With HTTPS Everywhere, Is Firefox Now the Most Secure Mobile Browser?

Peter Eckersley writes "Over at EFF, we just released a version of our HTTPS Everywhere extension for Firefox for Android. HTTPS Everywhere upgrades your insecure web requests to HTTPS on many thousands of sites, and this means that Firefox on Android with HTTPS Everywhere is now by far the most secure browser against dragnet surveillance attacks like those performed by the NSA, GCHQ, and other intelligence agencies. Android users should install the Firefox app and then add HTTPS Everywhere to it. iPhone and iPad users will unfortunately have to switch to Android to get this level of security because Apple has locked Mozilla Firefox out of their platforms."

Comments

HTTPS is secure?

[Anonymous Coward]

http://www.theguardian.com/world/2013/sep/05/nsa-gchq-encryption-codes-security

Re:HTTPS is secure?

[AmiMoJo]

It's not secure if they target you, but it massively increases the cost of monitoring you. Rather than just passively hoovering everything up they have to actually attack. If everyone does it their job gets hardware and more costly.

Re:Breaks some websites

[Anonymous Coward]

Sorry.. I have a Mac. Does the EFF have any comments about that? lol..

Get a new computer. Stop being a hipster and run OpenBSD.

Considering...

[Anonymous Coward]

People most likely don't type HTTP to begin with... I don't type http://facebook.com... just facebook.com. Google.com. slashdot.org. etc...

The S isn't just an extra S...

Re:unnecessary bloat cruft

[Rosco P. Coltrane]

If I forget to type the S, I like having the crutch.

It's a bit like automatic collision avoidance braking systems that are starting to appear on cars these days: you might say it's a huge crutch for people who are too lazy to drive properly and maintain distances, but you know what? it's a good idea I'd like to have it nonetheless, in case my concentration lapses.

Idiot

[Anonymous Coward]

It was highly illogical for you to blockquote all of that bullshit, Spock. I am unsure which one of you is the dumber one.

Re:unnecessary bloat cruft

[infogulch]

Nonsense. If you're browsing the web and following a bunch of links, you would have to long press the link to copy it, long press to paste it in the url bar, edit the url to add the S (this is mobile, so moving the cursor directly between the "p" and the ":" is non-trivial), and hit enter... for every link you follow.

You can't just click the link and edit the url after the page loaded because you've already given away the url path, url query, cookies, referrer, etc to anyone snooping your connection. And what if a site doesn't support https and instead redirects you to its' http variant? For some people they'd rather it fail to load than load insecurely. There are many reasons to use such an extension.

Re:Breaks some websites

[camperdave]

Hey allaunjsilverfox2. Some kind of malware has infected your PC and is spamming Slashdot over, and over again. Perhaps you should use a malware cleaner to clean it up. Might I recommend AVG or Spybot?

Dear MyCleanPC,

[somenickname]

I will admit that I was skeptical that a piece of software could cure my cancer, bring back my wife and prevent me from beating my daughter but, based on dozens of posts on Slashdot, I'm willing to give it a try.

Re:Dear MyCleanPC,

[stillpixel]

Wow. It could also be the end of Country music as we know it.

Re:Dear MyCleanPC,

[somenickname]

Well, nowhere in the glowing reviews did I see that it would also bring back my dog so, country music still has its place.

Depends on the threat model, doesn't it?

[fuzzyfuzzyfungus]

'Secure' isn't really something where you can just boil it into a number between 1 and 100 and call it a day. If you are worried about attackers sniffing the wire, a plugin that enforces SSL use is a major advantage. If you are worried about being hit with a zero day by the guy on the other end of the wire, it's entirely irrelevant.

Re:Depends on the threat model, doesn't it?

[Anonymous Coward]

Secure = 100;
itaday();

Re:Depends on the threat model, doesn't it?

