#

Had the Switch 2 pre-order in my cart but kept changing my mind, gonna skip it until there’s a game I really want. Hopefully won’t be too hard to get later. 🕹️

Manton Reece

24 Apr 2025 at 05:34

AI web search

 If you haven’t been following the latest AI models closely, you may have missed what is happening with integrating web search results into answers. It used to be that you had two options:

  • Use the model’s built-in knowledge, usually with a training cut-off of a year ago. That was extremely fast but it might hallucinate when it hit the extent of its knowledge.
  • Use “deep research” to let the AI gather info from the web and compile a comprehensive report. That took 5-10 minutes and the result was overkill most of the time.

Now it’s more streamlined. I’ve been using OpenAI’s o4-mini and it seems to work something like this:

  • Ask it a question that could benefit from searching the web to supplement the model’s built-in knowledge.
  • AI figures out a handful of queries for the web and feeds the search results back into its reasoning process.
  • In some cases it might use those results to go back to the web and search for more web pages.
  • Then it uses everything it learned to produce the answer.

This process takes somewhere around 30 seconds. It’s great for asking questions about coding with recent frameworks, or really anything that changes often.

In a longer post about this, Simon Willison writes:

This turns out to be a huge deal. I’ve been throwing all kinds of questions at ChatGPT (in o3 or o4-mini mode) and getting back genuinely useful answers grounded in search results.

He also comments on the downside to replacing humans viewing web pages:

This also means that a bunch of the potential dark futures we’ve been predicting for the last couple of years are a whole lot more likely to become true. Why visit websites if you can get your answers directly from the chatbot instead?

The results are so good that I’m now asking AI for simple queries that Google would be equally good for. Using AI essentially automates the workflow of getting 10 links from Google, clicking on 3-4 of them, then skimming the web pages to get your answer.

I don’t know where all of this is going. It feels like a pretty big shift, though.

Manton Reece

22 Apr 2025 at 18:28
#

I really like this post from Ashley Willis, about doubting yourself and the fear of what loud people on the internet might think of your writing:

I don’t know exactly when that changed. There wasn’t one big moment, just a slow fade. Something dimmed. I started second-guessing myself more.

Manton Reece

22 Apr 2025 at 17:00
#

Micro.blog’s backend is running so much more smoothly now after I addressed some memory issues yesterday. I often assume I know where problems are, and it takes actually digging in to discover I was wrong, the problem is fixable in a different way than I expected.

Manton Reece

22 Apr 2025 at 16:43
#

Nailed it. I’m too amused by wall outlets and light switches gone wrong, it’s surprisingly common. This one in the hotel stairway.

An electrical outlet is installed on the edge where two walls meet with the plastic needed to be cut, surrounded by a textured gray wall.
Manton Reece

22 Apr 2025 at 01:29
#

I’ll be speaking at EFF-Austin next month! An updated talk about blogs, social networks, the fediverse, and where I think the open web is going.

Manton Reece

22 Apr 2025 at 00:00
#

Micro.blog went off the rails today, for some reason the day I’m out of town an old memory leak decided to blow up into a much worse problem. I think I’ve got it under control now.

Manton Reece

21 Apr 2025 at 21:58
#

Speaking of Objective-C, when I see example code that looks like this:

NSMutableArray<NSString *>* paths

I simplify it to:

NSMutableArray* paths

I don’t think declaring types everywhere improved the language at all. Let a dynamic language be dynamic. 🤪

Manton Reece

21 Apr 2025 at 19:44
#

Experimenting with a special build of Micro.blog for Mac with Gemma 3 (4 billion params) running inside the app. Seems a good balance of download size and RAM, allowing me to run some AI magic on device that might be cost-prohibitive or wasteful on the server.

Manton Reece

21 Apr 2025 at 15:48
#

There’s no place I appreciate Apple Pay more than at the gas station air pump, instead of using quarters. 🛞

Manton Reece

21 Apr 2025 at 13:59
#

Yair Rosenberg writing at The Atlantic about the arson at the Pennsylvania governor’s residence. The suspect was struggling with his mental health and influenced by attempts to vilify Josh Shapiro over the war in Gaza:

But those struggling with internal demons don’t originate our external ones; they reflect them. In their confusion and pain, such individuals latch on to those already targeted by the broader culture and its preexisting pathologies, showing us not who they are, but who we are.

Manton Reece

20 Apr 2025 at 16:19
#

New laptop sticker, from Vintage bookstore and wine bar. Booksellers not algorithms. 🍷

MacBook Pro with various stickers, like Terrible Coffee, Mastodon, Bluesky, Books are Magic, and others.
Manton Reece

19 Apr 2025 at 14:26
#

Thinking about AGI. The big step that is missing is personal AI being able to learn when it answers a question. So if I use deep research and my AI goes off and spends 10 minutes researching an answer, all of that should be fed back into the model for later.

