Posts

Beta Software You Can’t Trust

The other day I decided to offer credit card payment capability.  I quickly narrowed it down to two choices—JPMorgan Chase, and PayPal.

Chase was professional and comprehensive, but required me to fill out forms, fax them in, then phone someone a day later. Uh uh, not me.

PayPal offered a cool online menu for creating customized Outlook-generated email with embedded credit card payment links.

I went with PayPal. Then the fun started.

Let’s just say it didn’t work.  I called PayPal tech support.

“Oh yeah, I see. Hmm, you’re right, it doesn’t work. Well, that’s beta software, and we really don’t support it.”
“If you don’t support it—why do you offer it on your website?”
“Well, some people like it. But it’s beta, we really don’t support it.”

Now, I like PayPal.  It’s a business I’d like to see succeed.  Nice technology.  Nice people, good customer service manners, cool techies.

But the end result—in this case—is flakey. Doesn’t work. Undependable.  Can’t rely on it. Untrustworthy in that sense.

And the big sin?  They think it’s OK!  After all, it’s just beta!  What do you expect?

Increasingly, I expect it to work.

PayPal is not alone.  I remember discovering Plaxo, one of the early contact management software updating programs, and thinking, “what a great idea.”  It was being offered in beta test format.

I tried it.  Spent hours on it.  Pissed off several friends.  I called to complain, and the nice/service-oriented/tech-savvy person offered me a free year if I’d come back in a month and try their update.

I did.  It was not updated enough.  I left them, and now here I am saying bad things about them online.  They blew two chances with me, and thought it was OK!

I think it started with Netscape. In its early days, academics and gurus were raving about this new business model that allowed customers to actually contribute to the design. User-friendly! Cutting-edge!

True. But also flakey. Undependable. Untrustworthy.  But they thought it was OK!

I’d be interested in hearing from others, but for my money, the trend has gone way past the equilibrium point. 

I don’t want to help design some trivial widget—I just want the stupid thing to work.

And I certainly don’t want to contribute to testing something I’m going to be depending on—like payment systems, Outlook add-ins, printer driver updates, data converter programs.

The tweaks just ain’t worth it. Dependability is worth it.

We are well past the invention of major new categories. We are into the ability to trust things like we used to trust dial-tone. That is to say, we want 99.9% dependability.  But too many developers still think selling mediocrity-in-progress is OK!

The new killer app isn’t email.  It’s dependability.  Software producers, you’re on notice.  I’m quite happy to pay full retail for a working product; I’m not falling for anymore “get it first” or shareware-priced new software.

I want software I can trust. Software that works.  I’ll pay good money for it.

But I’m done paying you money for the privilege of debugging your marginally-interesting flakeware.

Do your own damn beta.  It’s not OK by me anymore.