Internet Engineering Task Force 96, 20 Jul 2016

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: IETF96-REGIONALMENTORING-20160720-0845

Description

REGIONALMENTORING meeting session at IETF96
2016/07/20 0845

A

So we get started it yeah small group zone. That's fine! That's fine! We can have a small, a good discussion because we need to there's work that needs to be done anywhere. Someone as well. Okay,.

A

Everybody can just yeah stand here. A lot of regions are doing outreach. You know India Latin America. It was just over in the Africa region.

A

Doing outreach, so I think it's, let's share information, it'll be I, think good to have regional mentoring, I mean regions, doing labs or whatever other things, draft review teams and then come and then come back up because it's it's, it's bet, I think it's better, for you know to have that and then centralized there's a lot of stuff a bit that best practices. I know um region is doing a lot of things.

A

So if we can coordinate, I'll go I want to start an email list so that that we can, we can keep in touch I'm. Also thinking with the internet draft review teams.

A

um One thing that might work is for the region's to coordinate that and you guys get like two three four people together and once you've got a core group that works then come up because it's it's hot, but what happens is that we try to coordinate and somebody doesn't answer you know so, if you guys have already pre-qualified, if you know, if you know what I mean if you've pre-qualified the people who are interested- and you know I- have these five people for this working group, then you tell me- and that does that am I making sense with any of this.

A

So this is a small group. So let's just talk, but you know: let's talk on the microphone, because what Harish said is that he has some software that what if this is recorded, this is recorded. He has software he can convert. So we have meeting minutes okay, so.

B

I'm gilda company will it make sense, in fact, okay, I will be. I have some projects that may have some synergy on this. Ok, let me complemented this idea and I think we can. We can have a nice resulting in fact, okay, but the thing that the tools are ready to do it, such as to need to we need to to make them evolve, think that ready to to to be pushed or two to test them. Okay, we have developed array and make them in bulk and see how it behaves.

B

Then I think they have the two so to do this so.

A

You have some project started already in your region. Is that what you're saying let.

B

Me think is that the tools to do for the lapse, in fact, okay, so obviously I think first was we were discussing yesterday with christian, ok, the blood cold match here.

C

B

Side, okay and there's a project, an opium project called if I interrupt, which that project will be fit really on this side and I. Think that might be objective of this very.

A

Good, yes, I was at that that that that that, if the the IOT a one yeah, I was at that meeting to die. Think that's very, very good. If, after I do a couple of my things, harish is going to talk about what's going on in india, so he can tell you to three things there. Okay, so I.

D

Am reached from national internet exchange of india and what we are doing there is. We are collaborating with the different universities, and I am talking about the eye. I mean internet research, labs or so basically, the idea is to you know, based upon the internet drafts or the you know, best practices we are going to. You know create the test beds to tattoo test, how it is working in watauga, I mean shortcomings or limitations or watauga advantage out. I mean that test fine or it is not working correct.

D

You know so that we have a clear criteria through the implementation. The second is, we are also planning to have labs throughout India, so there it's one on IOT. We will discuss it further today with Karsten and on TLS on so that, because a few laps are easy to implement fewer, you know we have to discuss among the community how to implement and these kind of things we are planning, and we are also moving ahead very enthusiastically.

A

Let me and you go and then.

A

Okay, yeah so like we were just talking about, there's been conversation with six slow, sixth ish people and then also TLS. What we saw is, let's, let's not go crazy. Let's do two two things. We focus on IOT TLS, because one because I like like Irish, said there's a lot of work being done.

A

Iot in IOT already and TLS is pretty easy and I've started the connection with the some of the people in the TLS group who say: yes, we can think of things that you can test and then there's test TLS 13 server, so on. So all that work is proceeding.

A

This is this is all very experimental because there's some questions people are like. Is this going to take too much effort? We don't know, but we have to try because I think it's a very good way to bring new people into the IETF and also.

C

Please yeah fred baker when you say focus on tell us: what does that mean? Does that do? Do we bring in Steve hunters problem.

