Summer trip 2015 is in the books!


As time is passing I have to say that being more open with people around us is making me a happier person. Hell, being more open with me is making me a happier person.

Asking for what you want (consent is key after all) can make what you want happen.

Communication. With your partner. With your friends.

Not having fear about what people are thinking about you since chances are that they are too worried about what you’re thinking of them.

The list goes on and on.

Because what you want is a real thing and shouldn’t be ignored.

Want doesn’t equal need. But wants are valid regardless.

Laws and the Clerk

This is way old news by internet standards… but I have to say that I’m very happy with the outcome of the case against the clerk in Kentucky.

You decide to trade her freedom for a little publicity.

And not doing her job and keeping the rest of the county hostage.

It’s just like the people who still insist the earth is flat. (Yes, this is still a thing.)

And this is why we have the separation between church and state exists. The keep people in the government from compelling the citizenry from behaving by a certain religion.

Of course none of this affects non-gay couples. No one is being forced to get gay married.

But her sincerely held belief was preventing people from exercising their rights.

Which is exactly the point. One can’t force someone to be their religion.


Seattle has a problem with homelessness. We live in a temperate climate and amongst bleeding hearts that don’t know better.

When confronted by this problem on Reddit, the standard response is typically one of two extremes: “fuck ’em,” or “just give them a home.”

In my humble opinion: both a dead wrong.

If you look at the problem of homelessness you can take the first order approach of “well, they lack a home.” Alternately, you can treat them like rats. This misses the point.

The problem with being homeless isn’t the lack of a home, as crazy as that sounds. The problem boils down to not having the ability to have a home.

There can be simple and benign reasons (relatively speaking) like losing a job. This can be remedied by getting another job. Most people who are homeless due to this problem aren’t homeless after a little bit of time. Assistance in finding a job and temporary shelter while this process is playing out could very well help. Not being able to get a job can be mitigated with job training.

There’s another slice of the homeless that disproportionately affects LGBT youth: being thrown out of your parent’s home. Again, shelter is a good idea. Shelter, caring, and acceptance.

Those are the relatively simple cases. Simple because the people aren’t broken. Simple because they don’t want to be homeless. Simple because they want to be able to help themselves, but can’t due to circumstance.

Simple because by and large they are not the problem that everyone is pointing at as “the homeless problem.”

Then you walk down Aurora Ave. and you see all the meth faces. You see the heroin addicts shooting up in broad daylight in Denny Park. Simply giving these people a home isn’t solving the problem. The problem isn’t lack of a house. The problem is the fucking drugs they’re strung out on. You give these folks a house and I guarantee that half of them would strip the copper out of the walls and trade it for drugs. Helping these people get off drugs — and just as importantly to not commit crimes against people of property — would do a world of difference.

Or they could just be self-medicating. Which gets us to the other category: mental illness. We don’t like thinking about this much, especially after shuttering the institutions in the 1980’s. Treatment. This would do a lot of good for all of the mass shootings and so forth. Crazy is unpredictable. Crazy can get all stabby (Pioneer Square comes to mind). Crazy is what makes people take a wide berth around the folks sleeping on benches.

Simply giving a home isn’t solving the problem.

You need to solve the underlying problem before you can solve the visible problem.

VMWare – Why I don’t need Parallels any more

A few days back I was ranting on about Parallels.

I decided to move to their competitor. Not only is it roughly the same price, but at the same time they are coming out and advertising that the license is usable on multiple machines.

So there. That, my friend, is exactly why competition is a good thing. The market works in generally good ways for the most part.

All I had to do was import my existing images and like “poof,” I don’t need the old software any more. I have a few tweaks I have to do, like getting a static IP working on my dev Windows dirt, but that ought to be pretty easy.

Again, I have to stress that Parallels lost a long time (since around 2008 or so) customer. C’est la vie, eh?

Production Mode

There are different phases of delivering some software. You can be planning or writing the code. Eventually (I suppose if you’re lucky) you launch.

The process of getting things into production is very different from writing the code. It’s still work, but it’s different. At a place like Amazon where you have many other teams that you’re depending on this is a long process.

Then, after a long while of getting the product out the door is complete… you have to figure out how to code again.


