<?xml version='1.0' encoding='UTF-8' ?>
    <rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
      <channel>
        <title>Things Of Interest</title>
        <link>https://qntm.org/</link>
        <atom:link
          href="https://qntm.org/rss.php"
          rel="self"
          type="application/rss+xml"
        />
        <description>a website</description>
                  <item>
            <guid>699</guid>
            <title>Some consumer info</title>
            <link>https://qntm.org/consumer</link>
            <description><![CDATA[<b><a href="https://qntm.org/consumer">There Is No Antimemetics Division</a> »</b> Okay, so: if you see a book for sale with this cover, there are some things you should know before you buy it:


it is nothing to do with me, I didn't publish it
if you pay money for it, none of that money will go to me
this is V1, the old self-published edition of the story, which you can just read here, for free. It is not V2, the award-nominated new edition which you are (probably) actually trying to buy




The book above is published by Paladin Press. I don't know anything about them. My English-language publisher for There Is No Antimemetics Division is Penguin Random House. Covers for the official Penguin editions look like the ones below (US &amp; Canada on the left, UK and rest of world on the right). Buy these!










Paperback editions

In the UK and the rest of the world, the official paperback edition of There Is No Antimemetics Division will be released on 1 October 2026. It's available to pre-order now, and the cover looks like this:



We a...]]></description>
            <pubDate>Mon, 08 Jun 2026 12:45:13 +0100</pubDate>
          </item>
                  <item>
            <guid>693</guid>
            <title>There Is No Antimemetics Division by qntm– Conversation Tree Press</title>
            <link>https://conversationtreepress.com/pages/antimemetics-division-qntm</link>
            <description><![CDATA[<b><a href="https://qntm.org/antimemetics">There Is No Antimemetics Division</a> »</b>

There Is No Antimemetics Division will be published in a single state as a limited edition signed by the author and artist. It will be offered for sale when copies are ready to ship and feature:

Offset printing on premium paper.
Thoughtful interior design and typography.
Newly commissioned illustrations.
A durable, Smyth-sewn binding with a ribbon marker and head and tail bands.
A foil stamped cloth binding similar to our Collector’s Editions.
A slipcase at no additional cost.

...]]></description>
            <pubDate>Sat, 21 Feb 2026 10:43:25 +0000</pubDate>
          </item>
                  <item>
            <guid>691</guid>
            <title>qntm's leap second simulator</title>
            <link>https://qntm.org/files/simulator/index.html</link>
            <description><![CDATA[<b><a href="https://qntm.org/src">Code</a> »</b>
qntm's leap second simulator

During a leap second, Unix time cannot represent 23:59:60. So what does happen?
]]></description>
            <pubDate>Sat, 21 Feb 2026 10:34:16 +0000</pubDate>
          </item>
                  <item>
            <guid>690</guid>
            <title>How many valid JSON byte sequences are there?</title>
            <link>https://qntm.org/jsonutf</link>
            <description><![CDATA[<b><a href="https://qntm.org/jsonutf">Code</a> »</b> This essay builds on results from my previous essays "How many valid JSON strings are there?" and "How many valid UTF-8 (or UTF-16, or UTF-32) byte sequences are there?".
In that first essay I kind of dodged the most interesting part of the question by treating a JSON string as simply a sequence of Unicode characters, deliberately ignoring the problem of how the string was encoded on the wire. The task was still pretty interesting anyway. But now that we have the second essay in place we can combine the two.
There are three Unicode encodings here which I care about: UTF-8, UTF-16 and UTF-32. UTF-8 is the most important encoding but I'm going to start with UTF-32 because it's much easier, and then ramp up in complexity.

How many valid UTF-32 JSON byte sequences are there?

The UTF-32 encoding simply encodes each Unicode character using 4 bytes. The only valid UTF-32 byte sequences have length 0, 4, 8, 12 etc. bytes &mdash; there are no others. So it makes the most sense to ask, f...]]></description>
            <pubDate>Sun, 23 Nov 2025 23:59:07 +0000</pubDate>
          </item>
                  <item>
            <guid>687</guid>
            <title>There Is No Antimemetics Division FAQ</title>
            <link>https://qntm.org/antifaq</link>
            <description><![CDATA[<b><a href="https://qntm.org/antifaq">There Is No Antimemetics Division</a> »</b> "What is different between the original self-published version of Antimemetics (V1) and the new traditionally published version (V2)?"

