Why another game in this genre?

---

I'm a retired software engineer with 30+ years in the industry and, as software engineers often do, we write software to solve problems or fill gaps that existing applications do not address. In my case, I have a couple of hobbies that are not covered very well. 

For example, I'm an amateur skater and when I'm preparing for a competition, I have a handful of print outs for skate dance diagrams for dances I need to learn for the upcoming competition.  These are filled handwritten notes to remind me of goals and issues I need to address.

As an amateur French Horn player, I have a similar stack of sheet music that has it's own set of notes as well as MP3 files I use to practice various excerpts.

Over the last two years, I've written custom applications that allow me to organize and manage these 'paper' copes as PDF files, create workbooks of music excerpts to allow me to play specific parts of the MP3 file when I'm practicing an excerpt,  and annotate and manage them on my Windows notebook.

These applications worked well when I'm sitting in my office but not so well when I'm out and around so I started researching various applicaton frameworks that would allow me to ryn these applications on my iPad or Android phone.

These applications are quite complex and I needed something that was more constrained to allow me to perform a reasonably rich evaluation of the frameworks to understand their capabilities and issues.  

In the end, I decided a word guessing game fit the bill.  It had interesting layout possiblities, could be self-contained, and avoid platform-specific problems that would distract from the evaluation.

The vast majority of my experience is with .Net and C# so I filtered the candidates down to those that allowed me to continue working with these technologies.  After a couple of efforts with the frameworks, I ended up deciding on .Net Maui ( .NET Multi-platform App UI (.NET MAUI) | .NET (microsoft.com) and started focusing on the game design and prototype.

As I started prototyping the game, I put together a list of goalsf or the game. Most were centered around the game itself and how it played, while others are a result of poor experiences I've had with various games I've played in the past.

Primary Goals

• Consistent experience across platforms.
• A variety of word lengths.
• Provide a casual and challenging model (easy verss hard word lists). 
• Offline play - an internet connection is not required to play the game. 
• It should automatically save games in progress when the application is closed.
• Curated word list (no plural, no banned words, no proper nouns) 
• The ability to look up the definition of the word at the end of each round. 
• All words have valid definition(s) available in the various free online-search/dictionary engines, such as Bing, Google, and Merriam-Webster.

Secondary Goals

• No ads. 
• No data collection, sharing or third-party tie-ins. 

Below, I'll discuss some of these goals and how they impacted the design and implementation of the game.

At the end of the day, I want to be able to play the game without logging into a website or service and, if I don't have good connectivity, I still want to play the game and I don't want to wait.

This means the list of words is included in the game, not on some web site. Future versions might allow downloading other word lists but the game will remain an off-line game.

Caveat: The game provides the ability to look up the definition of the word after each round. If you are offline, the link will not work but the game remains playable.

Wordell!, like other casual games, provides a nice distraction from your day-to-day tasks. However, it should provide a level of challenge or it quickly ends up in the trash bin. 

Eventually, we all have put it aside and get on with our day work and I find nothing so frustrating as getting stuck on a word and losing all progress because I have other things to do.

To avoid this frustration, the game saves the current state of your game.  If you're in the middle of a game and have to close it, the next time you open the game and select the same option, the game will resume exactly where you left off, showing your previous guesses and even the letters you typed for your next guess. 

This saving is done for all word-lengths and difficulties. If you're part of the way through a 4-letter word, decide to try a 5-letter hard word, and go back to the 4-letter word, the game resumes where you left off.

Of course, if you're really stuck and close the game in frustration, that word will still be waiting for you.

I prefer playing the 5-letter word version while my wife insists on playing the 4-letter version. Who am I go argue :)

In practice, I really wanted more than just two choices, so I added a 6-letter version.  After trying out the intial prototypes, I quickly decided some of the words were just too esoteric(?) for casual play and I introduced the concept of a 'hard' word list.  Now, each word length has a 'causal' and a 'hard' list.

The 'casual' list contains words that the user would expect to encounter in normal day to day conversations, in a movie, or on a television show. 

The 'hard' list contains all the words in the 'causal' list plus words that are less common. The words are in use but are not necessarily in the average users typical vacabulary.

The decision to have the 'hard' list also contain the 'casual' was ensure the hard list was still playable by most users; otherwise, only the most hard-core word game players would ever try it.

For future versions I'm considering two other types of word lists, expert and random.  

Expert will be for the hard-core word game players listed above. Random will have a casual and hard option like the 4, 5, and 6-letter word list, but the word length will be selected randomly.

I find one of the most annoying trends with casual word games is the proliferation of advertisements in the game. While I can understand the need to monetize the work that goes into a game, the advertisements are often intrusive; especially when they disrupt the game play.

I started writing Wordell! with the explicit goal of having no advertisements.

