By the numbers: A front-page digg effect

Edit: Unfortunately, the images originally included in this post are gone, because of hosting problems in late 2009. My apologies.

This is mostly for my own benefit, but I suppose it could be of interest to some other folks. In a weird twist of events, my randomly jotted note about console applications for Ubuntu hit the front page of digg.com yesterday (no, I am not gloating). I say “randomly” because I honestly typed it up as I was leaving for work, more as a reminder for myself than for anyone else.

Of course, the mad rush ensued. Now that the hubbub has dissipated, the numbers are interesting, even if the implications … aren’t.

I usually get between 300 and 600 pageviews a day, with most of those going to the Edgy and Feisty speed tweak guides. On a good day, one or both of those pages might break 100 pageviews and I might get one or two comments on different areas. If I include links to the blog in my signature on the Ubuntu Forums, I get an occasional spike, which is usually traceable to a thread I was involved in. In all, it’s rather random, but at the same time, rather even-keeled.

Now here’s the pageviews for the last 24 hours.

Blog hits broke 11,000 for that 24-hour period, and the digg reached about 540 for the same time frame, which is respectable. The post was No. 2 for top posts for the day on WordPress.com, beating out the Scobleizer (a perennial No. 1) for a while and only losing the top spot to a post about learning fundamental Unix commands in 10 minutes (which I have to admit, was very interesting … much more than what I had written, if I have to be honest). Motho ke motho ka botho was the No. 2 blog of the day — again losing out only to the aforementioned blog — and was listed among the day’s fastest growing (which was not a new thing; that’s happened before when I get a spike from tuxmachines.org, or somewhere else).

I was actually watching my blog when the digg went to the front page, and the hit counter at the top right jumped more than 100 views in the first three minutes or so. Again, that’s about a third of what I get in a day. A few months ago I crossed 10,000 views and I thought it was an achievement. The digg effect added that much again, in the space of 24 hours.

As you can see here, referrals were almost completely from digg, whether from the front page or subsections.

Once digg picked up on it, there were some ancillary hits from stumbleupon.com and elsewhere. Unfortunately, those outlying sites were eclipsed by digg referrals so I can’t say where the other hits came from. I’m sure tuxmachines.org was in there (and had found the post much earlier, I might add ;) ), as well as some others.

Nearly 10,000 hits were for that post alone.

There was also a spillover, with the Edgy guide getting almost seven times its usual traffic. Other posts that had been listed in the Top Posts for that day were also heavily hit; the “Wikipedia Picture of the Day” and “Linux is bad for business …” both got as many hits in that day as they had for their entire existence up until then. I’m not sure why the “Edgy sans inittab” post is still popular; that too was just a random jot. But as you can see, it benefitted from the traffic.

I have noticed that the Top Posts counter seems to rollover at the end of the day, and whatever posts hit that list early in the day tend to be the popular posts for that day. (Does that make sense?) In other words, a random hit on a page immediately after the rollover puts it in the Top Posts list — which could put it into the Top Posts for the entire day, since people will see that post in the Top Posts category for the rest of the day. And click on it.

People leaving the site usually went to the Most Recent Desktop image, which is the xs.to link you see here.

After that, Gunnix’s lightweight apps page got plenty of traffic. And after that, the Ubuntu packages site was inundated. In fact, there isn’t a link on that list that wasn’t immediately visible from the post itself (or the comments) — except for the last image, which was the phony Vista screenshot from the “Linux is bad for business …” post. Again, spillover from traffic.

And as a final note, there was a tiny spike on my sister blog, with a few scattered visitors poking around here and there.

So what’s to be learned from all this? Well, I don’t know that there is much. The event itself is too unpredictable to me, since it depends so much on the whim of the digger, the whims of the digger audience and the chance that it will hit the front page at all. If you could confidently forecast the popularity of a post, I suppose you could anticipate the traffic load and perhaps even mold the site to glean more hits, but I don’t know that it would be worth the effort at all.

In addition, I don’t know that I even care for the overall effect. I subconsciously like to gauge the traffic against the posts I write, but a 30-fold spike in traffic completely ruins my perceptions on what people want to see or read.

As an example, it has been obvious since November or so that the Edgy howto was a popular and well-received post. It’s been a steady draw since it was written. That was part of my reason for jumping the gun with the Feisty howto, and testing it before it was ripe. Steady, regular streams of visitors are looking for something in specific, and something they can rely on. That’s what I try to offer.

But an unpredictable, one-day, 10,000-hit spike does nothing for me. It doesn’t give me an accurate clue as to what visitors want to read, since most of those people aren’t here looking for something in particular — they’re just looking. It’s like having a busload of tourists show up at your fruit stand, not really buying or looking for anything — just waiting for the driver to start up the bus again.

And on top of that, the demeanor of some visitors isn’t what I would normally appreciate. Off-color nicknames, random fanboy comments and playing host to flame wars aren’t why I do this. I even deleted a few posts that were just plain inappropriate.

So having that many random people suddenly show up and voice an opinion … well, to use another analogy, it’s like having the entire high school funneled through your living room, with kids shouting about how they like — or don’t like — your coffee table. Why is that any of their business? Are you here to check out the coffee table, or are you just here because everyone else is?

I think part of my dissatisfaction with the digg effect goes back to why I started a blog. I’m not a hub of information. I don’t pride myself on what I pry out of the Internet from the armchair in front of my desk.

These posts are primarily reminders of what works, what doesn’t work or what I learn from moderating. It’s what I experience with Ubuntu, and I use it to remind myself of what I did when I try it again. It’s primarily for me — it’s original, personal content that will help me, and might help someone else who’s having a similar problem. I’m not aiming for the flash-of-the-moment, digg-drawing spark of pseudo-news that pulls in 10,000 visitors in 24 hours.

So if you’ve read this far, thanks. If you were drawn here by the original post and you returned because you think the blog has valuable content, thanks doubly. Plan on digging something else? Well, thanks: I’m flattered. But if you really want to do me a favor, just jot me a note. I think I’d honestly appreciate that more.

About these ads

1 Response to “By the numbers: A front-page digg effect”


  1. 1 gouki 2007/03/04 at 11:21 AM

    Please keep the posts aggregated on UbuntuWeblogs more related to Ubuntu and don’t use ‘more’ tags.

    If you can, take a look at the rules (present on the submit page).


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s




Welcome!



Visit the Wiki!

Some recent desktops


May 6, 2011
Musca 0.9.24 on Crux Linux
150Mhz Pentium 96Mb 8Gb CF
 


May 14, 2011
IceWM 1.2.37 and Arch Linux
L2300 core duo 3Gb 320Gb

Some recent games


Apr. 21, 2011
Oolite on Xubuntu 11.04
L2300 core duo 3Gb 320Gb

Enter your email address to subscribe to this blog and receive notifications of new posts.

Join 405 other followers

License

This work is licensed under the GNU Free Documentation License. Please see the About page for details.

Blog Stats

  • 3,963,477 hits

Archives


Follow

Get every new post delivered to your Inbox.

Join 405 other followers

%d bloggers like this: