I went to put in my Visa info on a site, as shown in the picture. Note that the MasterCard radio button is selected by default… though I didn’t see that at the time. I know my name, of course, and even the card’s expiration date, but I [...]]]> First, the dumbness. Second, lessons from this dumbnosity.

I went to put in my Visa info on a site, as shown in the picture. Note that the MasterCard radio button is selected by default… though I didn’t see that at the time. I know my name, of course, and even the card’s expiration date, but I had to dig the card out of my wallet to put in the number. I did all that, filled in a few other items, and clicked OK.

The site told me my credit card didn’t match the type selected… and made me reenter my credit card number.

Dumb, inexcusably dumb on three-and-a-half counts.

1) You can algorithmically determine which card it is from the number itself; there is no reason whatsoever to ask someone to tell you want kind of card it is. (What, you think this will help cut down fraud? You don’t think fraudsters know the card algorithms better than any normal person?) Go here for an explanation of how it works. So stop asking for information you already have!

2) If you insist for some reason on a human being selecting one of these radio buttons, don’t start with one selected! Yes, that violates the normal rules for radio buttons… so see dumbnosity #1. Or don’t do it with radio buttons; make each clickable and put a big green border around the one selected, or something.

3) When it give me the error message, it erased my card number. So I had to dig back into my wallet, pull out the card…. You can use very simple on-page JavaScript to determine that the card and number don’t match, so don’t refresh the page because of this error. But this error shouldn’t exist to begin with (dumbnosity #1 again).

3.5) Why not make PayPal an option? I don’t want to keep giving all these websites my credit card info. Each time I do this, there’s an opportunity for a costly mistake, one more place a security breach can happen. With PayPal (or Google Checkout, I suppose), security is much more locked down. Sure, PayPal could have a breach also, but if I have 20 places that have my card number and can screw up vs. a single entity in that situation, the latter is a lot safer. (And it’s not like I’m increasing the risk of PayPal exposure, since I already have a PayPal account, as I suspect most regular Internet shoppers do. If PayPal exposes my data, the harm is the same whether I use them once or have 20 merchants that they serve.)

I do recognize that PayPal sometimes charges merchants slightly higher fees than the credit card companies, and that it’s a pseudo-bank that isn’t regulated like a bank. I’m not saying that merchants should absolutely offer a PayPal option, just that it’s well worth considering. It has pluses and minuses that I won’t go into here. As a consumer, I personally would like that option in addition to using a credit card. Indeed, I have backed away from buying at a few sites because they didn’t offer PayPal and I didn’t really want to give them my credit card… and I live part time on a small island in the Pacific Northwest with no retail stores other than food markets, a small auto- and boat-parts shop, and a lumberyard/hardware store, which means I need to either buy stuff on line or spend most of a day and $30+ to get on a ferry and then drive 20 miles to a shopping center. So I need to shop on line!

Lessons From Dumbness

1) Think first. If you’re asking for information from a user and the user believes you already have this information, the user will be angry and will trust you less. This principle holds whether you’re asking for data on the web, gathering IT requirements, interviewing (whether interviewee or interviewer), and so on.

2) Push back on dumbnosity asked of you. The developer who put together this page a) should have foreseen at least dumbnosities #2 and #3, and it’s not inconceivable he or she should have spotted #1. C’mon, even if you didn’t know about #1, didn’t you at least suspect that this was the case? Surely you’ve encountered sites that don’t ask you what kind of card you have! The developer should have questioned this requirement in the name of customer friendliness. Developers who don’t think beyond the strict confines of “I do exactly what the specs say” aren’t adding sufficient value in these difficult times, putting both their own jobs and their employers at risk. If you don’t push back when you spot little mistakes, you’ll have neither the practice nor the credibility to push back on the big ones. This isn’t just a problem for developers, either.

3) If you’re managing, remember to specify the problem, not the solution. Here the surface problem was “capture the dude’s credit card info,” not “step 1, put up three card logos and make the user choose.” But the larger problem was “find a secure way that the user can pay for stuff.” Stated that way, the PayPal strategy becomes an option in the solution space.

TTFN From My Island

Okay, it’s not my personal island, though often it’s quiet enough to pretend. But as I write this, the sun is out, I’m watching two sailboats make their way across the water in front of my deck, in the last hour I’ve seen a young (no white head or tail yet) bald eagle soar by less than 100 feet away, and I know I’ll get yet another spectacular sunset over the water this evening. And there’s no traffic, no traffic lights, no cell phone coverage or TV to clutter up the day, decent Internet bandwidth, and most of all a slight slowing of time, room for an “ahhhh” between some of the tick-tick-ticks. And now a raven’s calling. (They don’t say “Nevermore,” but rather sound like a crow with a cold.)

I only miss the city sometimes.

Have a great weekend.

It says Mailing Address in big red letters. Then it says Mailing Address in bold black letters. The cursor goes automatically to the first edit box.

So what do I — and I suspect most other folks — start typing? My mailing address, [...]]]> I was filling out an online form today when I came to this section:

It says Mailing Address in big red letters. Then it says Mailing Address in bold black letters. The cursor goes automatically to the first edit box.

So what do I — and I suspect most other folks — start typing? My mailing address, of course.

Designers, it may technically be part of your “mailing address” database table, but we humans on the other end of the screen don’t automatically think, “you want my name here” when we come to an area boldly and colorfully (and redundantly) labeled “mailing address.”

(By the way, this form was the gateway to a site whose core product is known to anyone who travels even occasionally; it’s not a one-off, one-person operation.)

Right now, WordPress — the tool I use to run both my blogs and my Lexician.com site itself — nags me at the top of every screen: “WordPress 2.9.2 is available! Please update now.” [Exclamation point in original.] There’s nothing that tells me [...]]]> Software vendors are invariably after you to upgrade. Should you? If so, when?

Right now, WordPress — the tool I use to run both my blogs and my Lexician.com site itself — nags me at the top of every screen: “WordPress 2.9.2 is available! Please update now.” [Exclamation point in original.] There’s nothing that tells me why I should update, other than their excitement as conveyed by that exclamation point: “We worked for weeks or days or hours on this version! Ya gotta do it! Now!”  Except…

1. It’s a minor upgrade to fix a small bug that doesn’t affect me.
2. The upgrade to the version before my current one had a serious bug that they failed to catch that led to the hijacking, corruption, and loss of the first year of NoCCrit, including about 250 posts and innumerable comments.

This message staring me in the face for a week, coupled with my beta testing of Office 2010 and discussions with a couple of client folks, got me thinking about when to upgrade. There are, I think, three different upgrade approaches vendors use:

The Big-Release Upgrade

Most server and desktop systems from large vendors follow the Big-Release methodology. Every two to four years, these vendors issue a major new release. Microsoft, for example, released Windows 7 last year, and they’re planning to release Office 2010 in a few months (June, according to word on the Web).

Should you upgrade?

Pros:

• These upgrades are generally safe and competently tested, despite the so-called words of wisdom suggesting you wait until the first service pack.
• They have new features that are likely to be useful, though you may not “get” them until you’ve worked with them. After all, what can Microsoft add to Word at this point that (a) isn’t already in there and (b) you might actually need? Surprisingly, there’s still a lot of room for growth, from little touches — moving the highly valuable Document Map to where users might actually find it and renaming it Navigation Pane — to new ideas such as the online/web version that competes directly with one of Google Doc’s strengths.

Cons:

• It costs money. (It’s a different calculus if you’re on an enterprise-wide license.)
• It takes time and effort, people think. It does take a bit of time while your machine grinds away, but you can do it during lunch. As for effort, Microsoft rolls out new versions to 50,000 people a night internally with virtually no glitches or hassles. If your corporate IT department can’t figure this out, suggest they talk to their Microsoft rep and ask how Microsoft does it. I assure you, there’s no magic, just an IT group that’s got this part of their act together.
• It will break some existing apps built atop the current version. If that’s the case — and I know it often is — then your IT team isn’t learning from experience. Stop doing this! Build on published APIs, not hacks. And then test with the beta versions. Even Microsoft occasionally gets this wrong internally, but it’s usually because a developer tried to do something extra-special and pushed into unsupported territory. That said, occasionally big companies change the underlying model in an unanticipated way, such as the file-formats change in Office 2007.
• You believe you know better than the users what they need. ‘Tain’t so.

The Annual Upgrade

Some companies do annual upgrades as an ongoing revenue stream. I use some music-creation software that used to dun me for annual upgrades, at about $300-400 a year in total. It’s an economic model users hate… but it’s a model that keeps the companies in business, instead of selling maintenance contracts or the like.

Software — mostly — isn’t free. If you have a solution you like that is free, great; go for it. WordPress, for me, is such a solution. On the other hand, I couldn’t have written my book Legal Project Management in anything other than Word (I tried Open Office and Google Docs to see if they’d work).

But often you pay in ease of use, or lack of testing, or intrusive (or hidden) advertising, or no support. There ain’t no such thing as a free lunch — TANSTAAFL, as Robert Heinlein put it. It’s a trade-off, one you should make carefully. Sometimes free is the answer; sometimes it isn’t.

If you don’t take an annual upgrade, you’ll likely pay considerably more for your next upgrade, almost as if you were purchasing new again. Sometimes you can find opportunities to join the “welcome back to previous customers” program and minimize those costs; keep your eye out for them.

The Quick-Turnaround Upgrade

The WordPress 2.9.2 upgrade falls into this category; so do many SaaS/hosted solutions, where they push upgrades out whether you want them or not. Many are insufficiently tested. If you can, resist them for at least a few weeks to see what problems other guinea pigs are turning up; check the support boards, Google for answers (can you Bing for them?), and exercise caution.

The other end of this scale is the semi-automatic upgrades pushed out by Microsoft, Adobe, and other large companies for your desktop software. In an ideal world, I’d put many of these off too, but we don’t live in an ideal world. We live in a world with sufficient bad guys who like taking on high-profile targets not just because it’s somehow cool to beat Microsoft’s engineers but because they are actively trying to rip you off if they can get past your computer security. Thus I think there’s little choice but to take upgrades in a timely fashion for popular software such as Windows, QuickTime, and Flash; it’s a matter of security.

These companies have reputations to protect and so they test the heck out of these rapid upgrades. That doesn’t mean they’re foolproof, but it does lower the risk so that taking the upgrade is safer, all told, than not taking it.

Finally, there are the antivirus companies that suck innumerable cycles from your machine with their daily upgrades. Again, the bad guys are truly out there, and so I think you have to accept these daily updates. My main cavil is that not all of these vendors properly throttle down the amount of bandwidth they use to push the updates, so that it can interfere with your use of the computer.

Other

Like any classification scheme, this one isn’t perfect. Where, for example, would you put the every-other-year major service packs for Windows releases? Still, it’s a good framework to think about how and when you want to upgrade, rather than leaving it to chance.

I’ve never actually seen the book, mind you. But I know it’s there. Why else would so many software designers follow the same rules so slavishly?

This rule is sometimes called NIH, “not invented here.” [...]]]> It must be written, in the Great Book of Software Design Principles:

• Thou shalt never leverage the usability/design principles discovered by thy rivals.

I’ve never actually seen the book, mind you. But I know it’s there. Why else would so many software designers follow the same rules so slavishly?

This rule is sometimes called NIH, “not invented here.” Developers — recall that I’m an emeritus member of that community — believe in two variants of this commandment:

• If thou in thy mind shall see a unique design that seemeth “better,” thou must then go forth and implement it, no matter what thy users and customers desireth.
• If thou shalt pass by a design of thy rival and not notice thy rival’s principles, then thou mayest not only ignore those principle but thou mayest claim that those principles, like grapes, are truly sour and an abomination.

Sometimes it works out for you. Most of the time it doesn’t.

