Mark Downie

When LLMs Remember Instead of Reason SWE-Bench Verification asserts to be ahigh-quality, human-validated subset of ...

A healthy dose of skepticism is warranted whenever you hear: “the most powerful LLM ever released across key benchmarks”

📚 "When LLMs Remember Instead of Reason" #AI #benchmark

www.poppastring.com/blog/when-ll...

I did not know there was a book, that is going on my list. Thx 👍🏾

YouTube video by Black Girl Nerds The Films We're Excited For (So Far) at TIFF 2025

The final schedule releases this week, but in the meantime, here's what we're excited for so far! #TIFF25

How much of Gaza is left standing? Analysis by independent researchers reveals the scale of the destruction

“The studies imply that estimated life expectancy [in Gaza has fallen by more than 35 years, to half the pre-war figure. In percentage terms…bigger than…during China’s Great Leap Forward; in absolute terms it is similar to the one in the Rwanda genocide”

This rescue kitten was abused by her previous owners, but her new foster mom knows exactly how to gain her trust... 🥺 #cat #catday #bluesky

Watching Sphere (1998) with Dustin Hoffman and Sam Jackson, the only good thing about it so far is a brief cameo by Queen Latifah #movies

Three screen caps from the movie Hackers with text. First screen cap: Protagonist Dade (Zero Cool) explains that “I’ve got a record. I was Zero Cool.” Second screen, his friends explain: Zero Cool crashed 1507 systems in one day. Biggest crash in history. Third screen, another friend says: Front page New York Times August 10, 1988

HACK THE PLANET!!! Happy Zero Cool Day to all who celebrate!

A post from the archive 📫:
A better vision for Central Ohio


https://www.poppastring.com/blog/a-better-vision-for-central-ohio
#columbus #musings

Multiple times this year I’ve had coffee with fellow immigrants who grew up in corrupt third world countries and the shared sentiment has been that Americans don’t understand how bad things are going to get.

The idea of the moral (billionaire) CEO is clearly a myth, but this still feels generous for some reason. My cynical view is that they cultivate and curate personas.

The moral theater is necessary because decency doesn’t survive shareholder scrutiny, and courage has no line in a quarterly report.

The shared memory of our culture does not seem to extend much beyond 50 years, shorter than a single modern generation.

A post from the archive 📫:
Making production diagnostics easier with Source Link


https://www.poppastring.com/blog/making-production-diagnostics-easier-with-source-link
#debugging #visualstudio #diagnostics

a) The president can't unilaterally federalize DC; Congress would need to repeal the Home Rule Act. Have fun trying to convince the parliamentarian that this doesn't actually require 60 votes in the Senate
b) Statehood please

There's enough horrible things happening in the world.

Please don't have your API return a 200 OK with an error code and add to that list.

spot art. text reads: Visual Studio Live @Microsoft HQ The hype is real! Streaming LIVE in just 2 days… Visual Studio Toolbox broadcasts from Microsoft HQ in Redmond

Tomorrow’s the day!

We’re going live with VSLive! and Visual Studio Toolbox—streaming straight from Microsoft HQ.
Is your reminder set? 👉 msft.it/63327sJ8R5

#VSLive #VisualStudio

I have a proposal.

A longer week with a disproportionately longer weekend. Giving us 4 day work week and 4 day weekend.

Reduction in the number of possible Mondays per year.

Thoughts?

A post from the archive 📫:
Validate your identity on Bluesky with your GitHub account


https://www.poppastring.com/blog/validate-your-identity-on-bluesky-with-your-github-account
#socialnetworks #musings

The #wnba has consistently outpaced the NBA in 2-point FG% since the mid-2010s, due to a higher concentration of post play and mid-range efficiency (the style of play I 💗).

Exclusive: Google is indexing ChatGPT conversations, potentially exposing sensitive user data Thousands of shared ChatGPT chats are now appearing in Google search results.

Well this is bad. Google is indexing ChatGPT conversations exposing sensitive user data

I tried a few quick searches. I found someone's chat where I can see their api key

I found some building their resume. Their name, email and phone numbers are exposed.

www.fastcompany.com/91376687/goo...

Let's go mvp! #lynx #wnba

Happy Golden Girls Day 🥳

spot art. text reads: Visual Studio Live @Microsoft HQ Streaming LIVE in just 6 days! Visual Studio Toolbox broadcasts from Microsoft HQ in Redmond

🎬 The countdown begins!

In just 6 days, we’re going live from Microsoft HQ with Visual Studio Toolbox at VSLive!
Get ready for expert insights, real-world demos, and dev inspiration.

📌 Set your reminder now: msft.it/63329sJ8EJ
#VSLive #VisualStudio #DevCommunity

A post from the archive 📫:
Collecting dumps anywhere with dotnet-monitor


https://www.poppastring.com/blog/collecting-dumps-anywhere-with-dotnetmonitor
#diagnostics #debugging

Adobe releases Windows on Arm versions of Premiere Pro and After Effects You can download beta versions of Adobe’s Windows on Arm apps

Adobe has released Windows on Arm versions of Premiere Pro and After Effects. Beta versions are now available to download, but they’re missing some features. Full details 👇 www.theverge.com/news/715769/...

only A'ja can get thru L.A. traffic 😤

Aspire 9.4 is here with a CLI and interactive dashboard features - .NET Blog Aspire 9.4 is packed with new features, integrations, and improvements

🎉 A new version of Aspire just shipped, 9.4 and it's one of the biggest releases yet. Very proud of what the team was able to accomplish.

📣 Blog post - devblogs.microsoft.com/dotnet/annou...

📝 What's new - learn.microsoft.com/en-us/dotnet...

#dotnet #aspire

1. Kid or Play ("head like a rocket...")
2. Michael Jackson
3. Angela Davis
4. Bobby Brown
5. Allen Iverson
6. Whoopi Goldberg
7. Tyson Beckford
8. Indie Arie (~2018)

UK to recognise Palestinian state unless Israel agrees Gaza ceasefire Keir Starmer has faced increasing pressure from ministers over Britain’s stance on the issue

UK to recognise Palestinian state unless Israel agrees Gaza ceasefire https://on.ft.com/4fddnYL

"A little bit of dereliction in an otherwise-polished city can be glorious" Derelict buildings don't usually get left alone for long but they have a unique value in today's cities, writes Jessica Furseth.

"A little bit of dereliction in an otherwise-polished city can be glorious" www.dezeen.com/2025/07/24/d...

Dallas just have it going tonight 🤷🏿‍♂️ #wnba