Book Review: “How Music Got Free”

41uqX+wVxbL._SX329_BO1,204,203,200_How Music Got Free: The End of an Industry, The Turn of the Century, and The Patient Zero of Piracy (Amazon, iTunes), by Stephen Witt, is a non-fiction book that takes a look at the creation of the MP3, and how that eventually cratered the profits of the music industry.

I came across this book pretty much by accident – I wasn’t looking at a book on technology or the music industry. And in some ways, you can ask yourself what is really the point of a book like this – don’t we know the story already? Music companies are greedy, they missed out on technology, pirates were just a bunch of kids, and that was that.

But this book goes into much more detail, telling a story that I don’t think most of us know. I, for example, am pretty steeped in technology, but didn’t really know how the MP3 came to be (and almost didn’t). While I knew of Napster and The Pirate Bay, I figured music was just being ripped by college kids, when in fact there was a deeply conspiratorial enterprise involving people in CD manufacturing plants who were sneaking out CDs (and later DVDs) before they were released, and this is where most of the leaks came from. And while the executives at labels were indeed stupid and greedy, there were those who were just blindsided by technology, and figured something like the MP3 could be beaten simply by churning out hit music (in other words, people would just want to buy it, like they always had).

This book focuses on three main characters who never met each other, but whose stories were intricately intertwined:

  • Karlheinz Brandenburg: The German scientist and engineer who worked for the Fraunhofer Instittue, and created the MP3 standard
  • Doug Morris: A record executive who worked at one time for all the major labels, and became the CEO of the largest label (Universal Music) as the industry imploded.
  • Dell Glover: A factory worker at the Kings Mountain, North Carolina CD manufacturing plant, who is probably responsible for leaking more music and movies than anybody in the world.

This book spans the late 1980s through 2013. The early focus on the book is on the technology of MP3, and the amazing politics that went on behind the scenes that almost killed it, yet through Brandenburg’s tireless efforts, eventually came to dominate (Brandenburg and his team was horrified by music piracy, yet because of piracy, they became rich through licensing their technology to MP3 player device companies).

The sections dealing with Dick Morris were less interesting to me, yet it highlighted just how obscene music profits were, and how they were completely blindsided by MP3 technology. The most interesting aspect of this to me is that the executives thought MP3 was inferior, because they were listening to their sound mixing engineers in the studio, who insisted compression was bad. Unfortunately for the executive, these sound engineers weren’t really “engineers” – they didn’t understand acoustic technology nor how it worked on the brain, so they just assumed compression meant lost information that you absolutely needed.

The most fascinating piece of this, however, was the story of Dell Glover, and how the pirate community grew. These were not people interested in riches, but rather just the excitement of being “first” to leak. The cat and mouse games of getting music out of the plant (or radio stations, or from store warehouses) and onto the internet were pretty fascinating. While the RIAA was suing individual users who really didn’t know what they were doing, guys like Dell Glover were completely decimating the industry, and not getting anything for it.

The book was very well written, not spending too much time on any one character, and finding natural places in the story to switch from one character to another. It got a little slow in the middle, and at times I felt it spent too much time on the recording industry, but it really picked up from just over halfway in the book, when the activities of Dell Glover and the other pirates took center stage.

It’s a pretty fascinating story – the rise of the MP3 seemed inevitable in hindsight, but the battles within the MPEG licensing industry almost killed it in the crib. And the laughable way that the recording industry behaved in response to piracy – how they completely missed who the real pirates were while they were busy yelling at Napster and single moms in Minnesota, was pretty amazing. And Dell Glover was a great character.

It is well worth the read.


How to Best Advocate for the Minimum Wage

minimum-wage-2It’s campaign season again (when is it never campaign season?), and Republican douche-bags are out in force running for president, casting aspersions on the minimum wage. Master creep Scott Walker just called the minimum wage “lame”.

Given the laziness of the national media, we are invariably going to get into a he-said / she-said about the minimum wage, with conservatives dissing it as some kind of communist take over of industry where we will all be wearing drab, olive-green uniforms singing hymns do Dear Leader if we raise it, and liberals calling anybody who doesn’t want to raise it a corporate fascist. CNN will put two talking heads from opposing ideologies on the tee-vee machine, let them yell at each other, and then say they gave equal opportunity to both sides.