[somenickname]

I loathe to say this but, HTTPS Everywhere is security theater. It makes your browser have a green icon where it otherwise might not but, that green icon is just an illusion of security. Considering recent revelations about the NSA, I would assume all SSL certificates are compromised. Like, literally, all of them. If the trust chain has been compromised by one party (the NSA), I would assume it compromised by all parties.

Re:Depends on the threat model, doesn't it?

[kasperd]

I would assume all SSL certificates are compromised. Like, literally, all of them.

No amount of snooping on the network would compromise a private key, which never leaves the server in the first place. Thus the only way you could possibly compromise the certificate, would be if you put an invalid public key in the certificate in the first place. Since that would be immediately obvious to any server owner paying enough attention, it is safe to assume that compromising all certificates cannot be done without being detected. If they literally compromised all of them, it would only take one single security aware server administrator to notice it.

This is why we should focus much more on protocols that are secure against passive attacks, but not against active attacks. Systematic passive attacks can be pulled off without detection. Systematic active attacks cannot. Protection against active attacks is much harder and is the reason we have the CA system. It is not that protecting against active attacks is a bad idea, it is just that it is so hard that much communication isn't protected. Opportunistic encryption with security against active attacks could be done without needing certificates. If on top of that you do perform certificate validation on the most critical sites after you have established a connection with only the opportunistic protection, then you do get protection against active attacks. In order for this to make sense, it is important that until you do perform an active attack, you cannot know if the connection has been secured against active attacks.

Both types of connections will benefit by being indistinguishable from the other to a passive adversary.

The connections with only passive security benefits because it is now easier to do encryption, and moreover being indistinguishable from the minority of connections with active security protects you from systematic active attacks. This is because systematic attacks are bound to hit protected connections once in a while, so they cannot go undetected.

The connections with the full active security also benefits because they are now attracting less attention. They will be a minor fraction of a much larger amount of encrypted connections. A passive adversary cannot know which of the connections have active security and are likely to contain the juicy stuff.

Re:Depends on the threat model, doesn't it?

[AmiMoJo]

Even if all certificates are compromised they are still worth using. Instead of passively collecting all that data the NSA/GCHQ has to perform a man-in-the-middle attack using a server that is geographically closer to you than the one they are spoofing. It costs them more time and money, limits their ability to spy on everyone all the time and requires them to maintain those servers. MITM attacks can be detected too, and in fact Chrome has made some progress on that with pinned certificates. I think there is a Firefox plugin that does something similar.

There are real and measurable benefits to using HTTPS, it's not just theatre.

Re:Depends on the threat model, doesn't it?

[jafiwam]

I loathe to say this but, HTTPS Everywhere is security theater. It makes your browser have a green icon where it otherwise might not but, that green icon is just an illusion of security. Considering recent revelations about the NSA, I would assume all SSL certificates are compromised. Like, literally, all of them. If the trust chain has been compromised by one party (the NSA), I would assume it compromised by all parties.

While this is true, chances are SLL Certificates still work well enough to keep the other nerd at the coffee shop from stealing your WoW forum account credentials.

No single person, ever, anywhere, has been able to single handedly defend themselves from the government of the place they reside. If the Government wants the account, they'll get it through twisting laws and sending the cops, not by snooping on it.

SSL protects against run of the mill crime. And, it does that well.

Re:Breaks some websites

[lister king of smeg]

Sorry can't be a hipster.. too damn old. I've been a Mac guy since errr.. like 20 years ago now.

The hipster movement is traced back to the 1940... you know back before it was popular.

NSA has the ssl keys

[hipsterdufus]

The NSA likely has keys from all the major SSL cert vendors, rendering this "spamvertisement" moot. HTTPS does not mean that you're secure from everybody. It means you've added a layer of security that will thwart MOST prying eyes, but those that really want to know what you're doing WILL know what you're doing.

What a silly thing to appear on slashdot.

Re:NSA has the ssl keys

[Jane Q. Public]

"What a silly thing to appear on slashdot."

Why is it silly? It still means that most people will be more secure, most of the time.

