Login
You're viewing the functional.cafe public feed.
  • Nov 17, 2026, 2:47 PM

    @JadedBlueEyes @Arusekk

    i don't agree with the framing that OIDC stands against big SSO providers, especially since the majority who've served on the board of the OpenID Foundation were representatives of big SSO providers. i think it serves a similar purpose to the big cloud providers standardising on Kubernetes:

    - shed the cost of each maintaining their own complete implementation
    - plausible deniability of corporate control through an "independent" foundation
    - plausible deniability of lock-in, even though they know in 99% of cases one of them will be chosen as the identity/cloud provider, and it's hard to switch away once users have chosen a provider
    - easier customer acquisition, since engineers are more likely to have spent their free time learning the standard if it's open, and the plausible deniablity makes them feel more comfortable using/recommending it
    - unpaid volunteers will provide tooling and language/framework bindings, and will do it with better coverage than the providers could manage themselves

    to expand on what I mean by centralisation, say you offer a service with a variety of identity providers, but the majority of your users all choose the same one. that provider then comes to you and says "hello, we found that your service doesn't comply with point 293 of our community guidelines. you have 30 days to remove the offending content, or we will block you. good day.", what do you do? you can either censor yourself or your users, or you can scramble to implement a migration flow, ask users to migrate ASAP, and then still lose a large chunk of your users because they did not do it ASAP. consider also the data harvesting potential for an identity provider that gets to see where all their users log on, and what information they share.

    1/

    💬 1🔄 1⭐ 1

Replies

  • Nov 17, 2026, 2:48 PM

    @JadedBlueEyes @Arusekk

    as i said, i don't have a problem with people using OIDC if it's a good fit for them. i see that it can be used without falling afoul of these issues, so i'm not here to tell anyone what they should or shoudn't use. if you want to know what *i* prefer, it's home-rolled auth. i hear you about the high-value secret, but i think the "don't store the password in plaintext" failure case of home-rolled auth is much easier to avoid than "don't fuck up your handling of these 4 magic numbers at any point in our 7-step process" in OIDC. and that's only considering the failure case on your side, you now also have to worry about failures and malicious behaviour from the identity provider.

    in my experience of using OIDC in the work project i discussed in the post, compared to the home-rolled auth we had before (granted some of these problems are due to the particularly poor choice of identity provider from higher ups):
    - now our service's availability is not only dependent on our own uptime, but a third party that has worse uptime than us, which we can't do anything about
    - when there is an outage, people can't switch to other tasks because all the services are unavailable, not just ours
    - since we handle national-security sensitive data, and higher ups had the great wisdom to choose a US-owned provider, we now have malicious token minting from our provider in our threat model.
    - the admins complain that the old system for issuing permissions was nicer to use
    - everyone except me is afraid to touch the OIDC code
    - our compile times have suffered
    - our testing setup is more complex

    even if you don't share my dislike of OIDC, i hope you can at least see why i think working on an OIDC library would be doing free legwork for capital interests, and that there are better things i could use my time and energy on.

    it's funny, whenever i write a blog post i try to guess what the most controversial thing will be and i'm always wrong

    2/2

    💬 1🔄 0⭐ 1
  • JadeJadedBlueEyes@tech.lgbt
    Nov 17, 2026, 9:19 PM

    @intarga @Arusekk Hah yes, it is hard to predict other people sometimes. I think you're operating in quite a different context to me - in either of the other spaces I'm very familiar with (Decentralised communications, directory single sign on), home rolled auth isn't an alternative like the context you are operating in (social login).

    From my context, OIDC is a component of a decentralised system that was co-opted. Many of the downsides you list (aside from the last three) are because ultimately social login is coupling two centralised systems, rather than connecting countless arbitrary systems like the former two areas where OIDC is used. In the end, I believe that adoption was primarily a technical decision, not a commercial one.