Imagine buying one of your favorite board games, opening the box and finding out you need a valid WIFI connection to play it. Then, before you can play, it requires you to provide your name, address, and your personal phone book and accept that it will share this information with companies you never heard of for reasons that are vague at best.  I imagine many of you would either toss the game in the trash or take it back to the store and demand a refund.

The practice of collecting personal data and sharing it has become an insidious practice used by many applications provided through the various online application stores these days.  While it's virtually impossible to avoid in our current online environment, I cannot justify participating in the practice.  

As a result, Wordell! neither collects nor store personal information and, since it doesn't collect it, it can't share it.

Caveat: The various application stores do collect information that allows the developer to understand issues with the application including things like the the device type, the operating system, crash files, and even application logs.   

Conscientious companies will do thorough reviews of their software to ensure personal information is not logged and restrict access to crash dumps to ensure inadvertant personal information is not accessed.

Wordell! takes this practice a step further but not requesting any access to personal information to ensure it doesn't access it in code or inadvertently log it.

In the future, I expect the game to include the ability to capture and send its own internal logging to the developer to aid in debugging an issue.  If I do so, I will honor the no personal data collection policy, the choice to send the data will be the user's decision, and the only user information will be the email to reply to and anything the user discloses in the problem description. 

I originally intended to provide the definition of the word at the end of each round but found the cost to be prohibitive for a free game.  I would need to charge up-front for the game or require a subscription; something I'm not willing to do for such a simple game.

Instead, I provide a link at the end of each round that launches the web browser to look up the definition using one of the freely available services.  The default is to open a web browser to the Merriam-Webster site.   The game's settings page allows you to choose between Merrian-Webster, Bing, or Google.

If I get sufficient user feedback, I may add additional choices.

Obviously, the word lists are a key component of the game and this effort is the bulk of the work. In fact, it's proving to be more work than writing the code for the game.  Initial prototypes used word lists found on the internet as a method for getting the game up and running. For the release, we needed a better approach to the word lists.

We have few goals we want to achieve for the word lists:

• Provide a 'causal' versus 'challenging' list.
• No proper nouns.
• No plural words.
• No offensive words.
• All words can be successfully looked up in an online dictionary, such as Bing, Google, or Merriam-Webster.
• The lists should be large enough to ensure words are not reused too often.

For the first attempt at generating a 'curated' word list, we researched both free and paid online dictionary databases.  The available free dictionaries got us part of the way, but the word lists were rather limited.

The paid dictionaries were more promising, providng a richer set of words and reasonable certainty that the words could be found in the various online search engines and dictionaries. 

Unfortunately, all have one severe limitation, no data retrieved from the dictionaries could be cached. This means that the word lists themselves could not be generated from the dictionary, since embedding the list in the game is a form of caching.

Many were rejected outright due to the cost of using the service; a free game simply cannot justify paying for an online enterprise targeted service.  Some even refused to provide a quote after reviewing the game's intended usage.

One dictionary was instrumental in pointing us to a final solution and is very appealing from a cost perspective: WordsAPI® - English Words API - Get Started Now for Free! (rapidapi.com).  The rates are reasonable enough that I'm considering paying for the service out of pocket to automate curation of word lists we create independently.

Additionally, the dictionary provides a metric that indicates the frequency of a word's usage.  While I cannot use the data from the dictionary due to their caching constraint, their support staff was kind enough to provide a link to a paper that describes the metric and how it was calculated. This paper sent me down a path for determining what words should be in the 'hard' list.

My original thought was to use the number of definitions available as a deciding factor, but the paper suggested the frequency of word usage (SUBTLEX) is a better approach. I eventually found a page from the Department of Experimental Psychology on the Ghent University site:SUBTLEXus — Department of Experimental Psychology — Ghent University (ugent.be) .

This solved two problems, it provided a word set that I could use to generate the 4, 5, and 6 letter word lists, and metrics to determine how to categorize the word 'difficulty'.  The final result involved choosing a 'reasonable' threshold, manually reviewing the results, and creating exception lists to filter specific words.

Curating the list to remove offensive words proved to be more tractable than I originally expected. I spent time researching banned word lists, reviewing the words, then creating an exclude list that automates removal when the word lists are generated for the game.

I'm currently on the fence about the goal of removing all plural words from the word lists. At first glance, it seems reasonable to include the word 'chat' but not 'chats', should the rule also apply to 'foot' versus 'feet', or 'cactus' versus 'cacti'?  My inclination is to reject 'chats' but include 'feet' and 'cacti' and use user feedback to help refine it.

As far as automating the list, it's an area for future research.

Removal of proper nouns is currently a manual review process since I have not found an affordable dictionary service that I can use to automate it. For now, I'm using an exclude list to filter words as they are discovered.