A cryptographer and privacy specialist Amandine Jambert, a.k.a. Cryptie, is our guest in the 18th Software Freedom Podcast episode. With her experience, Cryptie is the perfect person to tackle some basics about Security in Free Software. Join the FSFE community and support the podcast: https://my.fsfe.org/support?referrer=podcast
Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.
Transcript
From the show. Tap a time to play from there.
Welcome to the Software Freedom Podcast. This podcast is presented to you by the Free Software Foundation Europe. We are a charity that empowers users to control technology. I'm Bonnie Mering, and our guest for today's podcast is Amandine Jambert, also known as
Crypti. Crypti has been using Free Software for nearly 20 years now, and she has been a volunteer for the FSFE for exactly 10 years. Crypti used to be a cryptographer, I think that's where the name came from, and now she is a privacy specialist. Welcome, Crypti, to the podcast, I'm very glad to have you here. Thank you very much. First of all, I do have an open or free question.
What are your first thoughts when you're here to term Free Software? I didn't know what to do it, so I was like, I think the classical, what do you mean by free? Like, free, free, or free, free, so I think it was my first, you know, thing. But hopefully I knew it in French, and so someone just said to me, it's a logistical leap, and so it's like, okay, I know what it is, so it's okay, it's good, and in French
we use logistical leap, and so you already have the freedom in it, so then you just have to understand which freedom you have, and that's nice. Okay, thank you very much. And before we dive into the topic of the podcast, which will be all about cryptography and IT security, I would like to know how you got involved with the FSFE, because it's quite a long time now. In fact, I ended up on the FSCV website, and I was reading different stuff, and there
was this I love free software campaign, so it's just the best campaign on earth, and what it isn't translated into French, and like a lot of the website was in French, but not this campaign, and it was like, no, that's just not possible. And there was this link on the bottom of the website saying, hey, if you want to translate it, you have to do it, see this way, and so on, and so I sent it on email, and I said, hey, I would like to translate, so I love free software campaign, and some nice people
on Twitter, since it was GitHub, and yeah, never left. Okay, so, and how did you end up on the FSFE's website? Undesley? No idea. No idea. No idea. I was already a like a denominator for the FSFE at that time, so I don't know if I was looking for something, you know, looking for the FSFE website and ended up on the FSFE one, or if I was looking for an answer to a question, I don't know, and I just discovered that
a, we have an European version of the FSFE, like, you know, I didn't know yet, you know, how connected or not connected there, how separated there, and so on and so on, but it was like, you know, a good start, and yeah. Very nice. I'm glad you ended up on the website, and I'm very glad you joined the translator's team. Okay. So let's go over to the topic, if that's all right with you. My first question is, what does a cryptographer have to do?
Okay. So, I initially was a kind of cryptographer who designed algorithm, and so my specialty was designing protocol for privacy. So for example, one of the kind of stuff I designed was what we call a sensible signature. So it's, it's like, I will sign something, and you will, I will be able to say maybe like, bunny, you will be able to change some parts of what I signed, and the signature
will be still a correct signature of me, and everything, thing will be fine, but if there is a problem, we'll be able to say, no, since that wasn't the original money, it was a correction from bunny, and so that's one of the kind of thing I did, so that's kind of signature, but you can use this kind of thing in a more like global privacy way. So you have a lot of very nice cryptography tools that exist, and you can, for example, in choose, I don't know, I'm able to prove you that I am a certain age without giving
you my nine-post date, or I can prove you that I know, I'm a woman, I'm French, I'm any attribute, or any information about these attributes, and with a minimum information, and also a scan of things, you know. So like private information, or any information that is signed, then it can be like traceback to you. No, it's, it's even nice on the dots, it's, it's won't be, it's more that I can mathematically
prove something to someone, and it will just, you know, I will be able to prove them just the answer to a question without giving them any additional information. That's the cryptographic idea. You can apply it to anything, but that's the idea. For example, okay, so we are in a nice place here in a university, and so we have a room for this podcast. So imagine I want to prove you that I have the key for the drawer here, which is not
the case, but imagine it's the case, okay? Yeah, yeah. Then I can show you the key, okay, that won't be zero knowledge stuff. Okay. But how do I know this is the key for the room? Yeah, so that won't work. I don't want you to show me the key, because I don't want you to be able to copy it, I don't know, maybe you're, you know, I don't know, still my computer, but I don't know. But I can prove you that I have the key just telling you, okay, close your eyes, and I close the door, okay, I come back at the table, and then you can check and see that the door
is closed, okay? So I never give you any more information than to fact that I am able to close the door. Yeah. And so that's the zero knowledge. It suffices that I am able to prove you something without giving the information away. Yes. Okay, okay, I do think I understood it, we've worked on this, because this will be the base of our podcast for today. And now we have covered what cryptography is, and what is a privacy specialist?
