QA Hates You

If you’re working on a university Web site, there are different sets of rules than you would get in most real-world situations.

For example, if you want to register for the Marquette University MU Connect alumni directory, you see things you don’t see every day.

I don’t mean the warnings that some items on the page are secure and some are unsecure. You see that all the time.

I don’t mean the JavaScript error alert boxes:

Click for full size

You see that all the time, too.

No, I mean things like this:

Click for full size

The Gender drop-down list has the value “Unknown,” and the Web site selects that one by default. In a university setting, I guess this is acceptable, but many of your consumer packaged goods would not like to see it.

The serial, or Oxford, comma serves an important function in the English language, maintaining parallelism in structure so you logically can identify items with conjunctions in a series of items.

What do I mean? Here, look at this:

Click for full size

Jack Weatherford has written four other books:

• Indian Givers
• Native Roots
• Savages and Civilization
• The History of Money

Without the serial comma at the end, you might think it was only two:

• Indian Givers, Native Roots, Savages and Civilization
• The History of Money

Seriously, there’s a good reason to use it. The only reasons I can surmise against using it:

• Some AP editor with a hangover in 1947 didn’t know about it and didn’t put it in the stylebook
• Commas are so darn expensive

Since I have a good reason to use it and you don’t have a good reason to do differently, let’s just do it my way, okay? Thanks.

Newsweek explains how the decline of Digg is a cautionary tale for Web 2.0 companies and examines the claim that a site redesign inspired a backlash:

Digg’s wounds are at least partially self-inflicted. In August, Digg introduced a new design that users hated. Reaction was so bad that for a time some of the most popular stories on Digg were about how awful Digg had become.

Rose and the company’s new CEO, Matt Williams, who joined on Sept. 1, began scrambling to restore features that had been dropped, such as a “bury button” that lets users vote down stories they don’t like. “We very much disappointed our user community, and we’re deeply sorry,” Williams says. “We have very passionate users, and I know we can return Digg to a place they love.”

Of course, it wasn’t the only factor. It might not have been the main factor. But it was something.

QA is a collection of little polydactyl Dutch boys here to put as many fingers in the dike as possible to stop as many trickles as possible.

When I’m not reading the Riot Act, I like a good novel or three. Here are some of my favorites dealing with the corporate world:

	Then We Came to the End: A Novel This novel centers on an ad agency in the downturn immediately after 2000 and how the people within interact with one another and their own fears as they wait for the axe to fall.
	Lloyd: What Happened: A Novel of Business An executive without a project becomes the intermediary between his company and the one about to buy it. As he reevaluates his life and his relationships, he has to fend off the temptations associated with corporate power. Or yield to them.
	You Look Nice Today: A Novel An executive heading up a corporation’s Total Quality program is sued by his secretary for sexual harassment based on innocuous comments and her emerging madness.

All three are satirical in tone and read much like prose Dilbert, but with depth.

I think this is how a lot of designers get into the Web business.

It ain’t pretty when the pretty leaves you with no place to go.

A helpful message from the Springfield-Greene County Library System:

Click for full size

“Please note: Some users are getting a security certificate error message when trying to place holds or log in to their accounts on COOLcat. COOLcat’s security certificate is valid and we are working to resolve the problem. Please use the Internet Explorer browser to access your account until we have the problem resolved.”

That is, since we have a technical problem on our end, if a security message from Firefox bothers you, use a browser that’s not so pushy about valid certificates.

Thank you, that is all.

All publicity is good publicity? Ask Joel Jewitt of RapLeaf, who told the Wall Street Journal.

“We didn’t do it on purpose,” said Joel Jewitt, vice president of business development for RapLeaf.

The Wall Street Journal uncovered the fact that RapLeaf was storing Facebook IDs with data tracking information. Phantom bit of the header stored, perhaps, or maybe a thoughtless key written into the database.

Regardless, any day when your management team is not quoted in a major national news source pleading ignorance is a day when your management team should be thanking you, QA. Maybe you should let them know it.

An old Blockbuster envelope teaches us a valuable lesson about alternative methods of output:

Click for full size

So what portions of your application come out of the printer? Does it work right? Does it look right? Is it correct?

It’s not enough that you make sure the print dialog comes up correctly. You need to make sure that the extras that are often added to the printed page display correctly. For example, some maps add details such as the location, some Web sites put their names on it, and some applications use formula. To ill effect in this case.

If you want to be a real rapscallion, see what happens if you print to a file or to a PDF driver of some sort. Because someone out there in the real world just might.

