1. Since people who follow me keep dunking on this map, please note it's a map of existing and under-construction metros as of 2003. 
2. Moreover, it aims to imitate the structure of the London Underground map, not to make a generic map connecting world cities with metros.
3. If you want to dunk, note that the definitions get weird, e.g. Baghdad is included despite never having had a metro system.
4. There's also some uneven treatment of regional rail, e.g. Auckland made it but Brisbane didn't.
5. But it's not some weird erasure of Africa; there's very little urban rail south of the Sahara and north of South Africa.
6. African rail construction has been a total shitshow, e.g. the Addis Ababa system, which didn't exist in 2003, is plagued by blackouts.
7. Post-colonial urban rail (except in the Maghreb) has been generally bad, and poorer places (Dhaka, Addis) do even worse than, say, India.
8. Yeah, historical knowledge of how NY/London/Paris did things at comparable development levels would be valuable.
9. But nobody has such knowledge, because it is not useful in rich countries today or even middle-income ones like China.
10. And ex-colonies tend to want to be like the first world today under influence of both cultural cringe and global consultancies.
11. I can point it out the easiest in India because of its English-language media, but it must be even worse in Nigeria, Bangladesh, etc.
12. The Maghreb exception, as @AnanMaalouf points out, comes from French consultancies that have been there for a generation.
13. Moreover, once high costs are around, they seem to stick - India today is almost as rich as the Maghreb but still has high costs.
14. And India's fast economic growth hasn't tempered its self-hate; its high-speed rail is a turnkey Shinkansen with very high costs.
15. Urban rail is such critical infrastructure for major cities. And these countries can't build it affordably. That's the real crisis.
16. From what I've read about Lagos, I'm not sure there's a single important decision about its metro that it got right.
17. There's a huge shortage of good rail planning capacity in poor countries and I don't even know what it would take to build it.
18. What we're doing re construction costs could have some insights, e.g. re the use of cut-and-cover. But the keyword is "some."
19. The US vs. rest-of-first-world difference is in context where the American private economy is actually pretty good.
20. The Americans may still be using physical checks, but overall their economy is really productive, just like ours in Germany.
21. In contrast, in really poor countries, everything is bad; Addis's light rail failures are connected to its general poverty.
22. Low economic complexity -> people work in-neighborhood more than in city center. Blackouts -> unreliable electric rail service.
23. Weak state -> no ability to integrate fares with buses. @cegoldbaum used to live there and does not describe the system as a success.
24. What's more, people in these countries recognize how bad things are, hence the eagerness to imitate the first world (or China).
25. I genuinely don't know how to produce and propagate the required knowledge for efficient urban rail construction in poor countries.
26. Is it something the World Bank can do? Engineers Without Borders? Universities taking foreign students? I wish I had answers. /end
