[hackathon] Open Source Software and network protocol standards

[Stuart Cheshire <[e-mail]>]

Dear Hackathon crew!

I have a question about the relationship between Open Source Software and network protocol standards.

My intuition is that in recent years Open Source Software has had a significant impact on network protocol standards, both in the availability of easily accessible open implementations contributing to the success of network protocol standards, and, in contrast, the lack of easily accessible open implementations contributing to less success for those network protocol standards.

I suspect that many in the IETF Hackathon community share this intuition, but it would be interesting to back up that intuition with concrete data.

I’d like to build a list of compelling examples.

Can you think of cases where an Open Source Software implementation clearly helped a networking protocol become successful? Was that Open Source implementation driven by the people (or companies) actively working on the protocol standard? Or was it created by an independent community following the standards development process?

Are there cases of a networking protocol that did not become successful where, in retrospect, it seems clear that the availability of an easily accessible Open Source implementation would have helped?

And, of course, counterexamples are also interesting: recent protocols that failed despite having a robust Open Source implementation, or recent protocols that succeeded despite having no Open Source implementation.

I am not the first person to think about the growing influence of Open Source on protocol development. Indeed, the existence of the IETF Hackathon itself is, in part, an acknowledgment of this new direction in protocol development. And others have written about it in the past:

2015: <https://www.ietfjournal.org/open-standards-open-source-open-loop/>
2016: <https://datatracker.ietf.org/doc/html/draft-arkko-ietf-trends-and-observations-00>
2018: <https://www.ietfjournal.org/three-years-on-open-standards-open-source-open-loop/>

Looking back now, in 2022, how have those earlier predictions played out? Do we have clear winners we can point to? Clear losers? This will help inform companies how they might want to engage with Open Source. Should companies embrace Open Source for protocol development? Or should companies go back to the old way, with open specifications and private implementations?

Please reply on the list or to me privately, and I will compile a list of responses.

Stuart Cheshire