So cryptography is, you know, it's the art of mathematics to ensure that you can, you know, ensure confidentiality or ensures that you can identify something and so on and so on. But it doesn't mean to apply it to something in particular, so you can, you know, use it for cybersecurity in general, but you can also use it to protect as a privacy of individual.
So that was initially what I did during my PhD, so I was playing to privacy, and now I widened the part of privacy. I steal, I'm looking at a little bit of crypto stuff, but I also, I'm doing like IT security stuff or looking at how, you know, how you can apply also a privacy law we have in the EU to different technology, so maybe like in any technology, you can name it, and
then the question is how do you, how do you ensure that all the privacy law are correctly applied using this technology, because the law is supposed to be neutral, and so it's about protecting the data from people and making sure that the law is correctly enforced. Yeah, yeah. So it's also something with law in it and not only with cryptography anymore. Yeah, exactly. Now I want to like, in between both words.
And how do you find it to be in between? That's nice. Yes, yeah. I mean, cryptographer are already like, if you ask a computer scientist, he will tell you that we are mathematician. If you ask mathematician, we say, oh, no way, there are computer scientists. So now it's just like, no, it's another in between innovation. And in fact, when you are thinking free software, it's also something that is in between like your, you know, how do you mean, but all the time it's like we are speaking about low, and we are speaking about ethic, and we are speaking about computer science.
So we also are, you know, like in between technical and legal software at the time. Yeah. Yeah, no, that's the true thought. I'm talking about a tea and a tea security. What does a tea security mean for you? Like if you need to make a definition, how would you define a tea security? It's very tough thing, but it's, it's all the things that will ensure the security of
information, informatics system, so and also, you know, the different computer and stuff that we have. It's very tough to phrase it in a way that every auditorium gets it, but I think it's, it's the idea. So securing the systems, the informatics system, I would say. And our devices. Yes. So it's a very broad field. Yes. And what's your favorite part about it?
Favorite part. I think the privacy part, of course, I mean, IT security and privacy security is like, you know, 80% common and 20% very different, because it depends what you want to protect. And so the part I like is everything that is, you know, privacy related. So would you say that it's possible to have a secure system that protects all the privacy
or do you think this is like impossible to achieve? I know it's a mean question, but there's no 100% security. Yeah, perfect security doesn't exist. If you can have enough security considering your situation, maybe depending on your situation, but you have to, you know, be quite clear about what your situation is. But yeah, the perfect security is either not possible or just not like lifeable.
I mean, you would need to impose on yourself some very, very tough rules and tough assumption to ensure that you already, you know, have the highest level of security. And at the same time, you just want to, you know, live your life. And it's just no more practical. So it's perfect that you should not be for must be poignant in most situations, not something that you are aiming for. You should aim for something that is okay for you, like, yeah.
That you say, okay, I'm comfortable with this level of security. And I'm okay. And I'm very that some data might be leaked and that I'm not able to protect everything. Yeah, exactly. It's what we call straight modeling. So we are every one of us should, you know, decide what is the level of, you know, privacy, the need and their comfortable ways and try to help for it. Okay. I see. And talking about security, would you say free software is more secure than non free software
because I'm quite sure you have heard this argument before. As soon as you start talking with some people that are very intimate free software and quite cool because I'm also into free software, they start arguing that free software is more secure because of the poor freedoms. And I was wondering as a privacy specialist and as a person who has a lot of experience with IT security and cryptography, would you say that this is true or not true? I think it's more complex than that.
