Added Flattr buttons

Yesterday I decided to start using Flattr on a regular basis. The intention is to be mainly a supporter, but I certainly won't refuse if someone insists on sending a flattr my way ☺. As a consequence, you now see a "Flattr this" button at the bottom of every page on this website - you can use this to express your general approval of the site. If you like a particular post, you can also use the button on that post. This being a Drupal site, Flattr integration naturally is realized with the Flattr module. The documentation for the module is a bit sparse, so I wrote a few notes about its configuration on my wiki.

Little Go "sales" statistics

Little Go 0.12.0 has now been in the App Store since September 11, i.e. 4 days. I thought this would be a good opportunity to publish a few numbers from the iTunes Connect sales statistics:

  • Number of updates on the first 4 days: 976, 5537, 1785 and 1211. This shows that the majority of people are updating on the second day.
  • The total number of updates is 9509. I wouldn't go so far as saying that these people are all actively playing Little Go, but at least they haven't deleted the app yet, so it's probably reasonable to assume that this is the user base that thinks the app is not a total crap ☺. I expect quite a few updates more in the next couple of days and weeks - for comparison, when I published 0.11.1 in May 2013, the total number of updates in the first 4 weeks was 12724.
  • Since July, between 50 and 100 new devices have downloaded Little Go per day. I don't know how many new users that is, but it's certainly not too bad ☺. Apparently the app is still gaining some interest.
  • Looking at the monthly rates of new downloads, the picture is this: Between August 2012 and January 2013 the rate was steadily climbing from 1315 to a peak of 3415 new downloads per month. Over the last 7 months, the rate then slightly dropped to a monthly average of ca. 2800 new downloads.
  • In 2012 the top 6 markets were USA (3744 downloads), Germany (2678), France (1630), Thailand (1362), the Republic of Korea (1276) and Russia (969). Runner up on 7th place is Turkey (368).
  • In 2013 the picture so far is pretty much the same: France (4517), Germany (3884), USA (3242), Thailand (2234), Russia (1900) and the Republic of Korea (1348). As in 2012 the runner up is Turkey (559).

Little Go 0.12.0 released

Little Go 0.12.0 has been released to the App Store a couple of days ago. Since I plan to slow down my development activities on the project soon, the overall focus of the release was on improving the usability of Little Go, and less on adding cool new features.

The GitHub issue tracker has a list of all issues that I worked on for 0.12.0, and here is an excerpt of the changelog for the most notable changes:

  • It is now possible to configure the computer player with a threshold how quickly it will resign a game (#133). For instance, it is now possible to tell the computer player to never resign so that the game can be played out to the very end. This also allows beginners to play with a large handicap on a small board (up until now the computer player would always resign immediately when faced with an overwhelming handicap).
  • When an old board position is viewed, the intersection where the next stone will be placed is now marked with the letter "A" (#101). This can be disabled in the "Board position" settings.
  • Changes to the active profile are now immediately applied to the GTP engine (#123). For instance, it is now possible to change the playing strength or the resign behaviour of the computer player without starting a new game.
  • The user interface for changing the dangerous "Max. memory" profile setting is now vastly improved (#153). The maximum value that can be selected for the setting is now limited to a fraction of the device's physical memory, and the amount of physical memory that the device has is also displayed.

Finally, the GTP engine is configured to no longer recognize positional superko (#171). This is a temporary solution to bring the GTP engine's rules into sync with Little Go's rules. Little Go currently does not recognize superko, so this sync'ing fixes the problem that Little Go lets the user make a superko move which is then rejected by the GTP engine. Many thanks to Brid Griffin for emailing me a bug report that helped me with diagnosing this problem! Note that Little Go will officially support superko in 1.0.

And so they slowly fade away...

A few days ago, the news carried a short notice about Richard Matheson's death. I was a bit surprised at first, but then remembered that his novel I Am Legend had recently been remade into a blockbuster movie, so that's why the media know him. Just now I did some research while reading Isaac Asimov's "A Memoir", and so found out that another Science Fiction author, Jack Vance, has passed away a month ago, this time entirely without my noticing. Jack Vance is one of the few SF authors that I absolutely adore, and in the SF field he is rightly famous - but apparently not famous enough in the eyes of the general populace to warrant even the smallest notice in the papers. Just now I feel very sad.

