I have a busy day ahead of me, but not too busy for Pook-Emu Bee links. Before getting to links from around the web, I note that I just published the sixth and final part of my New Leaf Journal sixth birthday project, Be the Blogger You Want to Be (Or Read). Now watch me follow my own advice.

Photograph of the East River and Manhattan Skyline from the Brooklyn Heights Promenade, with Brooklyn Bridge Park in the foreground. The sky is foggy. Off in the distance, you can vaguely see a Portuguese Flag attached to a ship docked at the South Street Seaport.
I took this photo from the Brooklyn Heights Promenade on the morning of May 27, 2026. If you squint, you can see a Portuguese Flag attached to a ship by the seaport across the river.

1. Government says 5.7 trillion yen of Japanese media was pirated last year, thinks AI could help (Casey Baseel for SoraNews24. May 26, 2026.)

I assumed by “AI could help” that the Japanese government was thinking that AI could help track down copyright infringement. Not so: “However, one of the ideas being floated by the researchers is to promote the use of AI in production to lower costs and secure profits.” Well… depending on the uptake, I suppose that could also reduce piracy.

2. Magnifica Humanitas (Peter’s Path. May 26, 2026.)

It’s on my reading list, and perhaps my future article list.

3. Zubac remembers the shock of his first NBA paycheck being cut in half (Bruno Feliks for Basketball Network. February 9, 2026.)

A star NBA center on how he learned about California tax rates.

4. Freedom 250 Announces Free FIFA World Cup Fan Zone on National Mall (Jasmyn Jordan for Breitbart. May 20, 2026.)

This isn’t what General Washington fought for.

5. Linux On Android Provides Inexpensive, Powerful Computing (Aaron Beckendorf for Hackaday. May 26, 2026.)

I should really mess around with Termux at some point.

6. Beloved NYC bakery faced stomach-churning issue before suddenly closing (Nicole Rosenthal for the New York Post. May 24, 2026.)

I’m glad I didn’t buy anything from there before it closed (it was right by my favorite spot for buying coffee beans).

7. DuckDuckGo installs are up 30% as users reject being ‘force-fed’ Google’s AI Search (Rebecca Bellan for TechCrunch. May 26, 2026.)

Google’s AI search push is working very well for me so far. I have not seen a dip in Google traffic to NLJ, but I have seen a big uptick in the last week and a half in DuckDuckGo traffic. While Google is my biggest referrer by a good margin, DuckDuckGo has always been second and clearly sends much more traffic relative to the size of its user-base.

I was too busy yesterday to put together my daily Pook-Emu Bee links. I am busy today as well, but not too busy to take a link break.

Article Links

1. Six search engines worth trying now that Google isn’t really Google anymore (Amanda Siberlin for TechCrunch. May 21, 2026.)

This is a fair list but the article misses some nuance about the sources used by Startpage and Ecosia. I would add Mojeek to the selection.

2. Prosecutor charged for allegedly emailing self sealed Jack Smith Trump report veiled as cake recipes (Justin Dunleavy for Just the News. May 20, 2026.)

This scheme sounds like a lemon cake.

3. Jeff Bezos blasts Mamdani for casting billionaire Ken Griffin as a ‘villain’ in Hizzoner’s viral video filmed outside NYC penthouse (Carl Campanile and Anna Young for the New York Post. May 20, 2026.)

Mr. Bezos is not wrong. While he is on the subject of being right, can he right the wrong that is the Amazon chatbot on store pages? It evolved from Rufus to Alexa and I think it went from war crime to super war crime.

4. A before-and-after look at downtown Brooklyn’s new streetscape (Nate Berg for Fast Company. December 9, 2025.)

They did a nice job. But it is still Downtown Brooklyn. You can only do so much with all the traffic and ugly apartment and office towers.

5. Delivery Robots Take Over Chicago Sidewalks, Sparking Debate And A Petition To Hit ‘Pause’ (Alec Karam for Block Club Chicago. December 8, 2025.)

I’d welcome them to Brooklyn on the condition that they take over the bike lanes while keeping bikes off the sidewalk.

