Elektrine lite

← Feed

Heather

sphcow@mas.to

<p>Translating geek to human
 <a href="https://mas.to/tags/WomenInTech" class="mention hashtag" rel="tag">#<span>WomenInTech</span></a> <a href="https://mas.to/tags/DigitalIdentity" class="mention hashtag" rel="tag">#<span>DigitalIdentity</span></a> <a href="https://mas.to/tags/Internet" class="mention hashtag" rel="tag">#<span>Internet</span></a> <a href="https://mas.to/tags/Standards" class="mention hashtag" rel="tag">#<span>Standards</span></a> <a href="https://mas.to/tags/IDPro" class="mention hashtag" rel="tag">#<span>IDPro</span></a></p>

Posts

  • View post

    You know, normally I would celebrate a day with no meetings. But a day with no meetings means I can&amp;#39;t avoid paperwork. Women-Owned Small Business renewals, SAM renewal, various travel alert records... Meh.

  • View post

    What is OAuth becoming as software agents take on more authority? Recent OAuth WG discussions are probing proof-of-possession, browser security, agent identity, delegation chains, and whether some new authorization problems belong in OAuth at all. My latest looks at the bigger architectural questions behind the drafts. https://sphericalcowconsulting.com/2026/10/06/oauth-is-asking-what-authorization-needs-to-become/

  • View post

    Huh. They cut budgets and DOGE the staff, and then are complaining about being so far behind in PQC planning? Srsly? https://www.gao.gov/products/gao-27-108740

  • View post

    This has been a good morning. Did an hour-long ride on Zwift (MMOG) with a friend on the US East Coast and another in the UK. The fact that the tech exists reminds me that tech does have its positives. Sometimes it&amp;#39;s easy to forget that when I do my job, thinking about &amp;quot;what could possibly go wrong&amp;quot; with a standard or tech stack.

  • View post

    AI agent discovery is becoming a standards problem—and different groups are drawing the boundaries very differently. I compare the W3C AI Agent Protocol Community Group with IETF DAWN: where they overlap, where they diverge, and why scope may matter as much as protocol design. https://sphericalcowconsulting.com/2026/10/13/two-approaches-to-ai-agent-discovery/

  • View post

    Huh. My OAuth post last week had more podcast listens than written post viewers. First time that&amp;#39;s happened! I wonder what&amp;#39;s up with that?

  • View post

    Today&amp;#39;s received wisdom on a call: &amp;quot;Panic is the shorter version of deadline.&amp;quot;

  • View post

    PSA: time changes reminder This is a friendly reminder about the upcoming Daylight Saving Time (DST) changes: https://www.timeanddate.com/time/dst/events.html Most European clocks will move back one hour on Sunday, 25 October 2026. Most US clocks and some Canadian clocks will move back one hour a week later on Sunday, 1 November 2026. Between 25 October and 1 November, the time difference between the US/Canada and Europe will be one hour less than it is now. As a result, during that w...

  • View post

    On the road again. First stop, the Authenticate conference in Carlsbad, California, and an Identity Salon meeting. Then it’s straight to Dublin, Ireland, for the W3C annual TPAC meeting. And *then* a week-long vacation with 50 of my closest geeky friends in the Outer Banks, North Carolina. That was challenging to pack for, no two ways about it.

  • View post

    True or false? &amp;quot;The internet&amp;#39;s entire security model assumes a human is at the keyboard.&amp;quot; I disagree, but perhaps I&amp;#39;m missing something?

  • View post

    AI standards are proliferating, but “working on AI” doesn’t mean standards groups are solving the same problem. I looked at active work in W3C and IETF across content disclosure, AI usage preferences, bot authentication, agent permissions, Web capabilities, and agent identity. The interesting question isn’t just where work overlaps. It’s whether adjacent standards make compatible assumptions about each other. W3C, IETF, and the Emerging Shape of AI Standards:…

  • View post

    OH behind the stage: “Can you come here so we can coordinate the voice of god?”