    💬 1🔄 1⭐ 0
  • JadeJadedBlueEyes@tech.lgbt
    Nov 17, 2026, 9:25 PM

    @intarga @Arusekk Ultimately from what I can see of the historic context, OIDC was an evolution from the status quo (proprietary SDKs that could track users, OAuth 1.0 + custom shit) that incorporated ideas from OpenID, a standard for the blogosphere that strongly pushed decentralisation, to increase privacy for users and interoperability for developers. It did not need to happen because the commercial pressures of social login are so strong - and sometimes the big tech companies don't even care to let that play its course, see the forced rollout of Apple's social login.

    💬 1🔄 0⭐ 0
  • JadeJadedBlueEyes@tech.lgbt
    Nov 17, 2026, 9:36 PM

    @intarga @Arusekk Organisations are made of people, and different people within them have different motivations. Commercially, they did not need easier acquisition of customers - they already had an incredibly strong lever - and they did not need plausibility. The social logins were extremely successful, so they got resources, and those engineers likely wanted to do the best job they could as engineers - so they focused on security, simplicity, and standardisation (because open source reduces costs makes the world better). So we got many iterations, but in the end we got a much improved Auth component, perfectly shaped for fitting into their social login SDKs. Those SDKs still exist, but they aren't used as much for social login by nature of Conway's law and because they are less private still.

    💬 1🔄 0⭐ 0
  • JadeJadedBlueEyes@tech.lgbt
    Nov 17, 2026, 9:51 PM

    @intarga @Arusekk so you have an improved auth component! They've done all the work of standardising it, of ensuring it's secure and so on and so forth. So of course other spaces are going to use it! That is why open source is powerful. When you need a way for two arbitrary independant systems to exchange auth with each other it's great! Infinity better than earlier iterations, be that SAML or LDAP quirk stacks or earlier OAuth+proprietary tangles. This system funded by big tech has gone and helped other spaces, even those diametrically opposed to big tech. And I directly use the library that you decided not to contribute to in a project that is antithetical to their dominance. So yes, contributing to that project could help big tech, and people building stuff that enforces their dominance - but it also helps those whos existence is diametrically opposed to it! That's open source! And that's why that paragraph was something that I disagreed with.

    💬 1🔄 0⭐ 0
  • 💬 1🔄 0⭐ 0
  • JadeJadedBlueEyes@tech.lgbt
    Nov 17, 2026, 10:34 PM

    @intarga @Arusekk I suppose a part of it is while I agree with a lot of your article, I disagree with both your perception of open source as broken and your evaluation of OIDC as a technology structurally predisposed to centralisation. With that in mind, I think neglecting the libraries we all depend on is how open source is in this bad state - and as the beneficiary, your employer should be contributing by paying for your time to improve the commons. I could talk about this stuff for days and days - I know that it's complicated and subject to disagreement.

    💬 1🔄 0⭐ 0
  • Nov 18, 2026, 11:47 AM

    @JadedBlueEyes @Arusekk

    i don't think open source is broken, i only intend to put more thought into what i contribute, why, and its effects.

    while i'm not going to write an OIDC library, i would still be up for working on decentralised auth, but i would insist on revisiting the assumptions baked into the protocol, for example:

    - is a third-party identity provider strictly necessary? we do have examples of identity attestation without one, i.e. ssh
    - the core/standard claims can go, especially the gender one, with prejudice

    if the identity provider is necessary for a given use case:

    - does the identity provider need to be aware of which service the user is authenticating with?
    - can we build more redundancy and/or migration between identity providers into the protocol, so that any given provider becomes less central and powerful?
    - why is it necessary? i think many services use it as an assurance that a user is a real individual, and not part of a bot swarm. not all providers serve this need equally, and their answers for it typically lie outside the protocol. this favours big providers, and i think it's part of why they see OIDC as furthering their ends, not threatening them. is there something we could do in the protocol to counteract that?

    big providers benefit from these assumptions being baked into the standard and taken as given. by revisiting these, i think i could make something that has the positive effects i want while avoiding the negative externalities.

    the only reason i haven't started on this is that i think good implementations need to be use driven, and i don't currently have a use case. the work project is a non starter since there OIDC is not a technical choice, it's mandated by the provider, that's in turn mandated by management.

    💬 0🔄 0⭐ 0