It's like making pancakes. The first pancake is always going to suck. Just accept it and get through it so you can make a great batch of pancakes, or blog posts, or products, or...
I've heard lots of people say that, but it almost always just means the pan isn't hot enough! Wait an extra minute and the first one will usually be fine.
I'm not sure how that translates back to writing, though...
I pride myself on my pancake-making ability, and I completely agree with you: The pan is usually not hot enough. What I like to do is use Kerrygold butter, turn the heat on the burner higher than it should be, and wait until just before the butter starts to burn. Remove the pan from the heat, turn the burner down, and put your first batch in the pan. Also, make note of the temperature settings you use for next time.
The butter itself can also be your guide, the best temperature is a little after the butters smoke point. Too hot is better than too cold, as the cold batter will drop the temp anyway.
Especially when you use stainless steel pans. That shit gets stuck. Just heat up the pan, throw some oil on there, and watch it change. The oil usually looks a lot less viscous and that's when you know it's ready.
Woah, that's a fun one to play around with. Pancake must be commonly used in idioms. Try deleting each word individually, it's pretty hard to get it down to pancake.
Perhaps, but jumping straight into details without a general sketch of the "big picture" is also often counter-productive. What is the purpose of this function? What is the goal of this board game? These set a frame that allows the reader to set the details in their proper context.
Taken literally, the style is horrible. But as a concept, particularly for a complex work, laying out the structure in advance, and periodically touching base to detail what's been covered and what's yet to come isn't a bad idea. Of course, in a text, readers can accomplish this for themselves by skimming the contents, flipping to the end, and examining the notes and/or bibliography, and index (assuming a book-length treatment). I'll frequently skim an article, essay, or book first, to get a sense of its structure.
Reading and re-reading is another approach.
I get a stronger sense that such "here is the structure of this book" type introductions are born from initial outlines which guided the author. In many cases they can be significantly streamlined, but they're not entirely useless. Among technical books, the O'Reilly formula is generally pretty good -- it provides guidance but isn't overly pedantic.
In so many endeavors the hardest part is beginning. Starting a new job. Starting an exercise program. Quitting smoking. Leaving your comfort zone to do something new. For many people, starting a new software project can be intimidating: the open green field of endless possibilities can be paralyzing. Starting a machine learning class has made me feel dumb at first, so I quit.
The trick is to just start, and understand that you may stumble, and when you do know that you will learn from the failure and keep moving forward. Wanting perfection immediately without the failure first is where we usually go wrong.
As an extension of this I've also found asking questions helps with the growth mindset.
Often people are afraid to ask questions because they don't want to look stupid or admit ignorance, but explicitly admitting ignorance and asking what things mean or how things work in conversation actually has the opposite effect. You learn a lot faster and more often than not people feel more comfortable because you're being honest with what you don't understand.
The side effect is that you usually end up looking 'smarter' anyway - somewhat ironically not asking questions because of fear of looking stupid can lead to it becoming clear you don't actually understand something you're pretending to which then does make you look stupid.
I've seen two negative things that can come out of this - the first is the person that feigns surprise and says 'You don't know what X is?!". More common in college and around other people who are insecure about their own intelligence, best response to this is to admit ignorance again and re ask the question (usually becomes apparent that the person doing this doesn't know what they're talking about anyway and can't explain it well). I think this negative attitude is especially bad because it can disproportionately affect women in early CS because they are probably already worried about fitting in there anyway.
Second negative thing is that asking questions when you do know the answer to show how smart you are is not the same thing as admitting ignorance and asking questions to learn - it comes across as obvious and is annoying.
Long answer, but generally agree that growth mindset and treating negative experiences as 'learning experiences' is the way to go.
starting a new software project can be intimidating: the open green field of endless possibilities can be paralyzing. ... The trick is to just start.
This is bad advice unless you are a complete novice. Industry experience seems to weigh in on the side of focused analysis: you need to better understand the problem space and break a potential implementation down to a relatively simplistic first stage goal.
Relevant quotes:
Any attempt to formulate all possible requirements at the start of a project will fail and would cause considerable delays. - Pahl and Beitz, 'Engineering Design' (2007)
Sometimes the problem is to discover what the problem is. - Gordon Glegg, 'The Design of Design' (1969)
In practice, designing seems to proceed by oscillating between sub-solution and sub-problem areas, as well as by decomposing the problem and combining sub-solutions. - Nigel Cross
Success really depends on the conception of the problem, the design of the system, not in the details of how it's coded. - Leslie Lamport
You're not going to come up with a simple design through any kind of coding techniques or any kind of programming language concepts. Simplicity has to be achieved above the code level before you get to the point which you worry about how you actually implement this thing in code. - Leslie Lamport
It's also a question of how one approaches writing: pansing vs plotting. Some authors, like Stephen King, swear by pansing and begin by forcing themselves to produce a certain number or words; whether it starts with character development or the plot, they exercise their powers of putting words on paper. Some authors, like J.K. Rowling, take a different approach and begin with sketching, diagraming, etc. I don't think the two are mutually exclusive, though.
