Hacker News Reader

Comment by KVFinn | Hacker News Reader

If you're not just making slow progress but literally unable to make a single bit of progress, my goto strategy is similar to what writers call a vomit draft.

For writing it conventionally means means writing words without stopping to plan or edit, no corrections allowed, the rule is you just have to keep typing, no matter what. It's about something being better than nothing, creating momentum, and also avoids being too critical because you literally can not stop and make edits to old work.

Remember the only rule is keep typing. Even if it means typing random nonsense for awhile.

I do all that but I sometimes make it even more extreme. I make it the goal to produce truly terrible version of the the thing I'm trying to make. Full of cliches and tropes in writing. Amateur coding mistakes if it's a technical project. Not just bad but legit so awful that I would truly embarrassed if somebody else saw it. Like literally, what would so shoddy I'd be afraid to have someone look at my screen right now. I mean literally ask yourself what work is so bad you would be humiliated if your advisor saw it. Make that your goal.

But it still works. After you have something even it's an abomination, it gets your brain thinking about it and working on it, and it's so much easier to make the obvious improvements, and then more, and eventually you are just doing things normally.

View on HN

I cannot recommend this enough, it's what got me over the finish line after waaay too long, including a year with zero progress writing my thesis, just like the article describes.

Each day, it went the same: sit down, try to write about the work, find some meta problem to occupy myself (like "is this the correct order of topics that I'm planning to write in my outline" or "under which topic does this idea belong") before actually writing about the thing itself, cue an unproductive spiral of research into these distractions. After a couple of hours of not actually doing any progress, drift off into the internet "as a break", and only come back to the thesis for 20-30 minute intervals that are exactly as useless as they sound. Become more and more frustrated, but no more productive, as time goes on.

Vomit drafting suddenly gave me a productive focus. Suddenly it was clear that yes, where I want to put this topic is good enough, or there is this one other place where it makes a lot more sense. No more long internal debates over meta-questions. And I realized that yes, I do have a wide range of knowledge about my specialty, something I'm convinced I was subconsciously blocking on before ("what do I do if I try to write about this topic but realize I'm just too stupid to get it, and was just bluffing all the time?").

This is legendary Simpsons writer John Swartzwelder's technique, too:

Since writing is very hard and rewriting is comparatively easy and rather fun, I always write my scripts all the way through as fast as I can, the first day, if possible, putting in crap jokes and pattern dialogue—“Homer, I don’t want you to do that.” “Then I won’t do it.” Then the next day, when I get up, the script’s been written. It’s lousy, but it’s a script. The hard part is done. It’s like a crappy little elf has snuck into my office and badly done all my work for me, and then left with a tip of his crappy hat. All I have to do from that point on is fix it. So I’ve taken a very hard job, writing, and turned it into an easy one, rewriting, overnight. I advise all writers to do their scripts and other writing this way. And be sure to send me a small royalty every time you do it.

https://www.newyorker.com/culture/the-new-yorker-interview/j...

I think this a role tools like ChatGPT will fill: spit out a crappy version of something that humans can edit later. It gets you over the hump of looking at a blank page, but isn't itself a deliverable.

I plugged the elevator pitch of "my novel" and asked it to write the plot of that novel (I will never write it, I just like to think about the elements of it and have some notes and snippets) and it spat out almost exactly what I had written out as the longer form of the outline.

I laughed, because I knew exactly how original my thinking was going in. It was really funny (and depressing if I had a big ego) how close it was to my idea.

However, doing so allowed me to iterate a bit and I took a minor aspect of the plot (people don't always have enough to eat) and magnified it (the government is the only source of food and they use this to control people). It has made the story much more interesting (still not super original!).

Last week I was experimenting with ChatGPT for some of the sorts of trade pub columns I sometimes write. I wouldn't have given anything it created straight to an editor--though I've seen worse. But it gave me some bulleted paragraphs that I mostly agreed with and could certainly serve as a starting point for revising and fleshing out with more detail, maybe some numbers, quotes, links, etc.

