If the Twitter promotion algorithm works using machine learning, its code (not including the neural net) won't tell people much about how Twitter chooses what to show them. It also may not be much use to anyone.

Basically, the trained neural net is a part of the program. It can be considered source code, since (1) it isn't compiled from any other form of the code, and (2) changing it is possible (by retraining. But if you don't get that, you don't have the whole source code.

The UK immigration agency has infiltrated its agents into local governments, in agencies where people might look for help. Even in child protection services.

Immigrants will learn to avoid asking for any help for themselves or their children, and the agencies will say, "It's not our fault the didn't contact us."

A pilot program in the UK found that *supporting women asylum seekers in the community … improve[d] their well being at half the cost of [jailing] them.*

Ministers were not impressed, though. Perhaps because it contradicted their plans for a "hostile environment". The UK is already planning more immigration prisons.

Any one country which is so helpful to immigrants is likely to get far too many immigrants. But if many countries do it, it may be acceptable for all of them.

Elon Musk is citing a barely extant American "far left" as a bugaboo to excuse the powerful, menacing, election-stealing far right.

*Pacific Elders Say US-China Military Tensions Secondary to Rising Seas.*