Yes. I mean, do I believe that the most secure stuff are free software? Yes. Do I consider that some free software are really insecure? Also yes. To have something that is secure, you need it to be audited and checked. And you need to have some people who will check on it and will answer to the things that are being found and so on and so on. So you need a lot of things and a lot of conditions in order to have something that is, you know,
very secure at the end. And free software helps because it's a lot more people to do this, those audits. It's a lot to way more people to read the code and to discover that there is something, you know, I may stupid or intentionally, you know, bad in it and so on. And it also, you know, provides a system in a lot of cases where people will, you know,
will be used to answer to critics from outside, like they are used to to receive people saying, hey, you should change that or there is a bug there and so on. So they have the tool to hear when someone comes and say, hey, I have found a vulnerability on your system or on your software. And so they already have the tools to untware it and to correct it and so on. So it's more a question that free software have a lot of, you know, things that are very
good to provide highly secure software. But it doesn't mean that all software are secured because first, a lot of free software are just and will never be audited. Like nobody will look at the codes or they will look at some people will look, but that just some part to maybe add an API or maybe add a new functionality on some, but nobody will look at whether it's secure or not.
And if nobody will look at it, then probably there are a lot of pollinates, that's one thing. Another thing is, sadly, a lot of projects are under found. And so the people are very stressed and they may have a lot of things to do. And so they might, they might sometimes just say, okay, we don't have, you know, enough time or enough means in general to to check everything.
And they might, you know, have shortcuts and so not be as secure as they might be. And so it's, so yeah, so for me, the best software and the most secure one are free software. And but sadly, all free software are not as secure as we wish it to be. Yeah, that's true. No, thank you very much for the explanation and I do quite agree with this kind of complex
approach. Another thing I often hear is people argue that non-free software is more secure because people do not know how the code is written and what the code does exactly. And this is called security by obscurity. Okay, I'm a cryptographer, cryptography, security by obscurity has been like debunked for I will be nearly half a century.
At the time, we just, you know, realized that's the best crypto, best algorithm, but since whether one that's everyone will look at it and will try to either attack it or attack the proof because in some case, you have some mathematical proof beyond your, your schemes and so on and so on. And those are the cryptos that we consider as, you know, like, no, the most secure one. And it's the same with obscurity. I don't mean that you should, you know, give your private key to everyone.
It absolutely not the case. And even though you will use free software, maybe you will like not, you know, show to everyone what is your configuration, what's the parameter you are using and so on. So that is a thing, but there is two different things. It's, do you need free software and the way you apply free software and your system, do you need it to be like perfectly 100% transparent in the means that you will give every, you know, implementation data, every parameter you are using and so on.
That's two different things. You will be, in most cases, the one who, you know, brought the code you are using, you will, you know, use some software for someone, someone else. How can you be confident that this free software, this software, by default, in general? This software is first, doing what you say we say is doing. And third, that is doing it in a, you know, nice way.
Okay, if you can't look at the code, if nobody can look at the code, if nobody did not, then you're screwed like you can't know whether it's, you know, if it's really doing what it's supposed to do or not. And so, okay, so you will, you know, put all those black box in your system and then you will say that it's more secure. I disagree. I mean, I prefer to have a feeling, yes, I want, you know, be sure at 100% of every, you
know, every box I will put in my system, but I, I will be reasonably sure for most of that. And then, yes, I might not give you all my, you know, on the data of the infrastructure on which, you know, which thing I will be using and so on, but it's, it's more a question of how I will, you know, put it in my system, so it will stay. So there is very, you know, two separate things for me. Yeah, the one is like you do not need to be 100% transparent and you do also need, but
you need to be transparent on some level and that people know how to code works and let people know how they, how the software is coded and works. Yeah, because if you don't know how the software works and you don't know what it's, you know, what is doing with your data, what is doing with your, you know, your system, like, you know, that's, it's, in fact, you are implementing in your system, something
that might be a threat, like, you know, so, yeah, just free software. I think after we have played up all the prejudice against free software and IT security, I was wondering, you already mentioned that a bit, what do you think we need to make software more secure? We did do teaching when it people paid for auditing stuff. And in fact, both from an IT security point of view and from a privacy point of view,
because it's, other was saying earlier, it's 80% comment and 20% very different. And so we need more and more auditing and it means that we need, yeah, I know it's always the same, but we need money for that. But there, you know, we start to see some initiative from, you know, at the open level or in the US, you know, pushing for auditing stuff and like
auditing free software, so providing money for that. And I think if we called in an ideal world, all the biggest audits, not biggest in line of code, but in biggest in, in fact, there are the more used, liberate/software should be at least audited at least once every few years. And that will help everyone. And that is one thing. And the other is, you had some
very interesting academic work on developers. And I look specifically, you know, quite recently, about how they deal with passwords. And they study, they, you know, ask developer to write some, some codes to deal with passwords, some authentication codes. And for a part of them, they just had to do it. And for parts of them, they tell them, you have to, to ensure that it's secure.
