The Braindump Blog

Below are notes I took from reading the paper: How much food tracking during a digital weight management program is enough to produce clinically significant weight loss?

It’s known that weight loss can be facilitated by lifestyle interventions.

One of the strongest predictors of how much weight loss will occur is how often participants track the food and drink they consume each day, usually with a view to remaining within a set calorie goal.

However, tracking your consumption in detail is effortful. Most people do not find it possible to do so over the longer term. One approach to improving adherence would be to reduce the amount of effort required, for example by reducing how often people are asked to track food or reducing the amount of their food they are asked to track.

It is not yet clear how much food tracking is needed to get to certain weight loss milestones. Studies so far suggest that “abbreviated” food tracking approaches do correlate with weight loss, with the possible exception of tracking by taking a photo of the food.

Using data from subscribers to a weight loss program over a 6 month intervention period, the study aims to:

1. establish whether the abbreviated method of food tracking facilitated by by the program remains associated with weight loss.
2. establish thresholds of food tracking that are associated with certain weight loss milestones

Aim 1 was tested with both a nonparametric and linear regression methods.

There was a significant correlation between the percentage of days participants tracked food with the percentage of their original weight that the participant lost after 6 months, r = 0.4, p < 0.001. A 1% increase in the number of days tracking food associated with a 0.08% greater in weight loss.

A ROC curve analysis looked at showed that food tracking was highly predictive of the clinically significant 3%, 5% and 10% weight loss thresholds after 6 months. The optimum predictivity threshold was defined by the sum of sensitivity and specificity.

• For predicting achieving at least a 3% weight loss, the optimum threshold of food tracking corresponded to 29% of intervention days.
• For 5% weight loss this was 39% of days tracked,.
• 67% of days was the optimum threshold for predicting the 10% weight loss threshold.

Aim 2 added a time series clustering analysis method, dynamic time warping, to look for similarities between member food tracking behaviour.

This produced 3 clusters based on the number of days participants had tracked food in each week over the intervention period. The clusters differed by levels of food tracking, weight loss, participant age and marital status.

This study could not quantify the mechanisms why participants showed different food tracking behaviours. For instance, perhaps some had higher motivation or fewer barriers to food tracking and/or weight loss.

Limitations include:

• Participants were biased towards being female, with college degrees and higher incomes.
• This approach cannot fully establish that food tracking causes weight loss. There could be confounders.
• Whilst the optimum thresholds for predicting 3%, 5% or 10% of weight loss were established, there was no step-change in cutoff that perfectly predicted weight loss success. We should be cautious about interpreting the specific thresholds here.
• Food tracking was defined as tracking at least 1 food item on a given day; other definitions could be applied.

In conclusion, this study shows that participants can achieve clinically significant weight loss after 6 months even without perfect adherence to the recommended daily food tracking. In fact, no-one in this study tracked every day. Given that perfect adherence doesn’t seem to be possible for most people, and isn’t necessary for achieving significant weight loss, we should question whether it should be the default prescription for weight loss interventions.

Should one have the need to mass-download their Kindle books from Amazon without a ton of clicking around the Kindle Download Helper Python project looks to enable that.

I haven’t yet tried it, but it seems to work for other folk. Of course you’ll have to let it log in to your Amazon account which might be a deal-breaker for some, along with any risk associated with the fact that I can’t imagine it complies with Amazon’s terms of service. Although as it’s open source you can at least check in advance what the code is doing.

🎶 Listening to Endless Summer Vacation by Miley Cyrus.

Came to it via the ultra-viral hit Flowers which holds the record for being the fastest ever song to reach a billion streams on Spotify amongst other such achievements, taking 113 days if I did my maths right. It got 60 million streams occurred within a single week.

That song book-ends the album: the version you know at the start, a substantially differently-vibed demo version at the end. But there’s plenty of pop-rock in-between from the current phase of Miley’s constant re-invention. Bit of a mixed bag imo, but it’s a fun enough summer listen.

Somehow only today did I realise that you can use if statements directly within R’s dplyr library filter function in order to make conditional filter criteria.

For example if you want to filter a fictional dataframe to show only scores below 10, but only if the variable x has the value “filter” you can do:

filter(data, if (x == "filter") {
      my_score < 10
    } else {
      my_score == my_score
    })

A variable always equals itself, so the score == score bit effectively means “don’t apply a filter in this case”. You can of course use other criteria there if you want to apply different filters based on different conditions.

Inevitably some bright spark has developed a large language model AI chatbot whose explicit purpose is to assist with computer crime.

Based on GPT-J, WormGPT has no safety rails and has been trained on data from the darker side of the internet. So far its stated uses appear to be to write very convincing phishing scam emails, and to create and advise how to distribute malware.

Not the first area of life I’d have picked for needing to democratise and make accessible to all, but there we go. I remember listening to an episode of the Data Skeptic podcast last year where a researcher talked about the AI phishing email side of things so it’s not exactly a massive surprise that some 2023 malevolent entity went ahead and made it.

Tom’s Hardware quotes its developer:

This project aims to provide an alternative to ChatGPT, one that lets you do all sorts of illegal stuff and easily sell it online in the future. Everything blackhat related that you can think of can be done with WormGPT, allowing anyone access to malicious activity without ever leaving the comfort of their home.

A screenshot from PC Magazine:

All yours for 60 euros a month. Sigh.

Very unexpectedly, something about the apparent inanity / insanity of its plot makes me actually want to see the Barbie movie.

I’d assumed it’d be at least 80% problematic, but if it made Ben Shapiro spend 43 minutes burning his dolls in disgust at its wokeness then maybe it’s not so bad after all.

📚 Finished listening to The Monopolists by Mary Pilon.

As far as board game creation stories go, the story of Monopoly’s invention is a well known one. Not least because for a while it was included in the contents of the game box itself, at least in the US.

For the un-initiated: Stuck at home having lost his job during the US Great Depression, Charles Darrow had a sudden flash of inspiration and invented Monopoly. Firstly to amuse his family during that difficult time in American history and latterly, in 1935, to sell to the Parker Brothers company. Darrow ends up as an incarnation of the rags-to-riches American Dream story as hundreds of millions of copies of the game were produced and shipped around the world.

The game certainly spread. In the end millions of groups of gamers would go on to learn the basics of capitalist economics, the dog-eat-dog winner-takes all majesty of the free marketplace, with the winner of each game ending up with reams of imaginary real estate and their respective virtual incoming rents from their non-propertied companions. A true simulation of the American Dream, accessible to all, realised like never before.

Except, inevitably, that story isn’t true. Monopoly wasn’t really invented by Charles Darrow. Or any other individual entirely. Like many other of humanity’s cultural artifacts, what we now know as Monopoly is a pastime that evolved over space and time.

The author starts the story with the person that could perhaps most fairly be termed Monopoly’s originator, Elizabeth Magie. Born in 1866, she was very interested in both politics and inventing. This of course was certainly not encouraged for women of that era. Nonetheless, in 1904 she patented “The Landlord’s Game”.

Ironically, considering what it turned into, the game was intended to espouse her Georgist principles and demonstrate the perils of unbridled rentier capitalism; specifically how monopolies over land led to ruin, and how taxes could be used to ameliorate these outcomes. It was intended as part of a political argument against the massive concentration of land-based wealth in the hands of landlords, not a celebration of it.

In general, the Georgist movement was dead against the private accumulation of wealth from land ownership

From Wikipedia’s entry on Georgism:

…although people should own the value they produce themselves, the economic rent derived from land—including from all natural resources, the commons, and urban locations—should belong equally to all members of society

The Landlord’s Game could be played either with Monopolist rules, where players are incentivised to become monopolists and force other players out of the game - these will be familiar to anyone who has played modern-day Monopoly. But it also came with anti-monopolist rules, where the game is won only when every player has doubled their original money by virtue of communal wealth creation.