The book has received a total end-to-end editorial overhaul. The nature of the web serial format (which was used for V1) is that there are always things you're going to look back on and wish you had an opportunity to do over &mdash; to lay better setup for later payoff, to strengthen character arcs and plots, to add more consistent themes and pacing, to fix basic problems, to pick different words. And, in this particular case, to turn what was almost like a series of loosely connected stories, which read like an anthology at times, into a proper singular narrative, a novel. I had plenty of ideas of my own for the rewrite but my agent had many very valuable suggestions of her own, and my editors at Penguin also had plenty to add.

The backbone of the story is essentially the same but nearly every sentence in the book has been adjusted a little. The ...]]></description>
            <pubDate>Tue, 18 Nov 2025 00:06:10 +0000</pubDate>
          </item>
                  <item>
            <guid>683</guid>
            <title>How many valid UTF-8 (or UTF-16, or UTF-32) byte sequences are there?</title>
            <link>https://qntm.org/utf</link>
            <description><![CDATA[<b><a href="https://qntm.org/utf">Code</a> »</b> Quick simple one.
A Unicode character is... well, for the purposes of this enumeration, it is an integer from 0 to 1,114,111 inclusive, which we express in hexadecimal as U+0000 through U+10FFFF. The interpretations and interactions of these characters are extremely complex, but we can actually ignore these here. In particular, I'm going to ignore issues relating to surrogates.
There are several ways to encode Unicode characters as bytes (octets). Three such encodings are UTF-8, UTF-16 and UTF-32. Let's start with the most interesting and popular of these:

UTF-8

In UTF-8:


  Characters U+0000 to U+007F inclusive are encoded using 1 byte - there are 128 of these 1-byte sequences.
  Characters U+0080 to U+07FF inclusive are encoded using 2 bytes - there are 1,920 of these 2-byte sequences.
  Characters U+0800 to U+FFFF inclusive are encoded using 3 bytes - there are 63,488 of these 3-byte sequences.
  Characters U+10000 to U+10FFFF inclusive are encoded using 4 bytes - the...]]></description>
            <pubDate>Sat, 11 Oct 2025 14:33:50 +0100</pubDate>
          </item>
                  <item>
            <guid>680</guid>
            <title>How many valid JSON strings are there?</title>
            <link>https://qntm.org/jsoncount</link>
            <description><![CDATA[<b><a href="https://qntm.org/jsoncount">Code</a> »</b> A valid JSON string is a finite series of characters, each character chosen from an alphabet of 1,114,112 possible characters, which additionally conforms to the simple grammar described at json.org. We number the characters in hexadecimal from 0x0000 to 0x10FFFF inclusive.

These characters originate in Unicode, but aside from serving as a basic lookup table - character 0x0020 is a space, for example, and "x" is 0x0078 - Unicode itself is not actually massively important to this question. The JSON grammar doesn't say anything about the use or misuse of reserved Unicode characters, the proper matching of surrogate pairs or anything like that. This makes our job a lot easier, though it's still complex enough to be entertaining.

The various possible binary encodings of a Unicode string, such as UTF-8, are also not considered here. String length will be measured in characters, not in bytes. Again, this makes the job a lot easier.

There are 1114112N possible strings of length N ≥...]]></description>
            <pubDate>Tue, 23 Sep 2025 22:57:48 +0100</pubDate>
          </item>
                  <item>
            <guid>677</guid>
            <title>ChatSCP</title>
            <link>https://qntm.org/chatscp</link>
            <description><![CDATA[<b><a href="https://qntm.org/chatscp">Blog</a> »</b> Prominent OpenAI invester Geoff Lewis may be facing a conspiracy theory-related mental health episode brought about by extended conversations with ChatGPT.
I found this article to be very interesting. From my perspective, the most interesting angle here is the fact that ChatGPT is obviously just spinning SCP-esque containment documentation at Lewis.
The SCP project has had problems with readers taking its content too seriously since its inception. The very first question in the project's FAQ is,

Is SCP real?

and the answer to that question is,

No.

