Skip to main content






How Facebook Figures Out Everyone You've Ever Met - Gizmodo

I think tech people have been aware of this for a while, but it is worth exploring again

How Facebook Figures Out Everyone You've Ever Met - Gizmodo





Crypto and You

3 min read

So you want to learn more about the encryption debate.1 Well, take a knee, gang, its .

Flash back 10 years ago. No one gave a shit about cybersecurity unless you were in China or a ghostwritten Tom Clancy novel. Then, as people started using networked services in more places, the information leaks began. It was still not an issue for the government (particularly, law enforcement), because useful data was just as accessible to them as it was to nefarious agents, like hackers or Facebook.

In 2013, the Snowden leaks began to paint a picture of just how much our own security agencies relied on cybersecurity weakness in their day-to-day operations. Snowden had trouble finding a journalist who could figure out how to use PGP to read his heavily encrypted messages to disseminate this information.

The leaked information made Silicon Valley companies very angry. So they began to encrypt transmissions between their data centers, as well as building it into their email, messaging, and mobile operating systems.

Now, encryption is just lots of math. Ever watch a movie about code breakers in WWII? Encryption. The only difference now is that common computers can do a lot more math in a short amount of time. But before this point, it was never viewed as particularly important to consumer software.

Think of it this way. In the same way that a gun is an offensive weapon, encryption is a defensive weapon. It protects your information from prying eyes, whomever that may be.

So now law enforcement has a problem. This technology is widely available, even to enemies of the state. Their proposed solution is to break it. Or to put it in their words, make it work for some and not for others. Kind of like how if you point a gun at something or someone you like, it will not fire. Because it only works in certain situations.

This is obviously a farce. As the old argument goes, if we outlaw guns, only law breakers will have guns. Since enemies of the state are not likely to stop using encrypted communication if it is outlawed, the only people without it will be law-abiding citizens. Effectively the exact opposite of the stated goal.

Remember this when some idiot presidential candidate2 tries to tell you how encryption is bad because terrorists are bad. The only benefit of outlawing encryption is to spy on you.

  1. I know there is a new John Oliver monologue about this. I haven't seen it. I hope he makes some of these points better than me.


    via GIPHY



Beginning the move to Known

2 min read

Moving the blog to Known has been in the back of my mind for a while now, so I finally pulled the trigger. This has been a fun weekend project, now spilling into Tuesday. Wordpress maintenance has become too cumbersome for my needs, and Known has been a breath of fresh air. Some notes from my experience:

  • Invaluable installation instructions in the Known documentation, as well as here. The only issue I kept having was where to place the uploads directory, and how to point to it.
  • Really, reeealy looking forward to importing my old blog content here. I know that Wordpress import is on the horizon, and I forsee it bringing a large influx of personal sites to their hosted service.
  • I love that the functionality works out of the box.
  • Chrome handles it very well on Android, but I need to play around with it more to figure out how to reply. Almost needs to be its own app...
  • Facebook thinks I am a robot when I try to setup POSSE here, even after I answer a captcha and verify my mobile phone. So they don't get to see any of these posts.

I have not written many posts lately, as evidenced by my Wordpress page (currently at Hopefully this refresh will kickstart some creativity.


The Indie Web

2 min read

I have been griping about leaving Facebook and owning my data for a while now, but I may have finally found a solution.

The indie web, its called. Made up of the people who got tired of talking about these concepts and decided to start doing something about them.

I first heard about them on one of my weekly must-hear podcasts, In Beta (episode 90). Then a guest on another webcast I enjoy regularly - This Week in Google #241 - brought it up near the end. (TWiG actually just dedicated most of an episode to it - #266.)

I always viewed blogging (at least my blog) as a spot for thoughts that skewed toward longform writing, that could not be fleshed out in a short snippet. The indie web, however, encourages tweet-length thoughts as well as longer posts, which can then be syndicated to whatever social network you choose. The point is to not keep those thoughts siloed somewhere that may eventually shut down or change their policies, but to control your online identity, on your own terms. You can build your own tools, or you can browse the IndieWeb site to find something pre-built to use on your site.