But I’d like to detail and argument for the minimum wage that I think liberals should use, and while it won’t convince any conservative about anything, it will at least use language conservatives can understand.

Conservatives (and the libertarians in conservative clothing) make a point about the minimum wage that should be true. In their worldview, you have capital (those that are owning or running the business) and labor (those that are working for the business), and that these two groups will come to a naturally agreed upon wage that is an equilibrium between supply (labor) and demand (capital). Any intervention of government into the process just distorts this equilibrium.

When they make this argument to people in my income bracket (those making over $100k per year), it makes sense. People in my pay grade aren’t depending upon a minimum wage, and for the most part, we can pretty easily switch jobs if we want. We may choose not to because we have kids in school or we like the neighborhood, but that’s a personal choice not something forced upon us for lack of opportunity. My company has to pay me a “competitive” wage compared to other people in my field, and if I’m good, I can look around and probably demand even more than I’m making today from some other company.

However, this is not what happens as you get to lower wages. As work becomes less skilled, people are more easily replaceable (supply goes up). And these people are less mobile or less likely to be able to be mobile, which further exacerbates the supply. Capital, however, is always pretty mobile. We all know of companies like WalMart that set up shop in a city, and if they cannot squeeze the city into getting the tax breaks or other incentives, they move their store 5, 10, 15 miles away to the next town. The demand, therefore, hasn’t changed. WalMart doesn’t really give a flying fig that you have to travel 15 extra miles to shop there.

Which means through basic economics that wages will just go down. There are too many people competing for too few job positions. A natural equilibrium will still be reached, but it will be quite low. And this shouldn’t be a surprise. There is a reason that BMW opened its first US factory in the American south east as opposed to putting it north of San Francisco. Certainly, cost of land has a lot do with it, but BMW can be pretty sure their cost of labor will stay low – they moved into a depressed area with a lot of workers willing to work for less. It’s the same reason businesses abandon the US and open factories in China and India. We could eliminate minimum wage laws so the factories will stay here instead of moving, but is the point to really be paid one dollar an hour, or one dollar a day?

Labor and capital do not have equal negotiating power. Capital has much more negotiating power, and it always will. In order to equalize the negotiating power, you need a countervailing force. That force can be a union or it can be a government. It cannot be an individual worker applying for the job. That’s why we need a minimum wage. It sets a floor.

Naturally, this is going to have consequences. Anecdotes abound any time the minimum wage is raised as some business that was struggling to stay afloat now goes under, claiming the reason they are shutting their doors is because their cost of labor has gone up. But can you really say that this was the only reason? No. Plenty of businesses do quite well when the minimum wage goes up, and they do well across all sectors of the economy, so it isn’t like a minimum wage has damaged a specific sector. Perhaps that business that shuts its doors needed to shut its doors. Perhaps it just wasn’t well run.

Will this convince your crazy uncle who watches Fox News all day that the minimum wage is good? No. But it should at least be using language they can understand.



Follow-Up to Google Photos Review

In my previous review on Google Photos, I indicated that one of the problems it had was that it would seem to get lost and always show when photos were added, and I couldn’t undo that. This turns out not to be the case. When it was uploading photos from my Aperture library, it decided to grab photos that had been synced with my Facebook account, and those photos had no date on them (Facebook apparently strips out all Exif data). As such, the only thing Google Photos could rely on was the date when the photo was uploaded, and as soon as I had new photos from my iPhone camera in the library, I saw that those photos were now in front of the synced Facebook photos.

I would argue that Google Photos should not have updated photos from Facebook that are in an iPhoto or Aperture library, but that is a minor “bug”.

A second thing that has happened is that Google is continuing to run its algorithms on my now updated library. It has since created a new set of things called “stories”, which is essentially a smart photo album based upon when and where the photos were taken. As with the “places” smart search, the places are rather generic. However, the stories are actually pretty interesting. It created a story of a trip I took to Israel several years ago, and it did some interesting things. The photos were taken over 3 days, in three different locations in Israel. It separated each section into days, and also inserted a map before the section showing where the next batch of photos were taken. It even animated the map, showing the city at the end of one location, with a dotted line (looking like a plane trip) to the new location.

