Consistency and availability are competing requirements. It is trivial to keep the files in a SageMathCloud project consistent if we store it in exactly one place; however, when the machine that project is on goes down for any reason, the project stops working, [...]. By making many copies of the files in a project, it's fairly easy to ensure that the project is always available, even if network switches in multiple data centers completely fail, etc. Unfortunately, if there are too many users and the synchronization itself puts too heavy of a load on the overall system, then machines will fail more frequently, and though projects are available, files do not stay consistent and data is lost to the user.
and boldly summarizes
The architecture that we have built could scale up to a million users.
For example, each member of the House of Representatives is responsible for approximately 500,000 people. Assume that they spend on average two hours a day talking to their constituents and the rest is spent in committees, or on holidays (since we're talking about an average). That's 2628000 seconds per year, or around 5 seconds per constituent per year (10 seconds per term). If you want to have a five minute conversation with a representative, then you must find 60 people all willing to give you their time allocations. Or 300 all willing to give you 20% of their allocation. If you want to have an hour-long meeting, then that's 720 people who must give up all of their allowance, or 3600 who must give up 20% (or any breakdown).
It always amuses me when GPL'd software contains a clickthrough insisting that you press an "Agree" button, when the licence specifically says that no such agreement is necessary.
In fact, by placing the requirement that someone agrees to the license before using a derived work of the GPL'd software, they are violating the GPL...
Japanese products were initially low quality too. There have been a few interesting books on the subject of the change. In particular, several Japanese companies focussed very heavily on quality control processes for about a decade, which allowed them to dramatically improve their quality. Over the same time, the Japanese people who had been responsible for copying the designs became sufficiently familiar with them that they were able to initially improve them and then produce better ones.
The main factor stopping Russia or China going through the same transition is institutionalised corruption. It's hard to implement good quality control if you can't trust the people doing the inspections not to take bribes...
I'd have listed TalkTalk as the third large ISP, since they're the company that does the most LLU work. They install their own equipment at exchanges and only use BT for backhaul. There are quite a few smaller LLU operators, but BT dragged its heels to delay LLU rollout until they'd largely cemented their monopoly.
The problem with the split of BT retail and wholesale units is that there's no requirement for BT retail to make a profit. The wholesale part has to sell to BT retail at the same price that they sell to everyone else, but the retail division is able to operate at a loss and be bailed out by the rest of the company...
"I've seen it. It's rubbish." -- Marvin the Paranoid Android