While it can head you off down dead-ends, having some structure and words to start with tends to be easier than a blank sheet of paper.

“Homer, I don’t want you to do that.” “Then I won’t do it.”

I'm sure that a back-and-forth with Julie Kavner and Dan Castellaneta doing those exact lines would still give me the biggest side-splitting laugh of 1996.

That explains much of the Season 7 onwards...

This sounds like great advice. In an attempt to see if it works for me, I thought I'd give it a shot, so I set a timer for 2 minutes and started typing a response without a real idea of what I was going to say. I'd say it's pretty effective, but I do tend to fill in a lot of meaningless content as I think of what to write next. I guess the next iteration of edits can trim that out though. I could have used this in the past, and could definitely use it right now for writing some documentation that I am procrastinating on indefinitely. Thanks for the tip. This has been helpful. Not sure if I should actually post this, it's not really contributing much to the conversation, but this is an anonymous account of mine, so who cares. Timer is up.

Nicely done! Not sure I could tell this was not just another HN post if you didn't explicitly say so.

Voice dictation is helpful for this exercise even if you type fairly well.

I highly doubt it. It is the ACT of actually writing that forces you to think what you are doing. Using a voice assistant kills that energy. In fact: even if you -gasp- use pen and paper it would be even more effective than typing. Drop the PC, the smartphone and disconnect when doing this.

> It is the ACT of actually writing that forces you to think what you are doing.

This is incorrect. Everybody is different.

As an ADHD sorta guy, I write much, much better if I can walk around and get my blood moving while dictating compared to sitting with my butt going numb and my brain screaming out in oxidative stress agony as I try to think clearly.

You can take it even further by dropping the pen and paper and use a chisel and stone tablet. </s> While I agree about removing distractions, I would suggest using whatever your most efficient way of communicating is to get the idea or concept out of your head.

I love that this is a substantially nicer read than the kind of stuff GPT generates, and the amount of human effort to produce it is pretty close.

This is genius. Anecdotally I’ve seen a high correlation between procrastination and perfectionism. Once you have a vomit draft I can see how the perfectionist bent suddenly works in your favour.

I agree 100%.

When coding I have a rule: "first make it work, then make it pretty".

The MVP can be a complete spaghetti piece of crap, just make it work. When it works, you can iterate on making it elegant.

Oh, and if you work for a Corporation, don't tell them it's done when the crappy MVP finally works. You're still "working on it" if they ask.

Otherwise the spaghetti crap will end up in production.

  1. Make it work
  2. Make it right
  3. Make it performant (if too slow).

I disagree.

1. Make it work 2. Make it performant (always, but within reason) 3. Clean up your mess to prevent future mistakes

I claim that "if too slow" can't be judged because you almost never have the slowest possible target device.

I disagree with all of you.

Make it work. Then make it maintainable. Then polish it to make it more ideal.

When it works it can be spaghetti and lack testing and filled with hacks. After you make it maintainable it can still be less than ideal — some hacks here and there in leaf functions but things are localized and understandable and non-brittle (ie good testing coverage, design which has appropriate abstraction, refactors to promote test ability, no more spaghetti etc etc). Then in the final step you polish the stuff to gold (ie use better solutions or improve performance). It is the only way.

  1. Make it work
  2. Make it understandable
  3. Make it performant (if too slow).

Except I've seen too many spaghetti code craps in production to actually recommend this. There is a plethora of reasons why colleagues (or even you) will end up not waiting until code is production ready (e.g.: A new nice task comes in which you'd rather work on, you get reassigned, you get sick and someone else has to finish, you get a great opportunity to show off and say "hey I'm already done no problem", a tester sees and tests the features and misinforms the customer, etc. etc.) I am not saying your code needs to be perfect and then you only notice the one great flaw which requires re-engineering after weeks of hard work. I am saying find a good (backed by TDD) middle ground of fast work and fitting into existing / designing a decent architecture.

