Why Are Websites So Annoying?

 The UI design world has gone crazy.  A few arbitrary trends have become nearly ubiquitous despite being universally annoying.  Why is annoying your potential customers so popular?

1.  The most obvious feature of websites today is the inevitable popup that the designers shove in your face before you’ve even had a chance to view the rest of the site.  (Perhaps even worse is the variant where they give you 10 or 20 seconds to settle in and start digesting their content, only to interrupt you with the popup then).  Enjoying our site?  I just got here.  Give me a chance to form an opinion.  Like us!  Subscribe!  Get a discount!  Now or never!
   
   These popups are so infuriating I wish I could choose to simply never again visit any site that assaulted me in this way.  But if I did that, I’d need to give up using the web entirely.  The practice is nearly universal.  Why is there no innovation or differentiation in this area on different websites?  It would be a competitive advantage to offer a peaceful website that didn’t use popups.  There are plenty of other ways to offer discounts or solicit newsletter subscriptions.  I seriously doubt these popups are the most effective. 
   
   I know popup blockers exist, but they have drawbacks.  Sometimes they block functionality I want.  Sometimes websites detect popup blockers and refuse to work.  Plus, why should I need third-party protection?  Why don’t websites use techniques that actually make the sites appealing and easy to use?  The most successful websites in the world (think Google and Amazon) don’t use popups.
   
   Note that I’m not talking here about the “We use cookies” notices that every web site employs.  They have no choice about this because it’s legally required by Europe’s GDPR law.  This well-intentioned but ultimately annoying and useless provision has increased privacy and data control by exactly zero for the average user while degrading our web experience.
   
   

2. Hover everything.  So many websites do stuff when you hover over things on the screen.  Menus that expand on hover are the most prevalent, but often hovering over an item on a list is used as a sort of soft selection that causes additional information to be displayed.  Some video sites will even start playing the video if you hover over it for too long.  Perhaps most annoying is buttons that don’t show up unless you hover over the right area of the screen.  How were we supposed to know this option was available like this?  Do you expect us to hover over the entire screen to try to discover what commands are available?
   
   The main problem with hover everything is that it’s not safe to leave your cursor anywhere.  You have to actually spend time to find a safe place to rest your cursor so you can actually view the website without interference. 
   
   And since hovering isn’t even possible with a phone or tablet, why would anyone build in hovering as a major component of their site?
   
   

3. Flat design.  This disease affects the Windows operating system perhaps even more than it affects web sites.  In my opinion, Windows design peaked with Windows XP.  In those good old days, you could tell a button was a button because it had a 3-D appearance that looked like a physical button.  Old websites worked the same way. In this way, the computer was adapting itself to the way our minds work in our everyday world.  Nowadays, button-like UI widgets are distinguished from inactive elements by more subtle characteristics like fonts and layout.  The only way you can know for sure if a UI element is clickable is to hover over it and see if it changes in a way that says “click me”.  We are forced to learn the “design language” an OS or website uses in order to become effective with that interface. 
   
   Flat design not only makes interfaces difficult to learn, it can make them completely unusable.  Now that Windows title bars and borders are completely flat and colorless, it’s impossible to tell where one window ends and the next one begins.
   
   And since different UI elements are all flat and as colorless as possible, it’s not easy for websites to become beautiful and differentiated from each other.  Why is flat design so ubiquitous?  Why no innovation or differentiation?  

The “why” question about these UI design trends is pretty confusing.  My theory is that the designs are being driven from the technical side and decisions are being made based on what’s possible and easy to implement instead of from the design and marketing side driven by what would create a great user experience.

I think my theory is supported by the way “exit intent” popups appeared on the web all at once like a fit of collective hysteria.  These are the popups that appear when you quickly move your cursor towards the part of the screen where you can switch and close tabs.  Wait!  Don’t go!  Why don’t you like us?   Please like us!  We’ll do anything if you don’t leave!  You’ll be sorry if you leave now!  Please do free labor to tell us how to do our jobs better!

One day in 2012, somebody figured out how to determine exit intent and this mechanism spread like wildfire since it met an existing need in traditional marketing, namely customer retention.  They actually use phrases like “page abandonment” to describe user behavior.  Feel insecure much?

Does anyone actually respond to these desperate and disruptive UI features?  I expect very few do.  But it must be non-zero, or else nobody would use them, just like how spam and junk mail wouldn’t exist if zero people responded to them.  But is it really worth it to them to receive these few conversions at the cost of annoying everyone else?  I read a few sources that say popups do indeed work, but that they’re not worth the cost.  In addition to user annoyance, popups will lower your search engine rankings because Google penalizes the behavior.  So it’s possible that design decisions are being made at least partially from the marketing side, but by dumb marketers.