It's just that, historically, the people who suffer from this kind of problem, of being unable to discern authentic reality from fiction, have been, you know. Teenagers. Or at least, that's what I've kind of assumed this whole time.
I do not feel comfortable judging Geoff Lewis' mental state from where I'm sitting. Joe Wilkins, the journalist who originally wrote that article, feels similarly:

It's a very delicate thing to t...]]></description>
            <pubDate>Mon, 21 Jul 2025 18:57:39 +0100</pubDate>
          </item>
                  <item>
            <guid>674</guid>
            <title>HATETRIS world record history</title>
            <link>https://qntm.org/records</link>
            <description><![CDATA[<b><a href="https://qntm.org/records">HATETRIS</a> »</b> This page presents every known record in the history of HATETRIS. The game was originally released in April 2010. My personal best score at the time I released the game to the public was approximately 5 lines.

2010-04-04
Atypical's 11-line run is:

ϥقໂɝƐඖДݹஶʈງƷ௨ೲໃܤѢقҾחࢲටฅڗ௨ΡІݪ௨ళȣݹࢴටງ໒௨ஶໃܥ௨റІݮ௨ఴІݥذඡଈݹƍق๓অஒॴแђञඖЅи௨sǶɔۑడПݷޠقԩݹࠉൿຟɓతණງஈশ੬෪অࠑථධٽଫ൝ଆࡨশ૫СܭߜయլݚɶऋഭܭرɤธӃస൯


2010-04-09
An anonymous guest at the Speed Demos Archive achieved 17 lines with this run:

ۑටժݹਐටดݹமsරݪƐජଈݲ௨ණໃφذගדݶಒටܨݹসටѧݹ൭ඤדݜ௧ซະਨதԀໃڻಜʈະसѻගІѠ௧ซະऄமϺเݹߤඨVܭѻඳІʅઅගتףயҔзݢऊටȝधѻೲܨݷಗචЄࡨଫඝܘɚமʈฅ๐ષ෦ฅ൩Ԥ๗ཚޡதԻѣݪॳ౾ແߢࡃశ༩ܣறඤÐњ௬ගƫঋ୦ԟȠॾಭ


2010-04-11
Same person...]]></description>
            <pubDate>Tue, 08 Apr 2025 13:40:58 +0100</pubDate>
          </item>
                  <item>
            <guid>670</guid>
            <title>There Is No Antimemetics Division</title>
            <link>https://qntm.org/antimemetics</link>
            <description><![CDATA[<b><a href="https://qntm.org/antimemetics">Fiction</a> »</b> 

On sale 11 November 2025. Preorder now.
...]]></description>
            <pubDate>Mon, 03 Mar 2025 23:54:23 +0000</pubDate>
          </item>
                  <item>
            <guid>667</guid>
            <title>That string concatenation frisson</title>
            <link>https://qntm.org/concat</link>
            <description><![CDATA[<b><a href="https://qntm.org/concat">Code</a> »</b> If I see code along the lines of:


const ref = application + ':' + endpoint


then I'm going to request that you provide a rigorous proof that neither of the strings application or endpoint can ever contain a colon. In accordance with my belief that code should be obviously correct, not provably correct, your proof will look something like this:


if (application.includes(':') || endpoint.includes(':')) {
  throw Error()
}


This is because I have been down this road too many times.

There are two major scenarios where concatenating strings together in this way is dangerous. The first is if we intend to use ref as some kind of unique identifier. A key in a hash table, maybe. In this case, we are at risk of collisions, e.g. application = 'a:b'; endpoint = 'c' collides with application = 'a'; endpoint = 'b:c' as both result in ref = 'a:b:c'.

The second is if we are trying to send two strings through some kind of field which nominally only allows a single string to be ...]]></description>
            <pubDate>Mon, 03 Feb 2025 15:58:40 +0000</pubDate>
          </item>
                  <item>
            <guid>664</guid>
            <title>Developer philosophy</title>
            <link>https://qntm.org/devphilo</link>
            <description><![CDATA[<b><a href="https://qntm.org/devphilo">Blog</a> »</b> Amazing as it may seem after all these years, there are still junior developers in the world.
A few weeks ago at work we had a talk where senior developers (including me) were invited to spend around five minutes each talking about our personal software development philosophies. The idea was for us to share our years of experience with our more junior developers.
After the session, I felt that it might be valuable to write my own thoughts up, and add a little more detail. So here we are.
This listing is a little miscellaneous; it isn't intended to be an exhaustive exploration of the way in which I develop software. Also, if you are a senior developer already then obviously you might already be familiar with some of this. Or disagree! Software development is a famously subjective field. See you in the comments.

