I don't like generative AI/LLMs. They strip the enjoyable parts out of the work and leave only the tedium, and they cannot be relied upon. Never mind that they're essentially large-scale open source license violators.
My employer has been encouraging us to use LLMs in our coding work, and I've been resisting, but their encouragement is rapidly converting into requirement. We have to start submitting reports from Claude showing how much we utilized it. I've been desperately wanting to return more of my time to writing, and this LLM push has been the last psychological shove I needed to start moving away from the industry[1].
I'm sure I'll still code for fun, and for my own projects, but I think I'm going to be done with day-job coding unless AI turns out to be a bubble and/or the upcoming unsubsidized price tags throw cold water on the whole enterprise. If I can't make a living with writing, I may have to open a Cajun food truck or something.
Well, our unwillingness to be beholden to sponsors is just part of the reason we avoid advertising, but it's not an empty concern. We've done stories that big brands wouldn't like (the true story of Colonel Sanders, the New Coke debacle, etc) and we frequently share some unflattering stories in our Curated Links section.
The larger reason is just that I personally despise ads. I take great measures to eliminate them from my life--paying for YouTube Premium, using the SponsorBlock plugin, running aggressive ad blockers on every device, piHole, etc. It would be quite hypocritical for me to add ads to my own content, I don't think I could live with myself.
I know that some podcast providers wrap ads around everything anyway, but that's between the listener and their provider, there are absolutely ad-free alternatives.
Blog spam origins are murky at best, but I don't recall it being a prevalent thing until nearer to 2010. But of course recollection is an unreliable narrator. I'm also assuming you are using "predates" as in "preceding" rather than "preys upon", heh.
It's true that some kind of monetization change was inevitable, but I wished at the time they had executed differently. I just wanted fans who wished to see our posts to see our posts, no more, no less. The only options FB gave were to reach far fewer people (do nothing), or to push ourselves on an unwilling audience (boost posts into ads). No option to just pay a monthly fee to have it like it was.
So, while inevitable, I think it's still a good example of why one mustn't trust big corporations with one's work.
There are a few considerations. One is that the 'Give a Damn' system only aims to collect about $1,800/month, an approximation of our monthly expenses (not just hosting, but author payments, subscriptions to archives, research expenses, etc). And most months it barely collects the target amount. I've tried nudging the goal upward to see if I can slowly push it into actually-make-a-living territory, but it seems to have plateaued; increasing the goal there just means we usually miss our goal.
There's also the fact that the 'Give a Damn' system is supporting the site, to the benefit of all of its contributors, not just me. Changing that dynamic could be detrimental to morale.
There are lots of other small things making it very difficult to just update the existing system, not least of which being the same time shortage this fundraiser is trying to relieve. I'd need to make a bunch of changes to a very sensitive money-handling chunk of code, which requires careful coding and testing.
> What will happen if this experiment fails?
I have a series of decreasingly appealing follow-up plans if this one falls too short. But it's already earned enough to get me through at least a few months, so I'll have a chunk of time to take some positive action. Maybe that means I only get 6 months off before I'll have to resume full-time work, but I'll take what I can get.
Originally, in 2008, it was 100% Javascript, but a few years later I updated it to use CSS transitions for a much smoother animation.
That cycler simulator actually led to a very cool life experience. An administrator at NASA contacted me wanting to use it for a presentation. Of course I said yes! Then he invited me to be his guest to watch an early test of the SLS rocket that was happening about 90 minutes from my home. Of course I said yes!
I got to meet a bunch of astronauts, see a big rocket engine fire horizontally in the desert, and got a NASA SLS pin. The fellow ended up not using the animation, as that part of his talk got edited out, but nevertheless niftiness.
> I think moving to Substack (retain domain) would be a much stronger way to build in the long term
Hmm, the danger there is that one is putting a lot of one's eggs into a fickle basket. In the early days of Facebook we had a page with 20k+ followers, and we got a lot of engagement there, people followed us to be informed of when we published anything new. Then one day Facebook introduced 'boosting,' and overnight our posts were hidden from all but a fraction of our audience. Paying to boost each post would convert them into ads, which is not how we wanted to reach our readers. Our site traffic plummeted. I would have happily just paid FB a flat rate to retain access to our audience, but that option was not on offer.
I was already a proponent of the "own your platform" philosophy[1] (aka "Don’t build your castle in other people’s kingdoms"[2]), but that misguided reliance on Facebook really cemented it. It's nigh impossible to own everything we rely on, but I'm reluctant to give any company that much power over my project again.
> I thought you were recycling other folks / were blogspam.
It's understandable that you thought so, though the opposite is usually the case. There are a lot of creators who poach our catalog; if you compare publication dates you'll usually find that ours was published first. While I'm out there scanning the microfiche, reading the dusty old books, filing FOIA requests, and hiring researchers at the National Archive, these lazy creators just yoink the gist and earn 100x more than I do. It's a lot to grapple with sometimes. But I enjoy doing it, so I ignore the parasites most of the time.