6. Pocketbook Era With Koreader (tc3. September 21, 2025.)

I have a Pocketbook Color and InkPad Color with Koreader. I have almost talked myself into buying an Era for the side buttons, but it would be a bit expensive in light of the fact that I already have two working Pocketbooks. But let it be said Pocketbook comes with my recommendation for anyone who does not “need” the Kindle or Kobo stores.

7. Firefox Should Let Users Disable Searching from the URL Bar (Terence Towles Canote at A Shroud of Thoughts. May 20, 2026.)

I enable the search box on Firefox and agree with this take. But with that being said, I have reasons for usually searching from the address bar.

Videos

Natsu Basho (May Grand Sumo Tournament) Day 10: Atamifuji vs Wakatakage (May 19, 2026.)

I won’t spoil the outcome other than saying it was a big lift.

We are having a second consecutive 90+ degree day here in Brooklyn today’s Pook-Emu Bee links are warmer than the outside temperatures.

Photograph of an LED clock screen showing 35 minutes past the hour, indoor temperature of 84, and humidity of 51.
According to the Marathon atomic clock by window, it is 84.4 degrees F there (which is about 6 feet from my desk) and 51% humidity.

Links

1. Google Search as you know it is over (Sarah Perez for TechCrunch. May 19, 2026.)

The AIfication of Google Search is concerning but NLJ has been hitting new highs in Google Search despite my disfavoring AI cawlers via .htaccess and robots, so we’ll see if it has any effect on my numbers.

2. Japanese indie gay bar management sim Sodomy Fatale will have you serving blood cocktails to flamboyant homosexual vampires (Dorde P for Automation West. May 19, 2026.)

That title is a little bit more attention-grabbing than the one Victor and I gave our visual novel back in January-February.

3. Don’t let the old man in (Robert Birming. May 19, 2026.)

Too late. I already let myself in.

4. Fedora Retiring Its Deepin Desktop Packages (Michael Larabel for Phoronix. May 19, 2026.)

I was never tempted to try Deepin. That was apparently the right call.

5. Firefox 151 brings PDF merging, local profile backup, document PiP, and enhanced privacy (Paul for AlternativeTo. May 19, 2026.)

Local profile backup and the PDF features sound useful.

6. Matter of I-B-M-S-, 29 I&N Dec. 628 (BIA 2026) (Board of Immigration Appeals. May 19, 2026.)

Interesting Board of Immigration Appeals decision on adjudicating change of venue requests. The logic should be applicable to other motions. I have a feeling the Immigration Judge who issued decision below will be (or already is) looking for new employment.

7. LaGuardia Airport adds hologram ‘concierge’ Bridget in latest spooky invasion of AI (Brooke Steinberg for the New York Post. May 19, 2026.)

That’s not a hologram. That’s an AI generated woman on a screen. For real holograms, go read my review of A Midsummer Day’s Resonance.

Video

Natso Basho Day 8: Tobizaru vs Ura

I have been following the Natsu Basho grand sumo tournament (I’m up to date on day 9 of 11). This fight between Tobizaru and Ura on day 8 was terrific. Watch and enjoy. I watched the January and March tournaments as well and recalled the announcers discussing how popular Ura is (it was also obvious from the crowd). Ura struggled in January and March (maybe injuries?), but I got he idea. He has been doing well in May and is consistently producing exciting matches. I understand why he’s popular. Add me to the Ura fan club.

Replied to Testing suggests Google's AI Overviews tell millions of lies per hour by Ryan Whitwam (Ars Technica)
A new analysis from The New York Times attempted to assess the accuracy of [Google Search's] AI Overviews, finding it’s right 90 percent of the time. The flip side is that 1 in 10 AI answers is wrong, and for Google, that means hundreds of thousands of lies going out every minute of the day.

Ars Technica published an article by Ryan Whitwam titled Testing suggests Google’s AI Overviews tell millions of lies per hour. The article itself is about a New York Times investigation into the accuracy of Google’s AI overviews. The framing in the headline matches the first paragraph: “…1 in 10 AI answers is wrong, and for Google, that means hundreds of thousands of lies going out every minute of the day” (emphasis added).