As with my previous complaints with Google Photos algorithms, there is no way to modify the result to change the photos included, or to even make your own story. Maybe this is coming in a future update to Google Photos… who knows?

So far, though, of all the algorithms Google has run – people, places, things, “animation” and “stylized” being the others – this algorithm is working the best.

For a look at the story of the trip to Israel, click here.

Google Photos: A Review

As a Mac user, I have been debating whether to take advantage of all the new features of Apple’s Photos app, namely, the concept of cloud storage, and then storing all my photos in the cloud. I have nearly 200GB of photos from over 20 years of digital photography (and some scans of film). These are too many photos to fit on my hard drive, and in the past with using iPhoto, my mechanism was to have individual iPhoto libraries for every year, and then archive and back each library up (both locally and to a cloud storage service). If I needed to get an old photo, I would simply go to the library for the year I thought the photo was in, and look for it.

This was not ideal, because it is the process of going through some old photos where you get lost in the memory hole, and find lots of fun things to look at or share. Additionally, it meant that for each library, you had to run though “faces” to organize all the faces, basically starting from scratch.

However, as much as I think going to Apple’s Photos has merits, I’m naturally terrified of having my photos only in the cloud (with only individual photos synced to the local hard drive when accessed), and getting storage at Apple isn’t cheap. iCloud photo storage is, at this point, $3.99/month for 200GB (thus $50/year), and since I’m near that limit, I might have to bump up to the 500GB plan, which is $9.99/month ($120/year). You don’t have to keep your photos in the cloud only, you can keep them on your Mac as well, but I prefer my Mac to have SSD storage, and 200GB isn’t cheap, especially just for photos.

So, I got interested in Google Photos, released nearly at the same time. Like Apple Photos, you can store every single photo you own in the cloud, and access it on any device (a web browser on a Mac/PC as opposed to an application, but local apps on your phone or tablet). Unlike Apple photos, Google doesn’t really expect that this will be your only storage, because it recompresses your photos, and if you use RAW, it compresses them. So, you are expected to keep your photos on your own machine as well. However, it is free for unlimited photos. (You can keep original, unmodified photos on Google Photos, but it uses your Google Drive space. For me, to use Google Drive for this would mean I would have to be on the 1TB plan, for $9.99/month).

Reading the tech press on Google Photos also piqued my interest, because the algorithmic capability of the tool was highly touted. There were stories about how it would identify faces, and that the identification was so good that it would recognize a picture of your child at 1-year-old being the same child when they are 15 years old. It would tag photos such that you could type in “wedding dress” and it would find any picture that looked like it had somebody in a wedding dress. This meant, therefore, that you would have to do less manual work to get your library into shape, and would be able to find anything at any time.

So, I figured, well, it’s free, so if it doesn’t work as I hoped, no cost loss to me. And, if it worked, I wouldn’t have to worry about paying for Apple Photos. I would still have some extra work in terms of syncing new photos, though, because being an iOS user, things that were in my photo library might not automatically sync. But that trade-off might be worth it for all the things Google is giving me.

I’ve since been playing with Google Photos for about a week. So, how did it go?

Web Based… Really?

So, the first thing you notice is that Google Photos doesn’t have an application that runs locally on your machine. It only runs as a web based application. This caused me an immediate shaking of my head. Yes, by being on the web, it checks that magical check box called “cross-platform”, but seriously, who give a shit about this? Outside of geek land, nobody cares about cross platform. You either use a PC and always use a PC, or you use a Mac and always use a Mac, and you don’t care that it looks the same in one browser vs. another browser on a different machine.

Additionally, I’ve ranted incessantly on Facebook and other tech forums about how stupid web based applications are (Note to self: do a blog post on this), and Google Photos does nothing to dissuade me from the stupidity of web based applications. When people talk about the beauty of web based applications, they usually say two things: (1) that it’s cross platform, and (2) some phrase along the line of “yeah, it doesn’t do everything, but it’s good enough”. That last one is said about Google Docs all the time – it isn’t as good as Microsoft Word, for example, but it “serves my needs”.