But how will people see these posts if they are on or ? Well, they thought of that too. POSSE stands for Post (on your) Own Site, Syndicate Elsewhere. By syndicating, your non-IndieWeb friends will still get to see what you are up to. And using the webmention protocol, comments and replies will be pulled back into your own site as well.

If you are feeling less brave, one of the higher-profile indie web tools just launched a beta. It is called Known, and it has been treating me well so far. I look forward to their hosted/beta service adding a Wordpress plugin. But if I did my work right, you should see my post about this article down below.

TL;DR: if you see anything oddly formatted posts on my site or one of my social profiles, it is probably an attempt at implementing one of the Indie Web projects. Maybe if you are getting wary of Facebook and you're looking for a technically-inclined side project, you should check it out too.


Facebook's friend problem

1 min read

In the following days, even a passing meeting guaranteed that a friend request would pop up the next time I logged on to Facebook. I felt popular and informed, at all times abreast of what my hallmates, friends, and peers were up to each day....

Facebook's friend problem
from Pocket via IFTTT



Streaming Music Throwdown

6 min read

I have been experimenting with the various streaming music offerings, since it is the future of music consumption.1 Due diligence seems to be the only way to differentiate these rapidly-changing and -improving services, so here we go. Time for another table post.

ServiceGoogle All-AccessRdioSpotifyPandora
Price/month $10 $10 (decreases for multiple accounts) $10 $3
Import ability Upload (available for free) Matches2 Matches N/A
Radio recommendations Seemingly random Ok & adjustable Sparing Repetitive integration Third party3 Yes Yes Third party
Ads on free version No Yes Yes Yes, lots
Desktop No Basically a browser window Yes No

Current champ

Google's music app is pretty great, and when the All-Access part was released it seemed like the perfect complement to complete the service. In my opinion it is no longer doing what the All-Access part is meant to do - helping me discover new music. I guess in the most basic sense, I have heard some songs on there that I had not heard before (not memorable enough for me to bookmark them or add them to a playlist though).

The major feature of Google Music is that you can upload your own library of songs, but people forget that this is not part of the paid service; you can do this for free. Paying just adds the ability to play songs that you did not upload, and roll them into its sub par4 radio stations. Regardless of whether I keep the All-Access service, the music locker facet is invaluable and the best implementation I have seen.

Dark horse

Spotify is probably the biggest player in this space right now (I admit I did not give it a lot of credit on the first draft of this post), but it seems like a mess. I still cannot figure out how to add songs to my library without putting them in a playlist. This must come from years of managing a large library, but I do not want to organize my collection this way. Their radio offering seemed just OK - I tried it again for this first time in months (my starred playlist station) and hit the same 2 albums 5 out of the first 6 plays (The Suburbs and Yoshimi Battles the Pink Robots, but still).5


Pandora is kind of in a different boat, but I have used it several times over the past year. The algorithmic muscle of Pandora is great but the idea has yet to be fully realized. The fact that it can only pull from a library of 900k tracks6 limits its usefulness as a recommendation engine. The ads are a bit frequent, but not quite enough to warrant paying.

And the winner is...

Rdio seems like the front runner to me. It has many of the same top-flight features as the other two services, but with a little more attention to detail. The design is fantastic, and the organization is much more straight forward than Spotify. It feels much more social within the service than GMusic, but hitching to a Facebook account is optional rather than mandatory7. The radio recommendations can be adjusted between "familiar" and "adventurous," depending on what you feel like listening to.

There is one feature Rdio has that GMusic and Spotify definitely do not: the queue syncs between devices. This means when you start listening on another device, you are on the same song in your playlist or station. This detail by itself is nearly enough reason to switch.8 You can also mark items for download to your mobile device from the web player, which I know Google can't do.

The only gripe I have seen about Rdio is that it does not advertise its bitrate. Spotify and Google stream at 320 kbps on wi-fi, and Pandora is something like 160 kbps unless you buy a subscription. My argument is that if I hear something that I like but is low quality (on terrestrial radio or over a cellular connection), I am going to purchase it in a high quality format (CD, FLAC, vinyl - audio quality is a rabbit hole in itself). Things that are in a low bit-rate that I don't need to hear again are not a problem. Moot point.

