User Centered

Studying the design of everyday things

Subscribe to RSS feed

What day does the week start?

, ,

This post is meant to serve as a warning to web developers for United States websites.

I'm really excited about the new HTML5 Form Elements. Especially the new "date" input type: <input type="date">

Unfortunately, the "date picker" (the calendar popup that displays when selecting a date) is just a little too Euro-centric. Opera appears to have implemented the ISO 8601:2004 ($238.00!) standard that the "always reliable" Wikipedia claims is common in Europe**.

What does that mean? The week starts on Monday.

This site has an example graphic.

Unfortunately, weeks in the United States (I can't speak for you Canadians or Mexicans) start with a Sunday ("business week" starts on Monday). A few random unimportant like examples might be in order:

(okay, perhaps I've belabored it a bit)

Perhaps weeks starting with Monday are way better implementation? Perhaps they provide some unforeseen benefit but so-far as I can tell there's no profound improvement.

Now at this point, I should properly point out that I'd fully support the United States moving to the metric system of measurement but poor implementation in the past has been a major stumbling block (that and tools). There is a strong American sentiment for the "if it ain't broke, don't fix it." For the vast majority of applications for every-day users the old-fashioned way of doing things works well.

There is no compelling reason for them to switch (ease-of-use is hardly compelling when compared with the "hard-to-use hump").

In this case there are real issues with the spatial arrangement of the days of the week. An analogy would be if you're used to looking at a certain map North is always at the top then you might not immediately notice a change in such an arrangement. Getting lost sucks.

I routinely ignore the day-labels when doing "not today" operations with calendars -- such as "Oh, I need to do this next Friday," or "I worked on that project last Wednesday." In those situations, I don't know the date (nor is "today" marked on the calendar) so I merely mark the 'proper' position for a day.

So what does this all mean?

It means there is real risk for error in using an unfamiliar layout of something they feel they should be familiar with. In this particular instance I haven't been able to find any way of specifying how that date-picker widget works or styling it for "localization."

For now, it simply means using Javascript solution that puts the days in a layout proper for the locality. Though I'm hoping my Opera Forum posting or this post reveal some way of getting Opera to respect this particular locality usability issue.

Thoughts, ideas, questions?

Update:
* Added example of the input type-date syntax and link for clarity.
** Markus Kuhn's discussion on 8601 standard includes notes on the "week" - starting with the paragraph "In commercial and industrial applications" that also confirms common usage in Europe.

NTSB's Cell Phone Ban

There are many issues when it comes to the operation of a motor vehicle. The US National Transportation Safety Board, in a somewhat reaching opinion, have recommended the banning using personal electronic devices (any cell phone usage, etc.) while driving (other than those that support the task of driving, like GPS Nav).

Now the NTSB doesn't get to issue reports on driver safety at will but only as a recommendation after an an accident investigation. The reason they were involved in the investigation of this particular Missouri accident was because it involved school buses. As their report details, neither of the bus drivers were using phones.

In fact, the school bus accident had nothing to do with the driver of a GMC pickup truck who had been texting and struck the back of a tractor-trailer (empty, no trailer). The first bus driver had been distracted by a charter bus pulled off to the side of the road and didn't see the pickup truck stopped in front of him. The second bus driver had been following too closely and struck the back of the first bus. (The pickup truck driver and a student in the back of the first bus died)

Putting aside the odd roundabout way the NTSB chose to make it's opinion known on cell phones (after all, had the bus drivers been paying attention & driving safely, they would not have been involved in any accident) I wanted to consider whether an all-out cell phone ban was the answer.

My initial sentiment matches that of this NPR Report which included a quote about Prohibition, "it's unrealistic and it won't work."

Cell phones are not inherently bad for drivers. Distracted driving is. Do cell phones play a role in that? Yes, undoubtedly, but more important is the focus, training and concentration of the driver.

For example, the FAA bans unnecessary chatter in airline cockpits during critical flight phases. For the vast majority of flights this means lots of silence for 10 to 15 minutes of flight time at the beginning and end of flights, contrary to the nature of human beings; we are social. However, this cumbersome rule is observed because pilots receive heavy training and reinforcement.