While I am not a fan of “AI” replacing traditional search and have even published information about creating custom search engine shortcuts for DuckDuckGo’s No AI mode and Google Search, I am also not a fan of Mr. Whitwam‘s framing. Google’s generative AI implementation does not tell lies. It has no concept of truth or falsity. It returns purportedly responsive text to specific queries in accordance with its programming and index. It no more “lies” when the text is inaccurate than it “tells the truth” when the text is accurate. Assigning agency to agents confuses the issues.

(I will add as a separate issue that for those of us who support “traditional” search, framing the debate in terms of how accurate the AI results are is the wrong approach. Google and other traditional search engines often surface articles and posts with inaccurate information. As I noted in my prior writing, there are reasons for supporting traditional search and human authorship that go beyond assessing whether chatbots or the highest trafficked websites are more likely to spit out incorrect information.)

Quoted Just Want Links in Google Search Results Instead of AI Overviews? Here's How to Do It by Peter Butler (CNET)
To add a 'verbatim' restriction to your Web filter shortcut, append the attribute "&tbs=li:1" to the end of your custom search URL: https://www.google.com/search?q=%s&udm=14&tbs=li:1

I read read a good article by Hana Lee Goldin at Card Catalog titled Google Has a Secret Reference Desk. Here’s How to Use It. In the post, she ran through several tips for using Google Search effectively. I very seldom use Google Search proper, usually opting for alternatives. But I studied her post for ideas I can use in my research (Google does stand apart from most of the “alternatives” in  its breadth of search operators). One of her tips was to use Google Search’s Verbatim mode:

Google stops paraphrasing you entirely and returns results for exactly what you typed, stripped of personalization and synonym-swapping. It’s one of the most useful things Google has buried several clicks deep, and the fact that it takes three clicks to reach says something about how much Google wants you to find it.

This gave me an idea. I like writing about custom search engine shortcuts. What about an article on a shortcut for Verbatim mode? Alas, when I studied the Google Search URLs, I could not immediately ascertain what the shortcut would be. When I searched for an answer, I came across a comprehensive guide by Peter Butler for CNET. Thus, instead of writing my own Verbatim mode shortcut guide, I will cite to Mr. Butler’s as part of my learning in 2026 series.

First, Mr. Butler shared the Google Search shortcut for web search (excluding AI, instant answers, and other gimmicks):

The critical URL that you need to remember is https://www.google.com/search?q=%s&udm=14. The “%s” represents your search query, and the “udm=14” tag limits your results to the new Web filter.

For those who are interested, I wrote about how to create a custom search engine shortcut for the “no AI” version of DuckDuckGo.

Next, Mr. Butler shared how to add Verbatim mode to the web search shortcut:

To add a “verbatim” restriction to your Web filter shortcut, append the attribute “&tbs=li:1” to the end of your custom search URL: https://www.google.com/search?q=%s&udm=14&tbs=li:1

For those who want to use Verbatim mode without limiting Google to web search mode, I quote from Alex Barrientos’ guide on Gadget Review.

Create a custom search engine in your browser settings … with the URL
https://www.google.com/search?q=%s&oq=%s&tbs=li:1

Replied to Why I ditched Accelerated Mobile Pages (AMP) for my blog by Maddy MillerMaddy Miller (madelinemiller.dev)
Accelerated Mobile Pages, or AMP, is a web technology championed by Google to improve the browsing experience on mobile devices. Throughout this article, I'll outline a few of the issues that led me to ditch AMP.

I came across a 2021 blog post by Maddy Miller on dropping Accelerated Mobile Pages (“AMP”) support on her blog. Back when I started running The New Leaf Journal, I had little idea what I was doing and only vaguely knew what AMP was (note that my phone at that time was a BlackBerry Classic). The WordPress theme I implemented in June 2020 (and still use today) has built-in AMP support. I vaguely recall having researched into the matter before deciding against enabling AMP. Unfortunately, I did not keep a record of my internal deliberations. My decision turned out to be correct since Google stopped giving special favors to AMP pages in 2021 (see Maddy Miller’s article for a deeper discussion) and uptake has gone down over the years. I did however go through some pain in removing Yandex Turbo support (which I believe is similar to AMP) after having enabled it without fully understanding how it worked (I am glad I avoided the issue with AMP).