A

What if this is a separate discussion? Well, no, not not quite yet. What we mean by focus on TLS is will have tell us 1.3 server till has 1.3 client and I'm actually have been writing the Wireshark dissector for TLS 1.3 and actually tell us what is very good because we have set up in implementers email list where we can discuss it was that was in preparation for this kind of thing, so we in the lab we test it because I think there's two three prototype servers.

A

Already we implemented NSS, which is the one for Mozilla. You know we have that. So that's how we can we can. We can copy, we can look at the Wireshark dissector. We can create a set of P caps, the he's what fred is referring to is there one of the people on another project? I'm on has a problem with TLS 1.3. The problem he has is too advanced for this lab. So this is just pure. What does it work?

A

Because, because what I think- and you guys tell me if I'm wrong, if one I, if I'm reading a specification, if I don't see the packets, maybe some people can go through the specification understand perfectly the protocol flow I'm, not one such if I. I, if I see the packets it's much easier for me to understand, TLS 13 spec is 140 pages or whatever it is. You know so then I go back and forth. I say ah I see now what has happened to the certificate? Can't see it. You know ok!

A

So now I understand a little bit so I go back and read the spec, because I think this is how then, all of a sudden some clicks and you go. You know, there's a hole right here in the spec. That's, but that's my thinking, and maybe you guys tell me what you think and again, if we have regional mentors who are working with these labs working with the people, everybody in that group goes yeah you're right. There is a hole right here. So then you come up with something right to the email list.

A

Does this make sense to you at all what I'm saying Oh.

D

Actually of this kind of lab setup I, when we went to university and students were there, they are doing the engineering projects and also they are more inclined to see what is actually happening through the wash out packets rather than going through. 140 pages from first is to last one, so it will create the interest. You know lab setup and I know there are so many questions associated to it, but we all have in the community can give one or another answer and we can jointly move forward. I think.

A

Yeah, at least that's that's kind of the vision is that the labs will help people. We have the newest versions of things and and for me one of the very interesting things too is I wanted to write the wireshark dissector, because if you write the wireshark the protocol dissector, you understand the protocol. That's why I wanted to write it and I. Think people in the labs too I mean you have so many students.

A

If somebody says okay I'll write the wireshark dissector for this piece of code, you I think that's a very good opportunity to learn. I also know that people at wireshark very well and and they're very open. Oh, they can help too.

A

So now I'm going to have a irish. If you would, maybe you can talk about some of the things you guys are doing in india, so this is your presentation.

D

It's like me, the only one I can advance.

A

It yeah I know it's a it's a PDF, so so I just have to go to the next. You tell me what screen is the next one.

D

Most of things are a very experimental experimental and we have started it just to know what could be done to understand.

D

You know IETF in a in a better manner for the students and the faculty who are working in the university, so we have a project called India IETF capacity building program, in which the idea of IETF is you know main, is to how to implement the drafts and understand the whole process. So that is the main idea behind it, and this is more like the practical implementation. Next, please so during the Yokohama we had talked about, we have some ideas regarding the lab and then we jointly I mean we have talked to some universities.

D

So there is a university I e IE, st in seafood and dr. jaya sil. She is working on the NLP project. That is the to understand. This is related to a emailing list of IETF, because in a working group in there are so many emails. So when our new person is joining, he would not able to understand from where to start, because there are so many questions are linked. List is very long so just to make it easier for the new person there is a project that is like natural language processing.

D

So now work is started and it is in the process. So this is the example suppose there are so many. You know archive discussion list, and there are so many of you know threads inside it so from where to start who answered what how to understand things from the starting. So that is the main idea behind the project.

D

So, for example, I atif main discussion lists are scripted in which counts message by it's in center, and this is not particularly now. This is the version and please no process.

D

So now we have identified the problem area. That is the question. How do you decide when you get an email in your box? If you want to read it or not, so that is the main idea idea behind it: how to find out the I mean relevant emails in this, because there are so many males or so many I mean a big thread inside the emailing list. So from where to start, that is the main question and yeah.

D

So again, the secondaries when reading the email list after the fact for any thread, you get comment on comments on comments. It would be a good this I'm in deconstruct the chain says the comments were grouped by point within the topic. So that is how there are so many comments and we are just you know, constructing from where to start so.

D

This, this greater than sign is the first speaker and then response and response to the response, and then the latest comment so now this. This is how it will separate the you know different comments in that emailing list, so that we could understand from where to start, I mean a new person who is joining the any group, or this is the main idea.

D

So informational RFC spur IETF email list. Archival we have used are planning to use these RFC RFC six, seven, seven, eight and say seven, eight, four, two.

D

So main idea of ID defied the area and scope of experimentation, identify the type of experiment, ation and identify the mental group. Mainly, you know a subject, expert and identify the implementation group so that we can move further depending upon the requirements.

D

So these are the action points, identify the idea in scope and email. This analysis through the neck, natural language processing, identify the experimentation, and this is the new research and algorithm depending now. This is the very it's okay now so these are the four action points identifying the mental group and identify the implementation. Finally, the implementation, what we are doing is Nelly Jenkins with the doctor jaya, seal, inandhra, j me and the group of people are working on it, so basically mail. This is the whole new framework we are started with.

D

So this is the very wider idea about this methodology: how to define the solution domain then, depending upon the implementation, develop algorithm and finally, the implementation. This is the very garden you know generalized idea behind, so we can skip that one.

D

Basically, that is the again we are explaining the each and every step.

D

So now the future work is like how to you know solution so that we can I mean we are walking on the solution: domain, email, cuz, clustering and summarization techniques and in itm 96 we will discuss you, know more and we'll have more mentors and authors looped in while work towards the draft, so that this is the just an idea. I have represented that how to start with any mailing list and because there are several threads inside it and how we can make it better depending upon the discussion now.

D

This is the idea of IOT lab for, regarding that we are connectivity to TRG research group so that how labs can be implemented. So now in the Kaziranga University campus, we have done a memorandum of understanding with the kijiji. Onkyo is for the Kaziranga University for multiple activities. The mentor is dr. beyond, and the focus area is how to make the whole campus ipv6 enabled using the IOT devices, and this is the main idea behind ah of this experimental initiative.

D

So that is all about it. So this is the whole experimental thing.

A

So so, thank you so much Harry's. What what I think as good as is um so regions, are doing so much work. If we can collaborate each you know have have them. You know Latin America, India, cooperation, Africa, you know I mean all over I. Think that's very, very good because and I think it will be very good to say you know you guys have done this work week. We're doing this work. Do you want to help with us? That might does that, make sense to everybody too yeah, so yeah so I. Let me see.

B

Yeah so then yeah this is you go Duke, oven again well, uh but what we involved in in the indie development or experimental platform spiked with a long time to put 60 or 70 years? Okay, so the thing is I from the one side they explained that there have been in the last 60 k.

B

They have been a little revolution on the on the number of facilities, international facilities, general dresser facilities to to to to protest and to experiment, because of a first of all because of a research-oriented problem, meaning that you have to reproduce your experiment some way in a political way. That's what the first, the first idea we had over there and the second that you can let the people from whole around people around the world, in fact to test them in a you know, a variety of things. So you have you over your hardware.

B

You offer the interface you for the cell and management interface. In fact, so they can test your your things, your techy or algorithms are our test, your binaries, in different devices. The point is that there I've been working in fact on wireless devices. Then from 802 11 to IOT devices and the middle, so that's why the Hulk? What type may experience on and I think that is already available, gave us access, maybe 100 nodes. 55-200 knows it platforms and so on.

B

However, um lafourcade have their limitations and another another way of thinking is just to get um the platform in your pocket, meaning that what you have a very a device which is more which are available and you can bring to your lap. Ok on your you can you can reproduce this platform? Ok in your lap. Okay, without they come environment, will you have in very big platform, but you can start there ok and then scale that one of ideas- okay, so talking about that, ok, the the devices today, we will advise it today.

B

You can do that: ok in a small scale and manner, ok and test things by yourself, ok and then put them in the availability for Amelia universities of Mary places. So I think, is other. You have very big development sites, maybe installation killed, and then you have this small platform. You pocket things! Ok, that you have three four five notes when you can test small into a double interoperability on two sides on this, so d is not difficult in terms of hard word to do it. Ok, then, however, is available.

B

A hollow is commercially available and and and that way talking about devices such as you two devices their standard. Ok, there, you can test many things on them and when aside, as I mentioned before, km side of platforms and first two were no I hear about called much, they mean that may be ok and we will try to make synergy. Ok, I get with the other the other day feature up, so that provides what you mentioned before they.

B

It works directly on the on the packet captures so you're you have a 111 but capture from the one side to the other and says: okay, this works, and this does not work and so on. So it compares all the time that I have a sequence of things, and this project has also some if he drops, has some advantages net, because it's starting now, so they have some way. Preliminary skeleton, where things working now to show that it works and it will have some resources for those who want to compete for resources.

B

You want to extend this platform to provide new test Suites and to use the platform, so they have to resuscitate for the community case, just to compete for them. Okay, and to say the best project will win and have to work on them. So I think it's it's a way not only to work on that project, but to offer it and make it useful for the people hey.

A

What I'm thinking is, if we, you know, I, want to make sure other people have. You know we we have all our information in one place and other people have.

C

A

Maybe we can M the my my team, you know the the whole edu mentoring Directorate we're having a meeting on Thursday. So what I'm? What I think sounds good is, if you guys agree, is maybe we set up some webpages because I think some things maybe they can be test web pages to start with.

A

You know we can't make them open to the community immediately because we probably will make some mistakes start with and but at least, if we have consolidated, you know- or at least parts of the wiki I mean we start, maybe with a wiki and say this is what I'm doing. This is what I'm doing, because one of the things is like because I that there's a with labs there's a lot of security issues.

A

You know it's like who, if you open it up, who can login what, if you get a virus but but posting and yeah, of course, the same problem with posting binaries is like somebody posts about a man opposed to binary. There's a virus in it, it gets downloaded, is I ETF liable for the virus that just got you.

B

Know about that give it. The point is that uh you posted but you're the future. Either your implementation open-source mean that you're all you could provide yourself, I knew. Maybe you may compile it online, that's not for him. This is a binary. Choice. Is yours. We are just you tested, I, discover it we're going to publish it. Then the thing is: it's yours. It's yours, yeah.

A

Yeah we have to do the licensing, but that is the legal and all that framework, but I actually like that, because remote access and all that is very tricky but I think what I like a lot of those, especially the getting started with small device. Sure you download you get on started very exciting. Then you can post. This was my experience when I did that, so the next person goes: ha ok, ok, so then yeah and but I think and I think that makes for a rich environment to.

A

But so so my to do is to discuss the collaborative the web page or. However, we will do it and the legal framework so that I ETF is in no way liable if something goes wrong, yeah so yeah but I. Think yeah, ok, ok, but this is very good any any other.

E

So to think the two things so, first of all the press with made in England markets, I would say in the initial steps, because we don't know yet how's it going to evolve. So we are very flexible. So whoever comes with an idea which has support idea without many things, our support, it's just for willing for them to success, or something like that.

E

So the seven collaboration between the different efforts that are being started now, like like they'd 0 NT layers in our country's north america and yours in in india, will benefit chart on experiences and things to do, and so we can probably put them all together on a mailing list of summer somehow so they can learn from each other because not only six, an empty, it's a white patch it. So it's something that each one it's doing their way. Yeah.

C

E

Have an advice for them: we don't have any kind of support. We don't provide any kind of equipment, so we need to shape this idea somehow and if we put them all together, probably that's going to be the best way to do that and do mention bit the cosmic tool it. This is going to be useful for them, because it's a way to document the decode they are trying on the labs are implementing there and which ITF documents were used for that cold sure. So I encourage you to use it.

E

Let me know: if you want to be the we are not announcing it broadly, because we are on a slow start mode. We want to have some testers or initial users, but the toolies done reddit layer. It's online so or wants to use it it's more than welcome. So.

A

So yeah I think the email list, so we can do the collaboration, I think that's very important and we can already end in NM a website or a web page just to say this is being done. This is being done. This is being done and people can start the conversation. Whoever is starting I, think that and and that I think that's very good. um I know um what I get a lot is.

