Albert Wenger's posts and talks

Blockstack Browser and Token

Today Muneeb and Ryan from Blockstack delivered the morning keynote at the Consensus 2017 conference here in New York. They made two important announcements: first the availability of the developer edition of the Blockstack Browser and second the news that there will be a Blockstack Token. To understand the importance of both of these announcements, it useful to look at some of the history of the Internet. 

The Internet got going in the early 1960s but for the first quarter century its usage grew slowly. Broad usage really didn't take off until the creation of an easy to use consumer layer with the Web and the availability of developer tools in the mid 1990s. Now fast forward to today. We all use the Internet every day for nearly everything we do. But now we are seeing some critical problems that have emerged based on the basic architecture of the Internet. For instance there is a lot of blind trust into lower level infrastructure, such as certificate authorities that can result in large scale man-in-the-middle attacks (for example a recent event in Turkey which resulted in endusers believing they were connected directly to Google when they were not). Even more importantly, state in the current Internet is maintained in databases operated by companies such as Twitter, Facebook and Amazon which hold and control users' data. 

Blockstack aims to resolve these problems using blockchain technology and by building the developer tools and consumer layer to make this new decentralized Internet broadly available. The Blockstack system has now been up and running for 3 years and has a community of over 5,000 members. 

How does Blockstack address these problems? With Blockstack endusers (individuals and companies) control their identities, storage and payment credentials directly. This is accomplished by keeping the namespaces — both enduser names and addresses for applications — in a blockchain. The blockchain here is a virtual chain that currently sits on top of the Bitcoin blockchain but could be migrated to another blockchain in the future. In fact the chain has been migrated once already as it started out on top of Namecoin. The virtual chain layer requires only minimal guarantees from the underlying layer and allows for separate scaling. 

The Blockstack blockchain itself in turn only contains keys and pointers. All the enduser data is stored in existing storage systems, such as S3, Dropbox, Google Drive etc. But the way it is stored there is fully encrypted and distributed across these systems. That way each underlying storage system is used like a dumb hard drive and the use of multiple systems provides resilience at high performance. This is provided by Blockstack's Atlas Network and Gaia Storage components.

Now that all of these components are in place, how do consumers access these systems and how can developers build applications? This is where the Blockstack Browser comes in. By running a local application, consumers can use any existing web browser (Chrome, Firefox, etc) to access the new decentralized systems. Right from there they can do things like register names, make payments, configure their storage. This feels just like using an existing centralized system and requires no special skills or knowledge. Similarly, developers can now write decentralized applications without having to learn all the details of the underlying systems. Instead, they can write in Javascript and implement the authentication and data access APIs exposed by Blockstack. 

Finally, how does the Blockstack Token fit into all of this? For the Blockstack network to work there is some degree of scarcity that has to be maintained. For instance, there is a need for names to be unique. Whenever you have scarcity, you need an allocation mechanism. To that end Blockstack will introduce a token some time later this year. The Blockstack Token will be mined at the layer of the virtual chain and will power operations such as name registration. It will also be used to let Blockstack network participants vote on protocol changes, including a potential migration to another underlying blockchain in the future. 

So if you are a developer and want to start building apps for the Blockstack Network, head on over and download the developer preview.


Disruptive innovations often start out as worse versions of something that already exists. Worse along all dimensions except for one, but one that turns out to really matter because it unlocks a new use case for which the incumbent does not work at all or is not affordable. The PC was slower, had less memory and less storage than a minicomputer. But it was massively more affordable and could be set up and run without the help of the centralized IT department. That’s why the early adoption of PCs was in parts of companies that did not have minicomputer access.

Direct networking among consumer mobile phones is a disruptive innovation. It doesn’t give you the bandwidth of, say, LTE. If there is extra hardware involved it only works between participants who have that hardware. But it has one crucial advantage: it works even when the mobile network infrastructure is down (e.g., during an emergency) or where it doesn’t exist at all (e.g., when hiking in the deep countryside).

Extra hardware paired with your phone is exactly the approach that goTenna has been pursuing for phone-to-phone communication. They have been shipping tens of thousands of their first device which are in use by outdoor enthusiasts and also by emergency responders.

goTenna’s latest product, goTenna Mesh is about to ship and supports mesh networking. That means as more people in an area have the product, phone-to-phone connectivity improves for everyone. Including one super cool feature: with SMS network relay, if any phone in the mesh also is connected to the SMS mobile network then all the goTenna-enabled phones can relay regular text messages out through the SMS-capable device.

We are excited to be leading the Series B round for goTenna. The team has accomplished all of this on very little capital to date, including the development of the mesh protocol and of a hardened Pro device which will start shipping later this year. We believe this is a disruptive innovation in connectivity, and a perfect fit with the Access 2.0 portion of our current investment thesis.

If you are excited too, you can pre-order a set of the new goTenna Mesh units. And if you are truly excited you should check out the terrific opportunities for joining goTenna’s team.