Avoid, at all costs, arriving at a scenario where the ground-up rewrite starts to look attractive

It's generally pretty well-understood that the ground-up rewrite can be...]]></description>
            <pubDate>Mon, 03 Feb 2025 15:51:21 +0000</pubDate>
          </item>
                  <item>
            <guid>661</guid>
            <title>The Pale Blue Dot, But Time</title>
            <link>https://qntm.org/pale</link>
            <description><![CDATA[<b><a href="https://qntm.org/pale">Blog</a> »</b> 
On this page, every vertical pixel is equivalent to 100,000 years of time. The top of the page is the Big Bang. The letter "z" in this sentence is about a million years tall.
Get scrolling.

That's us.↓

↑During those 100,000 years, all of human history occurred.

This sentence is several million years after humans. We're gone now. We'll never be back. The universe continues.



...]]></description>
            <pubDate>Mon, 18 Nov 2024 14:01:55 +0000</pubDate>
          </item>
                  <item>
            <guid>658</guid>
            <title>A small adventure in WebAssembly with a mildly surprising ending</title>
            <link>https://qntm.org/wasm</link>
            <description><![CDATA[<b><a href="https://qntm.org/wasm">Code</a> »</b> We've started playing a new board game. An old board game, actually; it seems like it's originally from 1975 and has gone through a few iterations.
221B Baker Street comes with a deck of 75 unique, original Sherlock Holmes mysteries and a dense book of clues. At the start of the game, one mystery is selected at random from the deck and read out to everybody. Players then start at Baker Street on this board and roll dice to travel around the board to other locations, such as the museum, the theatre, Scotland Yard, the docks and the Boar's Head pub. There are 15 locations in all. At each location, there is a unique clue. When a player reaches the location, they open the clue book, secretly read that single clue, take notes, and then put the book back. The objective is to gather enough clues that you think you can solve the mystery... then swiftly return to Baker Street, announce your solution, and check it against the actual solution to see if you win or lose.
In my opinion, this is a ...]]></description>
            <pubDate>Sun, 17 Nov 2024 15:50:36 +0000</pubDate>
          </item>
                  <item>
            <guid>655</guid>
            <title>I have a publishing deal!</title>
            <link>https://qntm.org/publ</link>
            <description><![CDATA[<b><a href="https://qntm.org/publ">Blog</a> »</b> tl;dr: this is your last chance to buy Antimemetics V1! Less than one day remains!



In fact I have two publishing deals. Some of you may have already seen this news, and you may be able to guess where this is going.

The first deal is with Ballantine Books, a US-based publisher which is part of Penguin Random House. The second deal is with Del Rey, a UK-based publisher which is also part of Penguin Random House. Ballantine will be publishing my book There Is No Antimemetics Division in North America, and Del Rey will be publishing There Is No Antimemetics Division in the UK and most of the rest of the world. If all goes well, publication day will be in the third quarter of 2025.

This will be a fully overhauled, thoroughly rewritten version of Antimemetics. I'm calling this "version 2", and the original version "version 1". I'm still working on rewrites right now, but the draft of V2 in front of me is already (in my subjective opinion, if you value that) a mile beyond the ori...]]></description>
            <pubDate>Mon, 30 Sep 2024 10:25:06 +0100</pubDate>
          </item>
                  <item>
            <guid>652</guid>
            <title>I have a publishing deal!</title>
            <link>https://qntm.org/publ</link>
            <description><![CDATA[<b><a href="https://qntm.org/publ">Blog</a> »</b> In fact I have two publishing deals. Some of you may have already seen this news, and you may be able to guess where this is going.

The first deal is with Ballantine Books, a US-based publisher which is part of Penguin Random House. The second deal is with Del Rey, a UK-based publisher which is also part of Penguin Random House. Ballantine will be publishing my book There Is No Antimemetics Division in North America, and Del Rey will be publishing There Is No Antimemetics Division in the UK and most of the rest of the world. If all goes well, publication day will be in the third quarter of 2025.