Unfortunately, I don’t think there’s much we can do about these annoyances.  We’ll simply have to wait until some large company or explosive startup creates a new trend.  How about “clean design” or “intuitive design”?  Websites can do something called A/B testing, where they have two different site designs that are presented randomly to half of their customers.  They can then measure which design performs better and then roll that out to everyone.  Someday--soon, I hope--a major A/B testing result will show that these design trends provide an overall decrease in website performance.  Amazon has proved that great user experience sells, so eventually others should come to their senses.

How To Solve Wordle More Often and Possibly Ruin the Game

 

If you’re reading this, then you probably play Wordle or are at least curious about it.  There are many appealing features of the game, including the social aspect and the fact that it’s limited to one play a day.  If you particularly enjoy these features, then my advice on how to get better at the game may not be desirable.  But if you just want to get better at solving the puzzles more of the time, read on.

The first way to get better is to play more often.  Of the many Wordle clones and variants available, I’ve found https://octokatherine.github.io/word-master/ to be the best for just practicing.  It has the exact same rules as Wordle, but the tiles turn faster and you can play as many times as you want.  Just using this site will make you better at Wordle, which you can continue to play for its social aspects.  All of the examples in this post are from the WordMaster site.

Using this site, I’ve played many games and come up with some strategies that seem to help me solve the puzzle within six guesses at least 99 times out of 100.  Below are the main strategies I’ve discovered.

Opening Guesses

I always start off with the same two words, AILED and ROUTS.  This covers all of the vowels (except Y) and some popular consonants.  Using standard opening guesses like this may make the beginning of the game less interesting or exciting, but it produces more consistent results. Obviously there are many other words you could choose as your opening guesses, if you prefer to find out different information or just prefer different words.

The main downside of using two standard guesses to open is that your average number of guesses may go up.  With this technique, it’s less likely you’ll get lucky and solve the puzzle in three guesses or fewer.  I don’t personally care about this; I just want to solve the puzzle within six guesses virtually all the time.  Once in a while I’ll skip the second standard guess if the first one turned up, say, four hits.

Making the Most of Your Guesses

If you share my goal of solving the puzzle successfully (as opposed to getting it in fewer guesses) then you can get a different perspective on why you make the guesses you make.  Assuming you don’t play in hard mode (which seems impossible and not that much fun), you don’t need to use all of the information you already know in your subsequent guesses.  

The main purpose of each guess should be to maximize the amount of new information you get from the guess.  This often means choosing completely different letters from those that have already been revealed to be in the solution.  Of course these known letters will come back into play in your final guesses, but intermediate guesses should be fact-finding missions for what you don’t yet know.

As an example, suppose this was the result from your opening guesses:

You might think “Oh, that A probably goes in the second position” and be tempted to guess something like “games”.  Maybe you’ll get lucky, but probably not in this case.  If you think about it, there are a tremendous number of other words that also fit the same pattern.  For example: wanes, panes, manes, vanes, banes, bakes, makes, names, manes, wakes, cages, pages, wages, mages, sages, sakes, napes, vapes, paces, capes, caves, naves, paves, waves, mazes, etc.  Obviously, you don’t have enough guesses to try all of these.  If you get a thrill out of taking a big risk and trying to use your “intuition” in cases like these, then go ahead and take your best shot.  But there is a better way to solve puzzles like these.

The way to solve more reliably in cases where there are too many possibilities is to make guesses that identify which of the letters are involved first.  This will require ignoring what you already know, because guessing A in the 2nd place is probably a waste of time and guessing E and S at the end is definitely useless for giving you new information.  Instead, pick a word that covers as many of the possible letters for positions 1 and 3 as possible.  But importantly, you first just want to identify the letters, not their positions, which we already know are 1 or 3.  Looking at the list above, you can see that we’re talking about C, M, N, P ,W, K, V, G, S, and Z.  A lot of letters!  You’re doing well if you can find any legal word that has three of these consonants.  Four is ideal, but tough to do.  When I played this game, my third guess was CHAMP.

The H wasn’t really likely, but this was the best guess I could come up with, covering C, M, and P.  Note that using A in this guess wasn’t necessary, but does help provide confirmation of our expectation that A is in position 2.  We now know this with 100% certainty because positions 4 and 5 are taken and A has been ruled out of positions 1 and 3.

I hit paydirt with M, but that still didn’t narrow it down enough.  It could still be manes, makes, names, manes, mages, mazes, and maybe more I hadn’t thought of.  Still too early to be making final guesses.  I did another fishing expedition.  It was tough to come up with another guess that covered the other remaining consonants (N, W, K, V, G, S, and Z).  The best I could do was to cover K and N.  

Note that I used vowels that I already knew weren’t in the solution because all I was after was the identity of the remaining consonant.  I got lucky and found that it was K.  This made the solution 100% known:

Maybe you’re not impressed with a five guess win, but this one could have easily made you run out of guesses if you had tried to make final guesses starting on guess three.  Many of the puzzles are like this.

