Frequently Asked Questions: Technical

History, current size, and future of this site

This site, which I call , is constantly updated to present current data selected from my private database, which I call my Tunography. Later versions of will release to the public much more of the information that is stored in my database, including detailed notes for each recording of the tune. Since this project is an unpaid hobby of mine, I cannot give a projected date for when I will create those additional features.


I started the Tunography as a Lotus 1-2-3 spreadsheet around 1993 and began writing data manipulation and reporting routines (to manipulate the spreadsheet) using the database FoxPro ver. 2.6 around 1996. I converted the raw data to a Quattro Pro spreadsheet around 1997, and then converted the entire project to an Access 2000 database in late 1999. All data management operations have been carried out by Visual Basic routines within the Access database ever since then. At first, generation of all the report Web pages on this site were also generated within the Access database, and a few still are to this day.

The "Practice Machine" was born around 2000, initially as an Access database feature I built solely for my own use on my personal computer. It was inspired by the language-learning principles I had learned in the 1990s during my training to become a German professor. By April 2002 I was definitely relying on it for my daily instrument practice and was calling it my "practice machine."

This Web site was launched in February 2000 as a beta-phase, private project. In November 2000, interactive database query features began to be added using a free ASP server and VBScript. During February-March 2001, the site moved from its free university Web server and from the free (but very limited) ASP query server to my own rented NT server (, and I added affiliate links to CD and book retailers as an attempt to recoup some of my out-of-pocket costs. I bought the Web address in October 2001. During 2002 I undertook a major proofreading and proof-listening project. Then I was confident enough in the reliability of my data to finally undertake a public "launch" of the site.

I began the move from to in October 2002. I rewrote the entire Web site to run on a Linux server, using the php and MySQL languages instead of Microsoft ASP/Visual Basic/SQL. I also enlisted the invaluable help of nine generous volunteer beta testers to help me redesign the site, especially the Tune Search and Finder. Karen Cornelius, Shelley Gordon, Craig Heilman, David Moulton, and four other helpful souls were the beta testers, and our stimulating conversations are partially recorded for posterity in the archives (which died in late 2008).

I began redirecting traffic from to on 25 November 2002. Simultaneously I sent out e-mail press releases announcing the existence of this site for the first time ever, including inviting people to link to and publicly discuss and review the site. Over the next years more and more of this site's reports shifted to being generated by php scripts on the Web server, instead of by the Access database, and I gradually added more and more reports to pursue various musicological questions.

On 10 October 2004 I launched the "" features as a private beta-testing experiment shared with a few local friends. The initial core features were to track your own playlist and to use the Practice Machine. On this day Elizabeth Fine became the first user besides myself.

On 19 March 2009 I announced to the private beta testers that they now had access to the core elements of a social network among users. The initial features were the news feed with updates on recent activity of your friends, including e-mail notifications, and the user-profile pages. Building those features had begun earlier that winter, when it had become clear that Facebook was becoming the globally dominant social network, and that "regular folks" were adopting its use in their daily lives, so that I could assume that the concepts and tools of a social network would be relatively familiar to many Irish traditional musicians.

A public beta-testing phase of the "" features, which added the ability for users to store personal information about tunes and added social-networking features for users, began on 9 August 2009.

Beginning on 4 November 2009 one could use the full range of international languages to make notes and communicate on That was in fact only a prerequisite for the Google Maps integration that I launched that same day, which made it possible to collect the geographical location of every user, which I envisioned as opening up a vast new area of potential musicological research regarding the geographical flow and spread of repertoire, players, and influences. Simultaneously, the Google Maps integration required me to have begun using AJAX technology, which I implemented using the just-launched YUI 3 framework. Adding AJAX also opened up the floodgates to let me begin building a whole host of user-friendly improvements and features that I had been dreaming of for years.

In 2011 I finally felt the urgency and priority of a years-old ambition to include authentic audio recordings in the the experience. I devoted several months to building a rather custom and complex solution that finally resulted, on 11 November 2011, in "play" buttons appearing next to most references to a particular recording of a tune. Each button played the first 10 seconds of that exact recording. The major breakthrough was not so much the "play" feature but rather the infrastructure which almost entirely automated the task of creating, maintaining, and publishing over 10,000 little audio files, in a way that could painlessly grow to say, 100,000 such little files someday.