Re:NSA has the ssl keys

[dvdkhlng]

The NSA likely has keys from all the major SSL cert vendors, rendering this "spamvertisement" moot. HTTPS does not mean that you're secure from everybody. It means you've added a layer of security that will thwart MOST prying eyes, but those that really want to know what you're doing WILL know what you're doing.

Having the keys from multiple SSL cert vendors does not help a bit (and having the keys from many vendors isn't much better than having the keys of a single vendor). It does NOT magically allow you to decrypt SSL traffic from servers whose host key was signed against that cert vendor's certificate!

To decrypt traffic of multiple SSL websites requires you to obtain the private part of the SSL host keys from all the web-servers themselves. Note that web server host keys are signed via signing requests that do not contain a copy of the private key, so even when the cert vendors (CAs) are hacked, you cannot directly listen in on SSL communication. When the servers implement Perfect Forward Secrecy, then even obtaining a copy of the server's host key won't help as each connection uses a temporary key that's exchanged via Diffie Hellman Key Exchange, a method that generates a key shared between two hosts, that (somewhat counter-intuitively) cannot be deduced by sniffing the traffic between those two participants.

What you can still do is to set up a MITM attack: you set up your own intermediate server with its own host key and sign your host key(s) using one of the SSL vendor's certs that you obtained. Then you redirect all traffic to the servers that interest you via your server (i.e. proxying all SSL connections) and then obviously in the process you obtain the cleartext of all SSL sessions running via your server.

However, the MITM attack is much more difficult to deploy and scale than simple monitoring and recording IP data. Also skilled users will easily detect the MITM attack, as the host key's public part of the servers in question will suddenly change. There are firefox extensions to check for these signs of a MITM. Even SSL Everywhere has a checker built in (via the SSL Observatory). Or try Certificate Patrol.

who are we fooling?

[marienf]

> this means that Firefox on Android with HTTPS Everywhere is now by far the most secure browser
> against dragnet surveillance attacks like those performed by the NSA, GCHQ, and other intelligence agencies.

While I certainly think it is a good idea to encrypt traffic, this statement is highly misleading or naive: Since the CA
system is *flawd by design* and every one of those "authorities" in the long list of built-in CA inside
your browser can, by negligence or choice, supply any of these and other agencies with a valid certificate for
*any hostname in the world*, initiatives like these protect your privacy only from your local sysadmin/ISP, and also
do nothing against traffic analysis.

Should a US person/company trust that "China Internet Network Information Center" isn't going to create a cert for a
US bank or company to perform a MITM attach with? Should a Chinese company trust "Wells Fargo" not to?
Should the Greeks trust "TÜRKTRUST Bilgi letiim ve Biliim Güvenlii Hizmetleri A.. (c) Aralk 2007", or the
Turks "Hellenic Academic and Research Institutions Cert. Authority"? What on earth makes you think ALL of these
companies can resists pressures to misbehave? Yet all of them are built-in to your browser and "you" trust them.

Just go to any (Cloudflare, Akamai..)-accelerated site using https and check out the certificate used to see how that works:
They are issued certificates for the customer domains they accelerate, and hence have access to all the traffic.
In essence, they do exactly what a man-in-the-middle attack would do, except on a much grander scale (and with the collusion
of the actual domain holders). The agencies can carry out such attacks from within the ISP's, and your browser would still show "green".

The Cert validation in the browsers leads to a *dangerous false sense of security* at most. This is crypto, a weakest-link business
if ever there was one, folks. It's not ALL, or SOME that need to fail in order for PKI to fail, it's ANY of them.

Surely, we can do better than that: We should get rid of all centralised security illusions. Why aren't we signing contents using our PGP
keys that at least make multiple signers possible and habitual, and, and this is the essential difference, IMHO: That *you* have made a
conscious decision to trust or mistrust, to a certain degree, by reviewing a web of trust, as in informed consent as opposed to blind paternalism
of massivly built-in, pretrusted certificates by distant companies you really have no clue about.

