Infinite ALL the Scrolls

With the relaunch of Time.com I’ve had a number of people comparing it to Quartz. This happened a too when Daily Beast made updates last year, but its been more pronounced this time.

Either way, its something I feel that is still mis-categorized - especially when people bemoan the “infinite scroll”.

Well, call it “infinite scroll” if you like but to me its the furthest thing from that. I fully admit the difference may be in my mind alone. Let me explain a little bit of history and something I learned many years ago a The New York Times.

When a reader makes it to the bottom of an article there is a high tendency to lose them. Thats a general observation and not just specific to the Times. The common way people have found to deal with this was to add more links. Between the obligatory ads (seriously, why!), and recommended articles, callouts to compelling content, most popular, 3rd-party modules, and impassioned please to sign-up for a news letter you eventually end up alienating the reader and they bounce.

Gone. Disappeared. No conversion. No new tasty ad impression to help keep the lights on.

So it was at the Times. We’d see a high number of people bounce and wouldn’t it be great of we could entice them to stay a little bit longer.

In 2011 that gave rise to the oft copied “Coming up next” module. It flew out from the bottom-right corner of the page as you approach the end of the article. It was conceived and built by Tahir Khan and it was very very effective. The Bounce Rate dropped a very significant percentage.

Partial screenshot of Coming Up Next

To me the main conclusion was; readers are being paralyzed by choice. They’re overwhelmed with options. This module was clear and different from everything else and it solved one big problem - Do Not Make the Reader Think.

It didn’t try and guess from a list of curated items, or present a recommendation based on browsing history, it just said “go here” (arbitrarily the next article in a section) and a significant number of people did with one click.

It was genius.

Download ALL the Articles

In building Quartz this was one lesson I took with me. In our execution the idea wasn’t to implement “infinite scroll”. It was about taking the ”Coming up next” module to its logical conclusion - remove the need to click.

What most people do not know, is that when you load a page on Quartz - be it the main page or an article, you load ALL the pages.

That side-bar on the left side (we call it the “Queue”) - it typically has 18 to 26 articles present. When you open the page we load a big string of JSON and save it all locally. Its actually quite inexpensive and it means when you click on any Article its there immediately. Only images, if not already cached are left to load.

So as you approach the end of an Article there is no cost to “preload” that next Article. Its there. You don’t have to read it. It will never be counted as a Page View unless you scroll beyond a generous threshold (we were never ever sneaky about trying to abuse PVs - PVs are evil anyway but thats a different story…).

So, in the end:

Simple Reader Choice (effectively none) + Preloaded Data = Amazing Page Depth

This is the part where I disappoint people and provide absolutely no numbers to back all this up. Sorry! Its a combination of memory and also keeping promises.

However I’ve seen the Page Depth metric of a number of different web sites. Ours was several times higher - typically to a ballpark of 3 to 5 times.

This isn’t going to be a magic bullet. Even at Quartz there are things we didn’t plan for (a large ‘Engage” ad at the end of every article instead of every 2 or 3). Also consider that if a reader is skipping through thats a lot of content to push into DOM - see Mediums approach that avoids this).

So yeah, forgive me if I winch at the term Infinite Scroll. Its not quite that simple.

qzopen
qzopen:

It’s been a very busy year at Quartz. Out of almost 20 code repositories we have on Github, there are 4 that handle the core functions of the site and consume most of our development time.

Our most busy repo is qzapp, which handles core front-end application, styles, and other assets. It had 2,220 commits (as of this writing—still another week to go…), including the addition of 476,250 lines of code and the removal of 328,859.

Those numbers alone aren’t very meaningful (covers addition, movement, and removal of libraries and build files), but they reflect a high measure of effort and continuous iterations throughout the year that included:

- Refactoring of Javascript web app
- Offline mode, offline metrics
- Mobile navigation redesign
- Overhaul of analytics reporting
- Ongoing performance enhancing updates
- Twitter sign-in
- Annotations
- Highlighting

You can see more on our Version page.

And that was just one repository. It doesn’t cover all of our efforts spanning WordPress, ads, analytics, login and registration, internal tools, and backend.

There is even more in the pipeline for 2014. Speaking of which, we are hiring, so come talk to us.

qzopen:

It’s been a very busy year at Quartz. Out of almost 20 code repositories we have on Github, there are 4 that handle the core functions of the site and consume most of our development time.

Our most busy repo is qzapp, which handles core front-end application, styles, and other assets. It had 2,220 commits (as of this writing—still another week to go…), including the addition of 476,250 lines of code and the removal of 328,859.

Those numbers alone aren’t very meaningful (covers addition, movement, and removal of libraries and build files), but they reflect a high measure of effort and continuous iterations throughout the year that included:

- Refactoring of Javascript web app
- Offline mode, offline metrics
- Mobile navigation redesign
- Overhaul of analytics reporting
- Ongoing performance enhancing updates
- Twitter sign-in
- Annotations
- Highlighting

You can see more on our Version page.