My 2 cents only, and if you can make it work that's great. But I will never go ahead and recommend this to juniors or even younger seniors.

As someone who otherwise works somewhat similar to OPs workflow, I have to agree. One golden rule for me: The "vomit draft" version cannot be your MVP. It is never Viable for production.

That said, TDD (with emphasis on driven) is no silver bullet either, especially when working that way - the tests are equally affected and I've seen enouph code bases where the initial test surface was so off compared to what was needed/sensible that it ended up being solely a hindrance and stifling to actual rework/refactor - with people ending up throwing it away and rewriting the tests from scratch, or worst case, simply setting them to be ignored on build.

My variation on this. Make it work then make it slightly less horrible.

Then repeat until you are proud of the work.

My teammate says he does this. Writes everything three times, and is careful not to let _anyone_ see the first two, lest somebody tell him it's "good enough."

> Oh, and if you work for a Corporation, don't tell them it's done when the crappy MVP finally works. You're still "working on it" if they ask.

I would if I could, but schedules are schedules and stuff has to ship...

I'd go a bit further, it's more like we lack an evaluation compass therefore we can't even try because we don't know if we will be able to see a difference. You can't anticipate benefits of any path and most of the time it's anticipation that drives your effort (you "foresee" things). Dealing with that situation requires changing this habit and go wander, try to find anything that revives curiosity, gives you a clue, a perspective you didn't have.

Vomit draft can be great tool, but there's additional risk in science to keep in mind...

I've heard of a person who advocated this kind of draft, and then did a vomit draft themselves for a research project, with the draft including discussing stand-in results of an experiment that they planned to do.

Their colleague on the project saw the draft, and called out that the experiment had not been done. The writer said that they'd only written the draft to see what the paper might look like, once the experiment was done.

But the person who wrote the vomit draft had a problem doing the experiment...

Then that person submitted the paper anyway, with the vomit draft stand-in experiment results, and the paper got accepted. And without telling the other people on the project.

I understand that the story got a lot worse from there.

The vomit draft wasn't the only problem, but if you see someone vomiting drafts that are effectively scientific fabrication as they stand, I advise being uncomfortable with that, and emphasizing to the writer that this is a more dangerous practice than it might seem.

I think the problem here is that they submitted a fabrication. If someone is prepared to do that, I don't think how they drafted the document is really the issue.

I think, the problem here stems from making vomit draft a social technique instead of keeping it at a personal level. It is potentially harmful if other people can read it. OP describes it as a personal method and it may work by breaking some psychological issues, like a fear of under-perform or something like. But to do this one need to overcome all fears, to get rid of anxiety, and it needs a safe environment where you can write anything. Literally anything, to try it and to show to your mind that it is harmless thing, nothing bad happens, it is ok to write not good enough, because you can dump it later.

The problem is not the technique, but the inmorality of who publish it as something it is not.

DO THIS. I have a friend that was doing a PhD and procrastinated and they kick him out after the ten year time limit. True they gave him an MA for his work, but still that was ten years of his life wasted stressing out about procrastinating about not doing his PhD, that could have been spent at the beach with a martini and a girl in a bikini.

Wait, that's the alternative to doing a PhD??

where do I signup

The most common academic titles are:

Bachelors - Masters - Licentiat - Doctor

A Lic. is not use din the US, but very common in Europe https://en.wikipedia.org/wiki/Licentiate_(degree)

And is basically what you can take out after half a Ph.D. program. In many technical schools in Sweden you are expected to do Lic. on the way to Ph.D. or if you burn out you at least have a Lic.

I think GP is asking about the M&B, not the MA.

Licentiates certainly aren't commonly used in the States, but there are several programs that offer ecclesial licentiates like Licentiate of Sacred Theology or Licentiate of Canon Law, which are each required for work in certain areas even in the States.

In 30 years of academia in Canada -- I've never heard of this requirement....

Beaches...bikinis...martinis.