Educational as it might have been, it didn’t sell very well at the time. But it was taken up by certain communities - notably the Quakers and various socialist-leaning academic types - who passed various incarnations of the rules and materials of the game on to their nearest and dearest, sometimes adding or changing a few of the details. Over time it grew ever closer to the Monopoly we know today. Magie’s original version already had the property, railroads, utilities and jail squares familiar to us today, as well as idea of continuously circling a square board with no pre-defined end, a relatively rare setup at that point time. As it passed from hand-to-hand some of the properties got renamed and the Chance and Community Chest cards we’re familiar with today were developed.

What came next is a frustrating lesson in intellectual property law. After Darrow sold “his” game idea to Parker Brother, they went on to persuade Magie to sell them the patent to her original Landlord’s Game, misleading her into thinking that they would mass produce and market copies of it, with its original message intact, worldwide.

Well, have you ever seen “The Landlord’s Game” in a game shop? Of course not. In reality, they’d wanted to acquire the patent to protect themselves from the increasingly obvious truth that the real story behind the Monopoly game they were selling in such a profitable manner was not one that suited them.

If this book is to be believed - and it certainly seems very thoroughly researched - Darrow and Parker Brothers behaved somewhat deviously, probably fraudulently, to do their best to retain the sole right - the monopoly - to sell Monopoly based on its manufactured origin story. Whilst of course compensating none of the many other people who contributed towards the development of the modern-day institution of Monopoly over the decades.

Again layering on the irony, some of the truth of this came out as part of another intellectual property related court case, but one that Parker Brothers initiated. An economist called Ralph Anspach created a game called “Anti-Monopoly” in 1973 that was designed to help people understand the perils of real-life monopolies.

Parker Brothers didn’t like that the name involved the word “monopoly” and took him to court to prevent him selling his game. Preparing his defence led Anspach down a path of somewhat obsessive research into the history of Monopoly, with the resulting idea that Monopoly itself should be in the public domain with the copyrights and trademarks Parker Brother claimed to hold being invalid. A decade later a settlement was reached, but not before the case materials revealed a lot of the truth that Mary Pilon now, in turn, reveals to us.

Is it worth trying to increase your physical activity if you only have time to even think about it on the weekends? New research from Khurshid et al. suggests that yes, it is.

They find that getting at least the recommended 150 minutes of moderate to vigorous physical activity per week provides the same positive association with cardiovascular health whether you do most of your exercising in 1-2 days - their “weekend warrior” classification - or spread it out across the whole week.

…a weekend warrior pattern of physical activity was associated with similarly lower risks of incident atrial fibrillation, myocardial infarction, heart failure, and stroke compared with more evenly distributed physical activity.

Another upsetting outcome of the inevitable coming together of the generative AI infocalypse, enshittified subscription service monopolies, misguided algorithmic recommendation systems and the incentives of modern-day capitalism has been achieved. This time this victim is books.

According to Vice, many of the supposedly most popular entries in the Amazon Kindle Unlimited subscription service’s bestseller list recently were in fact fully AI generated nonsense.

A rash of books with similar covers, quite random titles and absolute nonsense inside them apparently hit the top 100 list, at least in categories such as Young Adult romance.

My favourite of the ones Vice reported on has to be the catchily named “Apricot Bar Code Architecture”. The introductory sentence, if you please:

“Black lace pajamas, very short skirt, the most important thing, now this lace pajamas are all wet.”

Now I don’t imagine too many readers made it all the way through these books meaning that whoever put them there probably isn’t raking in too much money per reader. It seems like Kindle Unlimited is paying something like 0.4 US cents per page read, if I understand Chris McMullen’s post correctly. But the fact that they got to the top of the rankings means they’re making some amount of money for someone, likely at the expense of the more traditional actually-makes-sense books.

