ON THE NET – Page 2 – rule 11 reader

On the ‘net: Mediocrates and LLMs

Might I suggest a name for anyone building yet another Large Language Model (LLM)? Mediocrates would be a far more honest name than the ones big tech companies choose.

Posted in ON THE NET

On the ‘net: Network Models at Packet Pushers

I’ve just started a new series on network models over at Packet Pushers. The first two installments are here:

I learned the Open Systems Interconnect (OSI) model way back in the mid-1980s, as a part of my basic networking education. Ever since then, I’ve used the OSI model in my day-to-day work as a network engineer.

First, models are not sacrosanct. A model is just a tool. If the model you are using is not working for you, feel free to modify it.

Posted in ON THE NET

On the ‘Net: The IETF at Packet Pushers

I’ve been writing a series about working within the IETF to publish a new standard over at Packet Pushers. The most recent installments are:

There are other seemingly mystical concepts in the IETF process as well—for instance, what is a “document stream,” and what is a document’s “status?”

You’re almost ready to submit a shiny new document to the IETF for consideration, right? Not quite yet—we still need to deal with mandatory sections and language.

Posted in ON THE NET

On the ‘net: Privacy and IPv6 Renumbering

One of the biggest advantages of IPv6, from a network administration perspective, is the ease of renumbering. While IPv4 networks can be renumbered using DHCP, the process of changing the address of every device on a network is always fraught with unexpected challenges. People (like me) have a habit of manually assigning printers and network attached storage devices a fixed address so they will be easy to find and use.

Posted in ON THE NET

Chatbot Attack Vectors

My monthly post is up over at Packet Pushers—

Machine learning systems “learn” from existing data pools and user interactions and are given “guardrails” by the system’s designers. Let’s look at some possible attack vectors and failure modes of these systems, specifically how training data, interaction with users, and the choice of guardrails might interact with security and privacy.

Posted in ON THE NET, SECURITY

On the ‘net: Privacy and Networking

The final three posts in my series on privacy for infrastructure engineers is up over at Packet Pushers. While privacy might not seem like a big deal to infrastructure folks, it really is an issue we should all be considering and addressing—if for no other reason than privacy and security are closely related topics. The primary “thing” you’re trying to secure when you think about networking is data—or rather, various forms of privacy.

Focusing on legal defensibility is the wrong way to look at privacy, or rather the wrong end of the stick.

What are some best practices network operators can follow to reduce their risk? The simplest way to think about best practices is to think about user rights and risks at each stage of the data lifecycle.

Posted in ON THE NET, SECURITY

On the Web: Privacy and Logging

In the last post on this topic, I concluded that IP addresses are protected information—operators should handle users’ IP addresses according to privacy best practices. But I also concluded that because IP addresses used for forwarding.

Posted in ON THE NET