People… “it’s good enough” is a justification for “this isn’t really good because doing this on the web is hard, but please use it anyway.” You want an application that works, not one that works “good enough”. Sigh…

Anyway, about using the web for this particular application. When you have a photo, and you’ve used a photo based application on the Mac or PC, what is one of the things you tend to want to do with that photo? Typically, you want to edit it or print it or crop it, or get quick info about it if it is in a grid with a bunch of other photos, and the way you do that is to right click on it. And right here is where browser based mechanisms sucks-ass. When you right click on a photo in Google Photos, it isn’t the “photos app” that is getting invoked with the right click, but rather the browser that is getting invoked. Instead of photo related actions that come up, you get things like “open in new tab”, “open in new window”, and “inspect element”. None of this has jack shit to do with photos… this is the same list you get when you right click on any web page.

Secondly, within browsers, the look and feel is not the same. When you see your photo library in, say, Internet Explorer on a PC, you will see a scroll bar. You can grab it and scroll to the bottom, which means the bottom of your library, right? Well, no. It’s the bottom of some piece of the library, and when you get there, the scroll bar might tell you more. Hmmm… Well, what about the Chrome browser or Safari on a Mac? Macs automatically hide the scroll bar so that you can have more window space to do, like, actual “stuff”. So you don’t a scroll bar right away until you start to scroll. In Chrome, you will then see the scroll bar, which you can grab and drag to the bottom, and again, this won’t be the bottom necessarily. And on Safari? You never see a scroll bar… at all. To get to the bottom, you have to scroll enough to see a black box appear on the right that has where you are (like, it might say, July 2014), and then you click on that and drag it down to the bottom, which again, isn’t the bottom. Once you get there, you might have more photos to scroll to.

So, right off the bat, the “cross-platform” wish isn’t cross platform. Can Google address this? Of course they can – a bunch of if-then statements in the application code to handle different browsers. But once you do that, you… aren’t… cross… platform… any… more.

Now, before somebody goes and continues to try to defend this practice and whine about how good cross-platform actually is, keep this in mind – Google Photos is not running in the browser on iPhone or iPad nor on an Android device (you can run it in the browser, but clearly somebody at Google thought an app would be a good idea). On “devices” (as opposed to desktops/laptops), Photos is an application downloaded from the store. Google, in other words, doesn’t even believe their own bullshit about cross platform, because they went and created an entirely separate code base for iOS and another for Android. They have 3 different code bases for Google Photos. They aren’t cross platform at all and don’t really care to be. Being cross-platform in the browser is just stubbornness.

How the Hell Is This Organized?

The next thing I noticed with the library is that I can’t figure out how Google is deciding to organize it. As I started using it, the organization was pretty obvious – newest photos (most recent date as per photo EXIF data) on top, older photos on the bottom. But you can click your mouse in the search box, at which point the screen changes to “people”, “places”, “things”, and “types”, and you can see “recently added” in types. After I clicked on that, the photos it showed were from the most recently uploaded photos. This was nice while I was uploading photos, because it helped me know if the uploader crashed (more on that in a moment), but once it got to that, I couldn’t find any freaking way to turn it off. It always put the most recently uploaded. There was no “type” to say “sort by photo date”. It was just “most recent”. Now, sometimes I could get it to be by date, but it usually meant closing the browser, or opening a different browser, and sometimes opening a tab. It was completely freaking random, and really pissed me off.

Additionally, “most recent uploaded” wasn’t always the most recent uploaded. Sometimes, it would not show the photos I uploaded “today”, only “yesterday”, and not the most recent from yesterday, but rather a snapshot of some point yesterday. And I couldn’t fix this, either.

And this happened on the iOS app, too. And I can’t figure out how to undo it. This has basically rendered the iOS app useless. I have tried killing the app and restarting it, but to no avail. There must be some database setting that got into the server when I went to “most recent”, and this affects every device the photos are streamed to.

The Uploader is Pathetic

Even though the mechanism for accessing your photos on a Mac is through the browser, there is still an application that you download for Google Photos, if you want to upload a bunch of photos. This is called “Google Photos Backup”.

(Pause for station identification as I point out once again that Google wrote Mac and Windows code for this, and is not, therefore, fully invested in “cross-platform”. You now have an iOS app, an Android app, a Mac app for uploading, a PC app for uploading, and a web facing application that doesn’t work the same on all browsers so will have to have “if-then” code inserted into it at some point).