In comparison, non-commercial and non-school motor vehicle operators receive a pittance of training (and, in fact, have no requirement for any formal training).

The root cause here is distracted driving, yet radios, clocks, passengers, food, water and makeup (girls, I'm looking at you!) have not been banned.

For me, this seems to fly right into the heading of "Usability" -- make these devices more usable and increase safety by decreasing the potential for distraction.

I have been a near victim of distracted driving. One of my closest calls, though, was the first day riding my motorcycle. I turned my head to the right to make the important determination of whether the girl on the sidewalk was as attractive as she seemed from behind unaware that a vehicle in front of me had come to a complete stop in an otherwise open traffic lane. I swerved, I was luckily in a position to miss and no one was in the other lane.

Should we ban short-shorts too?

Cell phones present a unique risk because of their ubiquity and their unique mental toll (time distortion from the mental toll of processing a message) but would that justify an outright ban? I can't imagine the thousands of people who sit in bumper-to-bumper traffic every day in Los Angeles, moving at 5mph would feel their safety margin had been increased by disallowing all mobile phone usage.

It's too broad a stroke and the NTSB is in error by applying the "trained operator" mentality to average consumers. They fail to take into account reality and that's bad for a safety organization.

Surely there are better answers? Or really should we attempt Cell-Phone-Driving-Prohibition?

The NTSB did recommend to Consumer Electronics Association, "Encourage the development of technology features that disable the functions of portable electronic devices within reach of the driver when a vehicle is in motion; these technology features should include the ability to permit emergency use of the device while the vehicle is in motion and have the capability of identifying occupant seating position so as not to interfere with use of the device by passengers."

Asking a lot and honestly, if we wanted to save lives by reducing accidents I think better driver training would accomplish it far more effectively.

Identification for Humans

One thing that annoys me about my industry is the insistence in many circles to use Social Security Numbers as standard identification numbers. Sometimes it's the whole number, sometimes it's the first 7, others the last 6.

In short, a very personal piece of data is splattered all over the place. (If you don't believe this is a bad thing then you should stick to the AOL Kids Only zone)

Now, I know I won't be able to replace this de facto standard but in working on my project, I can do something better. Unfortunately, I'm not a mathematician, so I'm not qualified to develop my own ID system but I do like to pretend. However, in this case, I wasn't able to find any "Best Practices" on developing individual human usable unique alpha-numeric identification.

More to the point -- quickly identifying people in a computer system for humans. (Names can be repeated, so can addresses, social security numbers are a no-no, and phone numbers are often used as the de facto secondary authentication method)

Here's some conclusions I've come to thus far:
  • Highly communicable in written and oral form (paper, email, text message, phone, vid conf, napkin)
  • Internal checksum or 'check digit' (like credit cards, barcodes, etc.)
  • Not generated based on any personal data point, except perhaps personal preference
  • No longer than 9 characters in length; preferably no longer than 6 characters (including checksum digit)
  • Total available IDs 8+ million (guesstimating)
  • Uses only alpha and numeric characters available as standard ASCII characters
  • Error resistant.
  • Not be mistaken for a phone number.
  • Limited vanity appeal - as dry as SSN.


To that end I've been generating some criteria based on various Internet sources, like base 32 and RFC4648 though not strictly related to my ultimate goal.

I am looking for feedback, but to kickstart and discussion I'll go ahead and give my first draft.

Yum's IDs
---------------------
ID Alphabet <- These Characters would decode (as an 'educated guess') to primary entrant
0	o	O	Q	
1	i	I	l	L
2	z	Z		
3				
4				
5	s	S		
6				
7				
8				
9				
A	a			
B	b	d	D	
C	c			
E	e			
F	f			
G	g	q	y	
H	h			
J	j			
K	k			
M	m			
P	p			
R	r	n	N	
T	t			
U	u			
V	v			
W	w			
X	x	Y	y	


Using this 27 character "alphabet" on a ID 5 characters long (+1 for a checksum), my math gives me 17,210,368 possible IDs. Though I'm strongly considering eliminating either P or R, V or W, X (awkward for mnemonic creation), J (because some people write it like a lower case I), and U or V (confusion and because, sadly, U is in 4 of the Seven Dirty Words).

