You can watch how much I rewrite here
twitter.com
twitter.com
I watched over his shoulder for a little while, but it feels intrusive to be standing there breathing down someone's neck while they think and write, and hey, I could just play it back later, so I left.
Memory is fallible after so much time but I think he told me later that he wouldn't keep using it because he was so used to vi. I want to add that he was going to try to get the Etherpad guys to implement vi shortcuts but that feels suspiciously like the sort of embellishment that creeps into a story years later.
Waiting with bated breath for the continuation. Not often you see dang write about things which aren't moderation.
edit: After seeing the continuation I'm just nodding along. Very few things could make me abandon Vim.
(I'd expect pg to know this already, but hey, here it is)
Now I'm wondering if there's a length limit to vim macros...
So I suppose you could just qq, write your entire thing, then "qp to get the whole history. I wonder if there's a tool that replays commands in slow motion, or some neat way to add sleeps to a command sequence.
Was that the intent of the feature, or something else?
It seems like this might be coupled to some kind of teaching program that helps you become a more efficient writer. Maybe something that could recap your session and offer a focused lesson on, say writing a paragraph from a prompt.
People change behavior when they are being watched. Even when I know I'm being recorded and no one is watching, I feel a bit of the spotlight effect going on. I don't know that I could actually record my honest self without being performative. I wouldn't be free to do anything that might be embarrassing. Even in pair programming or coding interviews. It's an incredibly intimate thing, to watch someone take their brain noise and broadcast it, raw.
Sure there’s an observation effect, but this is 1000x more informational than if Paul were to write an essay: “how I write essays”.
> People change behavior when they are being watched
Maybe they want to share this sort of thing, so they can practice not changing behavior while they're watched.
I don't use them often, but when I'm doing something new, these kinds of references are absolutely invaluable.
The impression you get from someone describing their process is totally different than seeing it. Especially iterative aspects, false starts, etc. All of those are thing writers AND readers gloss over.
For example, Vernor Vinge wrote many of his stories on early computers. He didn't have any way to record that, but his style of annotating the text as he wrote it and thought of ideas or problems turns out to be quite interesting. He eventually shared his annotated text for A Fire Upon the Deep; you can read it at <https://3e.org/vvannot/>.
Although he didn't ever intend to publish them himself, JRR Tolkien kept extensive (but very badly organized) records of how his stories evolved over time. Christopher Tolkien later published stories from those notes, but also successive versions of the same story from different stages of it's development. "Beren and Luthien" is especially fun because you can see Tolkien's mistakes, such as characters like "the appalling Tevildo", Prince of Cats who captures Beren in the earliest surviving draft.
The most surprising thing I've seen in 2009, courtesy of Etherpad - https://news.ycombinator.com/item?id=495336 - Feb 2009 (126 comments)
Watch Paul Graham write his latest essay - https://news.ycombinator.com/item?id=557191 - April 2009 (22 comments)
But each to his own. PG is obviously a prolific and successful writer, so for him this works.
For me, if I edit while I'm writing, self doubt gets in the way and I can't get a draft finished.
If I tried to just write it all and edit it after, it would feel like I forced myself to a conclusion. I wrote not just to express what I thought but to figure out what I thought. Putting the words on the page just has this way of clarifying my thinking, and I never end up where I thought I was going to when I started.
Why'd you stop? I always enjoyed your thinking.
For a writer, the computer (or not) is simply a matter of process just like bashing or swooping. And the world is entirely different for a full time professional writer than for ordinary people who write (I have made the statistical assumption that you are not a full time professional writer).
I mean the waste basket full of the morning's failed pages is a real thing, and if your goal is 500 good words a day, you're potentially in the novel-a-year range.
For what it's worth, the main ideas can go into an outline -- a step the computer encourages skipping. Here it looks like Graham was working from an outline via the ten idea structure. Maybe all first drafts are not equal?
It’s fascinating to see such a similar structure in so many things.
For example, in programming: breadth-first vs depth-first when navigating a tree structure, or in nature: the cycles of the slime mold looking for food.
I would never go back to that process, of course. It was painful then, and would feel even worse now. But it was possible.
Sometimes I discarded the nuggets as better ones came along. Sometimes the thesis subtly changed as I gathered these nuggets. Sometimes the layout changed drastically as I shuffled my index cards. All before I even committed to my essay structure.