This uploader is so bad I almost can’t comprehend it. First off, it doesn’t appear in the dock, only in the title bar. There are lots of apps that do this, so that isn’t a major problem. However, when you do this, you get into problems in that if the app has a problem, you can’t just go into “Force-Quit” menu to kill it. You have to bring up the Mac’s Activity Monitor to kill it. And boy, did this app need to be killed a lot, as will be noted later.

The uploader can be configured to find all photos on your hard drive (and external drives), and can also upload photos from iPhoto and Aperture libraries if you want. And the big problem with this is that it only seemed to recognize iPhoto and Aperture libraries when iPhoto and Aperture were running. If those weren’t running, it found no photos. And the huge gap is that it doesn’t, as of now, recognize Apple’s new “Photos” app as being a library to upload from. As all my 2015 photos are already in Apple Photos, this means I can’t upload any 2015 photos unless I first export them to a folder. Will that change? Maybe. It hasn’t as of yet, and Apple Photos has been out for a few months now.

It also seemed to not understand the concept of multiple iPhoto libraries, so if I was going to upload all my photos, I first needed to merge all these libraries into one library. That took, well, a long time. Too long.

Once that was done, we ran into the next problem, which is that the photo uploader app was apparently written by a junior high school kid in his spare time. The thing… just… died. All… the… time. The deaths were always silent. For example, it would say it had some number of photos left to upload, and then you would come back an hour later, and it had the same number of photos to upload. It hadn’t done anything in the meantime. It didn’t fail (there was no message saying it couldn’t upload), but it just hung there. You could pause it, and then restart, hoping that would kick it in the ass to get it uploading again, but no dice. The only solution was to quit it.

And quitting it… well, sometimes that worked and sometimes it didn’t. About half the time, quitting the app actually caused it to quit and then you could restart it. The other half of the time the app would turn into the spinning beach ball of death (meaning it totally hung), and that meant going into Activity Monitor and Force Quitting from there. And then restarting. I must have had to stop, force quit, and restart this app 30 times before all 188GB of my photos made it onto the site.

Google’s Algorithms? They Suck

Now, while this was very painful, and the idea of browsing 188GB of photos (over 22,000) on the web is pathologically stupid, the real reason I wanted to try this was because of the algorithmically capability of Google Photos. Google is, after all, an algorithm company with incredible search capabilities, right?

Google Photos has two levels of search here. First, it creates some categories automatically, like “people”, “places”, and “things”, and then there are searches you can do yourself. Let me tackle these individually.

“People”, “Places”, and “Things”

Google Photos has algorithms to search for people (their faces), which most photo browsers have now. (you can turn this off if you wish). And sure enough, Google Photos will find faces, and is pretty good at getting the faces right. But, there are lots of problems.

Let me start this faces rant by pointing out that my experience with facial recognition comes from iPhoto and Apple Photos, which does scan for faces as well. What it does, however, is scan for faces, and then give you the opportunity to “name” that face (your spouse, parents, kids, etc.) There is then a side-bar item for “faces”, and when you click on it, you get a gallery of face groups with their names, and then you can optionally “identify” faces, which brings up faces it thinks it has found but can’t identify, giving you the opportunity to assign names to them to help Apple’s learning algorithm.

Google Photos? None of this. It just groups people. The people have no names. And it doesn’t give you the opportunity to name them. I can’t figure out why this is. If you name the person, then couldn’t you go into search and type the name if you wanted? Why aren’t they named?!?!?!

The tech press was all ga-ga about Google’s ability to not just find people, but that it was smart enough to find pictures of your kids, and would recognize a child at the age of 1, 7, and 15 and treat them as the same person.

I have to say, I saw none of that. It didn’t do that at all. It seemed only good enough at finding faces that were within an age range – it was able to group my son from newborn to about 2, then again as a toddler, and on and on. But it thought they were different people.

Now, I’m not terribly upset about this. Apple can’t do this, either. This is hard. But I point this out because this is supposed to be something Google is “great” at, and it’s no better than any other facial recognition algorithm I’ve seen yet. The thing that pisses me off is not that Google can’t do it, but that they claim they can when they cannot.

