Judgement vs. testing
Fri, Feb 19, 10
John Gruber points to a short piece by Aaron Swartz who calls my earlier post Buzz launch wasn’t flawed, Google’s intentions are “dumb.” Here’s the passage he quotes:
Google is a $170 billion company. It employs thousands of engineers and developers. It tests, tests, tests, and tests more. In fact, its “designers” once unable to pick a shade of blue tested 41 variations of it. It’s ludicrous to think that the Buzz fiasco was simply a result of under-testing.
Swartz dismisses it with:
Yes, Google tests lots of minor details with lots of user data. How do they get this data? From actual users. How do they get actual users? By releasing products. So it seems totally reasonable to imagine them releasing something without heavily testing it; their whole culture is based around testing things in the wild.
I’m really not sure what exactly Swartz finds “dumb,” my claim or Google’s “whole culture” of “testing things in the wild”? Either way, I smell a “Rookie Designer 101″ engagement here.
First, some facts regarding Google:
- Google runs the third largest email operation on the planet.
- Google holds perhaps more personal data of various types than any other NGO in the world.
- Privacy of personal data has been a central issue for Google for the past several years, notably in the U.S., Europe and China.
- Google has repeatedly entered and failed the social networking arena.
- Google Wave, whose enduser functionalities overlap quite a bit with Google Buzz, has not been warmly received.
- Google is a $170 billion company with absolutely no shortage of resources (except strategic designers).
- Google did admit failure and backtracked on Buzz.
So, don’t listen to me. Google itself accepts there was failure. But of what? That remains the question.
First launch, then test and then call it beta?
According to Swartz it’s “totally reasonable” to release products and then test on actual customers. You would do that if you had no taste, no judgement and no experience in design and your customers accepted to be guinea pigs for you.
Exposing what was up to that point users’ private email network by default and then, faced with an angry userbase, suggesting convoluted series of instructions to remedy it is indicative of two equally sad possibilities:
- Google doesn’t have sufficient corporate leadership with taste, judgement and experience to stand up to the expedient Buzz product guys and say, “We’re the ‘Don’t be evil’ company, we don’t violate 175 million users’ trust in us for privacy for anything, period.” (If Google needs lessons on what happens when security/privacy becomes an afterthought in product design, it should look no further than the company it’s begun to methodically emulate in Redmond.)
- That the CEO of Google hasn’t yet apologized for the incident is frightening given Google’s ambitions, considering facts #1, #2 and #3 above. (Even the reclusive Steve Jobs personally apologized for MobileMe’s operational failure, gave rebates and made sure it was back on track.)
I don’t know which one is more disturbing, that every single one of the thousands of people at Google who tested Buzz internally lacked the taste, judgement and experience necessary to know privacy/security should never be opt-out or that if they did, management overrode their judgement for the sake of expediency to piggyback naked on Gmail by default.
You can’t test everything. Neither should you.
If every aspect of design was testable, there’d be no reason to have designers, pattern algorithms would automate it completely. (Perhaps that’s what Google would like.)
When designers do test, they do so because they themselves are not sure what works best in a given context. Testing informs designers, it doesn’t magically produce judgement. If Google needs to be taught in 2010 by testing on actual users without prior consent that the core tenet of a private/secure system design must not require opt-out, then shame on them.
That’s Not How or What You Test, Mr. Swartz, that’s how you expose your greed or incompetence. You be the judge.
The case against Opera Mini on the iPhone
Thu, Feb 18, 10
In 2002, Apple forked KDE project’s HTML layout engine KHTML and JavaScript engine KJS into WebKit which begot Safari. As usual, everyone thought Apple was either evil or ignorant for not choosing Gecko, everyone’s favorite non-IE rendering engine at the time. Apple rejected Gecko because KHTML/KJS offered less code, cleaner design, excellent standards compliance and faster speed.
Through its sustained commitment for nearly a decade not just to WebKit but also to WHATWG, HTML5 and various complementary aspects of the open web (canvas, H.264, SVG, etc.) Apple gave the open web community an increasingly credible alternative to proprietary platforms like IE, Flash and Silverlight.
Today, in the mobile business, every major platform but Microsoft’s does (or soon will) run WebKit as its primary web browser. The open source WebKit would certainly not be where it is today without Apple.
To Apple, though, WebKit is not merely a web browser. It bet its rich web-media rendering future on WebKit. Every Internet-connected Apple device, and likely Apple TV soon, runs on it. Even for the usual Apple anti-fans, WebKit, specifically its iPhone instantiation, is still considered the best browser on a modern, multi-touch mobile device.