My best guess from watching the video is that they are temporary place holders, breadcrumbs he leaves himself to return to, allowing him not to break his flow/train of thought. If that's the case, then the technique reminds me of Neil Strauss's "Power of the TK":
When you get suck on a word, or cant’ think of the right fact, what shouldn’t you do?
Don’t stop. Just type, TK.
Source: I was one of the founders of Stypi and we asked him when porting the data from Etherpad to Stypi.
https://techcrunch.com/2011/08/09/yc-funded-stypi-is-etherpa...
It's the complete opposite of the site shared in the post, but people seem to like it.
I write with font size setting of 8pt. It's too small to read comfortably (thus I don't edit), but it still lets me see the overall structure and what I wrote.
I remember a blogger who used to dim her screen as much as possible when writing, which was one of the ideas that inspired me to build this thing. This helps with maintaining flow, but doesn't really touch the overall structure of the text.
I'm a hobby writer and like your idea a lot! I love that you also make the text downloadable in a txt file.
I always feel overwhelmed seeing the text I have written, being aware of what will await me when editing. I would love to have a program on my computer like yours (I'm using windows). Nothing special just an IDE that works exactly the way your website works (I'm not a programmer and love HN for content and comments like yours) I would even happily pay for it.
Let me know if you're planning on doing an app and need a beta tester.
I am, will do!
What features does the web app miss?
I love that it has just four buttons: dark/light mode, download *.txt, eraser and full screen. Your piece of software made me instantly concentrate with laser sharp focus. Usually, I'm quite overwhelmed by my writing and the more I write, the more annoyed I get. So much to edit. Blah. The app achieves exactly what you've promised!
As a sidenote, something beautiful happened, I started playing around with long and short sentences. I even found some of my sentences beautiful. Something that had never happened before because I was constantly overwhelmed by all the other stuff, paragraphing, coherence and whatnot.
I was a bit irritated first that the program allowed me to erase multiple lines (it violates a bit the no-editing mindset) but later I found this option quite helpful when I needed to change the beginning of a sentence to steer in another direction that flowed better.
What I would love the program to have is, from the perspective of a writer, to know that the text I've written is somewhere safely stored on my computer like a backup file or something. That way, I wouldn't need to worry about the system breaking down and me losing all my work. For a serious writer, knowing that the text is securely stored somewhere in a backup file and readable with any program, would erase the fear and enhance the flow of writing. Furthermore, at first, I wondered if there is a limit of how many words I can write before the software breaks down or tells me I reached the maximum capacity. I think I'm allowed to write as much as I want, 100.000 words or 1.000.000 words, right? However, it would be super nice if you would communicate that.
Even though I love the word count because it gives me a sense of space and time, I would like to be able to turn it off, so I can enter a true flow state. Forgetting about space, time, and word count. https://endel.io/ for example uses infinite soundscapes to create flow states. That way, the human mind doesn't get jolted out of the experience, the science says. Maybe the same approached could be applied to your program, too? Anything that gives you a sense of time and space should be avoided to enable the writer to be fully immersed in writing. Because every so often I looked at the bottom right of the screen and saw that I've written 1000 words and started to wonder if I've already written enough, but in reality, I shouldn't have to worry about those kinds of things. I should just write.
Does this help?
Idea: animated documents. It starts with the core points. and if you have more reading time you move the slider and more detail appears, highlighted.
That sounds like visibility cycling in org-mode (?)
The thing is in certain fields it takes dekades to be able to do it in just 4 hours — or it takes dekades to do it at all. If you don't like that, go to your nephew and let him do it as an exercise.
It is worse the more people think "they could do it themselves" — e.g. graphic design is one of those fields where customer interaction can become absolutely exhausting if you have the bad luck of gaining the wrong customer base.
Except that in this case your clients are other developers.
The longer I work on a piece of code, the simpler, more easy to read, and less error prone it becomes. None of that work is visible in the end product.
It probably looked more impressive and complex half way through.
This engineer worked for a corporation for 40 years, and retired.
A few months into his retirement, his old company frantically calls him, begging him to come in, and help fix an issue with the system he worked on. Apparently, the new team had managed to hose it, and couldn't figure out how to fix it.
He comes in, sits down at a terminal, looks at it for five minutes, and says "Here's the issue. If you do this, it should be fixed..."
