When the EU recently ruled against Google on a person’s “right to be forgotten” and Google complied with the ruling in a reasonable and scalable way it was clear (to me anyway) that all hell was going to break out. It’s taken a few weeks …but it’s happening. The latest example of legal imbroglio comes from Hong Kong where a Movie maker is suing to have Google remove the autocomplete for their name as it links them to triad gang names. I guess it’s a bit like if you typed in 20th Century Fox and Google suggested the rest of the query should be “20th Century Fox Gambino Family” I could see how that would get annoying…whether or not there is any actual linkage.
This isn’t the first time Google has had problems like this, various EU jurisdictions have required they remove various autocomplete such as the one in Germany which linked “Scientology” to “Fraud”. I happen to think that one is perfectly fair comment…but that’s another blog…or see my reviews of some Scientology books on Audible http://bit.ly/1sgSBsx and http://bit.ly/1orvqI8 . In any event Google has always claimed that autocomplete is a purely algorithmic feature which they have no control over and thus no obligation to edit. Now the right to be forgotten precedent has been so clearly established it’s pretty simple to point to it and say if you can edit for that you can edit for this which has pretty much the same end result…in effect a right “not to be suggested”.
The search engines have sheltered behind the Digital Millennium Copyright Act in the US for the past 15 years. They have also relied on their enormous political sway to steer clear legal assault in the US. Clearly the rest of the world isn’t buying it and we can expect hundreds of thousands more requests to be forgotten and many other law suits with related issues all based on the right not to be remembered…or suggested.
There are certain mysteries in life which are ineffable…like where do all those solo socks go? why do I always end up picking up the tab? in the weird world of online there are several gaps which remain mysterious. Here’s a couple:
Tablets are divided between iPads (25%) and Androids (74%) with the Surface coming in a distant last at about 1%. The androids out number Apple because hundreds of companies make tablets with the Android OS on board whereas only Apple makes the iPad. What’s weird and (I have never seen a satisfactory explanation for this) is that the bandwidth usage of all kinds is dominated by iPad at about 75% with Android taking the rest. Given that all devices hook up to Wifi how is it possible that 25% of the users are consuming 75% of Tablet bandwidth. There are multiple theories out there…maybe iPads are favored by data intensive corporate applications of the devices, others suggest that the dominance of iTunes in movie streaming is a factor…but then Netflix runs on both….any way you look at it, it’s a bit weird.
The other gap which is causing waves in our industry is the search revenue per click conundrum. Most people are reasonably familiar with the concept that when we click on ads in search results somebody (mostly Google) gets paid. As end users have migrated to mobile devices at a rate far greater than anyone could have ever imagined the advertisers haven’t. Advertising agencies are notoriously conservative and (some might claim) in some ways dumb. Search delivered on desktops has been around for over a decade and they are comfortable with that medium. A few years back social media exploded on the market and agencies were caught completely off guard. A few years later they are just about starting to catch up with the opportunity. Mobile devices and delivering ads against those devices has left the industry scratching their heads…even now I am shocked and infuriated by major brands who still don’t have a mobile friendly site…let alone mobile friendly ad campaigns. The result is that although Google’s revenue continues to head north at a good clip they are achieving that growth with many more cheaper clicks. That steady decline has been happening over the last couple of years with no end in sight. In public statements Google has pinned their hopes on local advertisers paying a premium for mobile clicks to drive their local businesses. We are certainly seeing some pretty dramatic spikes in click prices for “hyperlocal” searches like “plumber West Oshkosh.” The problem is that even large numbers of local businesses smart enough to bid online (but often not smart enough to get good ROI for their efforts) don’t compare to the massive spend the big brands can move between platforms. This will no doubt level out over time….but it’s not going to happen overnight.