Report from the spam front: Greylisting remains successful!

It is now 2½ years since my last post on greylisting, and over 3½ years since I reluctantly started to use this spam prevention technique. I am pleased to report that greylisting remains as successful as it was on day one.

In fact, the situation has even improved since that last post 2½ years ago: The overall spam rate has dropped from 36 to 14 messages per day! It is unclear whether the reason for this is a world-wide decrease in spam mails, or a decrease in "quality" of spam mails, i.e. fewer spam mails make it past the greylisting wall. I don't need to know the exact reason, though, to feel happy about the result: Due to the low overall spam rate, on average only 1.4 messages per day now make it into my inbox. Wow, these days I can even leave my mailbox unattended for a few days without getting flooded - isn't that great?! (compare the current situation to when I only had Spam Assassin and, at the peak, my inbox was inundated by more than 40 new spam messages per day).

For more statistics details, see this wiki page.

Little Go 0.11.1 released

Little Go 0.11.1 has made it into the App Store yesterday. I decided to make this bugfix release now, and not wait another month for 0.12.0, so that I could get the fix for a very annoying bug out to "my" users. The full story can be found on the GitHub issue tracker.

Damon Lindelof getting close to the brink

A long time ago I decided that I would never ever watch another movie directed by Roland Emmerich. Not every movie on my menu must be super quality, plain good entertainment is usually OK, and sometimes I even enjoy to see trash films. But I always felt insulted by the stuff made by Emmerich because he seems to assume that the viewers of his movies (e.g. me) are dumbasses that are happy to see explosions, and never mind the story. So these days Emmerich movies are a no-go for me.

A few days ago I saw Star Trek Into Darkness. After maybe half the movie had passed, I felt how I got angrier and angrier at all the stupid mistakes in the story, until in the end I had to say: What a dumb movie! Then I happened to see the writing credits, and everything became clear: Damon Lindelof has done it again. As a writer this guy just seems to be abysmal, and now he is getting reeeeally close to the brink of my Emmerich hole...

Little Go 0.11.0 released

Little Go 0.11.0 has been published on the App Store on April 23 2013, almost 4 weeks ago. For some reason I completely forgot to write this announcement, and I also forgot to update the master and develop branches on GitHub. I just fixed this a few minutes ago...

For iPhone users this new release finally brings the long-awaited zooming feature, making it much easier to place stones on large boards. iPad users can also zoom, but this is of much less value, except maybe for the iPad Mini. It is now also possible to display move numbers and coordinate labels, things I originally planned to implement much sooner, but somehow there was always stuff to code that seemed to be more important. So in the end I am not sure how important these two features really are, but they are at least a nice graphical addition to the UI.

Something of which I am sure that it is very useful is the new ability to export and import .sgf files to/from other apps such as Mail or DropBox. Working on these features prompted me to propose this draft of a UTI specification for the .sgf file format, but so far the announcement on the sgf-std mailing list has not received any responses. Ah well, it's probably not the most important thing in the world ☺.

Little Go project page has moved to GitHub Pages

The project page for Little Go has moved to https://littlego.herzbube.ch/. The new site is hosted by GitHub Pages, accordingly the site source files can be found in this gh-pages branch. The only resources for the Little Go project that are still on my private server are the third-party software tar balls required to build Little Go.

Little Go 0.10.0 released

As of this morning, a new version of Little Go (v0.10.0) is available in the App Store.

A lot of work has gone into this feature release: Almost three months, much more time than I had originally anticipated. The biggest chunk was spent on adding the function for viewing board positions for moves played earlier during the game. The main problem I had here was designing the user interface: On the iPhone I had to add UI elements to an already crowded screen, and on the iPad I gave myself the challenge of revamping the entire "Play" tab.

Another thing that kept me busy were the graphic design changes: Adding artwork for the Go stones, and adding a wooden background to the Go board. With these changes the app suddenly looks much more attractive, which I find very satisfying because usually I am such a nil when it comes to polishing up a purely-functional UI.

No single release in the history of the project has more changes than v0.10.0, both visible to the end user and technically under the hood. On the one hand this is good because it tells me that the code is still malleable, on the other hand I am a bit nervous because many changes usually also means many new bugs. Let's hope that it doesn't get too bad.