This will be a fully overhauled, thoroughly rewritten version of Antimemetics. I'm calling this "version 2", and the original version "version 1". I'm still working on rewrites right now, but the draft of V2 in front of me is already (in my subjective opinion, if you value that) a mile beyond the original. There is plenty about V1 which I felt could be improved anyway, but it is frankly a...]]></description>
            <pubDate>Mon, 30 Sep 2024 10:25:06 +0100</pubDate>
          </item>
                  <item>
            <guid>649</guid>
            <title>I have a publishing deal!</title>
            <link>https://qntm.org/publ</link>
            <description><![CDATA[<b><a href="https://qntm.org/publ">Blog</a> »</b> In fact I have two publishing deals. Some of you may have already seen this news, and you may be able to guess where this is going.

The first deal is with Ballantine Books, a US-based publisher which is part of Penguin Random House. The second deal is with Del Rey, a UK-based publisher which is also part of Penguin Random House. Ballantine will be publishing my book There Is No Antimemetics Division in North America, and Del Rey will be publishing There Is No Antimemetics Division in the UK, the rest of the Commonwealth and most of the rest of the world. If all goes well, publication day will be in the third quarter of 2025.

This will be a fully overhauled, thoroughly rewritten version of Antimemetics. I'm calling this "version 2", and the original version "version 1". I'm still working on rewrites right now, but the draft of V2 in front of me is already (in my subjective opinion, if you value that) a mile beyond the original. There is plenty about V1 which I felt could be improv...]]></description>
            <pubDate>Thu, 19 Sep 2024 17:55:11 +0100</pubDate>
          </item>
                  <item>
            <guid>646</guid>
            <title>All I want for Christmas is a negative leap second</title>
            <link>https://qntm.org/leap</link>
            <description><![CDATA[<b><a href="https://qntm.org/leap">Blog</a> »</b> I just want to see it. Just once. I want to watch that earthquake ripple through all of global electronic timekeeping. I want to see which organisations make it to January morning with nothing on fire.



You know what a leap second is. The short version is that planet Earth is a terrible clock.

I love leap seconds. I love the unsolvable problem which birthed leap seconds, I love the technical challenge of implementing leap seconds, I love that they are rare and delightful and that they solve a problem, and I love that this solution is hugely irritating to a huge number of people who have more investment in and knowledge of time measurement than I do. It is a huge hassle to deal with leap seconds and I love that there is no universal agreement on how to deal with them. What should Unix time, for example, do during a leap second? Unix time is a simple number. There's no way to express 23:59:60. Should it stall for a second? Should it overrun for a second and then instantaneously ...]]></description>
            <pubDate>Tue, 02 Jul 2024 22:50:53 +0100</pubDate>
          </item>
                  <item>
            <guid>643</guid>
            <title>The Assist</title>
            <link>https://qntm.org/assist</link>
            <description><![CDATA[<b><a href="https://qntm.org/assist">Blog</a> »</b> At work we recently started experimenting with generative AI for assistance with programming. We have a new Visual Studio Code plugin which we can ask questions in English, and it spits back code. It was a really interesting piece of, well, mandatory training. I've formed some opinions.

The main thing I dislike about AI coding assistance is that I have to very carefully review the AI's code to make sure that it does the right thing and none of the wrong things. And I, personally, find that reviewing code is more difficult than writing equivalent code myself.

I don't know if anybody else has this experience, but I understand code better when I interact with it directly by hand. Either by bringing code into existence, or by refactoring code which was already there to do something new, or to do the same thing more effectively. I need to take things apart and put them back together. I need to play with the code.

Passively reading someone else's code in the dead form of a git diff ...]]></description>
            <pubDate>Fri, 05 Apr 2024 01:58:35 +0100</pubDate>
          </item>
                  <item>
            <guid>640</guid>
            <title>SCP-6X57: Demographic Anomaly</title>
            <link>https://qntm.org/shortfall</link>
            <description><![CDATA[<b><a href="https://qntm.org/shortfall">There Is No Antimemetics Division</a> »</b> This was written around March 2022, long after the completion of the Antimemetics tales on the SCP wiki.

