Watch pg write an essay in real time
code.stypi.com
code.stypi.com
It's especially interesting to note that there are few, if any, words in the entire essay that he didn't edit multiple times. A lot of times when you read something as well-written as PGs essays you get the feeling that the person writing them just sits down and hammers out brilliance. But in reality, at least in the case of PG - and I would venture a guess that most great authors are similar - greatness comes through sweat and repetition more than raw talent, until eventually that thing becomes second nature - or does it ever? I would kill to watch some of the great authors of all-time (a la Shakespeare, Hugo, Dickens etc.) write in real-time - it would be fascinating to know what their process was like.
“I have rewritten–often several times–every word I have ever written. My pencils outlast their erasers.” -- Vladimir Nabokov
But are you really trying to compare a bullet-point style blog post (not really an essay) to Shakespeare, Hugo, or Dickens?
You state that as if it's some universal truth but it's certainly not. We know that Asimov only ever wrote a first draft and a final draft and Heinlein wouldn't even create a first draft most of the time. Neither of them is a Nabokov or Hemingway but they certainly fall into the category of "everyone."
'I remember the players have often mentioned it as an honour to Shakspeare, that in his writing (whatsoever he penned) he never blotted out a line. My answer hath been, "Would he had blotted a thousand," which they thought a malevolent speech. I had not told posterity this but for their ignorance who chose that circumstance to commend their friend by wherein he most faulted; and to justify mine own candour, for I loved the man, and do honour his memory on this side idolatry as much as any.'
And what does the format matter? If it did, we would disregard haikus as poetry just because they can fit into tweets.
Script is a Linux utility which makes a typescript of a terminal session. Traditionally this was mostly of use for shell sessions which didn't involve full-screen (ncurses) utilities, as the cursor-movement characters would typically present a jumbled mash on your screen.
Using the '-t' (or '--timing') arguments sends timing between movements to stderr, or the specified file. With this, you can then replay a typescript file, with the timing information, to show what was actually presented on screen. An optional argument varies the replay speed.
I've made use of this to log and replay serial console activity (available on most servers through IPMI or related hands-off / lights-out management tools), which can be both instructive and helpful in documenting steps.
Further fun may be had by playing the output of scriptreplay through the phosphor(6x) X11 screensaver hack.
http://techcrunch.com/2011/08/09/yc-funded-stypi-is-etherpad...
I guess pg exported his text from Etherpad and gave it to the Stypi team.
Sometimes HN's date indications are less than useful.
It made me chuckle regardless.
I would imagine Facebook or another hosted blogging platform could get this data with relative ease.
While etherpad lives on as an open source project the original company/team was acquired by google and this resulted in google docs getting real time cursor movement and edits shortly thereafter (google docs originally updated via polling and often had collisions).
I think it's a shame that google never implemented more of the core ideas from etherpad, such as playback and contributor text highlighting. It's still the best online collaborative tool I've ever used. I had experiences writing letters to politicians in real time with dozens of anonymous strangers who we gathered via twitter, and collaborative grant proposal writing sessions that I can only describe as thrilling. It's also been fun to use etherpad to collaboratively take notes during conferences, and truly fascinating to replay and see how largely anonymous editors quickly start taking on certain roles without any prior discussion--one person gets very interested in fixing spelling mistakes, someone else gets really good at finding relevant urls for background information and links to slides, someone else just barrels ahead with whatever the presenter is currently talking about knowing that the others will fill in the gaps. A lot of the innovation happened in subset hardier, but etherpad made it way more accessible and this useful.
Etherpad deserved all the attention that google wave was receiving at the time because it really delivered on the promise of a truely flexible tool for collaborating and communicating more effectively than email.
> Etherpad deserved all the attention that google wave was receiving at the time
I find your post confusing because I thought that Google Wave did have playback and contributor text highlighting.
I suspect that the reason Docs never gained these features was that Wave was supposed to replace it, but AFAICT Wave never saw the kind of user adoption it needed (probably due to its sluggish performance) and was killed off.
Good to know I'm not alone. The more I read HN the more I find people, who are successful in their field and have the same behaviors as me. It's relaxing :-)
Or anyone else's
honestly been wanting to build this for myself forever
Founder of Y Combinator among other things
http://en.wikipedia.org/wiki/Paul_Graham_(computer_programme...
Occam's Razor explanation is that he just thought it would be cool to show.
To the contrary, I didn't think the answer to his comment was obvious. I don't claim to have the correct answer. And if that's the tone my comment conveyed, i.e., a dismissive one, then I probably conveyed myself improperly. That was not my intent.
http://www.xmpp.org/extensions/xep-0301.html
And the great demo: http://www.realjabber.org/real_time_text_demo.html
But still super interesting