Because Kindle Unlimited pays authors per page read, not per book read, if you can get 1000 people - real or imaginary - to read the first 5 pages of your effortlessly created clickbait book then you’ll get the same amount of financial reward as someone who can get 10 people to read the entirety of their award-winning 500 page novel - and more than if 5 people who read your 500 page novel valued it so much that throughout their later life they each read it another 10 times each.

I suppose it doesn’t take much for the same folk who set up fake ad-clicking farms to switch to fake book-reading farms.

New(ish) Conservative party faction just dropped. Let’s welcome the “New Conservatives” to the party.

They join the National Conservatives and a bunch of other factions such as the European Research Group, Covid Recovery Group, Net Zero Scrutiny Group, Common Sense Group, Blue Collar Conservatism, Northern Research Group and no doubt many others that don’t quite come to the top of my mind right now. One day there will be more factions than politicians if we lucky.

The New Conservatives faction has the extremely non-unique selling point of not liking immigrants. They did manage to come up with a 10 point policy, but every point is a unsubtle rewording of “immigration needs stopped”.

Although if they do ever fully exhaust our patience with anti-immigration cruelty they do have aspirations to move onto being - can you guess? Let’s all say it together: anti-woke.

🥱

I’ll let the i headline speak for itself: ‘They’re dickheads’: New Conservatives group of MPs will cost us the election, Sunak allies say

In particular they appear to be very concerned that the current Prime Minister - who constantly explicitly tells us that lowering immigration, at a time where in some ways the country might actually need immigrants more than ever even for purely selfish reasons, is one of his top priorities at basically any cost - even the cost of breaking international law - should think about adding lowering immigration as one of his priorities. How efficient they are to have achieved their goals before they even existed.

There’s a fun interview on the July 3rd episode of The News Agents podcast where the one of the high-ups of this pointless new faction really really tries to wriggle out of agreeing that what he wants to is prevent is even the most cruelly oppressed and tortured women stuck in Afghanistan from seeking asylum in the UK amongst other things, when that actually seems to be the whole point of his tediously unoriginal faction. Add dishonesty to the litany of sins.

Following some big companies such as Samsung banning or putting limits on their employees' use of tools such as ChatGPT, I wrote about some of the risks one takes when using generative AI systems to write code over here.

From the i:

Nigel Farage has criticised the Home Office’s decision to paint over murals of cartoon figures at an asylum centre for lone children, describing it as “a bit mean”.

Good grief. If Nigel Farage thinks your immigration policy is cruel that really should give you cause to question its morality.

But Sunak et al are still sticking to their story that wanting to see Mickey Mouse is a the main driver of the treacherous journey some of the most desperate children in the world make to our shores.

The Prime Minister’s spokesman suggested the decision was designed to “deter” asylum seekers from crossing the Channel as part of the Mr Sunak’s promise to “stop the boats”.

The latest deranged volley in the Conservative Government’s extremely inhumane and extremely ineffective “anti-immigration” policy has arrived.

It turns out that some asylum reception centres have cartoon images on display.

(image from The Guardian)

This is, of course, Too Woke.

From The Independent:

The Immigration Minister said pictures of cartoons and animals must be removed and that staff should make sure they are painted over, as they give an impression of welcoming, which Mr Jenrick didn’t want to show.

The staff, being humans, didn’t want to do this. But it seems like in the end Robert Jenrick got his way.

Presumably the concern is that millions of children are going to willingly risk their lives at great personal cost to themselves and any remaining family that they may or may not have in order to enter a country that is actively hostile to their very existence mainly because they want to see a picture of Mickey Mouse.

I’ve been looking at options for Network Attached Storage recently, to help me be more conscientious about backups, plus eventually reduce my dependence on various big tech cloud services.

For leaning about the Synology side of things I’m finding SpaceRex’s Youtube channel quite informative.

Frank Wilhoit succinctly encapsulates how at least a subset of conservatives - I suppose he would say all conservatives - appear to think:

Conservatism consists of exactly one proposition, to wit:

There must be in-groups whom the law protects but does not bind, alongside out-groups whom the law binds but does not protect.

