5 ms·

Six Principles for Making New Things (2008)

111 points by dayve 8y ago

QuadrupleA 8y ago
A little sad to read this 10 years later knowing that Arc hasn't really caught on. There's a clear optimism in the writing about the project's presumed future success - I think anybody who creates something public shares that feeling - but marketplaces are unpredictable and the odds aren't usually in your favor.
- shawn 8y ago
  Are you sure Arc hasn’t caught on? After all, we’re submitting comments via Arc. Some of the best work in history was done because the creator wanted it themselves. Even if no one else uses it, the work speaks for itself. Arc also played a crucial role in bringing attention to pg’s essays, which YC’s success was a direct consequence of. No essays, no audience; no audience, no founders, and no YC.
  - dayve 8y ago
    Interesting way to look at it.
  - JBiserkov 8y ago
    I guess what he means is that Arc is no Clojure.
- yason 8y ago
  I'd be worried if Arc had caught on already. Didn't he say Arc was to hopefully be a hundred-year language? It took Lisp several decades to inject its ideas into more common languages. It's now we've been seeing many of its features seeping into popular languages. Some non-Lisps even have limited macros. Being successful doesn't necessarily mean being popular but having impact. Can't tell yet about Arc.
  - cousin_it 8y ago
    Well, Arc doesn't seem to be solving an "overlooked problem" or "iterating rapidly" anymore. I think that's fine - not all seeds give birth to trees. Just find some new seeds to bet on :-)
- mmjaa 8y ago
  I think you're being a bit too diplomatic by calling it 'optimism' about Arc's eventual success .. to me at the time (and even more so now) it was more like hubris. Personally, one of the most important principles for creating new things that I care about, is this: "check yourself before you wreck yourself" .. this means, be sure you're aware of your own hubris before you foist it upon others. And I think the fact that Arc is relegated to the dustbin of failure demonstrates this even more so today. Hubris is a principle cause of unproductive behaviour in the technology industry, imho. Probably the #1 reason why things fail: people are not self-aware to the point of accurately recognising when their own hubris is getting in the way of success...
  - agumonkey 8y ago
    Maybe it was too ambitiously propped up. But it was also too early to get trendy. And it's a lisp, the amount of people that can appreciate arc's feature is probably a 4 digit figure.
    - jungler 8y ago
      With respect to programming languages, most software, and most creative works, all states of the lifecycle seem to be bad at some level: 1. It's just an idea/it's just a prototype. 2. It works but it doesn't support everything I want to do. 3. It supports everything but it's not the way I want to solve the problem. 4. I guess I'll have to reinvent the wheel to show them how it's done... The only real terminus to this cycle is when you've actually communicated the thing you want to communicate, which is quite difficult to do in the form of a programming language.
cammil 8y ago
I agree with this whole heartedly, yet somehow I don't often practice it. I wonder why.
cammil 8y ago
I wonder if initial crudeness actually has a _strictly positive_ contribution to the likelihood of success. Perhaps the form of the product has a role in setting expectations. If so this would have a role in attracting forgiving first customers (e.g. those most in need of your product) and repelling those customers expecting a more "complete" product.
Dowwie 8y ago
Can someone share info about Cezanne and Klee that PG was referring to? I doubt their approaches were as solution-oriented as PG believed them to be because of the disposition of artists, who usually create for themselves first and society a far second.
jamessun 8y ago
My favorite line in PGs essay: "I like to find (a) simple solutions (b) to overlooked problems (c) that actually need to be solved, and (d) deliver them as informally as possible, (e) starting with a very crude version 1, then (f) iterating rapidly."
- ai_ia 8y ago
  That is the Tl;dr of the essay. I looked at that and moved on.
weliketocode 8y ago
I'm on the fence about the message here. PG announces Arc and receives flack for it. Then writes an article noting that he's always received this same flack for everything that he's done. Sounds to me like the message is less about Arc or even the principles that he states but more about keeping your head down and working through unproductive feedback.

