June 2015 in review

This is the ninth of my month-in-review posts. I started month-in-review posts in October 2014. You can see my first post here and my previous month-in-review post here.

Some overall background

This was a very busy month for me at work, more so than most preceding months. The latter half of the month was in particular very crunched. The crunch ended on July 1, so I expect the workload in July to be more representative of the business-as-usual scenario (which might still be pretty hectic!).

Wikipedia editing

For background information, see my site page about Wikipedia

I created the following Wikipedia pages this month, both in the second week when the workload hadn’t picked up fully:

I also made updates to these pages:

Unfortunately, stats.grok.se, whose numbers I use to feed in to Wikipedia Views, is missing one day from the month of June. Therefore, the numbers reported here are about 3% lower than the actual numbers. As with all numbers from stats.grok.se, note that these numbers exclude mobile pageviews but include bots.

Comparison with forecast: I had forecast 160,000 pageviews for the month of June for all pages I created. The actual number of pageviews (138,745) was significantly lower, and even adjusting for the missing date (to get a total of 143,529 views), there’s still a huge gap. The gap was in the opposite direction to the direction of my gaps in previous months. For instance, for May my point estimate for the number of pageviews was 151,000, but the actual pageview count (excluding two days for which stats weren’t available) was 174,425. The steady underestimation trend had led me to up my June estimate to 160,000. However, the pageview count did fall safely within my 80% probability interval estimate of 125,000-215,000.

The key source of month-to-month variation appears to be the pageview counts for some highly popular pages. In April and May, the Wikipedia page about Internet.org got a lot of pageviews because of the topicality of the topic. There were no comparably topical pages in June, explaining the lower pageview count.

July forecast: My point estimate for the July pageview count is 150,000 pageviews, adjusted downward from the June estimate but still higher than the actual June value. For July pageviews, my 50% probability interval estimate is 135,000-175,000, my 80% probability interval estimate is 110,000-220,000, and my 95% probability interval estimate is 85,000-265,000.

Here is a list of Wikipedia articles I am considering creating for the month of July:

  • Shashank Khaitan, a Bollywood movie director whose directorial debut was the movie Humpty Sharma Ki Dulhania. I believe there’s enough material online on Khaitan to justify a Wikipedia page, and was mildly surprised that the page didn’t already exist.
  • Bored Panda, one of the top viral websites, broadly in the genre of BuzzFeed (before it became an uber-serious news outlet) and PlayBuzz. I was surprised to see that Bored Panda didn’t have a Wikipedia entry.
  • Form I-140, a United States immigration form.
  • Expedited removal, a concept in United States immigration enforcement where people may be deported directly without their deportation case going through a judicial process.
  • Arriving Alien, a concept in United States immigration enforcement for somebody who has recently arrived.
  • Immediate Relative (U.S. immigration law), a concept in United States immigration law for a type of relationship that makes one eligible to apply for family-based immigration categories that do not have numerical limits.

I am also considering making edits and improvements to these pages:

  • Cecilia Munoz, a member of Barack Obama’s administration who previously worked as an immigrant rights advocate but has recently come under flak for defending the Obama administration’s record deportation numbers. I’d like to update the page to reflect recent events, and also uncomment some portions in the Wikipedia page that were commented out due to a lack of diverse citations.


For background information, see my site page about WikiHow

I did not create any new WikiHow pages. The pages I had created in May got additional pageviews, as described below:

For the month of July I am considering making one WikiHow page, focused on how to reduce Amazon EC2 costs. This will be based on my experience at my workplace, but won’t involve any information specific to my job. I believe that laying out this information clearly will be helpful to everybody: users can save costs, and more efficient use of the computational infrastructure of the world means more value can be squeezed out of it. Whether I can do a good job writing that page is a moot point, and we’ll only know once I start trying.

Subject wikis

General background information: My site page about the subject wikis

I did not spend any time on the subject wikis. Traffic was within the range of expectation, with robust year-on-year growth (but a month-on-month decline, as expected based on academic seasonality).

Metric Jun 2015 May 2015 Apr 2015 Jun 2014 May 2014 Apr 2014
Pageviews 48,593 77,232 88,046 38,363 63,211 66,233
Sessions 24,136 37,834 43,968 18,695 31,225 34,310
Pages/session 2.01 2.04 2.00 2.05 2.02 1.93
Pageviews in sessions with at least 5 pageviews 16,314 26,016 27,991 13,065 20,727 19,837

Here are pageview counts for the other subject wikis:

  • Market: 11,808 pageviews, versus 13,477 for June 2014.
  • Calculus: 11,575 pageviews, versus 9,310 for June 2014.

Comparison with forecast: The total number of pageviews for Groupprops (48,593) was close to my point estimate of 46,000, and well within my 80% probability interval estimate of 35,000 to 60,000.

July forecast: In July 2014, Groupprops had a total of 34,973 pageviews. Based on the robust year-on-year growth trend, my estimates for July 2015 are as follows. My point estimate for the number of pageviews is 46,000. My 50% probability interval estimate is 40,000 to 52,000. My 80% probability interval estimate is 33,000 to 60,000. My 95% probability interval estimate is 25,000 to 70,000.

I don’t intend to put in any work on the subject wikis in July, beyond fixing reported errors.

Open borders

General background information: My site page about open borders

After a fairly long time, I drafted a blog post for Open Borders: The Case. The post, titled My reasons for skepticism of linking open borders to legalizing private discrimination, was published on the site on June 17, 2015. I also created the Wikipedia page on the Chinese Confession Program, which was an outgrowth of reading and research I was doing for forthcoming Open Borders: The Case posts.

I aim to spend more time on the site in July, resuming the site revamp as well as publishing more blog posts. I believe this will help with a general revival of the site after a significant decline in the publication rate of new material.

Other sites and forms of online presence

  • This website (vipulnaik.com) got 2,154 views in the month of June according to Google Analytics (and 1,871 views according to WordPress Analytics).
  • The views for Quora content created by me were as follows: about 9000 for questions, about 18000 for answers, and about 250 for posts.
  • Wikipedia Views got 1,050 views according to Google Analytics.

Friends’ projects

  • AI Impacts, started by Katja Grace and Paul Christiano, and managed primarily by Katja, got 3,705 views.
  • Cause Prioritization Wiki, started and managed by Issa Rice, got 2,914 views. Issa also set up web traffic minus spam referrals, and got a count of 1,087 views for the time period June 14-June 30 for the site, which extrapolates to 1,918 views for the month (note that the proportion of traffic that is spam referrals goes down as the overall pageview count of the site goes up).


I bought a few songs: Waka Waka (by Shakira and FreshlyGround), Blank Space (by Taylor Swift), and Samjhawan (unplugged) sung by Alia Bhatt (you can read the history of these songs, and get the names of the composers and lyricists involved, on their Wikipedia pages). I also watched the movie Humpty Sharma Ki Dulhania (my interest in the movie was initially piqued by the song).