The BBC understands that Buzz was only tested internally and bypassed more extensive trials with external testers - used for many other Google services.
Google said that it was now working "extremely hard" to fix the problems.
"We've been testing Buzz internally at Google for a while," Todd Jackson, Buzz product manager, told the BBC. "Of course, getting feedback from 20,000 Googlers isn't quite the same as letting Gmail users play with Buzz in the wild."
In yesterday's post, I speculated Google Buzz was only tested by its own engineers. Today, Google admitted to the BBC that was true.
Jackson told BBC News that the decision to create these automatic lists was borne out of the idea that Google "wanted to provide a great user experience straight out of the box".
Well, not quite. Google focused on providing a great Google employee experience straight out of the box. The article reports even when Google does "user testing", it uses "a network of friends and family of Google employees". That sounds like a privileged class, not everyday users.
Google needs to refocus its core products for the real users making up its market share, before insular thinking damages Search audience driven revenue.
While we mock those users, the simple fact is they haven't necessarily failed, something failed them. With all of our talk about the semantic Web and search engine optimization and tailoring search results to the individual user, there are thousands upon thousands of users performing the same simple search and following the same wrong road. If this were a standard traffic sign misdirecting this many people, it would have been pulled down long ago.
People won't use your product the way you want. They'll use it the way that works for them.
Last week's Buzz was around Google's spectacular privacy missteps in the launch of its Twitter killer, but should have been around the other illustration of Google's more fundamental failure: becoming Microsoft.
In 2009, Ballmer famously derided a Microsoft employee for using an iPhone. Encouraging engineers to eat their own dog food is a great way to find pain points and fix them. Discouraging engineers from openly using best-of-breed products the way their users will use them is a great way to stifle both innovation and the understanding of what real users want.
Google's engineers were likely delighted when Buzz automagically preconfigured their Follow networks. Then again, engineers are not known for social skills. Engineering was so far removed from how "real people" would react to seeing their most emailed contacts exposed, they missed how even engineering's most avid disciples – tech bloggers – would react.
For this mistake, Google took a massive hit in public trust. Down the road, Buzz will be held up as an example of why one company should not be allowed to control too much of our information. Google semi-apologized (saying it was "sorry for the concern", not sorry for the feature) and will hopefully consider its non-engineering users' concerns in the future.
While Google can afford this mistake with Buzz, it should worry whether the same lack of connection with everyman puts its core product at risk.
If Google starts getting Search wrong, it is in serious trouble – and the "facebook login" incident suggests Google's getting it very wrong.
A decade ago, I switched to Google because it offered me a search box that led to nothing but results. That's all I wanted, and all it did. This match made in heaven catapulted Google ahead of Excite, Lycos, Altavista, and Inktomi.
While the search box has stayed the same, Google's users have not. Today's Googlers don't know the difference between a URL and a search term, or even between a browser and the Internet. (I talked about these users in an earlier article about the iPad, commenting on how removed bloggers are from "most people".)
By focusing on features such as real time web search or categorized results for tech-savvy users, Google is stranding its mass audience – a mistake its advertising business model absolutely cannot afford. Google needs to cater to the folks who ended up on ReadWriteWeb through their typical use path, making Google work for them and get them where they're trying to go, instead of trying to retrain them to adapt to Google.
Google needs to refine the "I feel lucky" button until it's good enough to be the default, helping the Internet's least savvy users find where they want to be even if they're doing it wrong.
Touchscreen accuracy of the iPhone is much better than that of Verizon's Droid or Google Nexus One. When you're trying to tap a link, chances are you're going to be successful on the iPhone, and not on Android phones.
iPhones showed straight lines in tests with both light and medium finger pressure, while the Android phones showed zig-zag wavy lines across the screen."On inferior touchscreens, it's basically impossible to draw straight lines. Instead, the lines look jagged or zig-zag, no matter how slowly you go, because the sensor size is too big, the touch-sampling rate is too low, and/or the algorithms that convert gestures into images are too non-linear to faithfully represent user inputs. This is important because quick keyboard use and light flicks on the screen really push the limits of the touch panel's ability to sense."
Once again, comparing phones "feature for feature" doesn't tell the whole story.
Apple's uncompromising commitment to usability drives their engineering choices in ways that might not be obvious to engineers or even consumers seeing an ad, but are painfully obvious after you've experienced how the thing should work.