There can be few things more heinous than kiddy porn…nobody (including myself) could defend any aspect of it. However the recent story where Google tipped off the authorities that a Gmail user had child porn in his Gmail account is a little troubling. What makes it even more interesting is that the offending images were apparently attached to email presumably for storage. To discover the images one would have to open them, obviously it’s easier for Google to open and read. If I wanted to store content without even risking sending them over the net, I could simply attach them to an email and save that mail as draft.
What if I played entirely consensual harmless bondage or kidnap games with my partner and sent her pics of her bound and gagged in the trunk of a car. Those images might be clear evidence of a highly criminal event…or two kinksters sharing legally and consensually generated content. What if I sent a hate filled tirade against a racial or religious group perhaps calling for a mosque or similar to be bombed…I might expect the NSA or the FBI to come knocking at my door…but can I assume that Google will be dropping the dime on me? I’m sure it’s a complex legal area…but it’s also potentially a slippery slope. If I email my dealer and tell him where to drop off the kilo of black tar heroin…should Google be telling the FBI? Google is amazingly great at analysing all things data…even encrypted data…should the FBI be subcontracting them to scan all of our mail and pass along candidates for incarceration. I appreciate that I checked the box which said I allow Google to scan my mail and serve any ads they may feel appropriate but I don’t recall giving them permission to send me to jail as well?
For an incredibly rich and powerful company Google has a tin ear when it comes to the common touch. At the core of the problem is an apparent company wide belief that they are smarter than the rest of us and the rules don’t apply to them. The irony is that the billions of dollars of revenue they generate each year comes from regular people searching for stuff online. In comparison Apple does a much better job of fitting in with the common man. The iPhone and iPad are probably amongst the best designed and engineered things most of us possess. In comparison Google Glass and driverless cars are securely focused on the super rich tech insiders. One of their more public elitist debacles is now headed into the sunset. The Google barges were mysterious contraptions moored in San Fran and Portland harbors. Although their true function was never fully revealed, the consensus was that these were planned to be retail/display centers where the brightest and best could meet behind closed doors to become privy to and perhaps buy the latest and greatest from the Google stable. These barges have now been sold and are headed off to the breakers yard…or whatever else the new owners have in mind for them. Not only was this a very public attempt to end run retail zoning laws it was also a clear example of their “this is not for you” mentality. In comparison Apple builds huge beautifully designed retail outlets in most larger cities without velvet ropes to keep regular folks at bay.
Google is terrific at spotting a trend then jumping on it to use their wealth and power to make it both accessible and affordable to everyone. That’s probably a bit unsatisfactory for the Googlers who think the cools stuff should be kept for “people like them” but it’s where the smart money is. The democratization of technology has been a major driving force for the last couple of decades. As a rule of thumb, if a development’s users can be numbered in the hundreds and can only be found in the greater San Francisco Bay area it’s not going to drive revenue at scale.
Google has a black ops group which spends large chunks of speculative Google cash on ‘moonshot’ projects like self driving cars and dark fiber. Most of these don’t have immediate financial upside but neither did the original moonshot and we are still enjoying the benefits from that adventure. the latest project to come out to that group is one I can totally get behind. Ten years ago sequencing the Human Genome cost hundreds of millions of dollars. Nowadays you can get your broad genome markers worked up for $70 or the whole thing sequenced for about $1,000. This has led to a proliferation of DNA data for hundreds of thousands disease conditions. We know which mutations are closely linked to various diseases, we also have a pretty good handle on mutations which are implicated. Google has already partnered with Autism Speaks to help them digest the massive DNA data understanding that syndrome generates, now they are contributing in a different way.
Part of the problem of understanding the genetic causes of disease is that the majority of the data we have amassed so far has been from sick or very sick people. Even in cases where a very precise mutation is linked to a disease there may be other more subtle things going on. These things can be harder to spot when a person’s genome is already seriously impacted by disease. To help make comparisons easier and more accurate Google is setting out to come up with the base line DNA of a “healthy” human. That should help people studying disease more easily find other potential markers or contributing factors for their target condition. Initially Google will use the entire genomes of 175 apparently healthy guinea pigs. The data will be anonymous and they wont be taking samples from anyone they employ. Provided that the data is truly anonymous this shouldn’t generate any meaningful health concerns. Over time I’d expect them the use a larger and larger healthy normal dataset but this sounds like a good start. It’s good to see Google using their massive wealth and data processing powers for good in this case.