So there are my thousand-plus-word thoughts on the state of streaming music (Beats Music not included, because it is too new). TL;DR: Google Play's best feature does not require the paid version, and Rdio is the intuitive, good-looking underdog with a can't-lose attitude that wins my pick for best streaming music service available.

  1. The second footnote of this link mentions Plex, and while they have a media server, it doesn't support some basic music player functions, like playlists. I am sure this will be remedied in the future, but until then it is behind all the rest of the services listed here. 
  2. Matching with Rdio requires WMP or iTunes, which sucks because I actively avoid both. I don't think Spotify is as strict on sources. 
  3. This basically means no for Android. The only scrobblers I have tried read from the system audio player, which then catches all the podcasts I listen to as well. It would get the job done, but it is also really annoying. 
  4. The quality of the radio is a function of the feedback you give it, and only gets better over time. It did get better, but the quality definitely plateaued much sooner than I would have liked. This could be a function of what I listened to on the service, but Rdio and Pandora are both still improving their suggestions in my opinion, with roughly the same amount of feedback. 
  5. It does not help that their web player seems to be blocked at my place of employment. Desktop player is (or was) fine, but I have never successfully started their web player.  
  6. The largest library for comparison belongs to Spotify, which has over 20 million tracks. Google and Rdio are not far behind. 
  7. The best social analogy I have heard is Spotify : Facebook :: Rdio : Twitter. Yes, I have lots of friends on Facebook, but there are few that share my music taste. 
  8. integration is more that enough reason to switch, with this kind of history


'Virtuous' circles of software

8 min read

I started a list several months ago of iOS-first applications that had everything to gain from launching on  but seemingly just decided not to. Far from complete, and probably like the third post I had ever written on anything, but it was a start. Anyways, this morning on All About Android I heard about this guy who tried to elucidate the point in probably the most smug way possible. The tone makes me feel like he would carry hand sanitizer around just in case he accidentally touched an HTC One. So here is my attempt to FJM this whole thing.

photo: Andy Ihnatko via Compfight

1. "In the US, iOS market share is still extremely strong (even pre-iPhone 5s launch data showed Android having peaked, so Q4 data will be interesting with Apple’s refresh)

Well, since you aren’t going to bother with data, I guess I will: Android covers 52% of the US market to iOS’s 40%. But, if you want to go anecdotally too, I know someone who is dropping iOS entirely because of a gamebreaking iOS7 bug that neither AT&T nor Apple has remedied. This makes me doubt that will make up that 12% gap in Q4.

 Since the vast majority of innovative mobile startups come out of the US, Apple’s stronghold domestically has an absolutely massive impact on developer mindshare

Yes, there are a lot of US startups, because there are a lot of US VCs with a lot of disposable income. Apple’s “stronghold” doesn’t impact developer mindshare, it impacts VC mindshare, because they fall into iOS’s affluent demographic. It is a status symbol to them. That doesn’t make it better or easier to develop for.

2. All of my conversations over the past year with Android developers…  building and releasing on Android costs 2-3x more than iOS.

Ok, anecdotal again. Here’s an anecdote: Any.DO (just featured on The Verge for inspiring the design of iOS7 you love so much) started… on Android! DUN DUN DUN! It built up the hype there before hitting iOS, and I would say it worked out well for them.

3. The effort required to build and release an app is severely gated by capital-raising…

4. These structural limitations around capital raising for venture-backed companies…

*yawn* First world problems. You don’t even use the word ‘mobile’ (let alone ‘iOS’) until #5.

5. To build a mobile app with $1M in capital, a startup can roughly afford to hire one designer, one client developer (iOS or Android) and one back end engineer

Here is the bias. You know who had more than one of each of these (and definitely more capital)? Twitter, after buying Vine. They had nearly 3 months to at least add viewing capability for their Android client before its public release, which was already iOS-only anyway. But maybe that is a bad example. Again, Any.DO shows that you can do Android just as well with the same ‘constraints’.