Top Hat

Even back when I was in graduate school, I found the price of textbooks to be high and their quality to vary widely. Now that I have children taking college courses, I was shocked to find textbooks that cost over $200 and are still large physical objects that have to be lugged around! The high prices and lack of innovation are the result of a market structure which has become highly concentrated among just a few textbook publishers. That's why I am excited to announce that USV has led a new round of financing for Toronto-based Top Hat, which last year launched a content marketplace for higher education.

I first met Mike, the founder & CEO of Top Hat, shortly after he had started the company. He told me about his exciting vision for bringing innovation to the higher education market. But then he said he was getting going by replacing Clickers. For starters I didn't know what those were as they had come after my time in college. Once I figured out what a Clicker was, I admittedly thought going after those was, well, boring. But Mike was right and I was wrong. Starting with classroom engagement turned out to be the perfect basis for establishing a large footprint in higher education. We stayed in touch as Top Hat grew and then last year the team successfully used their user base to launch a content marketplace.

While it is still early there are many positive signs about the potential for the content marketplace that remind us of other successful marketplaces we have invested in over the years such as Etsy and Science Exchange. In addition to individual professors adding content by themselves there are also new behaviors emerging and we are particularly excited about collaboratively developed content. Much work remains to be done but the company is now well funded to execute on that.

Our investment comes from the USV Opportunity Fund, which we set up in part for this type of situation where we have developed a relationship with an entrepreneur over time. Also worth noting is that Toronto continues to impress us with its quality and diversity of companies. We now have five investments there, placing Toronto third as a location in the USV portfolio after New York and San Francisco.

Shippo and the Power of Abstraction: Introducing USPS ePostage

Our portfolio company Shippo today announced that it is now providing all customers with access to the United States Postal Service ePostage program. Shippo is the first company to have received such a license from the USPS since 1999. But best of all, the benefits of this license go to Shippo's customers without them having to do anything themselves! Such is the power of abstraction provided by API companies such as Twilio, Clarifai, Dwolla, Stripe and Shippo.

Wait, "power of abstraction" what does that even mean? When you are a Shippo customer, you integrate the Shippo API into your service, and Shippo connects on your behalf to the USPS and many other shipping carriers around the world. So: you as the customer have a single integration that is always the same and Shippo behind the scenes maps the many different integrations it has to this one format. Next time when your hear a programmer say "abstracting" you now know they mean hiding complexity and changes behind a well defined, easy-to-use "interface" (just like the graphical user interface on your computer hides all the complexity of the operating system underneath it).

Abstraction is powerful because it hides both complexity and changes over time. In the case of Shippo each carrier around the world works slightly differently, but Shippo's API gives you a single interface. Shippo does all the hard work behind the scenes to hide the complexity. And change: Shippo just got a much better integration with the US Postal Service. It allows Shippo to respond to your queries faster and achieve much higher uptime (by cutting a dependency on USPS systems). Because this change happens "under the hood" the service to you as the customer gets better without you having to do anything!

This power is why going forward I believe all software will be created by composing APIs. Need communications? Use Twilio. Need payments? Use Dwolla and Stripe. Need deep learning? Use Clarifai. Need shipping? Use Shippo!

Blockstack Funding

Walter Isaacson from the Aspen Institute (and author of the Steve Jobs biography) recently wrote a post titled "The internet is broken. Starting from scratch, here's how I'd fix it." His core contention is that the Internet at present lacks native support for identity, security and payment. And while that's correct, it doesn't necesserily follow that we need to rebuild from scratch. Instead, we can and should use the capabilities of blockchain technology to augment the existing Internet (and in parts supersede it).

There are a number of different initiatives underway to do just that. One of them is led by our portfolio company Blockstack (which was called Onename, when we first invested). The Blockstack team has been building a stack of open protocols on top of the bitcoin blockchain (hence the name) to support decentralized namespaces and applications. Naming is critical to building trust, which Muneeb points out in his recent TEDx talk. And Ryan in a blog post describes how this can then be used to address Isaacson's original points. 

We are excited about the progress that the Blockstack team has made, in no small part thanks to a growing community of contributors. To help grow this effort, we have led a new round of funding. You can read more about it on the Blockstack Blog.

Continuations : What’s Next?

Earlier this week at WebSummit in Lisbon I gave a talk about my favorite topic: our need to get past the Industrial Age. The result of the US Presidential Election is the strongest evidence yet how urgently we need to come up with a more coherent view of the future, or we will have demagogues lead us back into the past. 

Code Climate

Code is everywhere these days. There is code running in your watch, your car, your thermostat. You can't make a call, pay a bill, or book a ticket without code. Code is what allows companies to create new businesses and differentiated user experiences. "Software is Eating the World," as Marc Andreessen put it.