WKR,
-f

Re:who are we fooling?

[dvdkhlng]

> this means that Firefox on Android with HTTPS Everywhere is now by far the most secure browser > against dragnet surveillance attacks like those performed by the NSA, GCHQ, and other intelligence agencies.

While I certainly think it is a good idea to encrypt traffic, this statement is highly misleading or naive: Since the CA system is *flawd by design* and every one of those "authorities" in the long list of built-in CA inside your browser can, by negligence or choice, supply any of these and other agencies with a valid certificate for *any hostname in the world*, initiatives like these protect your privacy only from your local sysadmin/ISP, and also do nothing against traffic analysis.

Should a US person/company trust that "China Internet Network Information Center" isn't going to create a cert for a US bank or company to perform a MITM attach with? Should a Chinese company trust "Wells Fargo" not to? Should the Greeks trust "TÜRKTRUST Bilgi letiim ve Biliim Güvenlii Hizmetleri A.. (c) Aralk 2007", or the Turks "Hellenic Academic and Research Institutions Cert. Authority"? What on earth makes you think ALL of these companies can resists pressures to misbehave? Yet all of them are built-in to your browser and "you" trust them.

[..]

The Cert validation in the browsers leads to a *dangerous false sense of security* at most. This is crypto, a weakest-link business [..]

You suggest that MITM attacks on SSL are as bad as someone sniffing on unencrypted traffic. It is not! MITM attacks are active attacks and are much more invasive to carry out. That's not all: in principle all these MITM attacks can be detected: the host key of the Man In The Middle will differ from the host key of the original server (though your browser will accept the differing host key when it is signed by a rogue CA).

It is pretty dangerous for an adversary to carry out MITM attacks on a large scale, as sooner or later, this is going to be detected. The SSL Everywhere extension for example can (optionally) collect information for and check with the SSL Observatory to detect differing certificates that indicate MITM attacks.

There's also the Certificate Patrol Firefox Extension that persistently remembers certificates and warns when certificates changed for no apparent reason.

Re:who are we fooling?

[dvdkhlng]

You suggest that MITM attacks on SSL are as bad as someone sniffing on unencrypted traffic. It is not! MITM attacks are active attacks and are much more invasive to carry out.

Is "false security" better or worse than "no security"?

I really don't understand why everybody tries to reduce these encryption problems on the "false security" vs. "no security" dichotomy. No this is not about false security. This is about security against undetectable passive attackers vs. detectable active attackers. The amount of data a detectable active attacker is able to collect about my person are many orders of magnitude smaller than the amount of data a passive attacker is able to obtain. The active attacker will also only be able to obtain data from the point of time I was chosen as a target. The passive attacker will be able to go back in time and look at my communication (probably many years) before I became interesting enough to be deemed a target.

This is why implementing SSL, even if no protection at all against MITM existed, is much much better than no SSL at all.

Misleading much...

[raist21]

So what's with the uber pro-Firefox and Android spiel?

According to the web-site you can get the plug-in for Chrome as well. Albeit beta, but still.
And if that's the case, you can just install Chrome on your Apple device, it's in the itunes store, and install the plugin for it instead.

Re:Breaks some websites

[narcc]

AVG or Spybot? That's crazy. They pale in comparison to the obviously superior MyCleanPC!

Why do you think he takes every opportunity to spread the good news?

Re:What the frak?

[curty]

For the first time I see the death of /. coming around the bend.

You must be new here.

Not perfect, but it's a start...

[bogd]

So basically all this does is to force HTTPS requests instead of HTTP? (took me a while to find out - gotta love the fact that the "clever technology" link on their site, instead of going to a description of the actual technology, goes to... xkcd?! :) )

I see a few problems with this approach:
1)Not all content is provided over both HTTP and HTTPS. For multiple reasons, one being performance. Which leads us to the second problem...
2)A HTTPS session incurs a significant overhead for encryption. Which may be no problem for someone like Google. But for someone hosting his/her own (moderately successful) website on a small server, it might just overload said server.
3)Quite possibly the biggest problem with HTTPS is the fact that users have been trained over many years to just click "accept/install certificate" on self-signed certs. Not knowing that if you do this you are no longer secure.