Manton Reece

19 Apr 2025 at 14:07
#

Gus Mueller:

To me, Objective-C has always felt expressive and capable, doubly so when I first started using it.

After Swift became popular, I felt kind of guilty still using Objective-C so heavily, but I’m over it. Micro.blog for Mac is all Objective-C. I did two new releases this week.

Manton Reece

18 Apr 2025 at 22:42
#

Finished upgrading a server. Some things should be faster! (And some won’t be.) I’ll continue to look for places to optimize.

Manton Reece

18 Apr 2025 at 18:44
#

Coffee yesterday at Lazarus. ☕️

Iced latte with almond milk, barrels in the background, edge of MacBook Pro on the side.
Manton Reece

18 Apr 2025 at 17:39
#

Now that I’m using Hetzner in the EU, I’m having difficulty understanding Linode pricing. For example, dedicated 16 CPUs on both hosts:

Linode: $288
Hetzner: $110

This is a massive difference. Is Linode that much better? I feel like a fool for paying this.

Manton Reece

18 Apr 2025 at 16:52
#

Micro.blog for Mac continues to improve. Just released version 3.5. A new feature I like in this release: paste a photo from the clipboard directly into the Uploads section to upload it.

Also added a Preview button. I use this with command-shift-P but I’m sure some people didn’t know it was there.

Screenshot of Mac windows showing preview button.
Manton Reece

18 Apr 2025 at 16:24
#

I’ve switched my coding questions over to o4-mini. It’s very good and fast enough.

At this point, for me personally, not using AI for coding help would be like not using Stack Overflow or Google. I could go back to the 1990s when I had a printed reference open in front of me while coding, but why?

Manton Reece

18 Apr 2025 at 14:35
#

The Sam Altman interview at TED is worth a watch. Awkward and full of tension. But some good thoughts in it, probably the most I’ve seen Sam pressed on the big issues.

Manton Reece

18 Apr 2025 at 14:30
#

Didn’t sleep enough. Nightmare that I was arguing with the Subaru mechanics. Too close to reality. 🚙

Manton Reece

18 Apr 2025 at 14:15
#

Sad to hear there were two deaths in the FSU shooting. When news broke that it was a handgun, it gave me a small bit of hope that it wouldn’t be as terrible as it could’ve been. Imagine an AR-15 instead. That is why they should be banned.

Manton Reece

18 Apr 2025 at 04:16
#

Great post from The Fediverse Report about Bluesky complying with requests from the Turkish government and how a labeling service can be used to hide accounts:

Such a moderation layer allows Bluesky (the app) to apply moderation decisions that are only experienced by people currently geolocated in a specific country, more on that below. A few days ago, the Turkish moderation labeler became active, and the labeler started hiding accounts, making the accounts invisible for people in Turkey.

Manton Reece

17 Apr 2025 at 21:02
#

Taria & Como looks like a clever, crank-friendly platformer, part of the upcoming season 2 on Playdate. There’s a trailer on YouTube.

Manton Reece

17 Apr 2025 at 19:51
#

The Internet Archive is asking for help to defend against lawsuits:

A coalition of major record labels has filed a lawsuit against the Internet Archive—demanding $700 million for our work preserving and providing access to historical 78rpm records. These fragile, obsolete discs hold some of the earliest recordings of a vanishing American culture.

The Internet Archive is a unique force for good. Sad that they have to waste any time on this.

Manton Reece

17 Apr 2025 at 19:33
#

Matt Mullenweg blogs about the WordPress 6.8 release, how WordPress might use AI in the future, and the experience of being deposed at the WP Engine trial:

I really appreciated the due process and decorum of the rule of law, and just like code, law has a million little quirks, global variables, loaded libraries, and esoteric terminology. But wow, after a full day of that, I’m mentally exhausted.

Manton Reece

17 Apr 2025 at 17:30
#

I like the way Tony Stubblebine is running Medium. On sticking with diversity, equity, and inclusion:

Medium was built by and is run by a diverse group of people. This diversity is a raise-the-bar strategy. As a CEO, I feel confident that embracing diversity as a strategy increases the business, cultural, and intellectual capabilities of our company.

Unfortunately the acronym DEI has been poisoned. I think it’s fine to abandon the letters, while keeping to the principles.

Manton Reece

17 Apr 2025 at 16:50
#

Some interesting data here in the MIT Technology Review about energy use for AI data centers:

Electricity demand is on the rise from a whole host of sources: Electric vehicles, air-conditioning, and appliances will each drive more electricity demand than data centers between now and the end of the decade. In total, data centers make up a little over 8% of electricity demand expected between now and 2030.

There is so much noise around AI that it’s increasingly difficult to tell what is misinformation or just outdated.

Manton Reece

17 Apr 2025 at 16:37
#

Dave Winer has more thoughts on the “Twitterlike is a bad shape” post:

Each decision we make in developing our means of discourse shapes the discourse. And with the character limit and the inability to edit, and the incentives are all wrong (I can tap into your follower flow without your permision just by posting a reply) and makes almost all discourse spam or abusive or both. I’m planning a different structure for discourse in the World of WordLand.

Manton Reece

17 Apr 2025 at 16:01
#

I’m not a Gmail user or Notion user, but gotta say the UI and interactions in Notion Mail look quite good. I’ve watched a few videos for inspiration.

Manton Reece

17 Apr 2025 at 15:34

Twitterlike vs. Micro.blog

 This is a great post outlining many of the problems with Twitter-inspired social networks:

Twitter and its imitators have adopted a structural design that is fundamentally bad for people. This isn’t just a matter of who’s in charge; it’s a problem with the thing itself.

I’ve been saying this for quite some time. The design and features of social apps influence how we interact with it and the behavior it encourages. My book even had a chapter called UI impacts behavior.

The “Twitterlike” post is so well structured that I thought it would be interesting to compare it to Micro.blog, which leaves out some features on purpose.

Twitterlike Micro.blog
“Tight Character Limits” No limit. Defaults to short posts but can grow to any size.
“Anti-Link Culture” Links encouraged. Markdown for inline-links within text.
“No Edit Button” Blog posts can always be edited. Replies can be edited within 24 hours.
“Share Additions = Bad Shares” No built-in reposts. Can embed posts or use Markdown block quotes.
“Trending Feeds” No trends.
“Decontextualized Encounters” Can still be a problem in Discover, which is why we hand curate it and avoid divisive topics.
“No Host Veto” Partial support. Can hide replies on your own blog, but not in timeline yet.
“Inline Tags” No special support for hashtags.

The main argument against “Twitterlike” aligns very well with Micro.blog’s philosophy. There’s still more work to do.

Manton Reece

17 Apr 2025 at 15:13



Refresh complete

ReloadX
Home
(126) All feeds

Last 24 hours
Download OPML
A Very Good Blog by Keenan
A Working Library
Alastair Johnston
Anna Havron
Annie
Annie Mueller
Apple Annie's Weblog
*
Articles – Dan Q
*
Baty.net posts
bgfay
*
Bix Dot Blog
Brandon's Journal
*
Chris Coyier
Chris Lovie-Tyler
Chris McLeod's blog
*
Colin Devroe
Colin Walker – Daily Feed
*
Content on Kwon.nyc
Crazy Stupid Tech
daverupert.com
Dino's Journal 📖
dispatches
*
dominikhofer dot me
Excursions
Flashing Palely in the Margins
Floating Flinders
For You
*
Frank Meeuwsen
frittiert.es
Hello! on Alan Ralph
Human Stuff from Lisa Olivera
inessential.com
*
jabel
Jake LaCaze
*
James Van Dyne
*
Jan-Lukas Else
*
Jim Nielsen's Blog
Jo's Blog
*
Kev Quirk
lili's musings
*
Live & Learn
Lucy Bellwood
Maggie Appleton
*
Manton Reece
*
Manu's Feed
*
maya.land
Meadow
*
Minutes to Midnight RSS feed
Nicky's Blog
*
Notes – Dan Q
On my Om
Own Your Web
Paul's Dev Notes
*
QC RSS
rebeccatoh.co
reverie v. reality
*
Rhoneisms
ribbonfarm
Robert Birming
*
Robert Birming
Robin Rendle
Robin Rendle
Sara Joy
*
Scripting News for email
Sentiers – Blog
*
Simon Collison | Articles & Stream
strandlines
Tangible Life
the dream machine
*
The Torment Nexus
*
thejaymo
theunderground.blog
Thoughtless Ramblings
tomcritchlow.com
*
Tracy Durnell
*
Winnie Lim
*
yours, tiramisu
Žan Černe's Blog

About Reader


Reader is a public/private RSS & Atom feed reader.


The page is publicly available but all admin and post actions are gated behind login checks. Anyone is welcome to come and have a look at what feeds are listed — the posts visible will be everything within the last week and be unaffected by my read/unread status.


Reader currently updates every six hours.


Close

Search




x
Colin Walker Colin Walker colin@colinwalker.blog