I have not counted the IETF’s I have attended; I only know the first RFC on which I’m listed as a co-author was published in 2000, so this must be close to 20 years of interacting with the IETF community, and I’m pretty certain I’ve attended at least two meetings a year across that time, and three meetings a year in most of those years. Across that time, there has never been a time when I have not been told, at least once, “the IETF is broken.” And there has not been a single time I cannot remember agreeing with the sentiment.
So, how is the IETF broken? The trend that bothers me the most right now is the gold rush syndrome. A new technology is brought into the IETF, and if it looks like it might somehow be “important,” there is a “land rush” as people stake out new drafts, find use cases, find corner cases, and work to develop drafts and communities around those drafts. This generally results in a sort of ossification process, where there are clear insiders and outsiders, an entirely new vocabulary is developed, and the drafts fly so fast and furious there is almost no time to read them all. There are many problematic parts of this process. For instance, there is often a feeling that “this is important, no need to get the details right,” or “if you don’t understand, butt out of the conversation.”
A particularly troubling aspect of this is the wide desire to “be famous,” to chair a working group, to get your name on a draft, and ultimately on an RFC. This eventually becomes all important, carrying all practical considerations before it. The old ethos of “build small and flexible, code it, and let it grow where needed” is almost always lost in the shuffle of producing tens of drafts. Companies pay by the draft, or only pay for travel if you have a draft—both of which have a tendency to destroy the value of the community itself, and the way the community functions.
So that’s what broken. What’s right?
One night I was walking back from dinner with a couple of friends—Gonzolos and Joe—and I ran into Stewart Bryant in the hotel lobby. Soon enough, Paul Mockapetris joined the conversation. At some point, Dave Oran, Ignas B, and George Swallow joined the conversation. There are few places in the world you can get some collection of folks who had a hand in the creation of technologies like DNS, psuedowires, MPLS/TE, SMTP, IS-IS, IP fast reroute, and probably a dozen other technologies, standing around talking about “the good old days,” or where to go for dinner. Across this week, I’ve chatted with Tony Li, Tony P, Jeff T, Alvaro Retana, Russ Housley, Fred Baker, Alia Atlas, and… more than I can remember.
If there is one that is striking about all of these people, it is that they are all more interested in solving problems than taking credit. They all live by the old IETF mantra: “it is amazing what can get done when no-one cares who gets the credit.” None of them are obsessed with getting their names on drafts, or with inventing something new that will change the world. They see problems, they develop solutions; that is all.
This, then, is what is right about the IETF. People who care about the challenges users have with networks, and have spent their lives finding solutions. So people are what’s wrong with the IETF, and people are also what’s right. The point?
You can choose to participate in the IETF. In fact, I hope to see you at a future meeting. But if you choose to participate, be a part of the solution, rather than a part of the problem. Be someone who looks on the land rush with skepticism, who doesn’t care about getting their name on a draft, who just wants to help solve a problem that has been fairly explained and defined to the community. Don’t be afraid to work on small things, and to insist that solutions be small and well scoped, even if that means your name is not put up in lights.
Even better advice: carry this into all the communities in which you live in your life. We live in an age that values name recognition far too much, that worries too much about being left out of the latest gold rush, that worries too much about our “rightly deserved” fifteen minutes of fame. This goes far beyond network engineering, the ethos of the “old way” in the IETF. It’s a lesson we can all take away from this little community of engineers who have worked so hard across the years to build something on which we all rely every day—to the very formats of the packets which carry this screed to your computer screen, your email box, or however else you are reading it.
It’s been a busy few weeks in cybercrime news, justifying updates to a couple of cases we’ve been following closely at KrebsOnSecurity. In Ukraine, the alleged ringleader of the Avalanche malware spam botnet was arrested after eluding authorities in the wake of a global cybercrime crackdown there in 2016. @Krebs on Security
Reflection amplification is a technique that allows cyber attackers to both magnify the amount of malicious traffic they can generate, and obfuscate the sources of that attack traffic. For the past five years, this combination has been irresistible to attackers, and for good reason. —Carlos Morales @Arbor
For years, we’ve been pioneering the use of DNS to enforce security. We recognized that DNS was often a blind spot for organizations and that using DNS to enforce security was both practical and effective. Why? Because DNS isn’t optional. It’s foundational to how the internet works and and is used by every single device that connects to the network. If you’re considering using DNS for security, it’s important to understand the facts so you can combat the fiction. —Kevin Rollinson @Cisco
Attackers have seized on a relatively new method for executing distributed denial-of-service (DDoS) attacks of unprecedented disruptive power, using it to launch record-breaking DDoS assaults over the past week. Now evidence suggests this novel attack method is fueling digital shakedowns in which victims are asked to pay a ransom to call off crippling cyberattacks. @Krebs on Security
Amazon continues to improve the Consumer IoT space, introducing more — and smarter — WIFI-enabled gadgets. Good for us, but even better for Amazon: They get both our money and our data. —Jean-Louis Gassée @Monday Note
In December, Edward Snowden unveiled a new app called Haven, which turns your Android phone into a monitoring device to detect and record activity. Snowden has pitched Haven as a safeguard against so-called evil maid attacks, in which an adversary snoops through your digital devices or installs trackers on them when you’re not around. In interviews, Snowden was clear that one group he thought might use Haven was victims of intimate partner violence, who could use it to record abusers tampering with their devices. —Karen Levy @Slate
It’s my rather controversial view that the edge will, over the longer term (10+ years), eclipse what we call the cloud: the giant centralized hyper-scale data centers, which offer a progressive set of abstractions as a service for running applications and storing data. —Chetan Venkatesh
In earlier blog posts (Looks Like We’re Upgrading Again! Dual-Rate 40G/100G BiDi Transceiver and 40/100G QSFP BiDi Transceiver’s Backward Compatibility With 40G BiDi), we introduced the dual-rate 40/100G QSFP BiDi transceiver and described how Cisco uniquely offers 40G capability and backward compatibility. Let’s review why the QSFP+ 40G BiDi was such a big hit in the first place when it was released back in 2013, and how the BiDi value proposition still makes plenty of sense. —Pat Chou @Cisco
A large number of banks, credit unions and other financial institutions just pushed customers onto new e-banking platforms that asked them to reset their account passwords by entering a username plus some other static identifier — such as the first six digits of their Social Security number (SSN), or a mix of partial SSN, date of birth and surname. Here’s a closer look at what may be going on (spoiler: small, regional banks and credit unions have grown far too reliant on the whims of just a few major online banking platform providers). —Krebs on Security
I was asked by a reader to add categories and links for videos; I actually added three new categories, one for short videos, another for long videos, and a third for written posts. You can find these under the bottom menu item on the left. I am having a problem with the menu not showing up correctly, so I move the resources under the third menu item, as well.
Finally, I added a new archive page, which shows you all the posts in the “left” category across the three years this blog has been “in production.” I couldn’t figure out how to narrow things down so pictures and other stuff are not included, so there is more on the page than needed right now, but it’s a start.
On this episode of the Network Collective, we’re chatting with Miguel Villareal and Scott Wheeler about cloud connectivity.
On this episode of the History of Networking, we talk to Alia Atlas about the history of fast reroute and Maximally Redundant Trees (MRTs). Remember to send in your suggestions for guests and technologies.