Hasnain says:

“Let me finish with this, and it hurts: Most companies just need basic bash / make knowledge, a single instance SQL processing engine (DuckDB, CHDB or a few python scripts), a distributed file system, git and a developer workflow (CI/CD). Everything else is sugar and enterprise stuff. Not saying these last ones aren’t important, but the basics should be well covered, otherwise you’ll have a huge mess. I’m still surprised by how quickly we forget about the good practices we have learned over the years in software engineering (testing, deployment, collaboration, monitoring and so on)”

Hasnain says:

Susan’s powerful speech at Oxford Union the other day during which they passed a resolution declaring it apartheid (and I believe genocide). Worth reading in full. Powerful words. There is a super long segment I felt like quoting but it was a bit too long

“It’s clear to me that we’re not here to debate whether Israel is an apartheid or genocidal state. This debate is ultimately about the worth of Palestinian lives; about the worth of our schools, research centers, books, art, and dreams; about the worth of the homes we worked all our lives to build and which contain the memories of generations; about the worth of our humanity and our agency; the worth of bodies and ambitions.”

Hasnain says:

TIL lots of fun optimization math

"Taking a look back at all of the hardware-specific optimizations, we can see an orders-of-magnitude improvement over our initial naive implementation and stock NumPy implementation. Utilizing and exploiting hardware features has delivered performance boosts from 10 MB/s to 60.3 GB/s on Intel hardware, and 4 MB/s to 29.7 GB/s on Arm hardware. It underscores the absolute importance that specialized hardware acceleration libraries have on even the simplest of computational algorithms."

Hasnain says:

This is the most MBA speak sentence I've heard in a while

"Chris Hayward, policy chairman of the City of London Corporation, said the decision represented a "positive new chapter" for the markets as it "empowers traders to build a sustainable future in premises that align with their long-term business goals"."

Hasnain says:

"While talking to the CEO of Microsoft, Pratchett asked what would happen if a writer disseminated on the internet something atrocious and libelous, say a pseudo-academic work of Holocaust denial. “There’s a kind of parity of esteem of information on the net,” said Pratchett, “there’s no way of finding out whether this stuff has any bottom to it or whether someone has just made it up.” Predictably, Gates denied the threat of any sort of epistemological collapse. Without offering any mechanism for doing so, the billionaire told the author that “you will have authorities on the net… The whole way that you can check somebody’s reputation will be so much more sophisticated.” Google was three years into the future—Facebook would be founded in nine years—Twitter in eleven. If Pratchett seemed sardonic and cynical in 1995, then Gates’ pollyannish, Panglossian exuberance appears positively psychotic three decades later."

Hasnain says:

"HR veteran Steve Cadigan says his thinking—that PIPS are almost always a bad idea—was shaped when he worked at Cisco early in his career. “We did a five-year lookback at every PIP and found that 90% of people who were placed on a formal PIP, whether or not they survived it, left within a year of that warning. Which to me suggests there’s a fundamental break in the trust of that relationship,” says Cadigan, who went on to be the HR chief at LinkedIn and now advises companies on HR strategy.

So he sets out to avoid them. “We give you two envelopes,” Cadigan says. The first is a PIP. The second offers generous severance with a separation agreement and Cobra, or continued health insurance. “Seventy-five percent of the time people take option two. So we circumvent the whole PIP process and just say, for whatever reason it’s not working out.”

This approach is increasingly popular, particularly in tech. It’s used in some form, and sometimes on a case-by-case basis, at companies including Amazon and Meta."

Hasnain says:

Insightful read on promo packets. Once you’re in the know this seems fairly obvious, but it’s not obvious at all from the outside. Focus on impact, folks!

“Regardless of the narrative structure you choose, before typing a single word into your promo packet draft, try to tell someone (yourself even), verbally, your bunny story. How bad was your Dark Lord’s reign? What were your weapons of battle? And just how happy were your bunnies in the last chapter of your hero’s journey?”

Hasnain says:

“I do not quite know what is worse. The prospect that billionaire Washington Post owner Jeff Bezos dictated this absurd Editorial Board article, similar to how he forbade the Washington Post from endorsing during the Presidential election? Or, the prospect that the Washington Post editorial board actually believes the bloviating fact-less propaganda they just published. Either way, it does not bode well for a healthy fourth estate committed to holding the powerful accountable.

The facts are clear. The ICC is perfectly within its jurisdiction, right, and obligation to hold Netanyahu, Gallant, and Hamas leaders accountable for war crimes. That the ICC has not yet reached the point of filing arrest warrants for each of the despots mentioned by The Post is nothing more than an argument for increased ICC funding so it can fulfill its mandate of prosecuting war crimes. Rather than excuse war crimes committed by people The Post likes, perhaps its editorial board can remember that their job as journalists is not to capitulate in the face of atrocity, but to hold fast the call for democracy. If this is a task too difficult for The Post’s board, then maybe it’s time to resign and let actual journalists take the lead.”

Hasnain says:

“However, it is now clear to me that there is ongoing work on structured editing which either doesn’t know about incremental parsing in general, or Tim’s algorithms specifically. I hope this post serves as a useful advert to such folk that incremental parsing exists, and Tim’s and Lukas’s theses specifically, are absolutely worth the investment of time in reading.
I would love to see more progress in this area, and the more that those working on it know the breadth of our existing knowledge, the more likely we are to make meaningful progress. I hope to see more such progress soon!”

Hasnain says:

“An event happens on a date, at a time. The precise details of that time are very important: if you tell people to show up to your event at 7pm and it turns out they should have arrived at 6pm they’ll miss an hour of the event!

Some of the worst bugs an events website can have are the ones that result in human beings traveling to a place at a time and finding that the event they came for is not happening at the time they expected.

So how do you store the time of an event?”