And that was just one repository. It doesn’t cover all of our efforts spanning WordPress, ads, analytics, login and registration, internal tools, and backend.

There is even more in the pipeline for 2014. Speaking of which, we are hiring, so come talk to us.

44% of @CNN followers are Fake (and another 34% Inactive)?

According to ten minutes playing with “Fake Follower Check" there are a lot of big names on Twitter that actually have much lower influence then you’d think.

image

The big surprise for me is how low the ‘good follower’ count - many well below 30%.

@CNN has 10.175 million followers. This suggests that its true reach is 2.23 million followers. Thats a drastic reduction.

That leaves me asking; how inflated is Twitter?

So reality check… take this with a BIG grain of salt. A few critical points to remember:

- Many fake Twitter accounts follow popular organizations so they blend in
- This is based on criteria by StatusPeople. I do not know what that criteria is or how accurate. It may be way way off. No idea

If you have a few minutes to spare I’d love to gather more data on other Twitter accounts (which I’ll post here). Click here for more info or to help.

I include @qz, the company I work at, not because its a healthier rate but rather if I omitted it it would seem like I’m being selective.

Dave Winer says I’m rude

I’m wondering if its like email. You need to watch for tone and voice before you hit send. Anyway, Dave Winer says my comment on one of his posts was “one of the most rude messages i’ve seen in a while. congrats”. Perhaps its true?

Its deleted now.

But I do have a draft version. I no longer recall how close it is to the final version but not too far off I imagine.

I tried to quote from his post and be specific in replies so as to avoid broad generalizations etc. That didn’t help.

If it doesn’t read as “respectful” (as per his comment guidelines) then so be it. I was simply disagreeing with him not calling him names.

UPDATE: Its because I said “BS”. Well, I still think its was appropriate to make the point. His point on “new ideas” was way off and unsubstantiated imho.

His post is here.

My draft comment is as follows (pardon typos etc - its a draft):

===

NYT publisher Arthur Sulzberger said the mistake they made was not hiring more programmers sooner. I thought this was noteworthy — I think the exact opposite is true. I think the Times should have tried to avoid hiring programmers as much as possible. Before they had a lot of programmers it was possible to do deals with them, after the programmers came on, they had yet another set of gatekeepers, who as a side-effect of doing their jobs, kept new ideas from penetrating the Times.

I have to disagree with you there - and Mr. Sulzberger. There were developers. Quite a lot of them (for the time) but we were kept away over at 700 7th Avenue. We were not gatekeepers in the sense you describe simply because we were not part of the conversation.

What are these “deals” you speak of?

kept new ideas from penetrating the Times

Thats BS. Who introduced Twitter to the Times? Who pushed short-urls? New templating, TimesPeople (short lived social network of sorts), Times Wire, Times Skimmer, JSON feeds, Developer APIs, Recommendation Engine?

Developers. Developers pushed an pioneered those initiatives.

RSS. As such an unqualified success, it should have been used as the model for other good things that could have happened, but didn’t, because of gatekeeping


I’m not sure how - other than allowing full content in RSS feeds I’m not sure how newspapers acted as gatekeepers. I’d like to know more on your thoughts around that (apologies if you’ve gone in-depth on this elsewhere)

For example, I had a very simple Blackberry river of news for the Times in 2006. This was before it was known widely that mobile was everything (another thing Sulzberger said last night). But it couldn’t happen because the Times had its own internal effort to do a mobile app that was, imho, nowhere near as easy, fast or nice as the one I was able to whip up in a weekend because I didn’t have the time to make it complicated.

Its worth noting that most people do not want a river of news from the Times (or most other news sources). In the Times case people wanted the Homepage. They wanted the editorial judged placement on the lead story and what else was news of the day.

Also, the Mobile site of the time wasn’t particularly slow as I recall and to be fair the aesthetics level it was better then the version you built. Thats not a criticism - yours is much simpler but because its pure chronological and plainly formatted not something I would say is consumer ready. Thats just opinion on my part though.

Dear NYT,

I’m a big fan, honest, but please do two things:

1. Don’t run a full page ad (especially not this!) on the Homepage when there has been a shooting. Flip the ‘tragedy’ flag, I know the option is there.

2. Cookie it! I get the ad every time I refresh the page. 

Not to dwell on the first point too much, but you don’t exactly want to see someone with an axe kicking through your screen when the headline is about a murderous rampage within driving distance.

Not cool.

Dear NYT,

I’m a big fan, honest, but please do two things:

1. Don’t run a full page ad (especially not this!) on the Homepage when there has been a shooting. Flip the ‘tragedy’ flag, I know the option is there.

2. Cookie it! I get the ad every time I refresh the page.

Not to dwell on the first point too much, but you don’t exactly want to see someone with an axe kicking through your screen when the headline is about a murderous rampage within driving distance.

Not cool.