Sometimes, even in the same piece of writing, I bounce back and forth between the two strategies.
"Pantsing" is the correct verb. It means writing by the seat of one's pants, i.e. just making it up as you go along. It is recently coined jargon by the online writing community.
Something similar works when coding too: begin with ugly code that only solves a simple case. Then keep iterating and refactoring until it is good. Trying to get it right from the beginning just means you don't get started.
For me it really helped to use a simple editor. Focuswriter [1], Sublime with a simple theme for Markdown files [2] or just good old notepad (the digital or analog one). Or use WordStar, like Sawyer and George R. R. Martin.
Your good start sentence will come after "about 2-3 sentences in" it seems too optimistic, maybe for experienced writers that may be truth. I more pruned to believe the "Accept you will get it wrong the first nine times and only by the tenth time you'll succeed" - Edward B. Burger
Really: just write, then revise. Nothing special about the beginning. What hung me up as a writer for much of my life was not realizing I should get a rough draft down instead of editing as I wrote.
Summary: "just start typing. ... delete until the first sentence is suitable to begin the text."
That's a useful training aid, but shouldn't be necessary every time. After a few tries, starting an essay or post should be automatic. Writing a coherent essay is usually taught in high school. Strunk's classic "The Elements of Style" remains useful.
This problem afflicts some how-to videos. Starting with "Hi, how are you" in a one-way medium is pointless. How-to material should start with what you're going to do, go on to how to do it, and end with a recap of what you did.
> After a few tries, starting an essay or post should be automatic.
That's my experience for the kind of essays they expect in college, but I when I want to write something for public consumption it's going to take more effort. The coherent essay that you learned to write and high school and served well in college may be informative, but it's still going to be boring. Capturing interest is still hard even once you have the mechanics of essay structure down.
Or at least it is for me. I'm less than fond of Strunk & White. These days I'd recommend Style: Lessons in Clarity and Grace over Elements of Style.
As for the "hi", that actually does serve a purpose in one-way mediums. It takes time for our brains to adapt to the speech of a new speaker, so it's common to miss the first thing that someone says. But if everyone says "hi" first, then your brain has a split-second to adapt to new voices before hearing something new and interesting.
This tangles up a few seemingly related questions:
1. "How to begin writing a text." How do you get yourself going so you aren't stuck on a blank page?
2. "How to write the beginning of a text." What is the process you take to determine the first sentence of a piece of writing?
3. "How to begin a text." After you've gone through that process, what kind of opening sentence should you end up with?
The suggestion to just delete the first couple of sentences is an OK one. I think the important part is that it helps solve the first problem—how to get moving at all so you are past the blank page. And the key insight is to realize that everything is still mutable. Writing is not committing. Editing isn't either, for that matter. Only publishing is. Until then, it's all unfired clay and you can play with it to your heart's content.
I think deleting the first couple of sentences is a moderately OK suggestion for the second one. Deleting any sentences that don't carry weight is always a good thing, and the first few are likely to be unuseful. But there's a good change that the next few sentences need work too. In a first draft, every line is suspect.
I don't think the article tells you much about what a good first sentence looks like. I don't necessarily think it needs to have "useful content". When I write, especially for the web where attention spans are measured in nanoseconds, the entire goal of my first sentence is to get you to read the first paragraph. The goal of the first paragraph is to get you to read the first section. Then I start getting to the meat of the piece.
Often that means the first sentence is a hook. It may be deliberately emotional to get you excited about continuing to read. Or it may raise a question you want answered or claim to teach you something you want to know. Sometimes I just try to use interesting enough language or imagery to get you excited about the pure joy of reading.
Once I've done that, I've got a little more slack, a little more of your good faith, that I can then use to draw you through the rest of the work. If I'm trying to do something like explain how a garbage collector works, or describe a feature of a programming language you've never heard of, I need that initial burst of interest to get you to keep moving.
Then it's up to me to make the rest of the piece satisfying enough to be glad that I sucked you in.
'Start in the middle' is only half the answer. An introduction has to teach the reader enough about the topic to care about the main idea (i.e. exposition in fiction). You can start in medias res (a la the iliad) but you may end up circling back in a flashback (a la the odyssey, odysseus telling his journey to menelaos' court).
Starting in the middle is great if your reader already knows the backstory, but you alienate everyone but your core audience. That's expensive whether you're writing a blog post or a hollywood blockbuster.