Now comes Opera, mini Me
With amateur-time spectacles in airports that would make Adobe Flash evangelists proud, Opera now is daring Apple to reject its Opera Mini browser which it hasn’t submitted yet to the App Store:
“Opera is not based upon WebKit, no” confirms [Opera co-founder] Tetzchner “but it’s the world’s most popular mobile browser. Why would Apple not want the world’s most popular browser on the iPhone?”
So what’s not to like about Opera Mini? It says “50 million rides” in the ad above, it must be “the world’s most popular mobile browser”? If that’s the case, Apple should reject Opera for not being good at math: there are already more than 50 million iPhones/iPod touches sold, everyone of which actually relies on WebKit as its sole browser. Let’s consider the serious issues:
Proxy. It’s one thing for an app on the iPhone to query the web, talk to its own or others’ servers, but something entirely different for Opera Mini to proxy the entire web through its own proprietary servers. Yes, you read it right. Opera gets in between you and every single URL out there, from your bank to your school to your doctor’s office. You never communicate with any site directly, only through Opera proxy servers that first go to that URL, get a page, recompile it into its own markup language, compress and send it back to the mobile client that alone can understand it.
Security. When visiting encrypted pages, you have to allow Opera to get in the middle to decrypt and re-encrypt (via Opera Software), breaking what’s meant to be an end-to-end security chain. You need to ask yourself if you need another potential opaque layer of insecurity between you and, say, your bank account?
Scalability. In introducing the iPhone’s Exchange facilities, Apple stressed how it was more secure and scalable for a distributed system to deliver email than to rely on a single point of failure, like RIM’s centralized NOC proxies for BlackBerry devices, which have infamously suffered multiple failures over a number of years. Now imagine Opera, a tiny outfit in comparison, having a similar fate, except taking down the entire web for its users.
JavaScript. Opera Mini 5, the version which we assume will be submitted to the App Store, is currently in preview and is not the more capable one that requires Java J2ME runtime which Apple will clearly not approve. Apple gets dinged for not delivering the full Internet by excluding Flash, and yet I bet the very same Apple anti-fans won’t say a word about Opera not even trying.
In addition to recompiling each and every web destination into its own markup language, Opera also deliberately dumbs down its JavaScript support. After Opera gets a page from the real server, its onLoad JavaScript events are fired, but all such scripts are allowed only two seconds, and, for example, all interval and setTimeout functions are disabled. So if the original page was doing something time-related or took more than two seconds, too bad, the original page is compiled, compressed and sent to the mobile device incomplete. Consequences be damned. When on the device, there are only four events allowed to trigger JavaScripts, onUnload, onSubmit, onChange, onClick. Enjoy your full Internet, the Opera way.
Opera Mini may be a huge step forward for dumb WAP phones, but for a company that bet its future on the rich interactive features of HTML5 and JavaScript in WebKit, this Operatic intrusion is clearly a giant step backwards.
Persistence. One of the key capabilities in HTML5 is local storage, which allows web sites/apps to store varying amounts of data on the client side for personalization, preferences, data, syncing and off-line capabilities. Coupled with emasculated JavaScript support, Opera is looking into the WAP past instead of the HTML5 future here.
Interface. Pictures of the proprietary Opera Mini are not being made available, but the company said that the browser does not include such fundamental UI conventions on the iPhone as pinch-zoom. Apple has bet its fortunes on establishing the world’s first pervasive multi-touch UI platform, from Mac trackpads to iPhones to iPads. As we argued here two years ago, it would have been a mistake to allow Flash on the iPhone because it had no concept of multi-touch, regardless of any other issue. Nearly 100 million iPhone/Mac users are now familiar with Apple’s multi-touch gesture library, it’d be a travesty to allow a tiny player with a marginal interest in this platform and no experience in gestural UIs to pollute it now.
Unknowns. There are many of them:
- Will OS-wide copy and paste work as expected in Opera Mini?
- Will it allow multiple text and graphic items to be selected and copied with formatting intact to other apps?
- Will bookmarklets such as Instapaper or ReadItLater work as expected?
- What will happen to Opera when more and more device-specific functionalities (like GPS, accelerometer, background notifications, etc.) are integrated into WebKit?
- Apple’s said to receive $100 million/year for including Google Search in Safari. How does Opera interfere with that revenue?