Reduced to a 24 character "alphabet" for 5 character length IDs would leave me with 7,962,624 possible IDs. This would be reduced further if I implemented a rule against consecutive-repeating characters. (but 6 long = 191,102,976)

Also, I'm considering a start-character class (must be numeric or must be alpha or odd/even character; never starting with 0). The advantage being error-checking, disadvantage being, limiting the total pool of IDs.

[random tangent]My hope, is that mnemonic combinations can be auto-generated to assist in easy memorization of an ID. So if it gave you 2KE6P3 you might get "2 Kids Eating 6 Pizzas in 3!" (the more unique, the better - I've already memorized that code using the visual) - obviously there are a lot of caveats and those "mnemonically suited" combos would be exhausted quickly if that was the most popular option.[/random tangent]

More important than all that is the desire I have for an internal check digit to allow for quick, non-network dependent (but possibly computer dependent) verification of a valid entry. Something like Verhoeff Algorithm which can be used on numbers.

Using only numbers in a 5 digit ID, though, leaves me with only 100,000 IDs. I'm happy to apply the Verhoeff to my converted number.

(edit: removed stupid)

And now I feel the math here is safely over my head, so I'll stop.

Man, where is a bored math major when you need one?

Yum
(p.s. I relied heavily on this cutesy permutation calculator)

Technology for Non-Technologists

,

This older gentleman approached me today in the coffee shop. I had helped him a couple weeks ago connect his Android tablet (by Craig) to the wifi connection. Thus I labeled myself "free tech support." What's interesting is watching him interacting with his device and his profound confusion and misinterpretation.

Whoever sold him the device was obviously just trying to make a sale - he paid $100 for it and it's a constant headache.

We spent just about 40 minutes working on 'problems', in fact he just Columboed me with just one more thing.

I have to admit, many were quirky "I just don't understand the concepts involved here," but about 30 minutes in, as I was trying to gracefully extract myself from the situation I finally just concluded and said, "You should probably just get an iPad." In Apple's happy walled garden he could run free and after some initial setup things would just work because, often times, there is only one way to do something -- so once you figure it out, your set with many apps you encounter.

I'm not an Apple Cultist, I like the Apple products but I don't tell people to go out and buy them.

It was frustrating as he peppered me with questions like, "How do I get rid of Amazon?", "Why do my headphones not work right?", "Why are there songs but many of them in Japanense (or other asian script?", "How do I get my email to work?", "Why does it always say 'No network connection'?", "Why do my videos cut off in the middle?"

Some of this was clearly the inexperience with technology and the concepts involved but many of it was problems with the individual small interactions on the device - this is where Apple excels and why no one can threaten that market position because Android doesn't hold to those same standards of usability.

Yes, my grandfather had trouble with his iPad but once he figured out how to play music, access the NY Times, and check his email he was completely set and mostly things worked well for him. He used it all the time.

I don't believe anyone but Apple is willing to put the emphasis on usability necessary to actually compete with their products. My question is, why not? Why isn't simplicity of usage, usability, the number one priority?

Ideas?

Trash It

, , ,

Just watched my friend get a new phone. An Android.

He, like most, started installing new applications and having fun and then, like most, decided to uninstall some. He kept complaining to me that he couldn't get them to uninstall, he said he'd keep trying and they wouldn't go away. Finally, I asked him what was happening? He explained he was uninstalling it but they wouldn't go away.

So I watched him do it. He opened the home menu, grabbed an application icon by pressing and holding, the icon then "floats" as this functionality was used to drop icons on the "desktop" view. So he took the icon and dropped it onto the trashcan that appeared.

It was a priceless moment.

"Trash" is for getting rid of something and that's what he wanted to do. After all, on his windows machine or even an Mac you drag stuff into the "trash" or "recycle" to get rid of it.

Now, my fix for this would be a pop-up asking whether they wanted to simply delete the icon or uninstall the application but that seems a bit cumbersome.

...

A completely valid interpretation of the interface, in my opinion.

