This will lead to a superabundance of power-hungry compute power in the hyperscalers, and it's not entirely clear what can be done to consume it all and still run at a profit unless they manage to make ever greater gains for ever more compute-hungry models that cannot be run on consumer devices, unless they refresh their hardware at ever faster and more expensive rates.

The joke about data centers used to be that their core business was selling power at a loss; this may end up being true of the hyperscalers next.

According to the bookies, Binface is currently 8/1 to win (about 11% probability), and Farage 1/8 (about 89%), with the rest of the candidates nowhere. There's probably something Bayesian you could do to estimate the probability of various margins from that, but it's going to be uncomfortable for Farage no matter what - winning over Binface by an embarassingly small margin will not enhance his image, particularly if there is going to be another byelection called soon after with the financial issue continuing to be in the spotlight. The only way this works out favourably for Farage is if he scores a crushing victory over Binface with an increased voter share relative to his previous election result, and I think the odds of that are slim.

[0] https://en.wikipedia.org/wiki/RRS_Sir_David_Attenborough#Nam...

[0] https://news.ycombinator.com/item?id=48888331

I wrote it because I was too lazy to learn how to use KiCad's layout features properly, and thought 'how hard can it be?'. Several months later, I had this.

It's not intended to compete with Altium etc. but it certainly produces compact, valid and fully design-rule-compliant boards with much less work that doing it myself or using one of the low-cost remote labour platforms.

It uses constraint logic programming to solve the hard parts of the problem. Hierarchical decomposition of the circuit design helps reduce combinatorial explosion, which was a show-stopper for early versions of the system. Current indications are that I may be able to scale it further in the longer term to deal with more complex design scenarios and larger boards, without hitting the exponential cliff.

Binface, ridiculous though it may seem, is going to be the unity candidate for tactical voting, and that's deadly serious from the point of view of those who wish to get Farage out.

Even if Farage wins, but with a reduced majority, in an election where the alternatives are fringe or novelty candidates, it will be hard for his dignity to recover.

https://www.btwholesale.com/products-and-services/data/sogea...

I wish I had a link for the other thing, but it was deep inside some Openreach site that Google don't seem to have access to.

They very much want to cut that back to one; big cost savings. And there are eventually going to be (hundreds of?) millions to be made from tearing the copper out from underneath the pavements and selling it, unless copper thieves get there first.

It's taken British telcos years to plan for this, and it's been put off a couple of times to deal with practical problems such as situations where you absolutely can't put fiber-to-the-premises in in any reasonable timescale.

This time they really seem to be determined to make it happen, even if it involves bizarro products like SOGEA, and if I recall correctly a sort of exchange-hosted baseband-only single-line DSLAM for the most intractable cases such as elderly people with no access to mains power - but even then it will implement the standard Digital Voice protocols, not the legacy DSLAM stuff.

[citation needed]