This may feel contrary to the classic, now feeling extremely old-fashioned, idea that conservative politics tends towards the politics of law and order.

That sentiment was never straightforward to interpret. There is more to law and order than proudly caging the largest number of people you could possibly justify in an entirely inadequate prison system. But in Wilhoit’s telling it presumably can’t really ever be the case. At least not whilst the law purports to be something that applies equally to all people.

At any moment in time the conservative parties and surrounding movement might look to be extremely pro law and order. But this just means that the law is such that it currently promotes their own interests, usually at the expense of some other group’s interest. Should this stop being the case, so will the avid conservative fandom around the legal process.

These examples are probably so obvious as to not really be worth expressing, but we can see how quickly certain branches of the UK and US conservative movements reacted with wanton disrespect for the legal process and norms as soon as their political heroes - Boris Johnson and Donald Trump - were called to account for their illegal and immoral behaviour.

Johnson was more than happy to explicitly break international law to get his own personally-preferred type of Brexit. In the mean time, some of the conservative British media attempted to brand judges who were insufficiently enthusiastic about that same event as being “enemies of the people”. Trump’s supporters, well, some of them famously engaged in a violent insurrection.

In Wilhoit’s mind then, what would anti-conservatism thinking look like? Simply put, the idea that:

The law cannot protect anyone unless it binds everyone; and it cannot bind anyone unless it protects everyone.

As a sidenote, we’re not talking about Francis Wilhoit the political scientist, but rather Frank Wilhoit the classical music composer who just happens to share the same name.

The US Federal Trade Commission is investigating OpenAI, creators of ChatGPT et al, in order to establish whether they’ve violated consumer protection laws.

The Washington Post shared the Civil Investigative Demand Schedule that’s been sent to OpenAI requesting all sorts of information around how the company deals with the risks associated with their product.

The goal is to determine:

Whether [OpenAI] has (1) engaged in unfair or deceptive privacy or data security practices or (2) engaged in unfair or deceptive practices relating to risks of harm to consumers, including reputational harm…

Learned that you can work in two branches of the same local git repo at once without problems via the git worktree command.

Useful for if you want to quickly fix some code without having to faff around committing or stashing the stuff you were working on

git worktree add <path> automatically creates a new branch whose name is the final component of <path>, which is convenient if you plan to work on a new topic. For instance, git worktree add ../hotfix creates new branch hotfix and checks it out at path ../hotfix. To instead work on an existing branch in a new worktree, use git worktree add <path> <branch>.

PC Part Picker seems to be a good site for seeing a summary of what, surprise surprise, parts of PCs cost at present. There’s also a price history chart so you can see how it’s varied over time.

The Amazon price history isn’t shown but you could use Camelcamelcamel for that if you needed to.

It’s increasingly apparent that relying on the ability to reliably access or even link to internet content outside of the truly open web is a fool’s errand.

I had problems today building even this little blog, receiving a lot of “Failed to get JSON resource” errors. The culprit turned out to be my use of a Twitter shortcode, which my blog software kindly provides in order to make it easy to embed tweets in their full interactive glory.

To date it’s always been possible for someone who is not logged in to see an individual tweet and for tweets to be embedded on sites. There are sites that almost entirely consist of embedded tweets and some commentary on them, for better or worse (usually worse).

I’m totally guessing, but I’m going to assume that the sudden failure of this method of embedding tweets was a side effect of Musk’s “surprise” change to prevent not-logged in users from seeing anything on Twitter. As well as stopping logged in users, even the rare ones that actually pay for the privilege, from seeing as much as they might want to on Twitter.

This and the ongoing Reddit API fiasco makes me wonder whether Ryan Broderick was right in a recent edition of his newsletter to imply that the only future-proof option for the average user to share even very mainstream content these days is to post screenshots. Doubly so for any content on any site popular enough in in theory be bought by some ideologically incompatible billionaire. How web 1.0.

