I’ve been working on an internal web app for the past few weeks.

I’m not a “real” programmer, so as usual I just figure things out as I go along and make lots of mistakes, but I always learn new stuff. This is how I learned programming in the first place, got my first gigs building websites, then ended up at OpenPlans, Code for America and now USV. My favorite thing about the web is how it makes it possible to just pick a direction and learn as you go. It’s amazing, really.

My background is that I’m good with front-end development (html/css/js) and decent with PHP (have built a ton of wordpress websites, plugins, themes), but I quickly trail off from there. So I’m very comfortable in some places, and not so much in others (linux admin, deploying python apps, etc).

But for this project, I decided to try something new. I’m using tornado (a python web framework), paired with mongodb. The app is hosted at Heroku and my mongo database is running in the cloud at Mongolab. I’m using twitter auth for log-ins, and I’m also using the hackpad API to embed editable documents in the app.

What’s so cool about this is that pretty much everything is a cloud-hosted service, wired together in a really light way to make the app. There is a tiny amount of application-specific code. If we had used Brubeck instead of Tornado, there’d be even less.

It’s been a joy to develop in this environment. I like programming in python, and mongodb is a breath of fresh air. And being able to simply wire in things like hackpad (with some really great help from Igor) has been amazing. Deploying updates is as simple as git push heroku master and heroku does all the heavy lifting.

Much of the starter code for this came from Zach, our hacker-in-residence at USV, who’s been working on an update for USV.com. I don’t think I’d have been able to do it without his starting point — but looking at the Brubeck docs, which are quite good (kudos, James), perhaps that’s not true.

All of this reminds me of a post my friend Ian Bicking wrote a few years ago about What PHP Deployment Gets Right. For those who don’t know Ian, he is a rockstar python developer who understands the low level stuff better than anyone I’ve been around — among other things he wrote pip and virtualenv, which are primary tools for anyone developing in python. Anyway, the point of Ian’s post is that PHP deployment is, and always has been, so easy that it’s possible for people (like me, until recently) who are afraid of going deeper than the file system, to program and deploy apps.

Ian’s post is 5 years old now, and a lot has changed. It’s pretty sweet how the hurdles for developing web apps keep getting lower and lower.

Today I’ve got a post up on the USV blog about Senator Chuck Schumer’s Patent Quality Improvement Act, and the problem of software patents and patent trolls in general.

The PQIA would make it easier and cheaper to defend against frivolous patent infringement suits. This isn’t everything we need to fix the problem, but it’s a step.

Let’s give Schumer some twitter love for making this a priority and taking a crack at it.

The week before last, I attended the Mesh Conference in SF, which brought together a big group of folks working at the intersection of the web and the “new economy”: i.e., the “sharing economy”, “peer economy”, “connection economy”, “collaborative consumption”, “the mesh”, etc. As you might imagine, a large part of the discussion focused on the lexical problem that this “movement” faces, with all of these overlapping and somewhat competing terms that attempt to describe it.

I’ve written about this problem before — and it’s true that any discussion with folks working around this space has to struggle to avoid falling into the “what do we call this?” trap. Which is annoying, but I do think necessary. It’s clear that there’s an unfulfilled need to try and describe what we see happening here in a way that really connects. While the words we’ve tried so far do make sense, I don’t think they yet create an appropriately wide tent, and I definitely don’t think they resonate with folks outside the tech community bubble.

As I was thinking about this on my way home from the Mesh, it struck me that maybe there is a common theme that runs through all of this. At the heart of all of these platforms and communities is the fact that they are people-powered. They exist to super-charge people — to turn them into superheroes. They empower people and are powered by people.

They are powered by us.

At USV, we talk a lot about how “networks” are a fundamentally new kind of entity. Whether you’re talking about marketplaces like Etsy, Zaarly, or GoodEggs, communities like Tumblr or Indaba, or funding platforms like Kickstarter or Kiva, they all share a common theme: that they are people-powered. The platform is simply a lightweight architecture which allows people to connect and transact directly with one another on top of it.

Of course, you would be correct to note that all companies and industries are powered by people, whether you’re talking about school teachers, auto workers, or movie executives. But there’s a difference between traditional companies and these new people-powered networks. Buying a car from GM is about buying something from a company. Buying a lamp on Etsy is about buying from a person. Getting a record deal with a label involves people, but building a music career on a network like Indaba and getting your album funded on Kickstarter is powered by us.

In each of these cases, the companies and organizations (nonprofits like Wikipedia or the Freelancers Union) at the center of this aren’t traditional bureaucratic hierarchies — they are networks, and what they do is help people connect with one another directly in new, exciting, and groundbreaking ways. The connection is the service, and what’s being connected is us.

Nicco Mele calls this “the end of big” — and argues (correctly, I think) that this is not just a passing fad (like “social networking”), but a fundamental shift in society. And for all the opportunity it brings, it also presents new challenges as we work to hold on to the most fundamental societal values (safety, accountability, freedom, privacy). He’s right, and part of the puzzle here is to create a policy environment that honors and protects those values while leaving ample room for us to explore “powered by us” approaches that have never been possible before now.

After two centuries of de-personalization through industrialization, we’re now flooded with new ways to both re-discover individuality and community, while simultaneously tackling huge social and economic problems (education, health, energy, transportation). And just to make sure I include the word “awesome” in this post at least once, I’ll end by saying that this is all pretty awesome and exciting.

I am writing this post from the balcony of my Airbnb apartment in nob hill in sf.

Its by far the nicest place I’ve ever stayed in sf. Beautiful duplex penthouse, with an amazing view of downtown, for less than the price of a hotel room.

To get here, I hopped a ride via Lyft, with a super nice guy.

I guess this is my first travel experience fully powered by the peer economy. And is super awesome all around.

It just goes to show that you can travel and transport yourself via peer to peer means and have a totally normal, and exceptionally awesome experience.

This is something you can’t do in most places – sf is definitely the best testing ground for this kind of thing right now.

But for sure: the experience is more authentic and more enjoyable than would have been possible otherwise.

As much as all of this is mainstream in the tech community discussion, its still pretty outside the norm for most people.

But this weeks experience has definitely demonstrated to me that we have new and reliable ways of establishing trust with others, which are totally real and effective.

Pretty cool.

Matthew Yglesias has a good piece up this morning on the immigration debate in the wake of the Boston bombings. He points out that historically, as we’ve tightened our border lockdown, we’ve not decreased illegal immigration, we’ve just made the coyote industry more lucrative.

In my favorite line, he suggests that rather than tighten our lockdown, we should open up:

by far the best way to keep dangerous foreigners out of the country is to make it easier for nondangerous ones to enter.

In other words, the best approach is not lockdown, but antilockdown.

If you think about it, this counterintuitive thinking applies to lots of other issues, and particularly reminds me of the debate over piracy — e.g., the best way to decrease illegal downloading is not to make it harder to copy files, but rather to make it easier to buy & share them legally.

I am going to keep thinking of other issues where this kind of thinking applies.