is profitable not only because lead-acid absolutely destroys everything else in the market when you care about cost, capacity and safety but not weight, but also because control electronics for lead-acid are much cheaper than those needed for li-ion.

You mean like countless companies that sell VRLA and flooded cells residential backup batteries that have existed for decades?

If you want one, just get residential VRLA battery system. Minimum fuss with maintenance and control electronics, costs significantly less than li-ion for the same capacity, is more efficient than li-ion and there are plenty of companies that have well proven residential VRLA backup batteries in their inventories.

Or if you don't mind the fuss, just get the wet cells. Even cheaper and more efficient but require some regular maintenance.

Inefficient in comparison to what? Batteries that waste significant amounts of energy on being charged and then even more on discharge?

I can understand the comparison to things like large plants with burners rated in multi-megawatt range, but that's not what we're talking about, is it?