In the second half of 2013 a flurry of relatively "invisible" updates and improvements were made to features, partly due to the gradual demise of YUI technology (which I gradually replaced with jQuery), and partly due to user requests for features. The latter was a sign of increasing worldwide use of to manage player repertoires.

Most of 2014 was spent on following through on realizing the full potential of the 2011 innovation above, which involved manually identifying the precise audio location of all 20,000 individual tune recordings across some 12,000 audio files. Then at the end of 2014 I was able to build a significant upgrade to the Tune Search and Finder which not only offered those playable clips directly within the search results but also offered a completely new "responsive" design which meant that the rapidly increasing percentage of users using mobile devices had a much easier experience.

In 2015 technical work focused on expanding the"responsive," mobile-friendly re-design across the rest of the website, which includes various types of speed improvements that benefit all users. Finishing the years-old project of posting audio incipits for every single one of the 22,000 recorded sources was completed on 31 August. The personal tune tagging feature was added on 3 November.

Between 2015 and 2023, my work returned primarily to expanding data coverage and improving accuracy of the data.

In spring 2023, I was enabled by ChatGPT as my technical assistant to finally pursue an ambition I had been talking about privately for many years. I began building machine-learning analyses of the Tunography's data, particularly the leveraging the powerful potential of the Tunography database with its precise descriptions of – and verified associations with – almost 28,000 full audio recordings of tune performances. My goals are to enable future musicological analysis projects, create tools to accelerate my usual coverage-expansion work, and possibly to add new public features to


Development of new features on will likely continue indefinitely. Certainly I'll be expanding the tune data regularly - for current overall stats of the data coverage, see the reports listed on the front page.

You can follow the latest updates and features at the Facebook page for Or consider subscribing to the free Newsletter to receive only 1-2 e-mails per year.

All Tunography (and therefore ) data was entered and all programming was written and executed by me, leaving me with responsibility for all errors.

Credit for the image I adapted to create the "anonymous" personal-profile picture used in my. goes to: / CC BY-NC-SA 2.0.


I am sharing my ever-evolving to-do list for this site, ranked from highest to lowest priority. Feel free to give me private feedback on this list.

If you'd like to hear when I add new features, follow the Facebook page.

Can I use this site with any browser?

Yes. In fact, the entire site strives to meet global standards for accessibility to disabled people, especially for the sake of vision-impaired musicians who use this site.

I do optimize the site for Firefox (on Macintosh and Windows) and also make sure that everything appears as intended in whichever version of Internet Explorer on Windows is currently most widely used. In general you will see the intended visual layout with a browser less than 5 years old. If you haven't disabled your JavaScript, you'll get additional, non-essential features.

Nevertheless, I have designed this entire site to be usable by any browser, including text-only browsers such as Lynx, and for people who have limited bandwidth, such as in remote areas of the world without wired infrastructure.

How are search results sorted when using the Words mode of the Tune Search and Finder?

The following are quotations from the published documentation for the "fulltext" query feature of MySQL, the database I use.

"[. . .] the rows returned are automatically sorted with the highest relevance first. [. . .] Relevance is computed based on the number of words in the row, the number of unique words in that row, the total number of words in the collection, and the number of documents (rows) that contain a particular word."

"Every correct word in the collection and in the query is weighted according to its significance in the collection or query. This way, a word that is present in many documents has a lower weight (and may even have a zero weight), because it has lower semantic value in this particular collection. Conversely, if the word is rare, it receives a higher weight. The weights of the words are then combined to compute the relevance of the row."

"The MySQL FULLTEXT implementation regards any sequence of true word characters (letters, digits, and underscores) as a word. That sequence may also contain apostrophes ('), but not more than one in a row. This means that aaa'bbb is regarded as one word, but aaa''bbb is regarded as two words. Apostrophes at the beginning or the end of a word are stripped by the FULLTEXT parser; 'aaa'bbb' would be parsed as aaa'bbb."

See also: Which words are excluded from word searches?

Unfortunately I am restricted to using this less-than ideal search technology until can afford to upgrade its server to a $50/month level. Please consider supporting .