Another aspect of faces that is just terrible is that, once it’s identified people into groups, there is almost nothing you can do to “fix” what Google did when it makes a mistake. On the web interface, you can’t do anything – you can’t delete a bunch of photos from the person, you can’t merge one set of people with another, nothing. You are stuck. In the iOS app, however, you can do some things. You can’t “merge” people into each other. All you can do is remove people from a group, to the point where the group goes away. You also can’t go into a random photo and add a person to a group if it happened to miss that face (probably because the groups aren’t named!)

As for “Places”, there is not much to say here. You can’t create places, and the places it creates are pretty generic. For example, there will be a place called the name of the city you live in. Fine. But that could be my house, the mall, a restaurant, my kid’s school, the park, etc. Like with “people”, you can’t seem to create new places or move things between places if Google got it wrong.

“Things”? Even more pathetic. I was in NYC with the family recently, and took a bunch of pictures of the Statue of Liberty. Google photos created a “Statues” section and a “Monuments” section, and the Statue of Liberty showed up in both. But not all the pictures of the Statue of Liberty… only a sub-section of them. So, it was basically worthless. The same with the “Bridges” section – a subset of pictures of the bridge. I’ll have a little more to say on “things” in the next section – the manual search.

In general, Google’s algorithms here suck. They kind of work, and will probably get better over time, but for the most part, they don’t work very well, and I have no idea when they will start working well, and I can’t make it work better by giving it hints (i.e. grouping my own set of People or Places or Things). I’m at the whim of whatever Google decides to do, whenever it decides to do it… and that might be “never”.

Manual Search

Just like with the search categories Google automatically creates, Google Photos is supposedly working behind the scenes to tag your photos with interesting characteristics that make them easy to search. There are stories in the tech press about people typing in “wedding” and getting pictures to show up about a wedding, even though they never tagged the photo with a keyword of wedding. The photo just happened to have the bride in her gown.

I have to say that this… kind of worked. For example, I typed in “pool” to the search box, and it did find some pictures of my pool, some pictures of a kiddie pool that we had when the kids were babies, one picture of a pool table that my daughter had, one picture of a bathtub while the house was under construction, and another of a hold on a mini-golf course, probably because the putting surface looked like blue water?

Google Photos Search Results

Searching for “pool” returned 25 out of probably 1000 images of the pool

Now, at one level, this seems pretty impressive – I had never tagged any of these photos, and it found some. But on another level, this is a complete swing and a miss. You can dismiss the bathtub and mini-golf pictures for being mistakes. However, you can’t actually do anything about that – there is no facility for you to select one of these photos and say it shouldn’t have been part of the search. You would think Google would want you to be able to do that so that the search can get smarter, wouldn’t you? But, unlike with People, you can’t remove things tagged as that person. Additionally, my memory is good enough to know that I have way more pictures of people by the pool. It found 25 pictures. I can guarantee that there are at least 1,000 pictures of our pool. We did a lot of entertaining, and I took a lot of pictures with a DSLR on “sport” mode to capture kids diving in the pool. And Google Photos found none of these. There are also several photos of our pool table, but Google Photos could only find one.

Again, the justification one can make here is that this stuff is hard. And it is. But not working is still… not working. At best, what this does is point me in a general direction of dates for photos that might also match my search, as it does group those photos.

One last search I did was “wedding”, because that was a search that the tech press talked about. I expected it not to find anything as I couldn’t think of any pictures I had taken of weddings. It found one interesting picture that I can easily see as being tagged as “wedding” – we had taken a picture of our daughter in mom’s wedding dress. So, perfect (after seeing this photo, however, I remembered that there are at least 5 photos similar to this that Google missed). Unfortunately, it found 3 other pictures that had nothing to do with weddings – two were of a friend in a white jacket (maybe that was a wedding dress?) and one was of my wife and two others dressed in 1800s style frontier clothes doing one of those elementary school field trips to Sutter’s Fort. The clothing was kind of off-white, so maybe that’s why? But again, there were tons of photos on this day with my wife in that garb, and that was the only one tagged as wedding.

Quick searches for pictures of dogs and cats, where there must be a few hundred, found 5 pictures of our dog, and two pictures of our cats.

Sorry, Google. You suck at this.