Just a few words, like something just general. And both did, the bus group ended up doing something very different. I mean, the one where, the one who have been said, you know, to take care of security, did it in more secure way, but like very more secure way. Why the other one did something that works, but works like being very, very efficient, but not secure. I think in, from a lot of
developer, they are thinking about functionality, about insurance, I think our smooths are, you know, are working pretty well. And security is not considered as a functionality or not considered as something, you know, that's, you know, stains on mind. That's something that's, you know, they should think about. So I think it's also a question of, you know, like in the future, more and more people will, you know, will have them all those privacy and security questions in the back of their minds.
And then it will move on. So both a question of audits, you know, the two, you know, have all the, you know, so you change, you also think that the, the mindset of developers has to change a bit or it's changing. And because I see security has become such a huge topic that they think about that security by coding. I think, yeah. I hope so. I also think that we, in the EU, we are
lucky enough to have very strong privacy low. And I think that's the fact that we have those low and we are talking a lot about those low. And we have those big sanctions sometimes and so are, you know, like, you know, updating the moral software of people doing all those stuff because it's not just developers, it's groups, it's a community doing all those projects. And the fact that they are considering this is something important. And so I have a lot of hope.
For the future. Me too. Okay, my last question. You have already mentioned that I love free software campaign. Yeah. And I know that you have translated it. And so I'm quite sure, you know, that we have this every year on the 14th of February. And we use this day to say thank you to some developers or photographers or other people that contribute to free software. And I was wondering if you would like to give me the honor of saying thank you to a free software
project. Yeah, I will, you know, boss will thanks everyone that is not writing code because you're often forgotten. So I love people who write codes. It's not what I'm saying. I also, you know, thank you, all the one who, you know, tons of late stuff who ensures that you have very nice, you know, design interface and so on and so on. That's just so nice. So thank you very much. And if I have to pick one project, I will pick keep us. It's not the first time that I did it for
I love free software day. I know. But it's like, you know, I'm using it for like 15 years, I think. And it's, it's, yeah, it's something I can't, I can't survive without, you know, something to, yeah. So thank you. Thank you very much for your time, Crypti. You're welcome. And I hope to see you again in the future, maybe on the podcast. Yeah, up so. Bye. Bye. This was the software freedom
podcast. If you liked this episode, please recommend it to your friends and rate. Also subscribe to make sure you will get the next episode. This podcast is presented to you by the Free Software Foundation Europe. We are a charity that works on promoting software freedom. If you like our work, please consider supporting us with that donation. You find more information under FSFE.org/donate.
In our newest episode of the Software Freedom Podcast, Bonnie Mehring talks with Christian Nähle, Managing Director of Do-FOSS, an initiative promoting software freedom in the city of Dortmund, Germany. They discuss how...
It is Podcast Friday! For our regular episode of the Software Freedom Podcast, Bonnie sat down with Alexander Sander, FSFE's policy consultant to talk about procurement as its rules for the public sector is currently...
Fall is finally here, and so is our next episode for the Software Freedom Podcast! In this episode Bonnie Mehring meets with Dario Presutti, FSFE's project manager working on different policy and legal topics. Together,...
The Software Freedom Podcast for September is out! In this episode Bonnie Mehring meets with Carsten Agger, the FSFE’s Danish Coordinator. Together, they talk about the Free Software community in Denmark, digital...
The Software Freedom Podcast is not going on holiday this summer! In our 54th Bonnie Mehring meets with Jithendra Palepu to learn about the Android Developer verification and how it may affect the Android operating...