I have mentioned many times that Google is seen as public enemy #1 in many parts of the EU. It personifies American avarice and digital monopoly. Various EU courts have gone after it and occasionally they win…witness the recent goofy “right to be forgotten” hoopla. If there is any industry which has suffered at the hands of the digital revolution it’s the newspapers. They have fallen from mighty powerhouses of influence and money to business scratching out a living on the margins of the new digital world. Understandably they hate Google News more than almost anyone. From their point of view Google takes the content they wrote and deserves it up on a site which they don’t get paid for. Of course as soon as a Google news reader wants to read more than just the snippet published on Google they have to click through and that drives a ton of readers (and thus revenue) to the news site. A few years back the German government pass laws which essentially required that Google pay news publishers if they link to their content. Google responded by requiring that publishers agree not to pay for being linked to…if they didn’t agree they were out the index. Not surprisingly most signed.
The Spanish government has now passed similar legislation which taxes Google for linking to news sites. The wrinkle here is that they are trying to impose an “inalienable right” clause which they hope will prevent Google from solving this in Spain the way they solved it in Germany. On the face of it this is all very silly. The newspapers in question can stop Google from indexing them whenever they want by simply deploying a “robots noindex flag” on their sites…and they can lose the traffic which comes their way from Google by doing that.
If the Spanish try to make the weird “inalienable right” clause to trump the German solution (please don’t ask for an explanation of how that works I haven’t got one yet”) then Google could simply dump the Spanish news index entirely….then everyone loses.
This feels a lot like tilting at windmills on the part of the Spanish media dinosaurs who just don’t like the way the digital world turned out. Sadly that ship has sailed…Google does not and will not pay to index your content Señor Publisher, you may hate it and feel it unfair, but that’s the reality and in your heart of hearts you know that you need Google more than Google needs you.
Sleep is pretty important to most of us, and getting a good night’s sleep can be difficult for some. There are a slew of sleep-aids out there, but a recently successful Kickstarter campaign ($100,000 goal, so far over $300,000 has been raised) has pushed a company named “Hello” to create the module called Sense.
The Sense Sleep Tracker system consists of three parts; the sense module, the sense pill, and the app. The Sense module looks like a trendy rubber band ball, but it monitors things while you snooze along with the sleep pill. It’s a pill by name only, as it clips onto your pillow and tracks how much you toss and turn with a 6-axis accelerometer and gyroscope
What makes Sense interesting, though, is what else it takes into account. By tracking noise, light, temperature, humidity, and even air condition, it figures out a “Sleep Score” based on the potential for rest and how much you actually managed to get. Events such as car alarms sounding outside, or dogs barking are figured into the calculation as well.
The Sense module can record any rumblings it hears, but will also keep an eye on light levels, temperature and other disturbances near the bedside. Just want to relax? That funky module can put out white noise or other soothing sounds to put you to sleep and then wake you up at the perfect time with its Smart Alarm, which wakes you when you are out of your deepest sleep cycle.
The first units are scheduled to roll out in November.
This is a new one….for as long as I have been watching search this one is news to me. You may recall that Google has fallen afoul of the EU regulators who have declared that people have right to be forgotten and can request that Google remove content which is harmful under certain rather vague circumstances. It has subsequently been deluged with requests which vary from paedophiles attempting to cover their tracks to people seeking to protect victims of abuse from being identified. This was always going to be a tricky messy problem and it’s already getting worse. Google never agreed to do more than remove the content from the local domain of the complained if that domain is in the EU. That means that anyone can check on Google.com and still find the content. Indeed Google flags that they have removed something which is more likely to draw attention to the problem content. In addition sites have popped up which flag content which has been removed and notify news organizations to that effect. All this means that you are just as likely to find renewed notoriety than the obscurity desired.