Replied to Search engine shortcuts in Firefox by James (jamesg.blog)
Firefox has a feature that lets you map a character or sequence of characters typed in the Firefox address bar to a custom search engine. For example, you can map w to Wikipedia, so the query w coffee would take you to the Wikipedia search page for coffee ...

James at James’ Coffee Blog published a good post in 2024 titled Search engine shortcuts in Firefox, which regular New Leaf Journal readers will know is a topic near and dear to my heart. James demonstrated how to add search engine shortcuts to Firefox using the search box (which, readers should note, is distinct from running searches in the omnibar address bar (HT to yoasif for the correction; see source in Firefox docs) and provided a nice tutorial on how to assign and use shortcuts to custom search engines. I strongly encourage people to take advantage of custom search engine functionality. It can be put to creative use. See, for example, the following articles on the subject I have published on NLJ:

For fellow Firefox wielders out there, I recommend using Soufiane Sakhi’s Search Engines Helper extension for adding novel search engines (e.g., anything other than a search engine’s website’s default search). As I note in the articles, Chromium-based browsers make adding custom search engines easier and more intuitive than the Firefox family.

Replied to Google De-Indexed My Entire Bear Blog and I Don’t Know Why by James Zhan (James Zhan in real life.)
A month after I started my first Bear blog at blog.james-zhan.com, my blog was entirely de-indexed by Google for no apparent reason.

A month after I started my first Bear blog at , my blog was entirely de-indexed by Google for no apparent reason:



I have since migrated to (you ar...

I read a November 7, 2025 post by James Zhan about his blog (hosted on Bear Blog) being de-indexed by Google. As a threshold matter, I note that Mr. Zhan has a nice-looking and well-structured personal blog. He describes discovering on October 14, 2025, that his blog, which had previously been indexed by Google, had been de-indexed. He tried various forms of troubleshooting to resolve the issue, but it was not resolved as of December 7. Mr. Zhan did helpfully note that his blog remained indexed by Bing and Brave Search, suggesting that the issue was localized to Google (I will add it is good to see that being de-indexed by Google does not affect Brave).

Long-time readers of my work may recall that I went through a similarly annoying experience when Bing de-indexed The New Leaf Journal in January 2023 (see my initial report). Bing only began fixing the situation in July of that year and never provided an explanation for what happened. I came to suspect that the reason why Bing addressed my multiple emails at all was because I created a GitHub repository collecting Bing de-indexing stories and shared by GitHub repository to Hacker News (I received my first human contact from Bing within 24 hours of sharing my repo to HN). For whatever it is worth, I suppose that I was fortunate to only be de-indexed by Bing instead of Google. The Bing indexing hurt us more because it removed us from DuckDuckGo (which is effectively a Bing front-end) than Bing itself. DuckDuckGo is our second-biggest search referrer, but it typically comes in at 8-10% of our Google referrals. (Note this blog receives very little search traffic, and most of it is from the Bing family.)

Regarding Mr. Zhan’s case, I would not have recommended his decision to copy is blog to a new sub-domain (I doubt that will be the solution to the problem), although I approve of changing his domain registrar from GoDaddy to Porkbun (I am a Porkbun fan). The big tech search engines are fickle things, and the quality of Mr. Zhan’s blog in conjunction with the fact that neither Bing nor Brave de-indexed it suggests the problem was (and perhaps is) Google being Google. There is probably no good reason for the de-indexing and Google will never offer a reason.

Because I enjoyed the post, I submitted it to Hacker News (it probably worked for me and is known to have worked for others with respect to Bing). I hope his blog is re-indexed soon (if it has not been already).