While trying to track down the actual hyperlink to a post I found a screenshot of on a closed social network I was struck by how on an internet full of closed platforms, broken embeds, and crumbling indexes, the last reliable way to share anything is a screenshot. In other words, the camera roll is, at this point, the real content management system of the social web.

For anyone who is sick of either whatever the latest “rich person ruins social network” drama is or the thought of what their own younger self thought was just fine to post to the entire world, the software “Redact” looks like a free and versatile way to remove your posts from a huge variety of services.

It contains filters to allow you to selectively delete things. For example, perhaps you want to delete only Reddit posts you made last year that included the phrase “I love Reddit”, or only the last 7 days of your Twitter DMs. You can also set up a schedule if you want to repeat the delete every so often.

Looking at some of the commentary around the tool it seems like it occasionally stops working for some services. Presumably this is due to changes the social network companies make, including those targeted at stopping tools like this. I suppose you might be infringing certain terms of services if you mass-delete like this in many cases. But whilst the networks don’t spend much effort building their own tools to make it easy to manage and delete your past submissions to their services for obvious reasons this certainly doesn’t feel like a moral crime.

In order to delete things you are going to have to log into your account of the respective services through this software, so you will have to feel OK about trusting that the software isn’t doing bad things. They claim that Redact doesn’t store your info or transmit it anywhere, that their staff have no way of seeing your username of password. But it is closed-source so you can’t check the code directly, so caveat emptor I suppose. I haven’t seen anyone suggesting that anything malicious is going on.

The full list of services it can remove stuff from is long:

• Discord
• Twitter
• Reddit
• Facebook
• Twitch
• YouTube
• Imgur
• Deviantart
• Pinterest
• LinkedIn
• Tumblr
• Microsoft Teams
• Skype
• Instagram
• Slack
• Telegram
• Tinder
• Stack Exchange
• Email
• TikTok
• Steam
• Blogger
• Wordpress
• MyAnimeList
• Letterboxd
• Disqus
• Quora
• Github
• Spotify
• IMD
• Gyazo
• Vimeo
• Bumble
• Flickr
• Medium
• Yelp
• Pixiv
• RustleLogs

It’s currently available for Windows, Mac, Linux and Android. Support and other discussions about it mainly seem to happen on their Discord. They also tweet.

In “I wanted to be a teacher but they made me a cop” Adam Mastroianni makes the case that teaching someone is not the same as evaluating someone. The two are often in fact at odds with each other.

He suggests separating the two. Some people become professional teachers, others become professional evaluators. They are after all very different jobs even if we usually try to wedge the latter into the former right now, despite the (over)importance of grades et al. in today’s society.

Doing evaluation on its own would have a few major benefits. First, it would force us to take evaluation seriously.

…

Second, we’d see how hard evaluation is, and maybe we’d do it better.

…

And finally, if we made evaluation its own thing, we’d see how nasty it is, and maybe we’d do less of it.

It’s increasingly apparent that relying on the ability to reliably link to or access internet content outside of the truly open web is something of a fool’s errand.

I had problems last week building even this little blog, receiving a lot of “Failed to get JSON resource” errors. The culprit turned out to be my use of a Twitter shortcode, which the blog software kindly provides in order to make it easy to embed tweets in their full interactive ‘glory’.

To date it’s always been possible for someone who is not logged in to see an individual tweet and for tweets to be embedded on sites. There are sites that almost entirely consist of embedded tweets and some commentary on them, for better or worse (usually worse).

I’m totally guessing, but I’m going to assume that the sudden failure of this method of embedding tweets was a side effect of Musk’s “surprise” change to prevent not-logged in users from seeing anything on Twitter. As well as stopping logged in users, even the rare ones that actually pay for the privilege, from seeing as much as they might want to on Twitter.