The amount of code in systems tends to grow over time. Here is a beautiful illustration of the size of various code bases. As a first approximation, the historic growth in the lines of code is exponential, which is also confirmed by this chart of the growth of the Linux kernel. One fascinating aspect of exponential growth is how quickly the future outweighs the past. We have been writing code for about 75 years but it seems fair to assume that over the next decade we will write 4-5 times as many lines of code as we have up to now. Put differently, of all code that will exist in the year 2026, 75-80% will have been written between now and then (this is based on a 20% annual growth rate).

With that growth comes the question of whether all of that new code will work. We all encounter code that is buggy, has security holes, or doesn't run at all. An extreme recent case are the US F-35 fighter planes that cost $100 million per plane but have largely been grounded because their software isn't working.

When the question of code quality comes up, many people just shrug — they see bugs as an inevitable part of software development. Worse yet, there is a common mantra that you can have any two in software: fast time to market, low cost of development, or high quality, but never all three. This is also what people used to believe about manufacturing before the rise of techniques such as lean manufacturing and continuous improvement. As it turns out when you lead with quality in manufacturing you can in fact have all three: quality, speed and low cost. The same will be true for code, which makes assessing and managing the quality of code a key challenge for the coming years.

We are excited to be investors in New York City-based Code Climate, which provides tools to do just that. With Code Climate you can make quality improvement explicit, continuous, and ubiquitous, by incorporating source code analytics throughout the workflow of your entire development organization.

You can read more about the financing and the company's plans on Code Climate’s blog. Also: Code Climate is hiring.


Take a look around you, chances are, pretty much everything you see has been shipped, often multiple times, in order to get there. Transport of goods is a massive industry. Global parcel shipping alone is worth $300 Billion. With the shift to e-commerce parcel shipping is growing rapidly and consumer expectations for fast and cheap delivery are being set by the very largest players, led by Amazon.

Creating a compelling shipping experience for customers is hard. The industry is fragmented with many different carriers and service options. Information about shipping is difficult to find and pricing is based on multiple criteria that are not always clear. Individual carrier APIs vary widely and are often difficult to implement.

Today, we are excited to announce that USV has led the Series A financing for Shippo, a single API for all shipping needs. Shippo connects businesses to multiple shipping carriers, letting them compare rates, generate labels, and track shipments. Shippo brings access to shipping infrastructure, discounted pricing, and detailed data to businesses of any size with just a few lines of code

The complexities of the underlying infrastructure are abstracted away for developers, with Shippo taking care of the nuances and handling the edge cases. Just as Twilio made communications easy, Stripe and Dwolla payments, Shippo is doing the same for shipping. Businesses just getting started can even use Shippo without programming through a console and plugins for popular e-commerce platforms.

Shippo is building a network of customers and carriers to help optimize shipping for everyone that is already processing millions of packages to and from 230 countries. For customers Shippo provides access to more carriers and better rates and for carriers Shippo is a technology partner that allows easy programmatic integration by new and growing businesses.

We are thrilled to be backing Laura, Simon, and the Shippo team. You can read more about the financing and the company's plans on Shippo's blog. Also, Shippo is hiring.

Crypto Tokens and the Coming Age of Protocol Innovation

"Now, however, we have a new way of providing incentives for the creation of protocols and for governing their evolution. I am talking about cryptographic tokens. You can think of these like the tokens you might buy at a fair to get on a ride: different operators can have their own rides and set their own price in terms of tokens. You only need to buy tokens once (in exchange for fiat currency) and then can use them throughout the fair. With blockchains we now have a way of issuing and redeeming these tokens digitally (the underlying blockchain can be Bitcoin or Ethereum or possibly its own as in the case of Steemit).

"A for profit company can now create a new protocol and create value for itself (and its investors) by retaining some of the tokens. If the protocol becomes widely used, the value of the tokens will increase. For instance, think of a decentralized storage service (a la Amazon’s S3). Anyone can implement the storage protocol in whatever language they want to as long as they meet the protocol spec. They can then get paid in the relevant storage tokens. The original creator of the protocol will make money to the extent that it is adopted and to the degree they have retained some of the tokens (so they can sell them at a higher price later on). This is not hypothetical as there are a variety of such protocols out there, including Storj, SIA and Filecoin.

"I can’t emphasize enough how radical a change this is to the past. Historically the only way to make money from a protocol was to create software that implemented it and then try to sell this software (or more recently to host it). Since the creation of this software (e.g. web server/browser) is a separate act many of the researchers who have created some of the most successful protocols in use today have had little direct financial gain. With tokens, however, the creators of a protocol can “monetize” it directly and will in fact benefit more as others build businesses on top of that protocol.

"Given this new incentive I expect a lot of resources to be devoted to protocol innovation. That would be great as we have many missing protocols to make a decentralized data world work well. There is also a natural rate limit on how much of the wealth can be retained. Because the protocol is public (by definition) if a creator tries to retain too many tokens there is an incentive for everyone else to replicate the protocol with a new token none of which is retained.