This page is entitled How to Keep An Idiot Busy.

However, there is a way to trigger a “You win!” message. I assume you, QA people, will do so in under a minute.

The journey in this song reflects how many people end up in QA, I think.

This is the freak show, baby, anyhow.

I’m not just gobsmacked, I’m gobbeatenandleftfordead.

Remember the phrase fly-by-wire?

(Link seen here.)

Oh, yeah, I’ve seen this before. It’s the situation where your Web site requires a database transaction of some sort before even displaying any content to the user:

Click for full size

And something in the database has gotten so large that it crashes with an error before the user even sees your Web site. Good job!

You know what? You’re not going to find this in your test environment, probably, because you won’t hit the site enough to trigger it (well, not unless your developers are extraordinarily bad). Instead, this will crop up sometime in the middle of the day when some magickal unknown number is reached and your site goes down.

Now, here are some trick to look for it:

• Know the choreography of your application, and make sure you ask over and over again if there’s a log file or something whose limitation might cause the site to blow up. Did I say over and over? I meant over and over, again and again.
• Set your high priority customer sites as your browser home page. Forget your pretty iGoogle portal; make sure you see your Web site every time you restart your browser. If you do a lot of browser compatibility testing (and you should!), you can set each browser to a different site and test them all for this condition several times a day.
• You can run automated tests to check the presence of the Web site at the very least and maybe some functionality, too, while you’re at it. At the very least, you should send a request to the site and make sure it comes back with a page as often as you can stand. And by “as often as you can stand,” I mean daily, several times a day, or hourly, whatever you can get away with to ensure that the test does not skew tracking or hog resources.

If this error occurs, be advised that the common developer or tech team response to this is to clear out the database file and let it go. Which means the problem will recur. If you encounter this problem, make sure that your development or tech team handles it once and for all and handles it that same way for each new site and puts the fix in place for all the existing sites.

QAHY quick quiz! What’s missing from this e-mail?

Click for full size

Need a hint? Okay, if you fill out the form available here, the thank you page looks like this:

Click for full size

So what’s missing from these items?

UPDATE A couple of you guys got it right out of the box: the asterisk appears on both items, but the footnote identifying the limitations does not.

You not only have to look at what’s before you when testing, but you have to look at what’s missing. You get that with not only experience or business knowledge in the area of the software you test, but also from real world experience or business knowledge from outside the domain. The domain can be a blinder, and you have to look above it.

Good work, guys.

“Don’t behave as high as a mountain. Though a mountain is high, it will be climbed by animals.” — Genghis Khan

“Behave as volatile as a volcano. Though an animal may climb you, it’s in for a big surprise.” — The Director

UPDATE: I fixed my quote.

Do you look at the images in your Web site or application with your jaundiced QA eye? You should. Here’s why:

Click for full size

As you can see in the above image, which accompanies this story, the designer grabbed a number of images from the Web and put them together into a single file.

Unfortunately, he or she did not right-click and download the image file. No, in the interest of serving ignorance, this designer took a screenshot of the image while the cursor was over the image and didn’t notice the alt text displaying even when he or she created the complete image.

Not a professional designer, you say? Well, someone has a job making images for an Internet site here. And when you’re on your job harassing them, it’s your job to watch for this sort of boneheadism in the images.

Things to look for:

• Are the words spelled correctly?
• Are the client names spelled correctly? Note that in most cases in the 21st century, these are not real words and are ripe for misspelling.
• Do the lines align properly?
• Are the colors right? You can take a screenshot and get the color information through your graphics program (you do have something other than Microsoft Word and Microsoft Paint, don’t you? Why the hell not?).
• Make sure the image isn’t compressed into looking like something out of an 8-bit video game. If the happy couple using the product looks like Mario and Luigi, make the creatives try again.
• Get your minds in the gutter, QA.

And, of course, just pipe up if something doesn’t seem right. Like there’s alt text floating somewhere in the image that doesn’t belong to the image.

A little The Lonely Island to start the week.

Sadly, it’s the last lifeboat off this sinking project, guys, and QA ain’t on it.

As I said on Twitter:

Sometimes, QA just has to shout, “LEEEEROOOOOOOOOOOY JENKINS!” and test whatever dev gives them.

As a public service for you damn kids, here’s the World of Warcraft video I alluded to. Note that it has a lot of swearing in it, so play it loud only if you work in an interactive agency.