6. Almost zero startups are going Android-first under these constraints. Why? Because founders know they have an extremely high bar to prove traction on the primary platform, before they can raise additional financing and accelerate into two platforms

Again, this is a fallacy - see my answer to #2 again. Unless you think the “extremely high bar” is somehow lower on iOS. And it might be, when you have (generally) affluent guys paying (generally) affluent guys to create an app for a platform used by (generally) affluent people. It would be hard to get that kind of ‘vision’ to sync across stages of development on platforms with a more diverse base of users.

7. So it’s well known in tech circles today that seed round sizes constrain app development to a single primary platform.

Ok, that may be a point. My broken-record answer of Any.DO did start on only one platform, and most others start on the other platform. But in general, I would ignore any sentence featuring the phrase ‘it’s well known in tech circles’, much like ‘quantum mechanics.’


And startups are choosing to go iOS first not only because development is cheaper and easier, but also because money for in-app purchases and advertising is overwhelmingly skewed toward iOS 

This may be true. Once again, I will do your research for you, instead of relying on ‘tech circles’ - Play Store revenue is up to 35% of global app revenue market share, leaving Apple with a huge advantage. It may never be on par with Apple here, but it is hard to compete with an install base of (generally) affluent (generally) white users.

In fact, a recent study of Facebook ads shows ads were 1,790% more profitable on iOS. This is extremely incriminating for Android and is the worst kind of news for Google.

What do Facebook ads have to do with anything? I see Facebook ads on my Galaxy Note 2, and they suck because they are not relevant to me, so I don’t click on them. I only see them in the Facebook app, which is not made by a startup, and probably launched on both platforms as quickly as humanly possible. It is not bad news for Google, it is bad news for Facebook on Android.

You want incriminating? Ad revenue generated on iOS has fallen 11% in market share since last year, while Android has remained steady. I didn’t even have to use a ridiculously large number from a third party that has nothing to do with my overall point.

8. Since iOS better supports startups’ ability to prove metrics requisite for raising Series A rounds from institutional investors, the earliest most innovative services are almost always available first on iOS.

The metrics you gave in #3 were: significant traction, repeatable user acquisition strategy, early ideas toward monetization, etc. What do those have to do with platform at all? Download counts and active users are easily found in the Play Store and its developer console. User acquisition is definitely platform agnostic, and if anything, Android users can get the word out about a new app they are using with the built-in share menu that is light years ahead of iOS. And - repeat - see #2 to show it can be done.

9. … often these startups become acquihires for the top mobile acquirers (FB, Google, Apple, Yahoo, Dropbox, etc). Because they are almost always shutdown at acquihire, big companies often have some of the most talented iOS engineers and product people in residence for a 1 to 2 year earn-out period. Without a doubt, these employees skew toward iOS when they join internal projects or think up new ideas. And when they eventually leave, there is a good chance they’ll stick with iOS again. There is no doubt this forms a sort of virtuous circle of iOS-first talent in the startup community.

This is a symptom, not a cause. You can substitute Android for iOS in that paragraph, and it would be a symptom of Android-first, not the cause of it. And why is this ‘virtuous’? Did you mean to type ‘vicious,’ but your iPhone keyboard autocorrected it? A little Freudian if you ask me.

While in theory Android provides a very modern platform for mobile development,

Gravity is also a theory.

Startups simply cannot afford to bypass iOS and go Android out of the gate. One could even argue the gap is widening.

The gap is widening only in the minds of people like you. You perpetuate this stereotype, and this is the reason companies that are not startups, like Nike, completely skip over half the mobile market in favor of the first choice among the (generally) affluent ‘tech circles.’

The reality is that software innovation at the app layer is accelerating, and converged hardware / software development costs a lot of money.

It is accelerating because there is competition. And starting on a certain platform before moving to another does cost money, but that is not a reason to start on the same platform every time. The startup issue is a chicken and egg scenario, and nothing is going to change until VCs start laying their eggs on the other side of the proverbial fence.