Hacker News Reader

Wow, the http://www.telescopictext.org/ website is a very slick tool to create your own.

I created an example one here (made a few typos that I couldn't figure out how to fix):

http://www.telescopictext.org/text/AYeELHzYgVV8b

It has the feel of interactive embedded footnotes with positive reinforcement for opening them -- each time I clicked it revealed a few more features about you. Positive reinforcement is an important factor as you feel like you're discovering information hidden at a deeper level. Even if dozens, or even hundreds, have discovered that "secret" before it still feels like you've achieved something. The alt-text on XKCD is a perfect example of this.

Footnotes tend to convey additional information omitted due to necessity or lack of interest by the mainstream readers. I don't see them online a great deal, primarily as the web isn't composed of traditional pages so they don't translate well...

Has anyone else seen an interesting or novel way of doing footnotes on the internet?

(unrelated, but if anyone wants a fascinating example of footnotes in meta-fiction, the novel Oracle Night by Paul Auster has entire subplots and even another "book" hidden in the footnotes)

Slate magazine uses a nice way of presenting footnotes, especially in the series of long-form articles it published in the last few months. They use a small grey circle with a + sign in it; when you hover over the circle, a popup appears with the footnote. Some of the long-form articles have entire paragraphs in each footnote, which seem like they've been removed to make the article flow better but preserved for the interested reader who would like more detail. Others are simply corrections or clarifications added after the fact. I'd love it if more sites picked up this convention.

Another site which has a particularly unique way of writing footnotes is E2 ( http://everything2.com ). Like Wikipedia, editors ('noders') are encouraged to link articles ('nodes') together by hyperlinking words throughout the text. Unlike Wikipedia, noders can post anything: essays, fiction, poetry.

Some use the hyperlinks as commentary on the main text - if you hover over them, the title of the linked node can be very different from the text of the actual link, possibly changing the meaning of a sentence (for example, "I tell her that I'm [okay]", with [okay] linking to the node 'I am not okay'.)

Occasionally, noders link entire phrases or sentences which seem interesting, even if there's no actual node with that name. There is the concept of "filling nodeshells": writing new and interesting content to fill a node which was created by someone linking a phrase from another node.

Each node also holds a grid of 'softlinks': nodes which people visited from that node. Mostly this is composed of links from the text in that node, but it can also be used for readers' commentary via the readers deliberately visiting appropriately-titled nodes.

More info: http://everything2.com/title/The+perfect+node

My colleague (with some help from me) created an interactive version of the 1932 French poem "Nouvelles impressions d’Afrique" by Roussel, one of the early experiments in this vein: http://withhiddennoise.net/roussel/

Here Simon Collison has tried a decent approach for footnotes: http://colly.com/comments/redesigning_the_undesigned/

Search for "A few Easter eggs"

Arguably, hyperlinks are footnotes.

I see some deeper meaning in this - in that it is an explicit way to show how communication can be parsed down so simply. Many writers/communicators want to extrapolate every thought into word tombs - when they can very often be summated in something as small as "I made tea." Beautiful execution, here, on getting me (and hopefully others) to think about writing/communicating more simplistically - and then actually doing it.

Otherwise stated, "This makes me want to communicate more simply."

As I was playing with it, I was thinking about how it illustrates the way magnifying different aspects can give a different feeling to one's writing. Like: can I get better emphasis by keeping a particular facet short and direct, or does it pay to expand on it?

So I think it's not just about keeping it simple, but seeing where simple might be better, or where depth might be better.

The fully expanded text is:

Yawning, and smearing my eyes with my fingers, I walked bleary eyed into the kitchen and filled the kettle with fresh water from the tap, checking with my hands to make sure it was cold enough (The best tea comes from the coldest water). I glanced outside for a minute at the city mist. I could almost taste the grey. I plugged the kettle in and switched it on. As the kettle began to hiss, I looked for biscuits. Anything above loose crumbs would do. Thankfully I found some fusty digestives. For some reason, biscuits are always nicer when they've gone a bit dry and stale. I took the milk out of the fridge and poured some into a cup that I'd left out from having used earlier. The kettle began grumbling fiercely so I took it from the cord, threw a teabag into my cup and poured boiling water onto it. I watched brown swirls rise up through the muted white of milky water. A few minutes passed. I removed and squeezed the teabag, then flicked it into the bin. I picked up my mug and left the kitchen with a nice, hot cup of strong tea.

Thank you for the TLDR. I wondered whether the text would just keep expanding forever... In some ways I am sad that it did not.

Isn't the TL;DR "I made tea" ?

I think he meant, TL;DE (To Long; Didn't Expand)

;)

The only way to make it expand forever would be to support recursion. Would be interesting what texts you could come up with that. Kind of a textual Mandelbrot set.

It was a dark and stormy night. The captain and his men were huddled around a campfire. The captain turned to Jake and said "Jake, tell us a story.", so Jake began his story: "It was a dark and stormy night. The captain and his men were huddled around the campfire..."

Buffalo buffalo buffalo buffalo buffalo buffalo buffalo buffalo (http://en.wikipedia.org/wiki/Buffalo_buffalo_Buffalo_buffalo...)

Perhaps using some kind of Markov chain process? It could be amusing, but would be more interesting with some kind of understanding of parts of speech.

It might be nice to read news articles this way. Start with the headline and expand the parts that interest you the most. I can imagine it being difficult to write that way, though.

Or maybe just write the article, and present the most-important bits first, starting with the headline. Then, the first paragraph or so would cover the important points, and as you kept reading, it'd get less and less important.

That's genius. I wish there was an organization in which they wrote most of their articles this way. Perhaps they could even send these articles to other organizations for some kind of reprinting fee.

That's exactly how all news stories are written. Unless you were making a joke.