This and the ongoing Reddit API fiasco makes me wonder whether Ryan Broderick was right in a recent edition of his newsletter to imply that the only future-proof option for the average user to share even very mainstream content these days is to post screenshots. Doubly so for any content that’s hosted on any site popular enough in in theory be bought by some ideologically incompatible billionaire. So web 1.0.

While trying to track down the actual hyperlink to a post I found a screenshot of on a closed social network I was struck by how on an internet full of closed platforms, broken embeds, and crumbling indexes, the last reliable way to share anything is a screenshot. In other words, the camera roll is, at this point, the real content management system of the social web.

Australia becomes the first country to legalise the use of certain psychedelics in controlled medical environments for the treatment of certain mental health conditions.

…approved psychiatrists can prescribe MDMA for post-traumatic stress disorder (PTSD) and psilocybin for depression that has resisted other treatments

I’m very pleased the decades of research around this idea has finally produced a policy change.

TIL: The US isn’t the only country to officially celebrate American independence on July 4th. Denmark also does.

Much of that is thanks to Mr Henius, a Danish emigrant to the US, who in 1911 led a group of grateful expats in donating a swathe of Denmark’s land as a ‘place of homecoming for all Danish Americans’. There was a condition:

the park must hold a festival celebrating American Independence Day as a symbol of friendship between the two countries

…which still exists today: US flags, hotdogs and fireworks abound throughout the Rebild Festival.

Here’s a photo from RebildPorten.

📚 Finished reading 11.22.63 by Stephen King.

Jake Epping, a down-on-his-luck teacher, is recruited by a friend to kill Lee Harvey Oswald before he gets a chance to kill President J. F. Kennedy.

Why? His friend, Al, believes that if JFK had survived then the world would now be a much better place, not least because his practice of politics may have meant that the Vietnam War would never have happened.

Get rid of one wretched waif, buddy, and you could save millions of lives.

And how, given the book is set in relatively modern times, 2011? Well, as luck would have it, Al has discovered a strange portal in a corner of a diner. Walking through it leaves you in the same geographic location in the year 1958.

You can return back to modernity as and when you want. And no matter how long you spent in the past-times world, it’s only 2 minutes later when you come back to 2011. But it might be a totally different 2011 if you did something in the past that had substantial repercussions.

We know that Kennedy was shot in 1963. So Jake has a few years to get accustomed to discretely living in the past, to learn how the world used to work. Unfortunately for him, the past has a way of resisting changes to the original timeline. You might plan to do something that was never done in the 1960s only to be inexplicably diverted, distracted or incapacitated en-route. This tendency is not infallible, the past can change in some ways, but it’s not entirely straightforward.

A further complication is of course the mystery surrounding the JFK assassination. In our world there are plenty of people out there who do not believe Oswald was the real assassin, or at least not the sole person involved. These people include the large majority of Americans in a 2001 poll.

Would killing Oswald in cold blood really stop JFK being assassinated? Especially if not only is there a decent chance that he wasn’t the sole operator involved, but also that tendency of the the past to exert mysterious forces in the direction of recreating its original timeline.

Jake is not a natural murderer. Understandably he wants some certainty that there is some point to conducting this operation before committing such an alien act.

Of course that’s not the only unknown. Sure, Al believes JFK staying alive long would have dramatically improved the world. But would it really? Would there be no cost to pay at all? Few people would feel certain about that, and Jake is no stranger to doubt. So many time travel stories have conveyed to us the idea that altering the past could have intended consequences for the future.

Furthermore, living in any community for several years means you’ll struggle not to build up a lifestyle, habits and human relationships that might be hard to put aside for a small chance of possibly doing something that may or may not make the future a bit better. That’s even if you could navigate your way through an alien world that resists your every attempt to change it so as to be in the right place at the right time to put a potentially huge dent into standard-timeline history.

Well, I’m pleased to say that a mere 800ish pages later you’ll find out what Jake thought and did. Yes, it is a mammoth tome, so long that had it not come with such a strong recommendation I doubt I’d have considered starting it. But the recommendation was a good one. It was over long before I could ever have gotten bored of it.