Archive for February, 2008
When is a stack trace not a stack trace? When it’s merely parroting back a problem with the webconfig file.
Whenever I try a promotion, one of the first things I do is to submit an empty form. Sometimes it accepts an empty form, if you can believe that (and, if you’re a long time reader, you probably can by now). Sometimes, the page offers client-side validation to say, hey, you forgot something. Sometimes it relies on server-side validation. The Michael Payne $10,000 Exciting Windows! Makeover Sweepstakes promotion offers the last with a bit of self reflection perhaps.
Why, oh why, does your promotion or Web site require a special browser plug in? If you can avoid requiring a bit of gee-whizzery from an ActiveX control or special plugin, you ought to avoid them, as security conscious users (or those who are not security conscious but pretend to be so) won’t download them.
For example, I’ve noticed that the Web sites for HGTV and DIY require an XML plug in sometimes. A current example of this is the Great Garage Giveaway 2008 promotion.
Project Manager Frank Kelly guesses 4 reasons it sucks to be in QA.
I’ll rebut with one good reason about being in QA: Schadenfreude, all day, every day.
A recent article shows how the workarounds put in place in software impedes people. This article, entitled “Apostrophes in names stir lot o’ trouble“, alludes to a number of characters that software designers/developers mishandle:
It can stop you from voting, destroy your dental appointments, make it difficult to rent a car or book a flight, even interfere with your college exams. More than 50 years into the Information Age, computers are still getting confused by the apostrophe. It’s a problem familiar to O’Connors, D’Angelos, N’Dours and D’Artagnans across America.
To avoid SQL injection attacks, which can use the apostrophe, a lot of times developers just prevent users from entering apostrophes into the edit boxes on Web forms.
That’s not the best solution, as the people in the article attest. You shouldn’t let your developers get away with it.
I suspect that many designers really are art students that create “works” at home designed to shock the sentiment of the bourgeoisie, but since they cannot do that at work and remain employed, they find smaller ways to broaden the mind of the simpleton consumer. Such as flouting basic interface conventions, as in the example provided by the Xerox Incredible, Inc. promotion‘s sweepstakes entry form:
See the two checkboxes at the bottom? Ha, ha! You, too, are trapped in the petty, unenlightened mindset passed down to you by your ancestors. Free your mind, man! Those opt-in controls are radio buttons that are square and show checkmarks!
Throw best practices and de facto standards out the window and confuse your user. For his and her betterment!
In most cases, the basic control types won’t let you do this, but Adobe Flash allows designers to disregard the laws of physics and good interface design at whim.
Sure, we say it, and nobody listens.
Larry O’Brien says, in “Drinking the Dirty Water“:
When a system is conceived, continued Jitendra, it embodies the insight and experience of its conceiver, and the great achievement seems to be this conception. But even if we were to grant the conceiver absolute dictatorial power over the system, we would often see tears from his or her eyes.
As a system embodied in computer hardware, it has necessary characteristics. But the conceived system, that which is imagined and valued above all others, is a system not with computer characteristics, but with the characteristics of its originator. An extension of its creator, at least, if not an outright duplication!
That is, because they are the gods creating the world that is the software, it hurts their feelings when it’s proven that they’re just playing Populus in their own minds, and the customers and users often want something other than the perfection they’ve provided.
David S. Linthicum says, in “Why Some Developers Are Hurting SOA“:
However, another hard truth is that most developers don’t think about architecture, but focus more on the latest tools, languages and platforms. They are more concerned about how things are built, versus why they are built, or, more importantly, how software systems work and play well with other systems. That thought pattern needs to come to a quick end.
See? The developers are more interested in using the coolest things to craft their greatest fantasies, not quality, useable software.
Of course, I’d like to rail on about how QA says it, and nothing happens, but the cool writers with paying gigs in actual publications say it, and change occurs.
But I’m far to cynical for that. Perhaps a couple of high-level people who read the SD Times might slightly alter their thinkings, but for every one who cleans up his or her act, Java programmer factories turn out 20 more who will make the same old institutional mistakes at half the cost.
When you’re testing an application with any sort of security, you test the following as a matter of course:
- User with correct username/password can log in.
- User with incorrect username/correct password cannot log in.
- User with correct username/incorrect password cannot log in.
- User can log out.
- User who is not logged in cannot access protected functions.
However, in the case of some applications and most Web applications, the server has a time limit on user inactivity; that is, after a certain amount of time, the server assumes that the user is done and shuts off the connection. You better make sure that works.
Charles Hill finds an encouraging banner “ad.”
Charles Hill, he has the eye for QA if not the lack of heart.
Want to Web browse like an espionage agent? Bottlenose, a company that makes Web sites for wineries, lets you do just that!
Not knowing the sites, I cannot tell you how bad that is, but a missing handler on every event, well, seems a bit problematic, probably.
(By the way, the title is an attempt at replicating the sound of the bottlenosed dolphin for humorous purposes. I know, I know. Just because we’re good in QA doesn’t always mean we’re funny.)
You might think from the title that I’m going to address the proper way that you should hold social intercourse with your co-workers. Friends, I would not ask you to be graceful when belittling your inferior co-workers.
Instead, I am reminding you that your Web sites and applications are going to be out in the wild for a long, long time after you finish this firedrill project and move on not only to your next emergency, but onto your next job or career or the rest of your life. As such, you need to keep in mind the fact that someone in 1 year, 2 years, or 10 years might try to hit the URL you’ve specified somewhere on product packaging, in e-mail campaigns, or embedded in applications.
This bad Oscar appears on all of the registration pages and then the login page. Did someone forget to check all the servers when the code was updated? Or is the problem more inherent than that?
I wonder about the incomplete copy, though, because once you sign up in the registration, you get the ability to register your individual Nintendo products. I hope you know them by name because the pictures won’t help:
I have no pithy one liner with which to end this post. The continued disappointment I suffer when encountering obvious flaws at large companies’ Web sites depresses me beyond mirth.
Ah. A Web application has stood up to your rigorous testing in the original test environment. Hey, in a strange quirk of fate, you’re living in that fantasyland where the application is then deployed to a staging server that resembles the structure of a production server so you can run through it again on a final build. The application passes with flying colors, which means it hasn’t bled to death in stage from your repeated thrusts. No, some project manager or customer lover comes in and says, “Ship it.” One of your technical co-workers deploys the application to production sometime in the middle of a night on a weekend or, more likely, 20 minutes before the client expects it.
You, tester, have just one run through it and MillerCoors time, right? Hold on, little pardner, not so fast. Particularly if your application’s production environment uses a load balancer.
Face it: you can deal with a cranky QA staff, or you can deal with infamy.
All right, all right. I’ve finally enabled the e-mail address that goes with this blog. You can reach me at thedirector at this domain with your communiques (fellow QAians) or flames (heathen developers).
Dear e-mail campaign designer, please remember that smart quotation marks and smart apostrophes (as well as other special characters like em dashes) do not render properly in text only e-mails.
Ergo, say you’re putting together a tell-a-friend e-mail such as the one found on the Lost Sawyer’s Nickname Generator. You should probably take a little more care when copying the text from your word processor to make sure that you replace the apostrophe:
You could also do a little testing on it, too, but that’s probably too much to ask.
So I keep getting e-mails from Edward J. Correia, editor of Software Test and Performance magazine. Fan mail from someone who discovered his name in this post? No, he wants me to renew my subscription to the magazine, and I’m happy to oblige since I like the magazine a lot (it gives me food for thought and occasional jumping off points for posting here).
However, as you can expect, the online form has problems. Or else I wouldn’t bring it up.