Other Issues with Google’s Algorithms

I found a couple of other things with Google’s algorithms, and how they integrate into the application, that were baffling. Under “things”, for example, it has a section called “bowling”. The photo it chose as the cover photo for this was three friends, one of them holding a baby. Was the baby a bowling ball maybe? Then, when you click on it to see what else it could possibly have found as “bowling”, the result was “no results” – not even the baby picture!

From an application standpoint, going into a section like “people” was a pain. For the three pre-defined categories (people, places, and things), you are presented on the web site with 8 panels to choose from, with a link of “more” to open the section up for more things it found in that section. (Note: on the iPhone app, it is 2 rows of 3, and on the iPad, 1 row of 7 in landscape, 1 row of 5 in portrait. Again, cross-platform is not important) Anyway, on the web site, when you click on “more”, you go deeper into that section, but when you go “back” to leave a photo from something in that section, you are brought back to the main search, with the three categories – you don’t go back to just the section you clicked “more” on. “More” on? More like moron, amirite? (Thanks, I’ll be here all week… try the veal!)

A Strange Google Algorithm – “Animations, Panoramic, Collage, and Stylized”

There is one other thing Google Photos does, which is in a section they’ve labeled “Assistant”. As you add photos into Google Photo, it runs algorithms on them and tries to create animations from similar photos, panoramic images from things that look like they were taken with just a camera shift, collages of pictures from a similar time and place, and a stylized section – usually a black and white, or water-color inspired edit to a photo.

These things are “cards”, (because everything in Google is a freaking “card” thanks to Google Now), and you can add these into your library or dismiss them.

And, like every other damned algorithm Google has in Google Photos, it is complete crap. Some of the animations were a sequence of three images, which were taken so far apart in real life that the animation is the one character you cared about, with people popping in and out of existence as if by teleportation. Other animations were quite obvious, like when I had my DSLR in sport mode, so they were… “okay”. The panoramas were laughable. I had taken several pictures of our family in the cul-de-sac playing kickball, and I would tend to sit in one place and take pictures which could, in theory, be stitched together. But invariably, the panorama it created showed three kickballs at once in the panoramic image (because the source photo had been centered around where the ball was), and the same person showing up in three pieces of the panorama. The collages were OK, but to me these were “starting points” – I liked that it made the collage, but I wanted to be able to swap one of the pictures out, and there is no facility to do that. Lastly, the stylized photos were completely random (why did it choose this one and not that one?) and outside of one water-colorized image, were total crap.

Stylized Image

The one good stylized image Google Photos Produced

As these are cards, you can dismiss them and they won’t stay in your library. But this again points to why won’t Google let me customize some of these if it kind of did a good job, or why is there no button for me to force it to try to run this again?

What Is Google Trying to Do Here… For Real?

Given that I think the uploader is crap, the algorithms suck, and navigation is not easy, what is it that Google is doing here? It seems like it is a terrible customer facing thing. But… it’s free.

And maybe that’s what this is all about. Google wants to get better at recognizing patterns in photos, and the more photos they have the better. It’s free, so lots of people will happily dump their photos up there. Heck, even though I kind of hate how it works right now, I might even keep my photos up there. It was such hard work to get them there, and maybe the app and algorithms will continue to improve to where I’ll like it.

And this gets to what a free service is – if it is free, you aren’t the customer, you are the product. Google is going to use your stuff that you happily uploaded to improve their algorithms such that they can target you better with ads. In short, I’m not sure Google really gives a shit about your photos.

Google Photos: You are the Product

In Conclusion

Getting my photos up to Google Photos was a lot of work, and ultimately, the results were frustrating. However, I don’t think it is a total loss, because I look at this as further justification that it is worth paying Apple $120/year to store my photos. The interface for Apple Photos is much, much better, and I’m not giving up anything from an algorithm standpoint, as Google’s algorithms suck.

In short, McKayla is not impressed.

McKayla is Not Impressed


Book Review: “Losing The Signal”

Losing the Signal: The Untold Story Behind the Extraordinary Rise and Spectacular Fall of BlackBerry (Amazon, Apple) is a business biography of the Canadian phone maker that rose to prominence on the power of corporate messaging and e-mail to be a powerhouse, only to spectacularly collapse within a few years of the introduction of the iPhone.