111 points by dayve 8y ago

A little sad to read this 10 years later knowing that Arc hasn't really caught on. There's a clear optimism in the writing about the project's presumed future success - I think anybody who creates something public shares that feeling - but marketplaces are unpredictable and the odds aren't usually in your favor.

shawn 8y ago
Are you sure Arc hasn’t caught on? After all, we’re submitting comments via Arc. Some of the best work in history was done because the creator wanted it themselves. Even if no one else uses it, the work speaks for itself. Arc also played a crucial role in bringing attention to pg’s essays, which YC’s success was a direct consequence of. No essays, no audience; no audience, no founders, and no YC.
- dayve 8y ago
  Interesting way to look at it.
- JBiserkov 8y ago
  I guess what he means is that Arc is no Clojure.
yason 8y ago
I'd be worried if Arc had caught on already. Didn't he say Arc was to hopefully be a hundred-year language? It took Lisp several decades to inject its ideas into more common languages. It's now we've been seeing many of its features seeping into popular languages. Some non-Lisps even have limited macros. Being successful doesn't necessarily mean being popular but having impact. Can't tell yet about Arc.
- cousin_it 8y ago
  Well, Arc doesn't seem to be solving an "overlooked problem" or "iterating rapidly" anymore. I think that's fine - not all seeds give birth to trees. Just find some new seeds to bet on :-)
mmjaa 8y ago
I think you're being a bit too diplomatic by calling it 'optimism' about Arc's eventual success .. to me at the time (and even more so now) it was more like hubris. Personally, one of the most important principles for creating new things that I care about, is this: "check yourself before you wreck yourself" .. this means, be sure you're aware of your own hubris before you foist it upon others. And I think the fact that Arc is relegated to the dustbin of failure demonstrates this even more so today. Hubris is a principle cause of unproductive behaviour in the technology industry, imho. Probably the #1 reason why things fail: people are not self-aware to the point of accurately recognising when their own hubris is getting in the way of success...
- agumonkey 8y ago
  Maybe it was too ambitiously propped up. But it was also too early to get trendy. And it's a lisp, the amount of people that can appreciate arc's feature is probably a 4 digit figure.
  - jungler 8y ago
    With respect to programming languages, most software, and most creative works, all states of the lifecycle seem to be bad at some level: 1. It's just an idea/it's just a prototype. 2. It works but it doesn't support everything I want to do. 3. It supports everything but it's not the way I want to solve the problem. 4. I guess I'll have to reinvent the wheel to show them how it's done... The only real terminus to this cycle is when you've actually communicated the thing you want to communicate, which is quite difficult to do in the form of a programming language.

cammil 8y ago

I agree with this whole heartedly, yet somehow I don't often practice it. I wonder why.

cammil 8y ago

I wonder if initial crudeness actually has a _strictly positive_ contribution to the likelihood of success. Perhaps the form of the product has a role in setting expectations. If so this would have a role in attracting forgiving first customers (e.g. those most in need of your product) and repelling those customers expecting a more "complete" product.

Dowwie 8y ago

Can someone share info about Cezanne and Klee that PG was referring to? I doubt their approaches were as solution-oriented as PG believed them to be because of the disposition of artists, who usually create for themselves first and society a far second.

jamessun 8y ago

My favorite line in PGs essay: "I like to find (a) simple solutions (b) to overlooked problems (c) that actually need to be solved, and (d) deliver them as informally as possible, (e) starting with a very crude version 1, then (f) iterating rapidly."

ai_ia 8y ago
That is the Tl;dr of the essay. I looked at that and moved on.

weliketocode 8y ago

I'm on the fence about the message here. PG announces Arc and receives flack for it. Then writes an article noting that he's always received this same flack for everything that he's done. Sounds to me like the message is less about Arc or even the principles that he states but more about keeping your head down and working through unproductive feedback.