This SCP essentially covers the same narrative ground as the epilogue of Antimemetics, Champions Of Nothing. Now, redundancy in storytelling is generally something to be avoided. I have eliminated content from Antimemetics because I felt it was redundant before, Haku being the primary example of this. So you might wonder why I decided to write it, after Antimemetics was completed, knowing ahead of time that it would be redundant.

The answer is that I had a feeling that this particular story beat might actually work really well as a standalone SCP. Although it shares continuity with Antimemetics, I felt that this SCP would actually read much better to someone who was unfamiliar with Antimemetics and had no context or explanation for what the SCP reveals. I hoped that the SCP would convey a kind of harsh, implacable, simple horror, in a very concise Series I or Series II style. ...]]></description>
            <pubDate>Sat, 20 May 2023 20:58:39 +0100</pubDate>
          </item>
                  <item>
            <guid>636</guid>
            <title>Loops in HATETRIS</title>
            <link>https://qntm.org/loops</link>
            <description><![CDATA[<b><a href="https://qntm.org/loops">Code</a> »</b> The HATETRIS algorithm has always been perfectly deterministic, with no "RNG". When I originally created it in April 2010, the way it worked was to examine the current well state only and use this to determine which piece to return. If the well was the same, then the piece returned was the same. This design choice had two major implications:

It meant that replays could encode the player input only, omitting the AI's piece choices. This made replays shareable and it made it impossible to "forge" a replay which sidestepped the real AI to gain more lines.
It meant that if you ever found a way to return to a previously-seen well state &mdash; not necessarily the empty well state &mdash; you could repeat that same sequence of inputs forever to gain unlimited lines.

In the earliest years of HATETRIS' existence it seemed highly unlikely that a loop existed. Over the course of April and May 2010 the record score for the game rose steadily from 11 lines by Atypical to 30 by Deasuke, wher...]]></description>
            <pubDate>Tue, 06 Dec 2022 14:05:54 +0000</pubDate>
          </item>
                  <item>
            <guid>633</guid>
            <title>Valuable Humans in Transit and Other Stories</title>
            <link>https://qntm.org/vhitaos</link>
            <description><![CDATA[<b><a href="https://qntm.org/vhitaos">Fiction</a> »</b> 
"[D]elivers a refreshing dose of existential despair at the transhuman condition." — Charles Stross

qntm has been writing science fiction for most of this millennium. His works start from elegant, deep hypotheticals and wind entire universes around them, pushing science, technology, logic and possibility to breaking point and far beyond.

This volume collects the highlights of his short fiction, including "The Difference", "I Don't Know, Timmy, Being God Is a Big Responsibility" and the acclaimed "Lena".


Buy the book

You can buy Valuable Humans in Transit and Other Stories in the following formats:



Amazon (paperback):
US
UK
DE
FR
ES
IT
NL
PL
SE
JP
CA
AU

Amazon (Kindle edition):
US
UK
DE
FR
ES
IT
NL
JP
BR
CA
MX
AU
IN


Google Play (EPUB)
Gumroad (EPUB)


Paperback and hardcover editions are forthcoming, please give me a few weeks.

What's inside?

This book contains ten short stories. If you've visited Things Of Interest bef...]]></description>
            <pubDate>Mon, 07 Nov 2022 10:53:34 +0000</pubDate>
          </item>
                  <item>
            <guid>630</guid>
            <title>HATETRIS</title>
            <link>https://qntm.org/hatetris</link>
            <description><![CDATA[<b><a href="https://qntm.org/hatetris">Code</a> »</b> Play Hate Tetris.
This is bad Tetris. It's hateful Tetris. It's Tetris according to the evil AI from "I Have No Mouth And I Must Scream".
I have to be honest, this is not an entirely original idea. However, RRRR's implementation of the concept of a game of Tetris which always gives you the worst possible pieces leaves much to be desired:

 the keyboard interface frequently doesn't work
 the conditions for failure are ambiguous and inconsistent
 the playing field is only 8 blocks wide as compared to the standard 10
 the AI is either overly generous, or stupid, and frequently does NOT provide you with the worst possible piece

