Warning: don't click on the epub file expecting to see anything useful. Your browser will gobble all your available memory to show you a binary zip file.

I rejected your warning because I am a naturally inquisitive person, so I wanted to see exactly what would happen.

I shouldn't have.

Did you file a bug on github? Because it sounds like one.

Yes, you should Save Link As or something similar.

Yeah, thanks, already too late :/

Out of curiosity, what OS and browser? Opera on Ubuntu just displayed a big ass box of non-readable text but didn't gobble all my memory or anything.

I was reading the Startup Bible in PDF that I had on my iPhone and decided that such great writing deserved better presentation.

There are some formatting gotcha's, I put this together rather quickly so it uses some libraries that aren't too gentle with pg's circa 2001 HTML formatting.

I hope this doesn't violate pg's sense of copyright of his works, if so, will take down immediately.

Was hoping to see "republished with permission" somewhere before downloading. Does PG give open license to reproduce/distribute?

Well, he funds scribd, which has a substantial amount of illegal content. I don't think he should complain.

Hum, you could as well have said:"he's an internet entrepreneur, and internet is choke full of illegal content, so he shouldn't complain".

Just because he funds a website that doesn't tightly monitor what is submitted does not mean it is ok to freely distribute his work without permission. That being said, the articles are already available for free on his website, but I do not see a license on them.

Agreed - I have no desire to piss off pg considering the value his ideas have provided to me over the years. I considered keeping the collection private because of the copyright issue but one of his essays mentions that they're looking for people for their YC companies who are willing to push the envelope a bit :) Also, I thought about emailing him first but he says he's not responding to emails until September. (All not excuses I know).

That being said, I do wonder where the line is and will be drawn in the coming years. The ePub artifact itself may be controversial (not really sure) but I'm pretty sure I could distribute the code that generates the ePub or create a page that says "click this button to generate the ePub download" and have legal precedent on my side (from the VCR cases and since there isn't any DMCA workaround I'd have to get through for the content).

I recently saw on startuplessonslearned that they're selling their blog content for 29.99 (more than worth it IMO) but at the same time, their target audience could easily circumvent the need to do it either through instapaper, writing their own script or any other myriad of ways.

Sure, Eric is selling his blog content on Leanpub (http://leanpub.com/startuplessonslearned), but the key difference is that Eric asked us to do it, and he's getting money :).

And of course someone could get around the need to pay $29.99 by writing a script, but if you value your time at all it wouldn't be worth it.

We think it's a good deal for Eric (he makes money from something he has given away for free) and a good deal for the reader (it's much more pleasant to read a PDF than to click through posts in a web browser, and even more pleasant to read it on a Kindle.).

Agreed (I mentioned that it was worth it and I enjoy getting well-formatted eBooks as compared to instapapering everything).

My main point was that e-publishing seems like a tough business since the trend of information pricing seems to be heading to 0.

Seems to be missing one of my favorites: Maker's Schedule, Manager's Schedule.

I've been thinking about making a tool based on that idea: http://makerscheduler.com aka http://msched.com

icco Added a mobi version (Thanks!) and I put in an updated version that had a few bugs. There are definitely spacing issues etc but you can pretty easily understand the original meaning. Here's the main loop of code that extracts the text, if anyone has a better idea let me know:

  source = open(link['href']).read
  
  text = Readability::Document.new(source, :tags => %w[div p br font]).content
  xhtml_text = Sanitize.clean(text, :elements => ['a', 'div', 'pre', 'br', 'font', 'p', 'img', 'table', 'tr', 'td'], :attributes => {:all => ['class', 'id', 'src', 'href']})
  xhtml_text = Mustache.render(xhtml_template, :title => link.text, :content => xhtml_text)

where ePub formats expect something like

  xhtml_template = <<XHTML
  <?xml version='1.0' encoding='utf-8'?>
  <html xmlns="http://www.w3.org/1999/xhtml>
    <head>
    <title>{{title}}</title>
    </head>
    <body>
    <h2>{{title}}</h2>
  {{{content}}}
    </body>
  </html>
  XHTML

Instead of using this xhtml_template, you should pass the content through tidy with the -asxhtml switch on.

This will produce a valid xhtml file that passes epub validation (http://code.google.com/p/epubcheck/).

Awesome - exactly the information I was looking for, will set it up later tonight and post both a new ePub and the source.

Something isn't right. It doesn't work in Epub reader (FF extension), it doesn't open in Calibre, and it doesn't open on my PRS-505.