He then presents an invoice for $10,000.
The beancounters go "There's no way we can pay $10,000 for five minutes work! Itemize it!"
He sits down, scribbles a bit, and presents an invoice that says:
Time to fix bug: 5 minutes. $20
Six years of college, and forty years of experience, so I can fix a bug in five minutes: $9,980Hmm, okay.
If people wonder what I’m referring to, think about it in music terms: you don’t listen to a song in just one environment usually. There are speakers, headphones, your car, etc. so it’s a good idea to mix a track and test it on a couple.
It is my perspective that coming up in a print culture the best way to slow down and be critical of self produced work is to change the medium slightly. Thus the print out. It’s the same content but you are digesting it a different way. In this regard I would consider my approach superior to Paul’s and suggest you give it a try for your next important email, letter, or essay.
I find it quite sufficient to compile the text to HTML and read it in the browser, while editing the source. It gives the same shift of context from writing to editing, and it's more closely equivalent to hand-proofing galley prints, since it's in the form the audience ultimately sees.
I would love an iPad app designed around the old-fashioned proof editing flow, using the stylus. I've never seen one and don't know if there's a market for it, but I'd use it every week.
one observation: PG does plenty of local edits (reworking over and over one sentence), but not so much changing the order of sentences, or moving paragraphs around -- these are the things that I have always been told (such as in college writing workshops) I should be doing. maybe such "macro" edits are not needed in such a short piece. and truth be told I never found much need for them either, even for pieces that are quite long.
And a lesson I got from the edit that made me laugh: don't tell people what you're not saying, just tell them what you're saying.
(Or, more succinctly: Don't equivocate.)
Multiply insights like that by every edit he makes, and assume most edits are significantly less obvious, and you end up with some seriously interesting content.
Reminds me a bit of Hunter S Thompson famously typing out a The Great Gatsby in full, just so he could feel what it was like to write a masterpiece.
I always beat myself up that I am not a quick, efficient and articulate enough as a writer and I always to take some time to organize and reorganize my thoughts and sentences but when I see I am not alone in this and there are even native speakers who struggle with this aspect of communications despite being domain experts, I feel good about myself.
It's not clear that anyone has exactly, but it's certainly something that's been talked about [1], [2], including on HN itself 13 years ago when pg's etherpad writing session was discussed [3]. (Though I can't find any evidence of the threatened minor mode.)
[1]: https://emacs.stackexchange.com/questions/47495/how-to-play-... [2]: https://www.reddit.com/r/emacs/comments/tnhvs7/replay_macro_... [3]: https://news.ycombinator.com/item?id=495751
That's the cool thing about this; there's normally not much way to know how other people go about the process of writing.
Supposedly, Kerouac wrote On the Road on a manual typewriter fed with a continuous paper web.
So, yeah, different. Instead of composing a sentence in your head and then writing it down, you waste a lot of time editing it in place. It seems very disorganized to me.
While writing, I constantly let my IRD read what I have, as if it's a stranger reading for the first time. Then I adjust until it's out of complaints.
I don't know if this is a common technique, but it's really worked for me.
Here the case is, from the first sentence, it gets rewritten 20,000 times.
I should rewrite the last paragraph it's a bit wonky but I'm on mobile and I got kids to put to bed. Such is life.
Second - I have used asciinema to record various writing and/or coding sessions in Vim, but does anyone know if there's a native vim plugin for recording "replays"/playback sessions like this? Closest I found is https://github.com/chrisbra/Replay but it hasn't been updated in some time.
The addition I would make to this process is that I sometimes let comments sit, go do something else, and if I don't feel awesome about reading them when I've forgotten them, I delete them. Usually on controversial topics that I know I've been provoked into responding to, but also on other things. The key is to write something I don't cringe at reading after. Write in the voice of the person you would like to become, and then read it to reflect on whether you aimed high enough.
When I was done watching.... a wave of nausea hit me in the face.
Did this happen to anyone else?
I would love to see a playback of him writing a more exploratory essay, where perhaps the information wasn’t as concrete in his mind before hand.
https://byronm.com/13sentences.html
Avoids the senseless friction of the twitter link.
many of today's tech giants seem to thrive with the opposite approach: offer no customer service at all.
To be clear, this is speculation. I haven't actually looked at the code that creates the recording. Just pointing out that the process of recording often alters the recording itself.