Yes, that's the point.

Sadly, it's pretty trivial to disprove this by counterexample.

I discovered Joe Davis and TT a couple years (even going so far as to make my own editing tool which, I like to think, influenced Joe's own implementation).

I was thinking about the news article ideas the other day, actually. (E.g., a 24in60.com snippet that expends to a full article.) I think it would be really cool, but it takes a lot of work to make something seamlessly go from being a snippet of information to a full-fledged investigatory article. You'd basically have to write a series of increasingly more detailed articles using the same story flow. This might be something text analytics and computer-generated content could be able to do; if not today, then someday.

Why not go the opposite way and write increasingly shorter snippets from a full article?

This is whats shocked me about the authoring tool.

I agree, and did a little example of such:

http://www.telescopictext.org/text/3r3OhSm1fuXVT

I did a better go at it, but then realized I'd yet to sign up... I didn't quite have the mettle to do as good a job on the second attempt. But I certainly think it'd be interesting to have this on a site.

This would be a great way to supplement dense text for comprehension purposes and for learning new vocabulary. Learning moments are different for different people, this makes it one-size-fits-all.

Except, people with OCPD would click everything through, just to see if they missed anything, and really not make much headway into reading at all.

For people with OCPD, OCPD is obsessive–compulsive personality disorder.

http://eblong.com/zarf/zweb/matter/index.html

a choose-your-own-adventure kind of like this, released a week or two ago by Andrew Plotkin.

Nice, it's like having a slider to go from Raymond Carver to David Foster Wallace.

Nicholson Baker's The Mezzanine is a terrific novel about a guy who leaves his desk to ride down an escalator to buy shoelaces and a cookie.

To make your own, go here: http://www.telescopictext.org

Doesn't work with IE. I wonder what feature of IE9 isn't supported? Oh well. Maybe will have to write my own for IE.

This is great. All too often I've posted some rant, knowing it must be very concise to get the point across to soundbite readers, yet wanting to provide details to address obvious criticisms, yet having to somehow find a balance between minimal meme transfer vs. encyclopedic thoroughness overwhelming the basic meme. Hope this can be turned into a blogging/commenting tool.

There's got to be an easy HTML template for doing this without linking to an external website.

I can see this in terms of offering a Javascript for embedding in web pages.

Based on one of the comments below I couldn't help but wonder: Would trying to program an "ever-expanding" text like this yield any new ideas in NLP?

As someone in NLP I think I just came up with an idea about how it would be useful, but I won't say :D

I never thought expanding something would be so fun. Brings a new meaning to story telling if you do it right. :)

It was much more interesting than foldr.com.

I hope some one writes a javascript library that takes a suitable formatted text and has an expandTo(length) function so that the same info can fit into various screen sizes without having a scrollbar.

A reverse tl;dr. I've got an urge to try formatting my resume like this.

It would be nice if there was an "expand all" option, rather than forcing excess clicking to get to the end of the line.

It looks a good replacement for certain hyperlinks, if we can save/load the hidden text from the database.

Why the upper restriction in "Password must be between 6 to 20 characters" on telescopictext.org?

If you want to un-expand the text, hold the alt key. Very cool.

I saw something like this described many years ago (possibly by Ted Nelson, who coined the term 'hypertext'). The concept was like a volume control for text where cranking the volume control varied the amount of detail from one line summary through to a multi page article.

I've always really liked the idea but writing coherent content could be fiendishly difficult (the simple approach is just to provide a number of versions and cycle through them).

I wish 99% of all articles I read worked this way.

http://www.telescopictext.org/text/fCvNOyBqKSJAh

My "telescopic" thoughts on making this an interview question. When completely expanded, it does not have a grammatical flow in some parts because I couldn't edit text once created and I did not want to start from scratch.

This could be a very nice way of formatting a long description, such as the description of your software or startup.

Think about it: instead of greeting your visitor with a wall of text, you could have something like "XYZ is going to boost your productivity". The user would only need to expand that's interesting to him.

Yawning, and smearing my eyes with my fingers, I walked bleary eyed into the kitchen and filled the kettle with fresh water from the tap, checking with my hands to make sure it was cold enough (The best tea comes from the coldest water). I glanced outside for a minute at the city mist. I could almost taste the grey. I plugged the kettle in and switched it on. As the kettle began to hiss, I looked for biscuits. Anything above loose crumbs would do. Thankfully I found some fusty digestives. For some reason, biscuits are always nicer when they've gone a bit dry and stale. I took the milk out of the fridge and poured some into a cup that I'd left out from having used earlier. The kettle began grumbling fiercely so I took it from the cord, threw a teabag into my cup and poured boiling water onto it. I watched brown swirls rise up through the muted white of milky water. A few minutes passed. I removed and squeezed the teabag, then flicked it into the bin. I picked up my mug and left the kitchen with a nice, hot cup of strong tea.

This is the difference between my speech and my typing, my presentations and my speaker notes, and so on. Anyone know how to learn to do this as you go along, not preparing?

For a minute there I thought I'd be able to paste in some text and automagically get a TL;DR. Pretty neat nonetheless.

Now someone should make a tool that expands from specification to actual code like this. How would that work?

This is how cliff's notes should work.

Doesn't work on iOS Safari.

Nuts; that's my primary personal content creation platform now.

Type in console to cheat:

  $('._b').click()

Top tip: never put the milk in first! :-)

The text does not collapse back...

Is your refresh button broken?

That collapses all the way back. Where's partial collapse?

Hold down ALT and click a section (found this in: http://www.telescopictext.org/resources/reading-tips )

What the fuck is the point of this?!

Its 'fun 2 have'. :)

This is why I hate reading novels. And the fact they forced me to read several of them in high school when my English was rather poor.

Telescopictext