(UPDATE: there is also Bastet, which can be played online here, but is also far too forgiving.)
HATETRIS was an experiment to rectify those flaws. Also, every coder has to build Tetris at least once in their life. In addition, I also have a passionate hatred of JavaScript, but I felt that that hatred was borne out of ignorance, so this was an attempt to r...]]></description>
            <pubDate>Sun, 06 Nov 2022 21:58:46 +0000</pubDate>
          </item>
                  <item>
            <guid>627</guid>
            <title>Valuable Humans in Transit and Other Stories</title>
            <link>https://qntm.org/vhitaos</link>
            <description><![CDATA[<b><a href="https://qntm.org/vhitaos">Fiction</a> »</b> 
"[D]elivers a refreshing dose of existential despair at the transhuman condition." — Charles Stross

qntm has been writing science fiction for most of this millennium. His works start from elegant, deep hypotheticals and wind entire universes around them, pushing science, technology, time and logic to breaking point and far beyond.

This volume collects the highlights of his short fiction, including "The Difference", "I Don't Know, Timmy, Being God Is a Big Responsibility" and the acclaimed "Lena".


Buy the book

You can buy Valuable Humans in Transit and Other Stories in the following formats:






Amazon (Kindle edition):
US
UK
DE
FR
ES
IT
NL
JP
BR
CA
MX
AU
IN


Google Play (EPUB)
Gumroad (EPUB)


Paperback and hardcover editions are forthcoming, please give me a few weeks.

What's inside?

This book contains ten short stories. If you've visited Things Of Interest before then you may have already read most of them here, in one form or ano...]]></description>
            <pubDate>Tue, 01 Nov 2022 16:55:39 +0000</pubDate>
          </item>
                  <item>
            <guid>624</guid>
            <title>Mirror's Edge vs. Catalyst</title>
            <link>https://qntm.org/catalyst</link>
            <description><![CDATA[<b><a href="https://qntm.org/catalyst">Blog</a> »</b> 

Mirror's Edge came out for the PlayStation 3 in 2008 and it is literally the only PlayStation 3 game I own. I bought the console in 2011 because it was a top-of-the-line Blu-ray player and I guess Mirror's Edge was the only game for the console I had any interest in playing. I wrote a belated post-mortem of the game in 2012 and concluded that it had unique strengths and significant, fixable flaws, and pleaded for a sequel.

That sequel, Mirror's Edge: Catalyst, finally appeared in 2016. Last year (2021) I got back into gaming after a lengthy hiatus and this year I finally got around to playing it. So. What's different? Anything?

*

First, let's quickly recap the original game. The positives: unique high-altitude parkour gameplay; a shining, bright dystopia; crisp, bold, clean, white art design and presentation; great action set-pieces; great music, courtesy of ambient artist Solar Fields. The negatives: frustratingly slow-paced interior segments; janky controls; frustrating ...]]></description>
            <pubDate>Sun, 07 Aug 2022 00:44:20 +0100</pubDate>
          </item>
                  <item>
            <guid>621</guid>
            <title>Solving Tetris in C</title>
            <link>https://qntm.org/tetris</link>
            <description><![CDATA[<b><a href="https://qntm.org/tetris">Code</a> »</b> Exactly one of the following statements is true.

	In Tetris, you can always get at least one line before losing.
	In Tetris, a sufficiently evil AI can always force you to lose without getting a single line, by providing bad pieces.