Heh, I met a chap doing a PhD in maths at a decent university in Northern England (no names), he was in his 8th year: He was funded for the first 3 years, managed to get another year of funding when that ended and subsequently lived off his parents. By year 6 they were a bit sick of this and stopped, so he had no money for rent so started to live in the department, working until late in the evening and then sleeping in a cupboard. He'd been doing that for 2 years when I met him. Perhaps he's still there ...

> that could have been spent at the beach

As a procrastinator myself, this is actually a self-poisoning attitude that I've worked on removing, in the spirit of being kind to myself. (Procrastination tends to get worse when you have self-hate spirals.)

It's so easy to look back on wasted hours and think "I could have been doing anything better than that." I use to have days at work where I didn't accomplish much, and say to myself "I'm a horrible father because I could have spent today with my kids."

It's true that procrastination eats away hours of our lives, but there's no reality where most of us would have been able to say "I know I'll procrastinate the next three days, I'll take a beach vacation with those hours instead, and then do the job promptly the next day."

This kind of (common) thinking is no better than any other kinds of regrets in life, staying awake and thinking "what if I had asked that person out in high school?" etc etc.

But YOUR thinking is a self poisoning attitude (jesting a little with you btw :), because it creates a sense of complacency. If we regret those moments -- and we should -- we will be all the more vigilant not to waste them in the future.

Regrets are THE most useful lessons our brain can give us. If I HAD asked the person out in hughschool maybe I'd be happy now and rich. Or whatever. Next time you meet a hot chick that you love -- you won't make the same mistake twice. Ask her out!

This is definitely good advice for fleshing out the actual writing, I used this ontop of to the paragraph planning. So you build out a mindmap style plan with part > chapter > section > sub-section > sub-sub-section / figures / tables > couple words for a paragraph. Then you manage everything at the lowest level, you work out how many paragraphs you need by when. You pick paragraphs at random even if you dont have the results or good stuff to say you write the vomit draft of that paragraph. Then you pick just one thats already written and edit it. Repeat until you run out of time. It's the only way I managed to finish my thesis. The other thing I did was always do a writing warm up, so just something free flowing off the top of your head but still a bit technical. Maybe some instructions on how to make a cup of tea for an alien in low gravity. Without a plan though you can't validate you can actually finish, which is insanely important for motivation. You need the concept of % progress and time left. Also you need to know you've done enough work to actually write a thesis or if you need to do more work, and what work you actually need. The definition here is also a paragraph is about 3-5 sentences, a sentence being around 10 words. It is extremely hard to fail to write 50 words. That sense of failing to write is the mind killer. After a session even if that session lasts 20 minutes and you only wrote 50 words you update the spreadsheet showing vomit to edit progress. Until I did all this my progress was almost nothing.

This works well for writing. Writing, though, comes at a later stage after you've done at least some of the actual work (code, experiments, whatever). It's the work that people have most trouble with. The ideal trajectory is something like MVP -> write -> iterate. People usually get stuck in the MVP phase.

I don't have a good solution to this, except that at least from personal experience, prolonged procrastination is an indication that you don't like the work. And this can be difficult to acknowledge because maybe you liked it earlier, or thought that you did. And then ego becomes involved. Knowing what you like and don't like can also be quite tricky. Sometimes, the only real solution is to quit and do something else.

Yeah. I'm not a particular vomit draft or mindmap or whatever proponent for writing. But the original post seems to be more about actually doing the lab work for the thesis that's a prerequisite to writing something.

That was where I was with my undergrad thesis. There were dependencies on other work in the lab which wasn't progressing quickly and I just had a lot of trouble getting to the point where I had something to write about. The writing wasn't the big deal.

You know, I'm fairly sure that there is more than just procrastination at play. This feels like some human mal-adaptation where we do anything (or nothing) to avoid dealing with That One Terrible Thing, whatever it may be.

Throughout my career I've had this feeling or state of thinking multiple times, whether I had to work with legacy code, uncooperative individuals, or just a week ago, work with a bad DB schema and write complex queries against it.