Choose Opera interview with Opera UX's Yenny Otero

, , ,


Great interview on usability at Opera with Yenny Otero:
http://my.opera.com/chooseopera/blog/2011/09/09/opera-talks-yenny-otero

Also note, we have our round table interview from 2007, which Yenny participated in as well

Plus One for Google Plus

, , ,

Social Network websites are kind of a fact of life -- I try to escape them but too often they are as vital as a personal phone book for others. Especially in my industry (technology and/or entertainment), it's just a tool which causes people to go cock-eyed if you don't have it.

So with trepidation I've signed up for Google Plus (Google+?) with an invite from a buddy.

And right away I feel a touch of fear as I read the disclosures about how they'll share advertising info and Plus One info... and then I realize, they have a check-box next to them. I don't have to let Google+ share with other websites and I can change that option later. Hm.

Okay, so after that, I signed in and was greeted almost immediately with the warm glow and excitement of "Circles" -- there are times in your life when you have something just beneath the surface of the ripple of thoughts, you've been organizing it a certain way and all of a sudden its as if someone said, "Oh, yeah that's how I do it and here's how I improved it!"

In Facebook I struggle (and I do mean struggle) to organize my contacts, having finally arrived at:


Why?

A couple years ago a really embarrassing picture was uploaded and tagged of me - sharing it automatically with basically everyone! In hindsight, it's kind of epic and funny and awesome but my family and certain friends/coworkers/future employers/students/etc would not have approved (that was such an epic understatement I trembled a bit). So I needed to lock down my profile.

They had "Friend Groups" and after hunting through the site (I really don't know why it's so hard to find, it used to be on the sidebar then it went somewhere else... ugh) I was able to get these groups. The Aces, the Jacks, the Family, the "WTF" (seriously, I don't think I've ever met you but... uh... maybe?). Now they've changed it to "Lists" (I think).

And then, I had to go through the endless privacy settings on Facebook (It's like a game of whack a mole where I try and hit the right setting for the right group). Then whenever I go to post something or put up a picture or whatever I have to figure out how to lock it down. It also seems every few months Facebook changes their privacy policy/settings and I have to reconfirm everything and lockdown new "features" they've "helpfully" added for me.

In short, Facebook's mantra is "Privacy is Dead, Long Live the Share Everything!" (except if you're famous then you have to protect yourself.. poor, poor Zuckerberg) and I'm not alone in this belief: NYT Article. It wasn't always like this but I'll spare you the rehash.

I hate that attitude -- it's prevalent in everything Facebook does and it means that at a fundamental level, I do not trust Facebook. I may have to use it but it's with the same sort of begrudging, pride swallowing anger that you trudge through airport security, taking off shoes and belts in front of complete strangers and rude (not always but as a rule) TSA agents.

So, when I saw this:


My eyes widened and I was all "Whoa!" and "Yes!" - this is basically the default screen. No, we don't want to share every little detail of our lives with every person on this big rock! We do not. Facebook keeps pushing, prodding and even pulling us into their vision of "share everything!" It's begrudgingly they add any support for any "privacy."

So I can't review Google+ -- It may completely suck! It may be horrible! I have barely got past my Circles and Profile page but on those two pages I see something attached to almost every item (In fact, I haven't found one item it isn't attached to). "Share With..." or "Who Can See This?" - A drop down that lists "Anyone" "Your Circles" "Your Extended Circles" or "Custom."

That is a major win for me -- right there, it's RIGHT THERE! No hunting through the long-list of privacy items on Facebook's somewhat buried privacy settings.

Google+ acknowledges that we wear different hats in different settings and sometimes for really good reasons. I don't want to share with my Auntie Aggie that I was so trashed that I stumbled home at 1am on Hollywood Blvd singing "Start Spreadin' the News!" (which invariably pissed off some Angelenos). Nor do I want to share with all my coworkers that my little cousin just made it onto the Cheerleading squad and will be even MORE squeaky the next time I see her.

What's even more astounding? That such simple settings are lacking so completely that I get excited to have it.

I may find severe deficiencies over the next few days but for now +1 to Google+ on making Usable Privacy.

ICANN Be Greedy?

