kleinmatic

I wonder how much of the funding that led to this came from the Biden-era Chips & Science Act? I can't find a straight answer amid the AI slop and marketing hype about both of them.

Update: Looks like Trump admin converted billions in unpaid CHIPS act grants into an equity in Intel last year https://techhq.com/news/intel-turnaround-strategy-panther-la...

MLBT is terrific. Seems better than playball. I like having it open on a second screen.

“This patch series introduces multikernel architecture support, enabling multiple independent kernel instances to coexist and communicate on a single physical machine. Each kernel instance can run on dedicated CPU cores while sharing the underlying hardware resources.“

I got about halfway through before giving up on this article. I thought it was just taking a long time to get to the point then I realized one wasn't coming.

Thinking it was all a Duchamp-esque drollery, I checked to see if it was AI-generated and alas, it isn't.

Felt twinges of grief as I recognized some of these. I still miss Reader. But some weren't killed so much as became other products -- Urchin became Google Analytics, Google Mini became CSE (now Programmable Search), Google Refine became Open Refine, etc.

Pretty amazing how many people's ethics are less valuable to them than their product design opinions.

Google Reader, QuarkXPress 3.2, Lotus Improv.

"There is one drug simply listed as 'KNEES' and another as 'Foot and Ankle.'"

"The database includes three varieties of Clinpro 5000 toothpaste: bubble gum, spearmint and vanilla mint. One drug, CimziaCD, includes the notation "do not use" after its name."

They provide good detail about the software they used.

"The major software tools used for the Offshore Project were NUIX of Sydney, Australia, and dtSearch of Bethesda, Md. NUIX Pty Ltd provided ICIJ with a limited number of licenses to use its fully featured high-end e-discovery software, free of charge. The listed cost for the NUIX software was higher than a non-profit organization like the ICIJ could afford, if the software had not been donated."

Securiing their communications proved more difficult than free-text search, however:

"The project team’s attempts to use encrypted e-mail systems such as PGP (“Pretty Good Privacy”) were abandoned because of complexity and unreliability that slowed down information sharing."

Really, it's that publishing at the scale of a big paper requires separation of responsibilities. The workflow hasn't caught up to the new medium -- and neither has the publishing software. Newspapers are notoriously slow to adopt new tools. The New York Times was composed using linotype machines until 1978 -- about 20 years late -- and when I worked there in the late 1990s the pagination system (essentially, the CMS) was a green screen application running on a few rows of PDP-11s.

Jeremy's right -- publishing source documents is becoming a regular part of the process. DocumentCloud has accelerated that. At ProPublica we make a regular practice of it. Here's an index of documents we've published alongside stories:

http://www.propublica.org/documents

(Disclosure: I work at both ProPublica and DocumentCloud)

If something like this can slow your servers down, you're probably going to be in serious trouble when you get an inbound link from a heavily trafficked site. I'd recommend you put a caching proxy like Varnish in front of your web server process.

Varnish serves 404s quickly and happily, and will keep the storm from even reaching your apaches. You won't even notice it when the scrapers come in.

Thanks for the kind words about ProPublica. It's exciting to be around so many talented journalists and to get to work on great projects with them.

I run a small team (five of us) of developer/journalists. We're in a new-ish field of journalism called "news applications." Essentially it means we make software instead of using words or pictures to tell journalistic stories. There are crews like ours (some bigger, some smaller) at the New York Times, LA Times, Chicago Tribune, the Guardian in the U.K., and elsewhere.

Our work includes the Recovery Tracker http://projects.propublica.org/recovery and the recent "Docs for Dollars" project http://projects.propublica.org/docdollars that analyzes payments from pharmaceutical companies to doctors.

You can see pretty much all of our work at http://www.propublica.org/tools

I might have missed something in the video, but if I were an innocent student, the benefit for me in falsely claiming I cheated far outweighs the risk in defending my innocence.

The choices as I see them are these, whether you're innocent or not: 1) say that you cheated, and you get to retake the test as though you never took it the first time -- you don't even fail the test! -- but you never get to ask this professor of a lecture with 600 students for a favor. 2) don't admit that you cheated, get caught in some dragnet based on pretty flawed statistical reasoning (or better yet, a witch-hunt), and "not graduate." 3) Best case scenario: You say nothing, don't get accused of anything, and you get the undying loyalty of the professor, though that loyalty fails at the first try, because it doesn't extend to you getting out of a test you by definition shouldn't have to take in the first place.

I'm a bit stunned that only 200 students "confessed."

This project was done by ProPublica, the nonprofit reporting outfit where I work. There's more from the same investigation at http://www.propublica.org/dialysis including a leaked P.R. plan drawn up by an industry group to respond to our investigation -- and there's lots more to come in the next few weeks.