Where do web standards come from?

I have spent the better part of the past six years considering where web standards come from. Before joining USV, I was in the (now retired) urban technology incubator OpenPlans, in which, among other things, we worked to further”open” technology solutions, such as open data formats and internet protocols.

The two largest standards we worked on were GTFS, the now ubiquitous format for transit information, such as routes, schedules and real time information for trains and buses; and Open311, an open protocol for reporting issues to cities (broken streetlights, potholes, etc) and asking questions (how do I dispose of paint cans?) . Each has its own origin story, which I will get to a little bit under.

Last week, I wrote about”venture capital . community funding ” (i.e., the”cycle of disruption and domination “) — and really the purpose of the talk was the association between proprietary platforms and open protocols. My point in that post was that this tension is nothing new; actually it’s a normal part of the constant cycle of the bundling and unbundling of technology, dating back, nicely , eternally .

Given the development of bitcoin and the blockchain as an application platform, it seems like we’re in the middle of another wave of energy and effort around the development and deployment of web standards. So we’re seeing a great deal of new open standards and protocols being envisioned, suggested, and developed.

The crucial question to be asking at this moment isn’t”what’s an ideal open standard”, but instead,”how can these things are, anyhow?”

Joi Ito talks about the Web as”a belief system” up to a tech, and part of how I interpret that’s that it rests on the thought of everyone just agreeing to do things kind of the exact same way. Thus, we do not all need to conduct the exact computers, use the exact same ISP, or be members of a frequent club (social network) — instead, all we will need to do is adhere to a common protocols (HTTP, SMTP, etc). Nobody possesses the protocols (by and large) — they’re more like”habits” than anything else. It works because all of us agree to do more or less the exact same thing.

So when we are looking at all these new protocols emerging (from openname, to ethereum, to whatever), the question isn’t only”is this a great idea” but instead”how might everybody agree to do so?” . It is a political and social problem as far as a technical issue. And more often than not, there’s some type of”magic” involved that’s the distinction between”cool idea” or”fine whitepaper” and”everybody does it this way”.

Here’s a crack at bucketing some of the significant strategies I have seen for bringing standards to advertise. (These aren’t necessarily mutually exclusive, and are definitely not complete — would like to find different patterns and illustrations.)

Max Bulger makes a fantastic point on Twitter I have neglected here to incorporate the traditional, formal procedures of creating web standards — though standards bodies such as the w3c and the IETF. For this post, I would like to concentrate on hacks to that conventional procedure.

1 way to bring a standard to advertise is to just force it in, with your market position as leverage. Apple has been doing so for decades, most recently with USB-C, two years ago with the first USB.

Word on the road is that USB-C was less of a consensus-driven criteria body job and more of an apple hand away. Time will tell, but now that USB-C is the interface to beat all vents in the Macbook 12, it might become the only standard for notebook and mobile/tablet ports.

The Joyful Magnet Approach

I said the GTFS standard, which is now the principal way transit agencies publish path, real-time and schedule data. GTFS came to be due to work between Google and Portland’s Tri-Met back in 2005, as their alliance to get Portland’s transit data into Google maps — they made a lightweight standard as part of that. Then, Google used”hey, do not you need your info in Google maps?” As the joyful magnet, to draw different agencies (often VERY reluctantly) into publishing their information in GTFS also.

This strategy includes elements of the Brute Force strategy — you will need to have outsized leverage / supply to pull this off.

I recall talking to people at the time who was working on these other criteria, who were pissed that Google just drifted in and helped attract GTFS to promote. But that’s precisely the point I would like to make here: a route to market is often more is more important than an ideal layout.

The Wonderful Partner Approach

Not really knowing the entire story behind Creative Commons, it appears to me that one of the massive moments for that job was their partnership using Flickr to deliver CC licensed photos to advertise — giving photographers the ability to tag with CC licenses, and giving users the ability to search by CC.

CC was a little org, but they managed to associate with a huge player to get distribution and reach.

The Make-them-an-offer-they-can’t-refuse Strategy

The incentive mechanism is the key — bitcoin and comparable cryptoequity jobs have an integrated incentive to participate.

While the Bitcoin whitepaper might have been”just another whitepaper” (future blog article had on that — aka the open standards graveyard), it had a strong built-in incentive model that attracted individuals in.

The Bottom-up Strategy

At our staff meeting on Monday, we got to discussing how oAuth was. (for those unfamiliar, oAuth is the standard protocol for enabling one program to do actions for you in another program — e.g., let this program to post to twitter for me, etc).

According to the background on Wikipedia, oAuth began with the desire to assign API access between Twitter and Magnolia, using OpenID, and from there a bunch of open web hackers took the job on. First as an informal collaboration, then as a more coordinated discussion group, and eventually as a formal proposal and working group in IETF.

From being around the people working on this at the moment, it felt like a very natural, bottom-up circumstance. Less of a theoretical top-down need and more of a simple practical solution to some point-to-point difficulty that grew into something larger.

This one is tough. Produce a standard, and work really hard to get everyone to agree that it is a fantastic idea and embrace it. It is not impossible to do so, but it is not straightforward.

In 2009, John Geraci from DIYcity (a civic hacking community at the time) composed a letter to Mayor Bloomberg indicating NYC have an open approach to its 311 system (I worked on the letter together with John, as did a number of my coworkers at the time). From that point, Philip Ashlock from OpenPlans took the lead on turning it into a real thing — working for two years with towns across the usa, technology vendors big and small, and adjoining orgs such as Code For America, to develop the specification and get it deployed. I’d say that Open311 never had the”slingshot” or”magnet” it actually needed to become huge and impactful — it had been a slow grind. But Phil specifically gets tons and a lot of credit for making it happen.

In considering this, I looked to the background of foundational web standards like HTTP and SMTP.

This is Tim Berners-Lee’s original concept for an internet hypertext system, and here’s his more formal proposal to his managers at CERN to finance first work on the job. He asked for $50k in labour and $30k in software licenses.

This is John Postel’s first proposal for SMTP (the key protocol supporting email) to the IETF networking group.

I honestly don’t understand the politics of the way either of those went from whitepaper to actual, and I’d really like to hear that story from anyone who knows.

Another fantastic story is HTML5, that was started by a splinter faction from W3C (dodging the slow process there and the focus on XHTML), then finally merged into the formal W3C procedure.

1 large takeaway I have had from working on all this is that these things take time, and if you are playing with the open standards game, you will need the ability to be individual (in addition to getting a smart go-to-market hack). It’s hard to push a standard on a startup deadline. You’ll see that many the historic players here had full-time companies (CERN, Google, universities, etc) that gave them the stability they needed and the flexibility to dedicate some time to this kind of project.

And to reiterate the main point here when looking at emerging standards and protocols, we have got to concentrate on the question”how do we get there”, and think hard about which go-to-market approach to take.

Leave a comment

Your email address will not be published. Required fields are marked *