In my case, procrastination was just a consequence of wanting to do something but I very clearly also remember just sitting and staring at the task at hand, doing absolutely nothing, because I couldn't stomach how bad everything about the task that needed to be done was - bunches of redundant tables, illogical linking of data, lots of overcomplication, no documentation and no examples of helpful queries whatsoever.

> If you're not just making slow progress but literally unable to make a single bit of progress, my goto strategy is similar to what writers call a vomit draft.

In the end, just forcing myself to get started, writing out the dozen of different things I needed as a part of the query and then working backwards through everything, was what worked. It took hours of uninterrupted work, I felt miserable throughout it, but I got things done in the end, all because of that decision to actually work on it and deal with the pain and suffering, very much how someone would need to make the leap to dive into legacy code, or an issue for a project that doesn't have monitoring or instrumentation, or writing a thesis.

I think that's why techniques like Pomodoro also get recommended, because if you trick yourself into saying that you'll only do a bit of suffering (work on the horrible thing) now and will take a break later, it's more tolerable: https://todoist.com/productivity-methods/pomodoro-technique

I'm just writing this because to me it feels different from how people commonly view procrastination: just getting distracted and wanting dopamine, as opposed to being able to stare at the computer for an hour without doing anything, just because doing the thing would be horrible. The latter feeling makes you want to quit your academical program (which seems like what the author is dealing with a little bit), or maybe draft your resignation and leave the job market for a bit instead of dealing with the codebase or whatever (which I did, albeit for different and less negative reasons in my case).

I'm also experiencing this currently. I can sit in front of my thesis and stare into the screen for a whole good day. I also can distract myself with a smartphone but that's merely a distraction from the boredom. Around the topic of procrastination, I read "doing the thing would be horrible" for the first time and it just rings so true.

Doing this thing (this kind of work) feels horrible and I know I'll need to do it for many, many more days and weeks and these will all be horrible. There's no way around it. I often subconsciously try to make this time less horrible by experimenting with listening to music, have the TV running in the background or other things. This never really works. Or I just haven't found what I need.

At the end, whenever I do get a good chunk of work done, I feel really good about it. But I also acknowledge that it was horrible and I need rest now. And some dread builds up, reinforcing that the work is indeed horrible and the next chunk will require me to go through it again. The good feeling of getting work done does absolutely nothing to knowing that the work is horrible while you are doing it. It's the type of work (writing, editing) that I detest, not so much the content/topic.

I have no idea what strategy can help here (I welcome suggestions!). Strategies such as Pomodoro do not help me. A 5min break doesn't change that the work is horrible and I'll be doing countless Pomodoros throughout the days/weeks anyway. The amount of horrible work is not reduced and the little breaks don't make it less horrible, so it doesn't help me.

What works for me sometimes is tricking me a bit. Just change this one sentences here and... this one also looks really bad... and when I'm at it, this figure there could use an overhaul... and suddenly you are working on your thesis. Key point here is - I think - that you don't go at it with the intention of doing actual work (which you know is horrible). You just change this stupid sentence there because your inner perfectionist wants you to. Thinking about it like that, the horribleness associated with the work may be a state of mind I can work on. No idea how to though.

I think it's Cunningham's law at play.

Correcting something is much easier than coming up with it.

Reminds me : If you want an answer to a question on the internet, make an incorrect statement.

Yes, that's Cunningham's law. I generalized it because I genuinely believe it still holds, and applies nicely here. Sorry if my comment above was confusing.

So publish your vomit draft on the internet to have it improved for you?

Tangentially, this is where I see a ton of potential in these new ML tools like ChatGPT. If you're expecting it to produce finished work, you'll often be disappointed, but if it can even produce a bad version of what you need, the time saved could be enormous.

Having written 3 dissertations, a bunch of research papers and writing code daily, I'm a huge fan of this method. Sometimes I just create bullet points stating "here i need to talk about this and that" adding a bit of actual information whenever it comes to mind. If I have to write a piece of code that I really don't feel like, i use a mix of comments and meta-code.