Let me offer some examples:

1. The Book of Jobs sayeth, Apple did bring forth a mouse with a single button. Microsoft saw that it was good, but then saw that yea, a mouse that hath two buttons was better. I start with this one because it’s a great introduction to the debate, with good value on both sides. Apple chose to make the core action — clicking — simpler while making the secondary action — right clicking, which requires both hands on the Mac — more difficult. Microsoft chose to make more power available to users in exchange for a slight increase in complexity of learning. My personal vote is for the Microsoft method, but both are valid and have sane points in their favor. (Caveat: I was one of the first Mac developers in 1983 — the beta of the Mac — and 1984 and developed a general dislike for Apple for its treatment of developers during that period, so I am likely not objective here.)
   
2. The Scroll of Beraysheet sayeth, Microsoft didst discover that, traversing word to word, paragraph endings and punctuation were as equal to the starts of thy words; Mozilla claimeth thus that a word is but a word and that the cursor shalt pause upon solely the start of new true words. When you advance the cursor by word in MS-Word or the Internet Explorer/Windows edit tool (Ctrl+left or right arrow) and come to the end of a paragraph, the cursor treats it as a stopping point. Likewise, when you come to punctuation, the cursor stops. This is seriously elegant, because it recognizes that these are common editing points. When I move word-right from the last word in this paragraph as I edit this post in Mozilla Firefox, the cursor will not stop on the period, or the break between paragraphs. There is a technical word for this behavior. That word is “stupid.” (“Beraysheet” is the sound of the first word of the Bible in the original Hebrew, rendered in English as “In the beginning.”)
   
3. The Tale of Goog and MicGoog revealeth that Goog didst find value in a clean and simple search page, while MSFT didst clutter it unto infinity. And yea, Goog smote MSFT in the search wars. Picture the MSN start page, if you dare. ‘Nuff said.
4. The Book of Numbers and Letters sayeth that in spreadsheets shalt thou count the rows with numbers and the columns with letters; I am Kapor. Microsoft failed to obey this commandment in MultiPlan, the predecessor to Excel, using numbers for both, yielding the abomination R2C2 (row 2, column 2) for what any real spreadsheet would call cell B2. R2C2 wasn’t nearly as cute as R2D2, and eventually Microsoft got with the program and relabeled things Lotus-123-style. (Mitch Kapor used to run Lotus, makers of 123. Sorry, getting a bit obscure here halfway through my second glass of wine as I write this at night — though I assure you it’s sacramental wine. And the threat of a lawsuit might have had something to do with Microsoft’s choice; I was at Microsoft seemingly unto forever, but this battle truly was before my time.)
   
5. The Dread IE Scrolls claimeth that the righteous mayest move from browser tab to tab by commanding Ctrl+Tab, and that yea and verily thou shalt be able to alternate between two tabs by pressing Ctrl+Tab again. Mozilla and Chrome sayeth, Not so much. Actually, Mozilla Firefox sayeth, Heck No! Chrome sayeth, Ctrl+Tab shalt rotate through the tabs — a noble analog to the Windows Alt+Tab paradigm, except that they got the paradigm wrong! In Windows, if you hold Alt and press Tab a few times, you indeed cycle through your windows. However, if you press Alt+Tab, release it, and then Alt+Tab again, you alternate between your two most recent windows, an outstandingly useful model. IE implements the same thing with Ctrl+Tab. Chrome is out to lunch… and Firefox didn’t even realize the lunch bell had rung. Helllloooo! I’m sure someone thinks there is a rationale for this behavior, but like the Mozilla word-to-word issue, it’s dumb, to be blunt. Microsoft got this one right; copy it!

I could go on, but I hope you get the idea by now… and I’m running out of Biblical puns. (I almost cited #4 as stemming from the book of Corinthians, but figured that was way too obscure. A “Corinthian” is also a type of column. You know, like spreadsheets have columns. I think I’ll stop now. Jokes shouldn’t have to be explained.)

The first rule of software design is that if someone has a good user-interface idea, borrow it. A couple of 1980s court cases said it’s basically okay to do that (IANAL, of course), so get used to the idea that different isn’t better. Get over it. And get it right. Your users are counting on you… or cursing you.