It's a fair assumption that your reader came to you on the promise of the title of your piece. You have to fulfill the promise of that topic or at least give a tidbit soon enough that your audience doesn't give up and leave. You can tell a book by its cover because books with misleading covers don't sell.
Balancing exposition against delivering on your genre promise is a 'theory of mind' skill that the best writers have in spades. Think of the first 10 minutes of star wars: scary villain darth vader (check) strolling through a sci-fi spaceship (check). Gun-toting strong princess (check) needing a rescue (check) sends her sci-fi robots (check) to get help. Next scene: crappy teenager main character (that's your audience, check) buys same robots.
The order of the villain / princess / hero scenes are optimized to serve the 'message for help / hero to the rescue' exchange. If this were a blog post you'd still have that 'problem / solution' structure but in both cases you need to be checking all your genre points and delivering breadcrumbs of the genre or topic that the reader is here to consume as you inform the reader enough to care about your argument.
> You delete until you reach the first sentence
> that feels elegant and suitable to begin the text.
> Usually you will find it about 2-3 sentences in.
Judging by most articles I read, I would say it's more like the first 2-3 paragraphs in.
53 comments
[ 3.3 ms ] story [ 55.0 ms ] threadI'm not sure how that translates back to writing, though...
Perfect batches with no waste!
Shake on frying pan
If droplets wiggle and bounce -> pan hot enough.
(I know you know that...)
Interestingly, Google actually translates this to "better luck next time" instead of "the first pancake will be a lump/clod"
I love the Russian expression and we should steal it.
Perhaps that was the first cloddy attempt?
"First, tell them what you're going to tell them. Then, tell them. Then, tell them what you told them."
Following that structure leads to terrible, boring writing. The only useful part is the second sentence.
Taken literally, the style is horrible. But as a concept, particularly for a complex work, laying out the structure in advance, and periodically touching base to detail what's been covered and what's yet to come isn't a bad idea. Of course, in a text, readers can accomplish this for themselves by skimming the contents, flipping to the end, and examining the notes and/or bibliography, and index (assuming a book-length treatment). I'll frequently skim an article, essay, or book first, to get a sense of its structure.
Reading and re-reading is another approach.
I get a stronger sense that such "here is the structure of this book" type introductions are born from initial outlines which guided the author. In many cases they can be significantly streamlined, but they're not entirely useless. Among technical books, the O'Reilly formula is generally pretty good -- it provides guidance but isn't overly pedantic.
The trick is to just start, and understand that you may stumble, and when you do know that you will learn from the failure and keep moving forward. Wanting perfection immediately without the failure first is where we usually go wrong.
This is what the growth mindset talk is all about. If you absorb these ideas it can be transformative. https://www.ted.com/talks/carol_dweck_the_power_of_believing...
Often people are afraid to ask questions because they don't want to look stupid or admit ignorance, but explicitly admitting ignorance and asking what things mean or how things work in conversation actually has the opposite effect. You learn a lot faster and more often than not people feel more comfortable because you're being honest with what you don't understand.
The side effect is that you usually end up looking 'smarter' anyway - somewhat ironically not asking questions because of fear of looking stupid can lead to it becoming clear you don't actually understand something you're pretending to which then does make you look stupid.
I've seen two negative things that can come out of this - the first is the person that feigns surprise and says 'You don't know what X is?!". More common in college and around other people who are insecure about their own intelligence, best response to this is to admit ignorance again and re ask the question (usually becomes apparent that the person doing this doesn't know what they're talking about anyway and can't explain it well). I think this negative attitude is especially bad because it can disproportionately affect women in early CS because they are probably already worried about fitting in there anyway.
Second negative thing is that asking questions when you do know the answer to show how smart you are is not the same thing as admitting ignorance and asking questions to learn - it comes across as obvious and is annoying.
Long answer, but generally agree that growth mindset and treating negative experiences as 'learning experiences' is the way to go.
This is bad advice unless you are a complete novice. Industry experience seems to weigh in on the side of focused analysis: you need to better understand the problem space and break a potential implementation down to a relatively simplistic first stage goal.
Relevant quotes:
Any attempt to formulate all possible requirements at the start of a project will fail and would cause considerable delays. - Pahl and Beitz, 'Engineering Design' (2007)
Sometimes the problem is to discover what the problem is. - Gordon Glegg, 'The Design of Design' (1969)
In practice, designing seems to proceed by oscillating between sub-solution and sub-problem areas, as well as by decomposing the problem and combining sub-solutions. - Nigel Cross
Success really depends on the conception of the problem, the design of the system, not in the details of how it's coded. - Leslie Lamport
You're not going to come up with a simple design through any kind of coding techniques or any kind of programming language concepts. Simplicity has to be achieved above the code level before you get to the point which you worry about how you actually implement this thing in code. - Leslie Lamport
... more @ https://github.com/globalcitizen/taoup (which has a whole section of quotes 'writing with applicability to software')
It's also a question of how one approaches writing: pansing vs plotting. Some authors, like Stephen King, swear by pansing and begin by forcing themselves to produce a certain number or words; whether it starts with character development or the plot, they exercise their powers of putting words on paper. Some authors, like J.K. Rowling, take a different approach and begin with sketching, diagraming, etc. I don't think the two are mutually exclusive, though.
Sometimes, even in the same piece of writing, I bounce back and forth between the two strategies.
I'm rather perplexed by how useful the advice to start appears to be perceived.
[1] https://gottcode.org/focuswriter/ [2] https://github.com/revolunet/sublimetext-markdown-preview
That's a useful training aid, but shouldn't be necessary every time. After a few tries, starting an essay or post should be automatic. Writing a coherent essay is usually taught in high school. Strunk's classic "The Elements of Style" remains useful.
This problem afflicts some how-to videos. Starting with "Hi, how are you" in a one-way medium is pointless. How-to material should start with what you're going to do, go on to how to do it, and end with a recap of what you did.
Always engage brain before putting mouth in gear.
That's my experience for the kind of essays they expect in college, but I when I want to write something for public consumption it's going to take more effort. The coherent essay that you learned to write and high school and served well in college may be informative, but it's still going to be boring. Capturing interest is still hard even once you have the mechanics of essay structure down.
Or at least it is for me. I'm less than fond of Strunk & White. These days I'd recommend Style: Lessons in Clarity and Grace over Elements of Style.
As for the "hi", that actually does serve a purpose in one-way mediums. It takes time for our brains to adapt to the speech of a new speaker, so it's common to miss the first thing that someone says. But if everyone says "hi" first, then your brain has a split-second to adapt to new voices before hearing something new and interesting.
1. "How to begin writing a text." How do you get yourself going so you aren't stuck on a blank page?
2. "How to write the beginning of a text." What is the process you take to determine the first sentence of a piece of writing?
3. "How to begin a text." After you've gone through that process, what kind of opening sentence should you end up with?
The suggestion to just delete the first couple of sentences is an OK one. I think the important part is that it helps solve the first problem—how to get moving at all so you are past the blank page. And the key insight is to realize that everything is still mutable. Writing is not committing. Editing isn't either, for that matter. Only publishing is. Until then, it's all unfired clay and you can play with it to your heart's content.
I think deleting the first couple of sentences is a moderately OK suggestion for the second one. Deleting any sentences that don't carry weight is always a good thing, and the first few are likely to be unuseful. But there's a good change that the next few sentences need work too. In a first draft, every line is suspect.
I don't think the article tells you much about what a good first sentence looks like. I don't necessarily think it needs to have "useful content". When I write, especially for the web where attention spans are measured in nanoseconds, the entire goal of my first sentence is to get you to read the first paragraph. The goal of the first paragraph is to get you to read the first section. Then I start getting to the meat of the piece.
Often that means the first sentence is a hook. It may be deliberately emotional to get you excited about continuing to read. Or it may raise a question you want answered or claim to teach you something you want to know. Sometimes I just try to use interesting enough language or imagery to get you excited about the pure joy of reading.
Once I've done that, I've got a little more slack, a little more of your good faith, that I can then use to draw you through the rest of the work. If I'm trying to do something like explain how a garbage collector works, or describe a feature of a programming language you've never heard of, I need that initial burst of interest to get you to keep moving.
Then it's up to me to make the rest of the piece satisfying enough to be glad that I sucked you in.
Starting in the middle is great if your reader already knows the backstory, but you alienate everyone but your core audience. That's expensive whether you're writing a blog post or a hollywood blockbuster.
It's a fair assumption that your reader came to you on the promise of the title of your piece. You have to fulfill the promise of that topic or at least give a tidbit soon enough that your audience doesn't give up and leave. You can tell a book by its cover because books with misleading covers don't sell.
Balancing exposition against delivering on your genre promise is a 'theory of mind' skill that the best writers have in spades. Think of the first 10 minutes of star wars: scary villain darth vader (check) strolling through a sci-fi spaceship (check). Gun-toting strong princess (check) needing a rescue (check) sends her sci-fi robots (check) to get help. Next scene: crappy teenager main character (that's your audience, check) buys same robots.
The order of the villain / princess / hero scenes are optimized to serve the 'message for help / hero to the rescue' exchange. If this were a blog post you'd still have that 'problem / solution' structure but in both cases you need to be checking all your genre points and delivering breadcrumbs of the genre or topic that the reader is here to consume as you inform the reader enough to care about your argument.