The procrastinators among us have now an excellent way to turn their "vomit drafts" into a full fledged document. GPT-3 is able to transform bullet points into something that you can call a draft. I'm still not sure if that's a good or bad thing.

If a tool can help people create genuine research, then that tool is good!

However, I worry on both the plagiarism and correctness fronts. When you aren't writing it by hand, it might be easy to miss both.

This is going to sound ridiculous but I’m serious: if you can’t even type a whole bunch of stuff, type just a basic (terrible) outline and then feed it to chatgpt to expand on.

The output will be laughably bad, most likely, but you’ll know why it’s bad and you can fix the obvious mistakes and then go from there.

I was going to make this comment but you made it for me

> I do all that but I sometimes make it even more extreme. I make it the goal to produce truly terrible version of the the thing I'm trying to make.

You know, I've read and heard all sorts of advice on writing, and this is the first time I've seen this, and I think it's brilliant. Undergrad me absolutely should have done this, 'cause the perfectionism I grew up with took a long time to unlearn.

Do you have an opinion on getting drunk during the process to lower inhibition? Not a thing I've ever tried, but I've definitely heard of people doing it.

ADHD happens to correlate with a higher rate of addiction, substance abuse.

I would not recommend getting drunk as a tool for getting things done. The problem is it might work. And if it does, you don't want to make a habit of reaching for that tool whenever you find yourself procrastinating. That way lies aforementioned statistics.

I had a struggling Lit friend in college who I kept tabs on. She was a late adopter to blogging but she talked a lot about how writers find the motivation to write. It was an early part of my long process of discovering that everyone, everywhere is struggling with the same problems, we just call them by different names to feel special. Writers have tricks that are fundamentally Refactoring, possibly before we had a name for it.

I'm not a Hemingway fan, I don't enjoy his writing and I have the impression that he was a terrible human being, but his advice to other writers is apparently pretty good:

> Quit [for the day] when you know what happens next.

If you want to set yourself up for failure tomorrow, write until you're completely out of ideas today, then go do something else. In the morning, you'll dread going back to the piece because what do you have to say? Anything? Who knows. Maybe you've written everything good you'll ever write. Maybe you're a failure. Maybe sitting down at the typewriter/keyboard will prove your mother was right and you should have become a doctor. So much pressure. I'll go do something else and pretend everything is fine. Or I'll read about the writing process and hope that serves as sharpening my saw while I wait for magic.

When you stop with one item on your TODO list, you have at least one idea left, and having slept, you're more likely to pull on that thread and find two other things you want to write as well. You just have to get started and the creativity will happen. Starting is the hard part. Starting is what kills you before you've even tried.

It works for other things as well. I needed to create a „customer portal“ web application for a product I’m working on, and kept postponing it, because I don’t enjoy working on it. The pressure to have it functional became so great that I literally committed a generic dashboard template with just a couple text changes, truly ugly, vomit-style stuff. But it helped me overcome my blockade and start working on it in earnest (and actually enjoying it!).

First time I heard about this. And I think I really really like it.

I have a huge writers block for anything I'm good or within my specialized field, it's kind of strange. Especially for scientific work this is extra bad, not being able to write, no matter how much time I dedicate to it.

I'll take this approach to heart next time, thanks. I think this is right up my alley, including the funny name for it.

There's something there, corroborated by two examples in cinema and litterature: Finding Forrester (https://m.youtube.com/watch?v=zLBEFvMkQCo) and in Zen and the Art of Motorcycle Maintenance, with the rhetorics professor facing 'my students are stuck write even the simplest essay' problem.

Yes, write something, anything, then edit. These are two different states of mind, two completely different personality parts, and I wish we taught the first one better (and the second one too...). Shut down the critics part when writing. Should down the ego part when editing.

> my goto strategy is similar to what writers call a vomit draft

A "vomit draft" might be good approach to a writer's block. But I think the blog describes a burnout. You probably need actual help to recover from a burnout.