  • View post

    This has been rattling around in my head since the Global Digital Collaboration conference: We’re getting much better at explaining why wallets should exist and why credentials should be issued. I’m not sure we’re nearly as good at explaining why a relying party should actually want to accept them. Regulation can provide the stick. Fine. But the carrot should be something concrete. https://sphericalcowconsulting.com/2026/09/15/someone-still-has-to-accept-the-credential/

  • View post

    ICYMI: An IETF Internet-Draft, W3C Working Draft, OpenID Implementer’s Draft, and FIDO Review Draft are not equivalent measures of maturity. Because apparently standards were not confusing enough without overlapping terminology. A field guide to how four major SDOs actually work: https://sphericalcowconsulting.com/2026/01/13/sdo-process-field-guide/

  • View post

    Research is not just finding sources. It is deciding which sources deserve weight, tracking what you found, verifying details, and being honest about what remains uncertain. New post: how I approach research for defensible, decision-ready papers, including Google Scholar, NGO and international organization searches, source spreadsheets, database checks, cold outreach, and careful use of AI as an organizing aid rather than an authority.

  • View post

    I have been in a few of the many, many conversations this week at #IETF regarding if and how to built new Internet protocols for AI agents, and I have so many questions. So many, but maybe also just one. Are we ultimately moving towards building a separate internet (lowercase on purpose) just for AI? Is that what all these discussions lead to? FWIW, I think that’s probably a terrible idea. I don’t think I’ve heard anything that suggests the problems we’re attempting to solve are specific to age...

  • View post

    With over 100 active Internet Drafts on various topics around AI, a third of the available side meetings, and nearly half of the HotRFC lightning talks at #IETF 126, I’m feeling more than a little overwhelmed with all the heat and so very little light.

  • View post

    I am a Disney Princess. The view outside my kitchen door this morning. They are super curious about the noises coming from inside (coffee making noises).

  • View post

    A good process protects hard disagreements. But when the process is weaponized, it stops protecting the work and starts driving good people away. My latest post looks at process abuse in standards development, why documentation matters, and how to protect participation without weakening dissent. https://sphericalcowconsulting.com/2026/07/07/when-process-stops-protecting-the-work/

  • View post

    What the AI Vendor Landscape Reveals About Fragmented Identity Systems Heather explores how the AI vendor landscape reveals deeper challenges in fragmented identity systems and enterprise security architecture. By examining how tools function across identity, signals, policy, and enforcement layers, this episode reframes AI not as a feature but as part of a broader decision-making ecosystem. Understand why distributed decision systems create complexity, how… https://sphericalcowconsulting.com/...

  • View post

    Identity Systems Don’t Make Decisions Heather examines why identity systems do not actually make decisions on their own, and how enterprise security outcomes emerge from fragmented, distributed processes. By unpacking how identity, risk, and policy tools interact, this episode reframes modern identity architecture as a complex decision-making environment. Explore how deterministic systems, integration gaps, and inconsistent data interpretation affect access control… https://sphericalcowconsult...

  • View post

    Same user. Same device. Same request. Different outcome. Add AI on top and that inconsistency compounds. If decision logic is split across identity, risk, policy, and apps, where is control? My observations in the latest blog post. I&amp;#39;d love your take on it! https://sphericalcowconsulting.com/2026/05/05/identity-systems-dont-make-decisions/

  • View post

    #introduction (re-posting) Hello, world! I&amp;#39;m a freelance consultant working to support the #digitalidentity and #internetstandards fields. I&amp;#39;m not an architect or developer; I don&amp;#39;t design the standards. I help working groups of smart people figure out what to do and offer whatever kind of support is needed for the brilliant people who write the specifications to get it done. I&amp;#39;m based on an island outside of Seattle, with my sweetheart and two Very Fuzzy kitti...

  • View post

    Right now, we’re calling them “AI browsers.” But like most things labeled AI, that name probably won’t last. Soon they’ll just be… browsers. Which makes this question worth asking now: if a browser starts acting on your behalf, does it still fit the W3C definition of a Web User Agent? Maybe. https://sphericalcowconsulting.com/2026/03/24/when-browsers-start-acting-for-you/

  • View post

    Last time I asked whether AI-enabled browsers fit the W3C definition of a Web User Agent. Short answer: mostly yes. But that raises the more interesting question: What needs to change in the Web Platform Design Principles to make that answer a comfortable yes? Some thoughts. https://sphericalcowconsulting.com/2026/03/31/ai-browsers-and-the-web-user-agent/