In the current Information Week, Art Wittmann explains that process is to slay the Jabberwock:

One challenge for those in IT organizations is not to become so enamored of your own process that you lose sight of serving the business first and foremost. I saw this happen early on in my IT career after a boss went to see Edwards Deming talk about Total Quality Management.

….

Of course, it didn’t work that way. Certain things lent themselves to the TQM process and some didn’t. In particular, there are always those little exceptions, the small jobs that IT teams need to do without running them through whatever process you’ve chosen, because if you did run them through your process, you’d either miss the window of useful opportunity or you’d turn that small job into a big, expensive one.

We longed for some sort of 80/20 rule. We would happily use the TQM ritual for most of what we did, if only my boss would let us quickly deal with the innovation-oriented 20% of tasks that didn’t fit that model. It never happened, and rather than getting behind his TQM philosophy, we and business side peers rebelled, to the point where my boss lost his job.

The problem was that we were being asked to serve the process rather than serve our constituents in the business. Today, I see the same thing happening, often in organizations that are heavily committed to ITIL or some other definitional methodology.

Oh, I recognize that mindset. I saw it aplenty after sitting through process meetings where we discovered previously unknown gems of process, such as When the designers finish the design, they send it to QA, and QA looks it over before it goes to the dev team! Gems of process that hide in plain sight, but which we had to glean and polish in hours of process meetings.

Then, when we finalized the document and everyone signed in blood, the customer facing people, LOB in Wittmann’s vernacular, found that 20% in 100% of the cases that follow. “The client needs this now!” “We’re behind, so we just gave it to the development team with the misspelling of the client products on the buttons and everything, to dev so they could put something quick and dirty up,” et cetera and et cetera.

Then the personnel would switch (let’s just say the turnover at an interactive agency isn’t as bad as a Burger King, but probably worse than a Wendy’s), and we’d be off to make more flowcharts of proper process to ignore.

As I’ve said before, a proper process takes into account that the weasels in your organization are going to try to subvert or ignore the process whenever possible and to create alternative paths for those cases. If the client returns items x days/hours late or the x days/hours/minutes before or after the end of the milestone/stage, shorten the process by this. But only in those cases.

Otherwise your LOB customer-loving, over-promising people will continue to argue that the current effort should not be constrained by process just this once.

Since yesterday’s New Twitter bug proved so popular, let’s look at another one. You can recreate this one on your own using the following steps in Mozilla Firefox:

1. Click the Retweets link.
2. The Retweets menu displays. Click Your Tweets, Retweeted.
3. Your retweets, assuming you’re some fraction as clever as I am and someone else repeated what you said, display. Highlight one and notice the little more info subpanel caret that displays. I don’t know what Twitterians call it in the documentation, but it’s a > symbol:

   
   
   Click for full size

   Click that little button to display the panel.

4. The panel displays. Notice the caret changes to indicate you can close the panel:

   
   
   Click for full size

   Now, click the Twitter logo in the top left to return to the timeline.

5. The timeline displays:

   
   
   Click for full size

   I just tweeted something funny! Let’s see if anyone has picked up on it.

6. Click the Retweets link.
7. The Retweets menu displays. Click Your Tweets, Retweeted.
8. Your retweets display, but uh oh:

   
   
   Click for full size

   Notice that the caret is in the wrong position. To display the subpanel for this tweet, you have to click the caret twice. Once to get it into the proper state and the second to actually display the panel.

You have to have two things to work in QA: Wide eyes and quick eyes. These little tics are bugs and might not display on the screen for very long, so you have to see beyond where your mouse pointer or cursor is flashing. Additionally, a requirements document, if you even have one, probably isn’t going to outline precisely the state for this caret should be in each use case, so it’s probably not something you’re going to check off in your test checklist.

You just have to catch it.

Take a tweet, like this:

Click for full size

Now, say you have something clever to respond to the retweeter. Click Reply.

The reply box displays:

Click for full size

Note this is to reply to the person whose tweet was retweeted, not the retweeter. Well, that’s easy enough to fix.

Click for full size

In this case, I’ve replaced the reply with a sample (not to scale). Now click Tweet and….

Click for full size

Twitter tells me I’ve replied to the person whose tweet was retweeted, not to the value I to which I changed the @value.

Which sort of caused me a moment of panic when I thought I sent a senseless reply to someone I didn’t mean instead of an obscure reply to someone I meant.

Who would do that? I did it, didn’t I? The application should have recognized this possibility and either locked in the @name or picked up the @name or lack thereof for the success message.