One thing that works for me is to just start writing figure legends. That seems easy, it doesn't feel like "writing the paper". I end up basically writing the results section for that figure or table, so I just cut and paste most of it to the actual results later, and keep a summarized version for the actual figure legend. By coincidence, I'm trying to submit a PhD thesis today (Genetics/Computational Biology).

That's great advice. I do something similar with art. I draw 100 poor sketches/drawings that are meant to be really bad and then eventually the good idea pops out without trying.

I think this has something to do with avoiding triggering your amygdala, because once the stress sets in you get a fight or flight reaction and then creativity gets completely shut down. Dance around the amygdala and you'll be creating cool stuff.

I do a similar thing when I’m writing. Except I write on paper and allow myself to make revisions. Being on paper is pretty self limiting as far as edit spirals.

The other trick is when you are stuck instead of writing garbage write down why you are stuck and describe what would go there if you weren’t stuck. Sometimes this gets you to the solution, but if it doesn’t you still have your thoughts on it for latter.

I do something similar. There even used to be a lovely website that had a text entry box which would delete words if you didn’t keep up a certain words per minute. It sounds daft but it was actually useful to use at least a couple of times to train yourself to keep writing, even if it was garbage. I think now I might be tempted to use dictation to reduce barriers ti getting words down even more.

Sounds like The Most Dangerous Writing App! https://www.wired.com/2016/03/sadistic-writing-app-deletes-w...

direct link : https://www.squibler.io/dangerous-writing-prompt-app

"There’s one solution that each and every book on writer’s block offers: write five words. Any five words. Follow this advice, Mr. Ashbery, and you’ll never have writer’s block again."

(From "Uncreative writing: Managing Language in the Digital Age" by conceptual poet and "word processor" Kenneth Goldsmith)

Edit: Oh, this historical paper is slightly different angle, but also relevant, helped me quite a bit:

Upper, Dennis. "The unsuccessful self-treatment of a case of “writer's block”." Journal of Applied Behavior Analysis 7.3 (1974): 497. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1311997/pdf/jab...

I'd like to try this, but it's tougher when you're collaborating on a shared document.

For example, I collaborate on papers that always start out as a shared Overleaf doc. When I'm starting a new section, I just want to toss some ideas down on the page, knowing that the prose is terrible. I can really enjoy this phase if I know it's free of judgment. However, it's so much harder when I know that someone might open the doc in an hour and judge my writing before I get to edit it - or worse, if I see that they've opened the page while I'm in the middle of it, or god forbid see their cursor click around near where I'm typing. It's a lot harder to concentrate until they leave or start editing somewhere else. I end up taking a while to write each paragraph, planning & editing as I go.

So write it locally in an unshared document and paste it into overleaf when you want someone else to see it? I don't like overleaf to begin with, but to feel this tied to it is a bit wild.

I've tried, but it only works well if it's the beginning of a project. Later on, it gets annoying to sync back and forth between copies as others work on it, since I like to see what I'm writing in the context around it.

Amazing advice. This technique is what got my PhD thesis across the line.

Word vomit and just concatenating together all my current work into a single LaTeX project. Once it existed as a single document and I could start to “see” it as a thesis, even though most of the writing was terrible, the momentum was there.

Currently writing my MS thesis. "Concat all my existing work into a doc" was my initial move as well, and I discovered I had written a little on almost every topic already. In some cases, had written lots.

That let me move to outlining and shuffling things around almost immediately, making writing about filling gaps instead of staring at a page. That in turn let me evolve my writing along with the codebase; now, in late January, I'm well ahead of schedule on written material (knocks on wood)

Also known as "write drunk, edit sober" * - where your choice of drunk may be defined as 'producing text unsuitable for others to read in that form'.

* Hemingway may have coined this term, along with "make sure you are 100% offline because the Internet never forgets".

The equivalent of this today is getting high and using GPT3. I don’t think there is a name for it—yet.

