> The hyperscalers have already outspent the most famous US megaprojects

[aeva@mastodon.gamedev.place]

@cwebber imagine all the things that could have been done with that money if it was spent on stuff that benefited the general public instead

[musicman@mastodon.social]

@cwebber the data center bubble is clearly dumb, but I think the railroad line might be the most interesting/misleading one on the graph.

Or maybe it is the Interstate line that is misleading...or potentially misleading? Idk, no graph is ever going to be perfect, but yeah...

anyway...

It seems to equate rail spending and road spending, but the Interstate system is not the entire federal road system.

[ohmrun@indieweb.social]

@cwebber
Mirror, mirror on the wall, who's the over-leveragedest of them all?

[seanmiddleditch@hachyderm.io]

@aeva @cwebber Benefit the general public!? That's absolutely not what the people who invented money wanted it used for.

[smilingdemon@mastodon.art]

@cwebber utterly insane.

[balcanquhal@tldr.nettime.org]

@sixohsix @cwebber perhaps, but all of the numbers the hyperscalers are associated with are partially/entirely made up

[elexia@catcatnya.com]

@cwebber luckily we will never need another data center ever again

[sixohsix@icosahedron.website]

@balcanquhal @cwebber so we're agreeing that this chart is misleading and meaningless but in different ways...?

[flashesofpanic@ruby.social]

@musicman
I've been working my way through _Railroaded_ https://app.thestorygraph.com/books/ac561c89-bdeb-453c-be0a-348eb2f288e5 and if anything it makes me think the railroad line is *too steep* - railroads in the second half of the 19th century were massively over-leveraged and overbuilt (especially in the American West), and a very large fraction of the investment went into the pockets of the financiers rather than into construction. Which... sounds familiar, now that I type it.
@cwebber

[jpkolsen@social.data.coop]

@darkuncle @cwebber It would basically be indistinguishable from the x-axis

[shane_kerr@fosstodon.org]

@cwebber Marx said that capitalism requires periodic destruction of capital. Just sayin'

[aspensmonster@tenforward.social]

@shane_kerr @cwebber Crisis of overproduction here we come

[gimulnautti@mastodon.green]

@cwebber It feels like we’re creating a new ”oil”. On purpose.

A global dependency on compute cycles, with control in very few private hands.

Ready to hold the world hostage line Iran in the Strait of Hormuz..

This is genuinely scary shit.

[akamran@indieweb.social]

@cwebber this is a mind boggling waste of money

[pjakobs@mastodon.green]

@cwebber are those inflation corrected values? Apollo dollars were about 4x of today's dollars iirc

[npars01@mstdn.social]

@flashesofpanic @musicman @cwebber

Niall Ferguson has observed that financial crashes precipitate wars, big ones

The bankruptcy of the corrupt Canadian railroad private-public partnerships in the early 20th century started a domino that cascaded into WW1

List of defunct Canadian railways - Wikipedia

(en.wikipedia.org)

Canadian Government Railways - Wikipedia

(en.wikipedia.org)

The story of Canadian National Railway

Explore Canadian National Railway’s history, freight dominance, privatisation, Bill Gates’ stake, pandemic impact and why investors value rail infrastructure.

(freetrade.io)

Historian Sarah Paine talks how the wealthy's response to the corruption & crash of 1929 led to WW2 & 90 million deaths.

Corruption, crash, war -- a not-lovely cycle

[neilk@xoxo.zone]

@cwebber Adjusted for GDP, it is still pretty respectable

Not quite the Marshall Plan, way behind railroads https://x.com/finmoorhouse/status/2044985359281381690?s=20

Still it just seems impossible. Does this represent orgs with committed budgets or actual spend? We can make warehouse style buildings almost instantly but we can’t make GPUs that fast.

[outfrost@mastodon.social]

@cwebber the railway situation makes a lot of sense in this context

[screwlisp@gamerplus.org]

@cwebber
What if you norm this graph by the net worth of the world's richest person at the time.
@akamran

[tabaer@iosdev.space]

@cwebber

Much of this will turn out to be wasted