I think a lot of people know the basics of the story, and in some ways assume that the fall was pre-ordained. After all, Apple has come in and disrupted industries before, Steve Jobs was on top of his game in the mid 2000s, and the iPhone was simply revolutionary, so of course this meant somebody like BlackBerry would fade away.

But the book tells a more complex story. A well-written narrative by authors by Jacquie McNish and Sean Silcoff describes how BlackBerry managed to create a data messaging network at a time when data networks were terrible and expensive, and how BlackBerry managed growth through a two-headed CEO, one of them the technical genius (Mike Lazaridis), and the other a cut-throat businessman (Jim Balsillie). These are brilliant guys who created something amazing.

What is even more important than the rise of BlackBerry, however, is how it fell apart. Lazaridis and Balsillie were just as smart before the iPhone came out as after. But the things that allowed BlackBerry to exist and thrive – it’s stinginess on data usage, and its long battery life, simply became unimportant to people. Data networks improved while the cost dropped, and people didn’t mind having to plug their device in more often, especially if the device allowed them to play Angry Birds.

Missing this was a failure of Lazaridis, who was still insisting that 4G networks weren’t worth implementing, and who felt that customers would flat out reject a device that didn’t last multiple days on a single battery charge. However, to be fair to Lazaridis, he had gone through this experience before, telling carriers they didn’t need 3G as quick as they thought, and pushed them to a technology dubbed 2.5G, and Lazaridis was proven right. It was not out of the question for him to think he was right again on the 4G transition.

Similarly, on the business front, Balsillie rubbed many people the wrong way, and BlackBerry’s insistence on monthly network fees to use the encrypted data service was something that carriers were constantly trying to get out of. As soon as BlackBerry faltered (with the ill-conceived touch-screen Storm device after the iPhone came to market), the carriers began to flee to Android. Was BlackBerry a better, more robust, and encrypted service?  Yes. Did it matter?  No.

The book also highlights one of the things that is common with startup businesses that become successful. Many times the person or people who found the company are not the people who should be leading it after it makes its mark. BlackBerry never learned this. The co-CEOs appointed a compliant board that didn’t ask questions, and when market realities changed, there was nobody to question the founders when they continued to play by the same technical and business rules as before, when Apple and later Google changed those rules completely. Essentially, the game changed from baseball to football, but BlackBerry was trying to sell better pitchers mounds and more efficient means of putting the bases on baseball diamonds.

One issue I had with the book is that secondary characters – friends of Lazaridis or Balsillie who were also very important to BlackBerry’s success, sort of fade in and out of the narrative, only popping in to give the after-the-fact 20/20 hindsight account of how a decision Lazaridis or Balsillie made was wrong or could have been better. It became a bit difficult to remember all of these names, and since they weren’t considered essential characters, it is hard to understand whether the failures and successes were all on Lazaridis and Balsillie, or if the actions of these people also contributed to the rise and fall.

As an example, Lazaridis would give incredibly difficult, probably impossible, product deadlines for a new device. Up until the iPhone release, the engineering teams consistently pulled these off. Afterwards, it was consistent failure (starting with the Storm). It would have been nice to get more information as to why this was – were teams burnt out? Were they just lucky before and the law of averages kicked in? Did the scope of the impossible change to something that truly was impossible whereas before it was just hard?

Another issue I had was that this book didn’t contain any pictures of the people or products. Usually biographies have some photos in them, and their absence was strange.  The book would go to great detail to describe what the first BlackBerry looked like (the 900), describing how revolutionary and addicting it was due to its keyboard. Additionally, lots of time was spent describing the Storm and what was wrong with the technology, but no picture of the Storm itself. I found myself jumping out of the book to look these devices up, both to get an understanding of what was being described, and to take a trip down nostalgia lane as I remember many of these devices. Putting these in the book seems like a no-brainer.

Despite these shortcomings, it is a very well written book, and a quick read. It doesn’t get too deep technically, so you will not get lost trying to understand the technology, and the business stories, especially the one where Balsillie basically conned Blackberry’s way into Europe, are quite entertaining. There is also an interesting epilogue, showing what the two now-departed CEOs are doing with the riches they earned.