I could name many more unknowns, but the fundamental problem with Opera Mini is this: other than a few geeks who want everything because that’s what defines them, have you ever met an iPhone user who complained about its browser? Most iPhone users I know bought an iPhone primarily because they loved the browser on it. Sure, who wouldn’t want a faster browser? If the iPhone got 3X faster this summer, will people stop wanting it to be even faster? Of course not.
I have no idea if Apple will reject Opera. But before people start jumping up and down about how evil Apple is, I hope they consider some of the issues posed by Opera, because so far the only advantage they promise, but haven’t yet publicly delivered, is speed. And speed isn’t everything, if the Wintel saga hasn’t taught us anything else.
WAC’s most whacky adventure
Wed, Feb 17, 10
You heard the story: Two dozen Rip Van Winkles wake up one day and decide that it would be a good idea to band together to defeat the evil monster in the forest. They lounge about, wring their hands, have a beer, perhaps two dozen kegs of beer, and then, in a moment of sheer brilliance, name their holy alliance Wholesale Applications Community (WAC).
The first word of their first collective action, wholesale, ought to tell us all we need to know about the likelihood of their success. Like shoveling commodities to the hapless consumer, “without distinction” as the dictionary would put it. Because for the Rip Van Winkle alliance, software is just like sacks of commodities. You load them up on a truck, send them to a depot someplace and you’re done. They call it, “Load’em up once, ship’em everywhere.” After all, rice is rice “without distinction” whether you’re in India, China or Japan, right?
It gets whackier at WAC’s manifesto site: wholesaleappcommunity.com, and yes, that’s the URL.
If you believed the fantastic and utterly bogus numbers generated by VCs and the peddlers of the first generation of mobile phones about how many billions of WAP phones would eclipse traditional wired access to the Internet just any day now, then you’d be very comfortable with WAC’s claim:
Will establish a simple route to market for developers and provide them with access to a customer base of over 3 billion customers.
Some access. Just how many of these 3 billion phones are in fact capable of running, say, smartphone-class applications is of course a minor detail, as we’re dealing with wholesale commodities after all. Here’s a thought: either these carriers “own” these 3 billion customers in which case developers for whom WAC was supposedly created get shafted or developers own those customers, which we know won’t be the case as there’s no incentive for the carriers.
Then comes the biggest bludgeon only a giant Rip Van Winkle can murderously handle:
The alliance aims to unite a fragmented marketplace by involving players from all related industries to create a community based on openness and transparency to the benefit of all.
Yes, you read it right, “openness and transparency to the benefit of all” by phone carriers. It is a terrific idea to bring into this pitch phone carriers’ biggest asset: openness and transparency! But that’s not enough. To rally openness advocates something else has to be bolted on:
We believe our model presents the most compelling format on the market where developers will thrive and customers will reap the benefits of greater choice.
Of course, choice. By the traditional champions of choice, phone carriers no less! Because, truth be told we just don’t already have enough choice in phones or mobile OSes.
WAC’s ambitions know no bounds:
…access to the latest and widest range of innovative applications and services to as many customers as possible worldwide.
This alliance will deliver scale unparalleled by any application distribution ecosystem in existence today.
It’s wholesale in case you forgot. This could go on. Forever. So let’s cut to the chase:
In practice this means that developers will only have to create one version of their application and this can be used on multiple types of devices and operating systems (such as Symbian, Android, Windows etc) which is not the case today.
Here’s a challenge to WAC: if you believe in this fantasy, or more likely if you want developers and users to fall for it, why don’t you design a mobile operating system that works on all devices not named iPhone or iPad and be done with it. Kill them all and replace them with your grand unifying OS. Until then, you can wade through search results for why “Load’em up once, ship’em everywhere” never works…on your 1.5-inch phone screens.
Buzz launch wasn’t flawed, Google’s intentions are
Mon, Feb 15, 10
Last week Google exposed private aspects of Gmail accounts by default in its introduction of Buzz and then backtracked to offer what can only be described as user-hostile instructions to remedy it. Even the generally knowledgeable analysts are being naive about how this could have happened:
Google is a $170 billion company. It employs thousands of engineers and developers. It tests, tests, tests, and tests more. In fact, its “designers” once unable to pick a shade of blue tested 41 variations of it. It’s ludicrous to think that the Buzz fiasco was simply a result of under-testing. Indeed, it was not an implementation snafu at all, as often described. It was a reflection of the strategy with which Google has decided to capture the enormous territory left up for grabs by the decline of Microsoft.