It’s always nice to get back into code again. (Doubly so since the system I had been dealing with wasn’t even freaking code the way you normally think about it!)

Schematic Capture

This is a strange rant I’m queuing up.

This is about circuit design and the process both before and after it.

The general process of making a circuit board that’s all populated and ready to buy goes kind of like this:

  1. Come up with an idea
  2. Draft of circuit idea
  3. Schematic (sometimes on a napkin)
  4. Choose components
  5. Prototyping
  6. Enter schematic in a design tool
  7. Board layout
  8. Spin board and populate
  9. Testing
  10. If you need to go back to step 2

Eventually you’ll finish. (if you’re lucky)

The issue I have is that going from 4 to 7 many times you have not one, but a number of similar components to choose from. Even beyond that, sometimes the same component comes in different packages.

Of all the tools I’ve been dicking around with there seems to be a bit of an inversion. You need to pick a package before you drop the part on your schematic. This isn’t as much of an issue with a lot of ICs, but when you get to passives or jellybean ICs you have choices. Unless you have other design constraints, if you spec out a 1nF cap, you can get that in any number of different sizes.

Why do you have to artificially constrain that at schematic capture time?!



Seattle is not known for spectacular weather events. Sure, it has the reputation of rain, but it’s never portrayed as a violent rain, but rather a gloomy sedate form of wetness.

This isn’t really true, but it keeps people away. (In the summer it is locally known for spectacular weather)

Today was a weird one for sure though. In fact we even got a warning about it from the weather service. It was forecast for rain and wind. Wind for real this time.

Ok. So this isn’t midwest wind, or hurricane wind.

Seattle is soft.

We got a downed power line (well, power feed to a house) in the trees.

(Only 240V 2-phase… somewhat high energy, but not too dangerous)

The funny thing is that this passes for severe weather.  :-P

Some trees lost branches and parking lot trees (with shallow roots) tumbled. But that’s really the extent of the event.

So… windy.

Easter Eggs

This video dropped on Wednesday as we were launching Amazon Underground.

Within the first few seconds we got an Easter Egg.

I’m not going to say what exactly, but if you guess I’ll tell you if you got it right.  :-P

The Lull

Walking in this morning was kind of strange.

Bunches of people were up since 2AM yesterday. Maybe earlier.

Crash. Hard.

This morning there was almost no one.

Helium balloons barely stood upright having descended overnight.


Check service dashboards.

Check social media for mood of the product (good! :-) ).


Amazon Underground!

I’ve done a few minor little product launches at Amazon in the past. I’ve had big launches outside of Amazon before. It’s all good.

Today we had a real launch. Something I was a part of.

Amazon Underground.

This is actually a pretty cool thing for Android folks and their developers. Basically you pull developers out of the freemium type of game where you it feels you have to pay to have fun.

Up until this point a developer could have ads in their apps, have in-app-purchases (IAP) to make money, or simply charge money for the app itself. (There are some other ways too, but these are the common ones) Ads are annoying and paid apps have a lot of difficulty to gain traction.

This leaves the pay-to-play apps like Candy Crush and such. It’s either a stupid grind to get to the next level or you pay to get there quicker. Either option is annoying and not really fun. That said, this is the way to get the most money in many cases.

What we have made is a system where the user gets the app for for free (all of the app for free!) and the developer gets paid for the time the users use their app. Amazon foots the bill for it and we make up for it with occasional ads and getting more people into the Amazon ecosystem. This leaves developers free to make a game that is simply fun.

I do have to say my preferences don’t really lean towards Android, it’s just not my cup of tea. This isn’t enough to make me skip off to Android. But, other folks like Android, and for that it’s a pretty compelling thing for both users and developers.

Launching it. Watching the press and generally positive vibes. Getting a “Jeff Letter” on our own page that different, and dark and non Amazon-y. Being proud of what we did. Each of us did little parts of the whole. But Amazon is a big place with lots of systems all over the place. A big team. A big launch.

Yeah, you should check it out.  :-)

<disclaimer>I’m writing this as George, not as a representative of Amazon. My own opinions, not theirs.</disclaimer>  (yes, they asked people to say that… sigh)

%d bloggers like this: