On 10/27/2017 09:46 AM, Alia Atlas wrote: > On Fri, Oct 27, 2017 at 12:41 PM, Marc Petit-Huguenin <petithug@xxxxxxx > <mailto:petithug@xxxxxxx>> wrote: > > On 10/27/2017 08:40 AM, Michael Richardson wrote: > > > > Gonzalo Camarillo <Gonzalo.Camarillo@xxxxxxxxxxxx <mailto:Gonzalo.Camarillo@xxxxxxxxxxxx>> wrote: > > > As you know, ISOC supports the IETF in several ways (in addition to > > > providing funding). Please, think about how ISOC could help support the > > > IETF in potentially new ways or how the current programs could make an > > > even larger impact, and let's have a conversation in Singapore. Thanks! > > > > My take is (in this order); > > * longer, more involved Hackathons, > > * open source reference implementations (%) > > * cheaply and easily reproduceable test bed/skaffolding for complex protocols > > * facilitating conformance testing, particularly for open source implementations. > > If we are going in that direction then I think it about time that the IETF > starts using formal methods to verify protocols, so instead of partially > checking that a protocol works (which is the best that hackathons or testing > can bring to the table), we have a guarantee that they do work. > (self-serving too, as I am working since a couple years on yet another > markdown language that does exactly that). > > > The wonderful and frustrating thing is that there isn't a single activity that > will capture all of what would be useful. Sure, but it would be a good start to have something in the ISOC mission statement about promoting protocol checking (hackathon, testing) and formal verifications of the IETF standards. > Luckily, these can be bitten off in smaller chunks. How would either of you > experiment with creating the efforts your are suggesting? > What support is needed? Why would folks be motivated to help or able to see > what incremental success looks like? > > Incidentally, do check out the Code Lounge at IETF 100 as an effort to encourage > Hackathon-style coding to continue... > > Regards, > Alia > > > > > (for instance, it's particularly difficult to create complex enough > > infrastructures to test routing protocols such as BGP4, but this also applies > > to SIDR, S/MIME, OAUTH and even some IPsec setups) > > > > Looking up, I see a theme which is really about getting from Proposed > > Standard to Internet Standard faster and in ways that engages more pieces of > > the vendor and operational communities. > > > > (%)-you may say this is self-serving, and I agree. So I'll just make my > interest explicit. > > > > -- > > Michael Richardson <mcr+IETF@xxxxxxxxxxxx > <mailto:mcr%2BIETF@xxxxxxxxxxxx>>, Sandelman Software Works > > -= IPv6 IoT consulting =- > > > > > > > > > -- > Marc Petit-Huguenin > Email: marc@xxxxxxxxxxxxxxxxxx <mailto:marc@xxxxxxxxxxxxxxxxxx> > Blog: https://marc.petit-huguenin.org <https://marc.petit-huguenin.org> > Profile: https://www.linkedin.com/in/petithug > <https://www.linkedin.com/in/petithug> > > -- Marc Petit-Huguenin Email: marc@xxxxxxxxxxxxxxxxxx Blog: https://marc.petit-huguenin.org Profile: https://www.linkedin.com/in/petithug
Attachment:
signature.asc
Description: OpenPGP digital signature