A

This is all very good, but I ETF is not a university. Our job is to do standards, and so in the end, this has to come to there's a standard and we tested it and or we implement or whatever and there's a hole right here. So it goes back to the email list for comments. Is that.

E

And there's not a benefit for the ATF it the reason that we are doing that in writing, Americans that it helps us to engage more people. So it's more on the diversity objective of the ATF and not on the work itself. That's why we are supporting sure.

A

Sure no I think outreach is very important. What is I think I was talking in another another working group. I say you know sometimes with IETF. It's like.

A

If you have a house and the first step is, is five feet high and only the people who can do a standing jump of five feet and one can come in you just you need a few smaller steps and then many more people can can come in, but it's just but as I said I'm just under saying you from my perspective, everybody says you know, people say to me: you all. You spent all this time on this that where's the output where's. Can you show me some metrics?

A

That say you know you're not just talking, and you know, but I think its long term, it's a longer term effort and it- and I think, it's very important and and I think the collaboration between different parts of the world, that's also very, very important, so yeah I think this is a shocking. We have. Consensus in this group is a very select group so and any any other things we should. There is somebody on me to echo, but not yeah yeah it needs anything else.

A

Should we have I, see I, don't know um I'm thinking, we should have meetings at every IETF just to say what is our progress for some other projects? What I do is like a monthly summary of what has happened. um Does that seem useful to say just to say you know this month. We have this and then you guys around the email list. um Yeah.

E

Christian, that's useful, but I wanted. What that you mentioned a web page or something that can collect information. I can also offer. We went through the experience of creating a tool that it's part of the tracker, but it's not official ITF like code much so.

C

E

Also help with some tool if it's required to reserve document or whatever we may need with with the labs or with anything else. We have already the connection to the database that the trucker's or we can extend or add additional applications without going through the tools, team and or the which the develop it for editing, that it's faster and and we, the small community, can create and support them.

E

The downside, it's that a it's not going to be a school, that's the one that is created an official from the teens tool. It's something that it's more product I mean yeah, you start with. Actually they will take it take that, but in the meantime, eats it works yeah.

A

No I think that's fine, because I think may I the one of the things you may have missed the thing about the internet draft review teams, because one of the problems we have is so many people sign up for internet draft review teams and out of 5. You know in the end. Maybe only one comes to the webcast so so and and we need I- have somebody who's going to help.

A

Do the code to match people already so that I don't have to do it all by looking and then and that's so he can work with, but he can work together. But the other thing is: if the region's, if you already say I, know I, have these two three people who are very serious and then you send to me and then we match them up with people from other regions. That will be very helpful because I really believe in this draft review team concept, but I think that's.

A

You know, because because I think that's an app for regional mentoring, I think that's another way to grow training. You know awareness the than the review groups and then the labs I think, let's not do more, that already yeah. That already is a lot. If we can show success, that's why I say show success throughout I'm. I keep if I get a very good contribution.

A

If I get a very good comment on the email list, as a result of some of these review teams, I like I, know, Edwin has been involved with some and then then I send I send um to could my I have iesg. You know is d, of course, the value you have to either you know I can't just do like this. I have to say what did at what happen. So if I get a very good comment, I send up to them and say: look this. You know in the this.

A

This new person commented and look. This is very nice and, and people look at that, that's because, in the end our job is to create very good standards. So so that just you know it can I I, don't mind taking a long time. Not I mean not it's not forever, but it is not like a company where you every quarter. You have to show this is not the right attitude, but a year two years Indians something has to be very productive, so I think, let's I think the focus is good.

A

I like that, and I think frankly, small team is much better. If there's a hundred people here will never get anything done so good, very, very good. It ever yeah. Thank you. Everybody yeah.

F

Sure sure where's that little sign up.

C

To be like, oh.

F

F

F

C

F

I'm also be happy to achieve I'm. Looking too.

F

A

No, that's very good because we also okay, there isn't theirs. Thinking. Let me give you my card too. There's an effort. Also going on in yeah I mean behind the scenes. It's like you always have to try to do a charter, because.