"Get high and unleash the AI"

I was thinking something like “the cyberbong draft”

> even more extreme

I'm a musician & amateur home producer, and this extreme approach of yours is exactly what I recommend to people (and what I do myself) to overcome the feeling of "I can't get started because what I start on sounds bad and I have an overwhelming fear of making something that sounds bad." Turning it on its head by purposefully trying to make the cliched, trite, and generally whack song I can, can be immensely freeing.

When I wrote my Master thesis, my vomit draft was written on paper with pen. Safe to say it was horrendous and even full of rant and swearing. Yet ultimately I managed to write much more words than directly on LaTex.

I find that I get stuck in an endless loop of trying to make something perfect and it massively slows me down. I can spend hours changing a paragraph a million times… it’s extremely frustrating!

That's what I did as well.

The problem often is fighting the urge to correct things. For me the only thing that helped was sleep deprivation, but probably alcohol has a similar effect.

I would love this to be turned into a text editor. You set up a target, e.g. 20K chars and start typing away. No formatting, no distractions. The program always just shows the current line, or even just the 10 characters, while previous lines fade out into black with no way to reveal them. Only after hitting the target is the entire text revealed.

Optionally with some reward animation/sound after every target/10 characters.

I used a somewhat similar technique after I got stuck in my PhD for more than a year: I sat down and wrote a chapter explaining why my research was doomed, because XY could not be properly measured and quantified etc. It turned out that once I had my concerns spelled out on paper instead of tossing them around in my head, proceeding to the actual measuring and quantifying was anything but impossible.

This works wonders, especially for collaborative documents (which a thesis should be, a collaboration between you and your supervisor/other people). A first "vomit" draft allows you to at least discuss the concepts, and it forces you to align the thoughts in your head. It is how I have written all my papers/thesis, and I cannot recommended it enough!

This might be dangerous advice, but consider interactively using a language model in writing "vomit draft". I tried copilot recently, and even though some of what it suggests is "no, no, NO!"-material, it really helps. I hadn't known how much simple writing block - struggling to write the easy parts - kept me back.

I took a similar approach with my thesis, I used voice typing while writing my first draft. The other approach I use in combination is writing every heading and subheading to create a document outline, then fill the gaps. The power of these approaches are breaking down the work into smaller manageable chunks.

This is also recommended by Donald Knuth in one of his technical writing guidelines [citation missing] so +1

You have my curiosity. I am trying to isolate what would make this work.

1) Do backspaces count towards corrections?

2) Would you say this also applies to coding?

3) What seems to be implied is, that this generates something I feel lends itself to "obvious improvements" instead of just throwing away and starting over. Your thoughts?

For this, I highly recommend The Most Dangerous Writing App: https://www.squibler.io/dangerous-writing-prompt-app

It deletes all your work if you don't keep typing. Good motivator.

Yes! Hard approach to take for a perfectionist but it’s the only way that works for me too. GitHub receipts for latest project show a lot of regressions. Embarrassing. But better than nothing, and finally got the MVP done. It was your approach that made it possible.

This right here is "worse is better" in practice!

A thing in the world may well be the worst possible example of that thing that can exist. Yet it exists, so has a couple of key properties:

1) it does something (albeit poorly), not nothing

2) it can be improved upon.

"[discussing his last exam] Last time I only failed by the narrowest of narrow margins"

"You what? You walked in there, wrote "I AM A fish" four hundred times, did a funny little dance and fainted!"

- Red Dwarf

You don't even have to type. Text to speech is fine for this and the barrier to getting started is nil. Make a rough outline of what you want to cover and have at it.

That’s basically evolution vs design. Design allows you to be stuck in planning mode forever. It’s noble even, because you want to do it right.

And if you're so stuck you don't even know what to do, you can always clean the space around you.

Wow ! That sounds like a very useful technique to me. Will definitely try to adopt that. Thank you.

Forgot who said it but 'shitty first draft' stop thinking start doing

"Write drunk, edit sober."

Then have ChatGPT do the editing :)

Reply on news.ycombinator.com