Which one?
Prove it.
The setup
I created HATETRIS in 2010 because I wanted Tetris players to suffer. It was something I created for my own amusement and as a programming challenge and to investigate the use of JavaScript as a gaming platform. HATETRIS is a version of Tetris which always supplies the worst possible piece.
HATETRIS explores only to a depth of 1. It looks at the immediate future and chooses the single worst piece for the present situation, ignoring the possibility of the player using the current piece to set up a situation where the second piece forces a line. After the first iteration of the game, I started using a bit mask to express each row of the playing field ("well"), which made testing the insertion and removal of possible p...]]></description>
            <pubDate>Mon, 25 Apr 2022 19:10:24 +0100</pubDate>
          </item>
                  <item>
            <guid>618</guid>
            <title>Control</title>
            <link>https://qntm.org/control</link>
            <description><![CDATA[<b><a href="https://qntm.org/control">Blog</a> »</b> Control is a 2019 third-person action shooter developed by Remedy Entertainment. The game takes inspiration from a variety of sources, but its most prominent and notable influence is the SCP project, which is very important to me, for obvious reasons. (I don't like to call it the "SCP wiki" these days... I think this gives people the false impression that individual SCPs don't have specific authors, when they absolutely do. The site has a chronic problem with attribution and this is part of that problem.)
This essay is, I guess, a review of Control. Specifically, it's a review from an SCP contributor's point of view. Technically there are some mild spoilers for Control down below, but if you read it anyway, and then just wait a few weeks, you'll forget all the important specifics. Probably.
Control is probably the closest thing to a AAA SCP game which is ever going to exist. It's the highest budget that anything SCP-related has ever had. It's definitely the best that the SCP project ...]]></description>
            <pubDate>Tue, 29 Mar 2022 23:08:24 +0100</pubDate>
          </item>
                  <item>
            <guid>615</guid>
            <title>&quot;Lena&quot; isn't about uploading</title>
            <link>https://qntm.org/uploading</link>
            <description><![CDATA[<b><a href="https://qntm.org/uploading">Blog</a> »</b> Quite a lot of science fiction isn't about what it's about.
"Lena" is about uploading, but uploading isn't real. It doesn't exist.
It might exist at some point in the future, but that just seems pretty improbable to me. As I understand it, right now, to "accurately" simulate the behaviour of just a handful of neurons requires a building-sized, custom-built supercomputer, running at around 1/1000th of real time. A human being has one hundred billion neurons, so the distance from here to there is something like nine orders of magnitude of processing power, and the distance from there to the "Lena" scenario is several orders of magnitude again. Moore's Law has to tap out at some point, right?
And the really hard problems probably aren't problems you can throw more and more transistors at. The real problems will be biological factors I know next to nothing about. What about the scanning process? How do you simulate the rest of the body and what happens when you don't? What happens when ...]]></description>
            <pubDate>Wed, 26 Jan 2022 13:52:51 +0000</pubDate>
          </item>
                  <item>
            <guid>612</guid>
            <title>Absurdle</title>
            <link>https://qntm.org/wordle</link>
            <description><![CDATA[<b><a href="https://qntm.org/wordle">Code</a> »</b> Absurdle is an adversarial variant of the charming and (currently) viral word guessing game Wordle.
The mechanics of Absurdle are identical to Wordle, or as close as I can make them. Among other things, they use the same word lists. (There are actually two word lists in Wordle: one is a list of 2,315 words which can appear as answers, and the other is a list of 10,657 rather more obscure words which are allowed as guesses but will never appear as answers.) The major difference is behind the scenes.
Wordle picks a single secret word at the beginning of the game, and then you have to guess it. Absurdle gives the impression of picking a single secret word, but instead what it actually does is consider the entire list of all possible secret words which conform to your guesses so far. Each time you guess, Absurdle prunes its internal list as little as possible, attempting to intentionally prolong the game as much as possible.
The easiest way to explain this is with a worked example.
Sup...]]></description>
            <pubDate>Mon, 10 Jan 2022 00:42:06 +0000</pubDate>
          </item>
                  <item>
            <guid>609</guid>
            <title>Selecting content for a short story collection</title>
            <link>https://qntm.org/so</link>
            <description><![CDATA[<b><a href="https://qntm.org/so">Blog</a> »</b> So I think I've mentioned a few of times that the next thing I want to publish is a book of short stories. Due to [gestures at the world] all of this, and some other things, progress on this has been extremely slow over the course of 2021. Currently I am hung up on selecting content.
At present, the book will definitely contain:

	"Valuable Humans In Transit" - very slightly rewritten
	"Gorge" - somewhat rewritten
	"I Don't Know, Timmy, Being God Is A Big Responsibility" - almost exactly as is
	"The Difference" - almost exactly as is
	"Lena" - as is
	"28 Millennia Later" - lightly rewritten probably
	"The Frame-By-Frame" - somewhat rewritten

This is, I would say, enough content to fill about half of a book. In total it comes to around 25,000 words. This would be a good book, I think, but too short. There's a floor on how much I can charge for a book, and I also want to give people their money's worth, which means I would like to put more content in.
Other than this, I have...]]></description>
            <pubDate>Sun, 19 Dec 2021 16:55:07 +0000</pubDate>
          </item>
         
      </channel>
    </rss>
  