Not how but why
As mentioned in Google Buzz: The Big Misdirection, Google introduced Wave last year in its much abused “beta” form to a yawning public. After much hoopla prior to its introduction, Wave has virtually disappeared from public interest:
Wave remains an over-engineered and under-designed product that was poorly prepped for general introduction beyond a small developer base, which has become an all-too-familiar Google ritual. This tweet cleverly captures what unites Wave with Buzz:
Google clearly feels pressure from Facebook and Twitter in terms of social networking, personal data access and real-time search, as well as location info from myriad geo-apps and smartphones. Unfortunately, it hasn’t been able to deliver products that have captured the imagination of broad segments of the consumer market lately. From Nexus One to Latitude to Talk, Google is in danger of being relegated to servicing geeks.
Unsure of its ability to successfully roll it out as an independent product, Google must have then decided to force feed Buzz through its Gmail user base of 175 million. Google executives likely reckoned that in a single day Buzz would garner more users than Twitter has been able to in two years after all that celebrity publicity. That really is why Gmail users woke up one day to find their private account details exposed to the public, unannounced and unprepared, because without such default exposure Google executives likely didn’t believe they could deliver a critical user base for Buzz. That’s not “improper testing,” it’s a platform strategy. And the fact that Google reacted quickly to public pressure doesn’t negate the fact that its arrogance was thoroughly exposed. The correction isn’t significant, the exposed intentions are.
Why would they do that?
Microsoft became the largest technology company in the world primarily by leveraging its two widely used platforms to enter into new product areas. But having missed search, advertising, online services, digital media, smartphones and a host of other 21st century phenomena, Microsoft is in slow and steady retreat from most of the lucrative new consumer markets.
The only other company that can fill this evolving void is Apple, but Apple is not interested in commodity businesses. Google sees a great opportunity and has decided to pursue it, mostly by imitating Microsoft’s leverage strategy: if you want free mail, you (also must) get social traffic (because we need your personal network data graph). You’re welcome, enjoy your Buzz!
This leverage strategy can indeed let Google harvest more social territory, at the expense of Facebook and Twitter…but only for a time. Eventually, what Microsoft is going through now is what will happen to Google, even if Google thinks it’s immune to Microsoftdom.
Ugly mutation
Google has a tradition of experimentation. It routinely introduces products and features, tests their viability and culls under-performers. Such speed and mutative variation may be seen as a competitive edge: release early and release often.
Marissa Mayer, Google’s vice-president for search and user experience, says 60-80% of Google’s products may eventually fail. Unfortunately, the few that survive are neither all that inspiring nor market leaders. What Google lacks is not infrastructure, engineers, money, time or even great ideas. It’s the ability to delight users. What Google is missing, in other words, is strategic design.
Can they handle the truth?
What’s urgently needed at Mountain View are senior strategic designers with sufficient experience, clout and guts, empowered to stand up to geeky top management, MBA-driven product guys (Jonathan Rosenberg), left-brained quality assurers (Marissa Mayer), Microsoft-bred (Vic Gundotra) and countless other dominating engineer-managers to boldly demonstrate why pulling a trick like Buzz is short-sighted for Google’s long(er) term business interests.
Let’s not mince words: Google is not very good at design. The cacophony of its recent designs in Wave and Buzz are proof positive that Google’s single most valuable contribution to strategic design, its sparse search page, is but a distant memory now. Welcome to the Microsoft Ribbon-land.
For a public that doesn’t even know what a web browser is, what Gmail lacked was not a bolted-on Buzz that further complicates what’s already a poorly designed email reader. What’s needed is not a knee-jerk reaction to Facebook and Twitter that would make Microsoft proud, but a fundamental rethinking of the presentation of Google’s sole cash cow: search. In 2010, the design quality of its search results is a disgrace for a company as ambitious as Google.
Google needs to think deeper
It’s become fashionable among geeks to paint Apple “evil” for its steady control of the user experience. Google pretends to be the champion of the “open” web, but it surely is not, as explained in The Big Misdirection. Apple has no pretense at “openness” but, unlike Google, it thinks deeper when designing its products.
In its urgency to offer a me-too product, Buzz confuses the read/unread email paradigm with real-time messaging stream like Twitter. It adds insult to injury by co-mingling various cognitive spheres like blogs, photos, videos, status, etc into thin soup delivered through an unceasing firehose. The final blow is the embarrassingly unfocused layout: the complete absence of visual hierarchy and progressive disclosure, overabundance of visual cues/links for action, and clumsiness in using white space to strip away meaningful information density.