, ,

With the announcement of the generic top-level domain free-for-all I admit I'm skeptical of the claim by the President & CEO of ICANN:

Originally posted by ICANN Website:

"ICANN has opened the Internet's naming system to unleash the global human imagination. Today's decision respects the rights of groups to create new Top Level Domains in any language or script. We hope this allows the domain name system to better serve all of mankind."



Really? Open gTLDs are going to better serve mankind? What, are they going to help world-hunger, require less energy or decrease drunk driving? Seriously?

What ICANN seems to fail to address or, at least, appreciate is context.

201-555-1225

How do you know that's a phone number? (a little ethnocentric example, admittedly) Context is implied in the form.

Likewise google.com and yoursillywebsite.org or even minecraft.net are all easily identified as websites. Sure, they've added to the complexity lately -- .name, .info, .aero... even .museum -- but they are all used much less than the big three.

If anything, it's going to increase the ambiguity and confusion. At the very least web developers (from websites to web browsers) are going to have to keep an up to date list of gTLDs to try and properly parse URIs. I see few, if any, advantages -- except for a pay raise for everyone who works at ICANN for that hefty $185k fee. (they may be a non-profit organization but that doesn't prevent their executives from personal paycheck profits)

Perhaps Google desperately wants "search.google"? Would Opera decide they want "browser.opera"? or, much more likely... "yourname.facebook" or "yourname.fb"?

There is a hard limit of 1000 new TLDs per year but that's still a lot and I still fail to imagine how this makes the Internet better.

Maybe someone has better ideas than me?

google maps infuriates!!!!!

I've decided to write this post while I'm in the midst of trying to solve my problem.

Here's my problem. I need to give directions to an EXACT spot on a map. I live in a city, addresses can not only be confusing but misleading. Thus, I prefer to give directions that will get people to where they need to go -- not wandering about looking for the exact spot.

This seems like a damned simple problem. We have Longitude and Latitude which allow us to specify exact spots -- or you can do like GPS and use that notation. All I care about is, right now providing the exact location for the next 3 days that this event will be taking place.

Apparently this is not an easy task with Google Maps as I've spent 20 minutes (not counting the time I've taken to write this post... or the weird Opera browser crash) trying to figure out just that!

At first I clicked Google's "Edit" button on the map, thinking it would let me edit what I want to edit -- but it takes me to some weird mapping application which is 200% of what I don't need/want/etc.

So then I searched, "get latitude and longitude from google maps." I was able to get the longitude and latitude fairly easily by searching for the general vicinity and clicking on the point on the map. Then I clicked the pushpin and it gave me Lat/long.

So now I have the point -- but no link to share this point.

Okay, back to Google Maps! You'd think I'd enter the magical Lat/long voila!... you'd think wrong -- Google refuses to display just the point -- it insists on linking to a damned address... WHICH IS WRONG.

So the only link it will give me is to this screen, the image of a location with a red dot labeled A (The wrong address) and a Green Arrow pointing to the location I'm talking about. This would be okay if it were, at least, the right address -- then I could give instructions but it's wrong!

No matter what I click, I cannot get rid of the incorrect address link (if it'd let me at least link to the right building, I'd be happy) nor can I get it to just display the green arrow.

I'm so annoyed!

And just as I typed it... an idea came to mind. I'm going to try Bing.

And the sad thing is... in about 2 minutes I now have my link -- with my own annotation of the location. (Granted, it wasn't completely obvious because I had to click "Email" to actually get the link) It now loads up the view I want, marking the exact location and with a note about what it is.

Wow.

And that my friends is ... well, I guess, how Bing Maps just made a friend.

google is rethinking the address bar

, , ,

....for better or worse, Google Chrome is considering cleaning up the URL display. There are certainly a lot of technical concerns with abstracting the address bar, phishing and relative paths etc, but I'm excited to see someone consider the idea.... I wrote a couple thoughts in the forums back in 2005 on the matter and I've seen some evolution in labeling secure sites, showing certificate stats etc, and highlighting/bolding domain names...but not much else. If google is successful, this is a significant abstraction of the "technical" aspects of how a browser works, it will be interesting to see the reaction.