Keeping Notes

I’ve found it can be helpful to use the game as a way to “take notes” about what you’re thinking.  For example, suppose you got the puzzle below.  I had three letters known from my opening guesses, so I made a “final” guess on guess number 3.  Okay, DINKS is a weird word and probably not the solution, but I thought D was probably in position 1 and just made the guess.  

 

Clearly, D is in position 4, so it’s just a matter of making guesses to determine the letter in position 1, right?  But before doing that, it’s good to take a moment to figure out how many possibilities there are and if you have enough guesses left to guess the possibilities one at a time.  

This is where using the game to take notes comes in.  You can enter each possibility for position 1, as follows:

Obviously, this isn’t going to form a valid guess.  It’s just keeping track of which possibilities there are and how many possibilities exist.  I came up with four (FINDS, MINDS, BINDS, and WINDS), and I entered the starting letter for each of these possibilities.  Since there were (at least) four, this told me I didn’t have enough guesses left to guess them one at a time and be sure of solving the puzzle within six guesses.  Time to go consonant fishing again.

Taking notes like this on a consonant possibility also helps you come up with a guess that covers as many of the possibilities as you can.  After staring at FMBW for a while, I came up with WOMBS, which covered three of the possibilities.  It hit on W, after which the solution was 100% known.

When You Get Stuck

Sometimes you are confused by a puzzle and can’t come up with even a single word that fits with the known information.  You can always guess something like CHAMP or PINKY that just covers a lot of letters, but this is a last resort.  What can help you think about more likely possibilities?

This is another way I use the game board to help me think.  I come up with things I’d like to know, or possibilities I’d like to try out, and I enter these on the board with X standing in for unknown letters.  

For example, with this puzzle, I had three of the letters, but no positional information known even after three guesses.  I decided to try seeing if E could be in position 5, so I entered the pattern below to stimulate thinking along these lines:

After a while, it became clear that this wasn’t a likely pattern.  Where did the R go?  Obviously a vowel was needed in the middle of the word, but E had already been ruled out in position 3.  Was there an E in position 2, or was Y used as a vowel?  I couldn’t think of a single way this pattern could work, so I tried a different one.

This pattern included all three known letters in positions where they hadn’t already been ruled out.  So if I could come up with even one word that fit this pattern, I would likely get the positional information I needed.  I did find one word that fit the pattern, and this turned out to be the solution.  It was much easier to come up with this somewhat unusual word by staring at the template pattern with the Xs.  

Things To Remember

It’s easy for me to forget that the solution may have repeated letters.  It seems like a waste to guess a second copy of a letter when you already know it’s in the puzzle.  But repeated letters are a real possibility, so it’s important to include possibilities with repeated letters when you’re coming up with lists of possible solutions.

I also struggle to decide when to guess Y.  It’s rarely used as a vowel in the middle of the word, but sometimes is.   When it’s used at the end of a word, it can be preceded by a lot of things like L, N, D, or repeated letters like PP, TT or NN.  When I do guess it, it’s usually in position 5.

The main time I guess Y is when there doesn’t seem to be enough vowels.  However, when there’s not enough vowels, it’s more likely that a repeated copy of a known vowel is part of the solution rather than Y.  I guess Y only after repeated vowels have been ruled out.  

The other thing I need to remember sometimes is to double check my guesses and make sure that I’m not guessing something I already know to be false.  It’s a good practice to scan up the columns before entering your guess.  Look to see if the letter you’re guessing has already been guessed in that column.  Sometimes you’re forced to make a guess you know is false, but it’s a waste of a guess to do it by accident.

Summary

To summarize, I’ll just give you one more example which combines the techniques described above and shows why you should focus on gaining new information rather than guessing the solution prematurely.  

Suppose you got the following information from your opening guesses:

Now playing A again in position 1 would give you zero new information.  We need positional information on R and E, and we need information on what other letters are part of the solution.  

Here’s what I guessed:

Why is this a good guess?  We know it’s not a possible solution, since it doesn’t include A in position 1.  E in position 5 doesn’t seem likely either, since we’d need a vowel in position 2, or more likely 3, and it looks like we have all of our vowels already.   

Note, however, that this result tells us exactly where E goes with 100% certainty.  It must be in position 3 because 1 and 2 are occupied and 4 and 5 have been ruled out.  

In the meantime, this guess allowed us to try out two new consonants and gave us the position of R.

What was the answer though?  I entered the known information about A, R, and E to see if any words would jump out at me.  

Words that start with vowels are difficult for me.  I couldn’t think of any candidate solutions at first, until I remembered to consider possible repeated letters.  Then the solution came to me.  

I hope these tips help you be more successful at solving Wordle and related puzzles and that you feel smart as a result.  I still play Wordle and have fun with it.