I’m sure Google executives don’t think these are critical, as long as Buzz is free and can be leveraged through Google’s other widely used properties. If Buzz was a startup product, it would have died shortly. But when you expose it by default to 175 million users, who needs to worry about design and delighting users!
If this takes you back to the ’90s, to a place called Redmond, you’re not alone. Buzz wasn’t an accident. Get used to it.
Google Buzz: The Big Misdirection
Fri, Feb 12, 10
Google has introduced another platform. Well, maybe another product. Okay, perhaps another feature in Gmail. Surely, it must be another piece of Google beta spaghetti on the wall, no? No.
Google Buzz didn’t percolate in Google Labs, where laymen could try it out at their pleasure. It suddenly appeared in your inbox. Edward Ho, tech lead for Buzz explains the big idea:
Five years ago, Gmail was just email. Later we added chat and then video chat, both built right in, so people had choices about how to communicate from a single browser window.
Trust me, I didn’t lift this from the Microsoft court documents in the illegal product bundling trial. But if you ineptly launched a product, okay a platform, with underwhelming aesthetics, poor interaction design, overabundance of functionality but with little focus in the form of Google Wave and received more jokes than traction, what would you do for an encore? Why of course, you’d force feed it over a product, err, platform with 175 million users. Why wouldn’t you want to use Google Buzz within Google Gmail running in Google Chrome browser on Google Chrome OS powering a Google tablet? Everything in a “single browser window” indeed.
A nation of whiners
Now that they shifted their attention from the “big iPod touch,” geeks are divided on Buzz. Some are stuck in details. Jolie O’Dell asks in RWW: “In addition to creating ‘best guesses’ for who to friend and follow using Gmail & Google Talk, why doesn’t Google simply use Twitter OAuth and Facebook Connect to import existing friendships?” Like, “Why doesn’t Apple just allow Flash on iPhones and iPads?” Right.
Most jumped on Google’s appalling attempt to strong arm its way into unsuspecting Gmail users’ inboxes by automatically injecting Buzz into private email workflow and then making it very obscure to get rid of it. It takes a few steps, okay maybe a dozen or so steps, to re-privatize your follower list so that you don’t inadvertently expose your hitherto private Gmail life to the rest of the world. Obviously, if you were looking for design and interaction clarity in Google products, let alone personal data privacy, you are barking up the wrong Silicon Valley pole.
So Buzz received a harsh welcome in some quarters. There’s always that nagging question, now that Google is in a mad rush to Microsoftdom, is it hell bent on devouring all other social networks in its path to online domination?
Has Google become a startup killer?
According to many, with Buzz Google’s trying to kill Twitter. And Facebook. And FriendFeed. (Alright, FriendFeed was acquired last year by FaceBook so this is a double whammy.) And Yammer. And Yelp. And Foursquare. And Gowalla. And BlockChalk. And Yahoo! Updates. (Yes, I know, you can’t kill the undead.) And Jaiku. (Oops, Google already owns it.) And whatever Microsoft may introduce in 2011, or later.
However, the company that couldn’t make Orkut, Google Video, Latitude, Google Base, Knol, Google Catalog, Dodgeball, Google Answers and many other products successful may have a hard time killing off its competitors. Of course attention whores with conflicts of interest like Jason Calacanis already declared Buzz the winner: “Facebook is going to see their traffic get cut in half by Google Buzz.” Well, that settles it then.
The Big Misdirection
Why exactly did Google create Buzz anyway? According to Google’s Open Web Advocate Chris Messina:
This is a downpayment on where we’re going with the open, social web…
Yes, the open web. Where you could go tinker with a product’s source code openly and freely. Like you could with Buzz. What, you can’t? How about its parent Gmail? No, not open either? You could use Buzz without Google Profile, no? No choice there either? Confusing?
For what business reason would Google offer Buzz then? To shift most online activity to its unified platform so that it can make money from search/advertising, its sole cash cow, you might answer. Surely, Google’s search/ads business must then be open, no? No? Isn’t Google the most prominent champion of the “open web”? They made Android open, didn’t they? You could go tinker with the source code of what makes an Android phone attractive and meaningful, right? Like Gmail, Google Maps, Google Apps, Google Voice and so on? No? The crown jewels of the Android and Chrome OS kingdoms, the “open” alternative to the iPhone and iPad, are not open? Something’s wrong.
Now imagine if Microsoft or Apple did this. The blogosphere would have exploded with self-righteous indignation. But we already know Microsoft and Apple are evil. Google, the “Don’t be evil” company, obviously, is not.
Magic acts are entertaining only if you can’t spot the misdirection.