The EU is already unhappy with the situation and has called Google, Bing and Yahoo in for a good spanking. It’s hard to imagine that they will readily agree to the EU essentially censoring indices which Google/Bing would claim they have no sovereignty over. The new angle the EU legislators are taking is that by leaving the information in other places to found the search engines are guilty of “data pollution”. Leaving aside the weirdly Kafkaesque concept of information pollution its’ an interesting take. If a US oil giant pollutes an EU coastline they fall under the sovereignty of the country involved and can be fined/regulated accordingly. Why is information different…if it’s harmful…it’s harmful…clean it up world wide. As I said this feels like an entirely novel approach and I have no idea if it will have legs enough to impact the situation…but it will be fascinating to watch.
For several years the enormous political lobbying power wielded by Google has meant that it has been pretty much impervious from any legal assault in the US. Add to that the intense fear which all California lawmakers have about offending or scaring off the Silicon Valley Golden Goose. Even in cases where they exploit their clear monopoly the Google ship of state has sailed on regardless. So it was with some surprise that I read about the decision they lost this week. The case is incredibly inside baseball. Back in 2012 in an attempt to catch up to the cavalier but very successful approach towards end user data used by Facebook, Google consolidated their various end user terms and conditions. In so doing the complainers allege that end user data was compromised leaving users open to identity theft or harassment. Was there any actual harm…I doubt it…but that never stopped anyone exploiting our legal system. The judge did kick out a bunch of smaller sillier provisions but the central complaint stands and Google will have to answer it in court.
On a cursory examination of the facts it certainly looks like Google fumbled the ball by making the changes without asking for opt in permission from their users. At the very least they should have told us clearly…but who reads the small print nowadays…lord knows what I’ve agreed to over the years . For all I know Apple my have rights to my first born by now. So now the case grinds forwards. Will it eventually fall apart because the harm involved was so small and so hard to prove that another judge will eventually put it out of its misery? Perhaps…but I really wasn’t using my end user privacy anyway.
I’ve been traveling much of this week and I have barely powered up my laptop all week. It’s likely that you (like most of us) are increasingly experiencing most things online on your mobile devices rather than the desktop. That’s great for you and me…but not so great for Google. Google just released it’s Q2 results and they were pretty darn good…revenue up but also significant increases in spend, many in areas unlikely to yield revenue for a while. While the suits in Mountain View are investing in “moonshot” projects like wearable tech and self driving cars the part of the business where the bacon gets made is under pressure. The long and short of it is that the advertisers have not followed the users in any where near enough numbers to keep the average value per click sold from declining…again…a decline which has dogged Google for the past few years.
Mobile devices don’t have as much real estate, are harder to track and the advertising folk involved are still more conservative than the users. So although Google’s revenue has continued to grow investors are looking at a strategy of moonshot projects in the context of challenging key metrics in the core business with less than enthusiasm. In the same week that Google announced its numbers it also announced the departure of the guy officially running their ad business. Nikesh Arora departed to SoftBank. It was interesting to see him address this very issue of click decline a few months back. He commented then that he sees local advertisers as the silver bullet solution to this problem. We have certainly seen the impact of increased interest in online advertising by local businesses result in significant click inflation. The problem Google has is a relatively small number of local businesses currently spending like drunken sailors on clicks they don’t necessarily feel the benefit from isn’t a long term solution. Local businesses don’t have the marketing fire power enjoyed by agencies and large companies….they want leads not clicks. That’s exactly where we come in, we turn those clicks into the leads they need. There is a cultural ravine called “local” which Google will have to cross or fill in to turn the tide on click price decline. Interesting times indeed.