> FWIW, in this new age, patronage might be the only way. Allow people to pay on a sliding scale, with an uncapped upper end.
That's not far from what I'm attempting with this fundraiser experiment. There is a modest goal for the year, but no cap on contribution, so if someone(s) with vast resources is inclined to make a generous contribution, they are able to do so.
One problem is that I don't know how to reach such people apart from this omnidirectional signal. Another is that I would not be amenable to string attachments. Maybe I'm broken, but I'd rather shut down the site than allow a wealthy benefactor to call any shots, and most wealthy entities won't like that (I expect).
> And give them access to a tightly moderated, thoughtful community.
We kind of have that in our comments sections, but a more unified place might be an interesting exploration. I'll ponder that, thank you for the suggestion.
It's funny you mention those three in particular. We've collaborated with Stuff You Should Know several times, and we almost worked with both 99PI and RadioLab on separate occasions (i.e., I pitched ideas that got some traction, though they didn't ultimately materialize). I still think my pitch for RadioLab is a good one, though it's a throwback to when they did episodes with three stories around a central theme (e.g., Blood).
> I was more than happy to throw some bucks at it.
Thanks! If this fundraiser works out, you're part of the reason we get to keep going.
Heavens to Betsy. My server alerted me that traffic was far exceeding a median Thursday. For the record, I am a fairly active user on HN, but I was not responsible for sharing this link here, directly or otherwise. I have little stomach for self-promotion (probably to my detriment).
One thing you can do is store a word profile in your database, which is a string of all of the letters in the word in alphabetical order. So the word profile for 'apple' is 'aelpp'. Then you can just include all of the words that match the profiles of the real words. For example, 'quiet' and 'quite' both have the profile 'eiqtu'.
I've built a some word games over the years (e.g., wordwhile.com, omiword.com), and one thing I've discovered is that players find it very unsatisfying when they enter a real, valid word and the game rejects it. I imagine that the timer would amplify that sense of unfairness.
Granted, there are also people who get annoyed when the game seems too accepting of unusual words, but if you can point them to a valid dictionary definition for that oddball word, they usually accept it without argument.
As a non-fiction author, I find myself reading a lot of non-leisure material for research--books, old newspapers, and that sort of thing. A few years ago I noticed that my leisure reading was on the decline, so I decided to delete all of the social media, news, and gaming apps from my phone, and replace them with an e-reader app.
Now, when I have a few idle moments away from my computer, instead of checking something like reddit, I read a few pages from a book. It's great, I recommend it! I'm back to reading a couple of novels per month, and I don't have so much of that queasy wasted-time feeling that social media tends to give me. I've even learned how to pause reading mid-paragraph and resume easily later; that part took some practice.
Here in Salt Lake City, a prolonged drought has made it a tinderbox. Home fireworks are prohibited, and there's signage up all over the place warning of the danger. And yet, the firework stands popped up in grocery store parking lots like they do every year, and people lined up. Bananas.
Luckily no new fires started around here, but we also have the local "Pioneer Day" on July 24 that starts up a fresh round of pyrotechnics.
I was in my car with my seven-year-old when the Paul Simon song "You Can Call Me Al" started playing. She saw it pop up on the infotainment screen, and wanted to know why anybody would want to be called A.I. (our household has a low opinion of A.I., largely due to the unending firehose of slop aimed at children on YouTube).
This is reminiscent of Coley's Toxins[1], an early experimental cancer treatment using Streptococcus pyogenes bacteria. Coley and other practitioners claimed some substantial success, but the mechanism of action was not understood, and the treatment varied so much that scientific verification was elusive.
(Note: I am affiliated with the linked website, but I didn't write the article in question)
And they got Rick Moranis out of retirement to reprise his role as Dark Helmet, which is just delightful. I hope we see more of him in movies now that his children are grown.
I've been building sites on WordPress for over 20 years, and it's fine, but if I were building a new site today, I'd probably use ClassicPress[1]. It's a fork of WordPress that eliminates a lot of the cruft that WordPress has accumulated over the years.
When I was kid, I was playing Nintendo games with my cousin, and my very straight-laced Mormon mom kept calling for me to come upstairs for dinner. I kept replying, "Just a sec!" as we tried to finish the stage. After a few times through this loop, she yelled "NO MORE SECS!" It's been a running joke in my family ever since.
One recalls the Pentium FDIV bug, where an estimated 1 in 9 billion floating point divides with random parameters would produce inaccurate results[1]. A small subset of Pentium users would ever encounter the problem, yet it was considered a major controversy, and Intel ultimately replaced $475 million worth of CPUs (in 1994 dollars)[2].
* @[email protected]