AlternativeTo reports that Mullvad VPN is shutting down Leta, its soon-to-be-former privacy-centric Google Search proxy, on November 27, 2025. I have been a Mullvad customer since 2021 of 2022. I recall noting when Leta was launched and trying it, but I was never too impressed with how it displayed results (setting aside the whole issue of using Mullvad VPN credentials to “log in” to a search proxy. Instead, I followed the advice Mullvad is now giving Leta devotees:

For those seeking comparable levels of privacy, Mullvad recommends combining a virtual private network (VPN) with a privacy-focused browser as an alternative solution. According to Mullvad, this approach can deliver similar privacy protections by masking user IP addresses and reducing tracking across the web.

That sounds about right. Moreover, while I have not written about search in a while, I still recommend keeping a healthy diet of search indexes.

I read an interesting Hacker News comment by user dumbfounder (great username!) on Hacker News today. I excerpt the pertinent part below:

I created a search engine that crawled the web way back in 2003. I used a proper user agent that included my email address. I got SO many angry emails about my crawler, which played as nice as I was able to make it play. Which was pretty nice I believe. If it’s not Google people didn’t want it. That’s a good way to prevent anyone from ever competing with Google.

dumbfounder

I have had problems with bad crawlers (especially bad AI cralwers) on my sites. At the same time, dumbfounder highlights the reverse side of the coin. Many sites block good crawlers such as robots.txt-respecting crawlers for indepdent search engines. While all webmasters are free to control access to their sites as they see fit, allowing Google and other select big tech search crawlers while excluding small and independent search crawlers both limits search diversity and prevents people who may rely on small or niche search engines such as Mojeek, Marginalia, or Seznam from discovering potentially interesting writing. I previously published an article on this issue advocating for webmasters who want to support an open web and search engine diversity to ensure that good crawlers from independent search engines and directories can access their sites.

Replied to Google vs HN at the NLJ (1) by Nicholas A. FerrellNicholas A. Ferrell (The Emu Café Social)
As of the evening of January 18, 2024, we are on pace to have more Google referrals this month (according to Koko Analytics, which I also use here) than Hacker News referrals.

I have been busy with New Leaf Journal and law work as of late, but I return to The Emu Café Social with a fun New Leaf Journal update. As of the afternoon of January 29, 2023, we have more Google referrals at The New Leaf Journal than Hacker News referrals despite our fairly high Hacker News page 1 appearance earlier in the month. It took a record Google month to do it, but is another sign things are looking up in terms of New Leaf Journal notoriety after struggling for much of 2023 with our arbitrary Bing ban.

Over at The New Leaf Journal, my January 5, 2024 article, which was about drug-enhanced professional cycling results, became the sixth New Leaf Journal article (going back to February 2021) to appear on Hacker News page 1 (only after a stop in The Browser newsletter). It reached as high as 11th on HN and was our fourth-strongest Hacker News article of the six that made page one. The page 1 run dissipated, but it left behind one encouraging point about NLJ’s notoriety. As of the evening of January 18, 2024, we are on pace to have more Google referrals this month (according to Koko Analytics, which I also use here) than Hacker News referrals. It will take a record Google month for us to do it, but we should make it by the 29th or 30th. While I am not a Google fan and only use Google Search via Startpage, I am a proponent on making it possible for people to find writing of interest with whatever search engine or front-end they prefer.

Favorited marcinzm (Hacker News)
The money from showing a pancake ad make them do that. It's not a search problem. They know it's not a good experience but they also know it makes them more money than not showing it. The goal is money. Your experience is merely a means to that end.

I starred this very insightful comment by Hacker News user marcinzm about search engines. The commenter noted the obvious, that big tech search shows ads because its purpose is to generate revenue through ads. But I appreciated the commenters putting it in terms of means and ends. Ideally, the user wants a search engine to be responsive to his or her query. That is, entering the query is the means for achieving the end of finding a useful result. However, commenter marcinzm noted that the end for the search provider is money and the search experience is a means toward creating more ad revenue. While this is not a novel point, it is phrased well here. One good anecdote is site-specific search, but this is an area that needs more work.