Silicon Cloud

Yesterday we posted an article to Digg for the first time not expecting the results of what happened next. Here we analyse the life of a Digg article and what people are calling the ‘Digg effect’.

Our story starts yesterday early evening. We’ve had an article on the site about how inexperienced web-masters ruin the appeal of their site by making some basic mistakes. We thought the article was relatively amusing and decided to put it on Digg.

Step 1 was to post the article into the Digg site. This was fairly easy as we already had a Digg account. Once our article was in digg on the diggall list we sat back and watched what happened next. Quite quickly a few people ‘dugg’ the posting and within about 15 minutes the post had 10 diggs and appeared as the next level of popularity in the cloud view. Things were going well. All this was helped by the first comment received on the article which was almost as funny as the article itself. Thanks James.

A cup of coffee later and the posting had nearly 40 diggs and was shown in the BIG BOLD text in the cloud view to show that it was gaining popularity. The post had been on for about an hour now and a couple of people were providing comments. Traffic to the site had increased initially in the first 15 minutes, but had levelled of since.

At this point we had to leave the office for a couple of hours. It was when we came back that the term ‘digg effect’ dawned on us. I thought I’d check what was happening, as I was quite excited about having my first post getting some comments and the site receiving some traffic.

The first thing I noticed on my return was that the post was on the front page, in the top 5 most popular articles of the Digg site. A couple of pleased expletives were uttered in wonder at how the post had done so well. A quick check of the statistics showed that nearly 900 people had dugg the article, providing over 70 comments on the digg site and close to 50 comments on the article itself. Wow. I then checked the server logs to find that the site had received 16,000 visits serving 35,000 pages in 4 hours. Unsurprisingly at it’s peak of interest the server could not cope with the demand that the digg effect was having. For some users the pages were simply un-viewable. Ironically an article about methods to annoy and frustrate visitors to a web-site had by it’s own success and the bandwidth it was needing had made it become so slow and un-usable that it was annoying itself!

After 5 hours on Digg and a good spell on the front page, some people decided that it was a spam article and it was removed from the list. I was slightly disappointed by this as 900 people had thought the article good enough to digg it and plenty of comments were being left. Before you all start commenting about how I was only after revenue from advertising and adsense, know this. The people who read Digg are generally net-savvy users who either have advertising banners blocked, or are experienced enough to know not to click on one. The 35,000 visitors to the site barely generated enough click-thrus to cover our server costs.

Anyhow, the site article had now been removed from the Digg front page. Phew, we could breathe again and our server could recover. However the traffic did not stop. It reduced by about a third but we now found that we were in the top 10 posts on the del.icio.us popular page and also the front page on diggdot.us. Although the Digg effect had now officially ended as we were no longer in the index, in the 8 hours that passed we received a further 5,000 views and 10,000 page reads.

The last time we checked, the article had 942 diggs, 92 comments on Digg.com and 58 on this website, with a total of 47,000 page views. Traffic has finally begun to slow down.

The lesson of the story is that when a page makes it on Digg then don’t be surprised if your server cannot cope. Be prepared for some negative comments. When 150 people comment on your article someone’s going to post something negative amongst all the positive responses. Above all have fun and enjoy the ride.

Just to prove I haven’t learnt my lesson. We also put this article on Digg… but don’t expect the server to survive if we hit the front page again.

Update: The traffic on the original posting has really started to slow down now. Still seeing clicks coming from some sources, I expect that it’s bookmarks and track-backs that people are now browsing. In the last 5 Hours we’re seeing about 20-30 hits an hour on the posting and a few more comments but nothing on the scale of yesterday.

Update: This posting has just made it to the top of Digg as their most popular post. Two in 24 hours that’s some going. So far the server seems to have held up. There seems to be less traffic on this post than when the other post went ‘front-page’ yesterday, it is the weekend after all. However it’s still fairly early in the day in some parts of the US so we’ll see if it holds.

Popularity: 11%