The Spy War: How the C.I.A. Secretly Helps Ukraine Fight Putin →

Now entering the third year of a war that has claimed hundreds of thousands of lives, the intelligence partnership between Washington and Kyiv is a linchpin of Ukraine’s ability to defend itself. The C.I.A. and other American intelligence agencies provide intelligence for targeted missile strikes, track Russian troop movements and help support spy networks.

But the partnership is no wartime creation, nor is Ukraine the only beneficiary.

It took root a decade ago, coming together in fits and starts under three very different U.S. presidents, pushed forward by key individuals who often took daring risks. It has transformed Ukraine, whose intelligence agencies were long seen as thoroughly compromised by Russia, into one of Washington’s most important intelligence partners against the Kremlin today.

How Liberals — Yes, Liberals — Are Hobbling Government →

Nicholas Bagley is a law professor at the University of Michigan, a former chief legal counsel to Gov. Gretchen Whitmer and the author of a fascinating paper called "The Procedure Fetish". In it, Bagley argues that liberals — liberal lawyers in particular — have helped hobble the very government they now need to act swiftly and decisively..

Some great points in this interview on the role of rules:

What’s a concern is what we put in place to try to protect against those risks. And I think far too often, we’ve defaulted to lawyerly, procedural rules to try to get agencies to do better. And I think we haven’t been reflective enough about the possibility that, look, if you’ve got an agency that isn’t living up to your expectations, layering a bunch of procedural rules on top of it and subjecting it to judicial review is probably not going to help all that much.

And if we’ve got problems, we want to be thinking much harder about institutional design, about adequate budgetary support, about making sure they’ve got a clear statutory mandate, about getting good leadership, about fixing management problems. There’s a whole host of things we could do to improve the functioning of the administrative state.

On running an agency:

[I]f you are asking a group of lawyers how better to run a big institution, you’re asking the wrong group of people. [...] They’re not very good at running organizations. [...] It’s not something you learn in law school, how to run a government agency or — it’s just not part of the skill set.

On government legitimacy:

You write that, “Legitimacy is not solely, not even primarily, a product of the procedures that agencies follow. Legitimacy arises, more generally, from the perception that government is capable, informed, prompt, responsive and fair.”

And on public/private partnership:

[O]ne thing that’s become clear to me is that the private sector depends on an effective public sector in order to achieve its goals as well.

So right now, we’ve got a ton of money coming in, both private capital and also public capital for new investments in semiconductors and electric vehicles and renewable energy facilities, so all this money sloshing around. But all of these facilities require permits from the state. They require financial support, in some cases, from the state. They require literally parcels of land that the state often assembles.

Not to mention, they depend on a work force that the state has educated. They depend on a robust infrastructure, like transportation infrastructure, on water. I mean, if you are a private company looking to expand, looking to build these new businesses of the 21st century, you need a functional state. And crippling the state is bad for business.

This is You on Smiles. We All Have Photographic Memories Now →

We’ve ceded many of our remembering duties (birthdays, schedules, phone numbers, directions) to a hard drive in the cloud. And to a large extent, we’ve now handed over our memories of experiences to digital cameras.

How Democracies Must Compete with Digital Authoritarians →

Artificial intelligence (AI) may still hold the potential to solve some of the world’s most intractable problems [...], but when it comes to risks to privacy and civil liberties, AI already has been a game changer in favor of authoritarian states. AI-enabled tools have turbocharged every pre-existing form of repression including: mass and targeted surveillance, censorship, and the spread of propaganda.

Retired Ambassador Eileen Donahoe outlines that democratic nations must:

1. Develop and showcase a democratic model of digital restraint at home
2. Exercise leadership in international tech diplomacy and standard setting
3. Invest in tech innovation as a means of advancing democratic values

Offense will win some battles, but cyber defense will win the war →

Changing the calculus on defense remains the most important way to prevent attacks, even if it is not as attention-grabbing as offensive efforts.

The wonder material graphene may have found its killer app →

[G]raphene remains largely unemployed. Certainly, no killer application of the sort predicted when the stuff was discovered has emerged. But that could be about to change. Concrete is [...] an important material and of great concern to those attempting to slow down global warming, because the process of making it inevitably releases carbon dioxide. And graphene may hold the key to reducing that contribution considerably.

The 10-Year-Old Tweet That Still Defines the Internet →

[T]he best-known @Horse_ebooks tweet, posted 10 years ago today, was astounding in its clarity and salience. It described both the internet and our entire human world. "Everything happens so much," @Horse_ebooks tweeted on June 28, 2012.

EvilSln: A New Exploitation Technique for Visual Studio Projects →

Running arbitrary code from the Internet is a well-known risk, but did you know opening it might be?

Using logs to build a solid data infrastructure (or: why dual writes are a bad idea) →

Say you have this tangle of different datastores, caches and indexes that need to be kept in sync with each other. Now that we have seen a bunch of examples of practical applications of logs, can we use what we’ve learnt to figure out how to build these systems in a better way?