And the more we keep forcing HTTPS, the more webmasters will use self-signed certs. Not many people want to go through the hassle of obtaining (and maintaining!) a valid SSL certificate for every single website they run, even if that cert is free. Which will only exacerbate the problem...

Re:Not perfect, but it's a start...

[heypete]

I see a few problems with this approach:

1)Not all content is provided over both HTTP and HTTPS. For multiple reasons, one being performance. Which leads us to the second problem...

True, which is why HTTPS Everywhere only enables HTTPS on sites that support it (they are specifically whitelisted by the extension devs).

2)A HTTPS session incurs a significant overhead for encryption. Which may be no problem for someone like Google. But for someone hosting his/her own (moderately successful) website on a small server, it might just overload said server.

While HTTPS does incur some overhead, it's surprisingly small for modern servers. Google, for example, was able to add SSL/TLS to all Gmail connections with no new hardware, no additional servers, and SSL/TLS accounts for only about 1% of their CPU time (see here for details).

Pretty much any server will reach other bottlenecks before the slight overhead of SSL/TLS becomes an issue. Using Perfect Forward Secrecy is important for security and using DHE-based ciphers do incur a moderate overhead compared to non-DHE ciphers (a factor of about 3). Using ECDHE instead makes the increase in overhead only about 15% rather than 300%. See here for details.

3)Quite possibly the biggest problem with HTTPS is the fact that users have been trained over many years to just click "accept/install certificate" on self-signed certs. Not knowing that if you do this you are no longer secure.

And the more we keep forcing HTTPS, the more webmasters will use self-signed certs. Not many people want to go through the hassle of obtaining (and maintaining!) a valid SSL certificate for every single website they run, even if that cert is free. Which will only exacerbate the problem...

[citation needed] Getting a domain-validated SSL cert from publicly-available CAs is the work of a few minutes and, as you point out, often available for free or very low cost. Many hosts will automate the generation of a private key and CSR, making the process one of copy-paste for the customer. Other hosts handle the entire process of generating a private key, getting it signed by a CA, and configuring things correctly.

Sure, some sites use self-signed certs, but these are usually for personal or internal corporate purposes and not for the general public. The scary warnings in browsers aren't likely to go away anytime soon, so I doubt that any webmaster of a website meant for public use is going to be using self-signed certs (other than those catering to specific, tech-savvy audiences).

Re:What the frak?

[Cenan]

You're confusing User with Customer. We're the users, advertisers are the customers.

"Can remove spam" and "will remove spam" are not the same thing. They absolutely, trivially could prevent this kind of spam - but why would they? Nobody at Dice cares! In all the years I've come here I've never seen the admins do anything remotely resembling administration of their site.

Another comment on a thread, no matter how trivial or spammy, enforces the illusion of a site that is still alive. This illusion is used to make the search indexers think that something of relevance is going on at the site, and rate it higher, which in turn exposes yet more of Dice's advertising. The key to proper SEO is novel content, the trick is that the content doesn't have to be at all relevant or even coherent, it just has to be new and Google will swallow it like a junior at the prom with the star jock.

Whenever some moron codes up a new incarnation of retardo-bot and launches it in a flurry of masturbation, a whole host of /. users will flock around and comment on the spam. It's a viscous cycle and Dice has no incentive to stop it.

Short story even shorter: Dice runs the site. Dice profits from not removing spam posts.

Re:Breaks some websites

[alex67500]

Q: Why did the hipster burn his mouth?
A: He ate pizza before it was cool!

Why doesn't Slashdot do SSL?

[crow]

This doesn't work with Slashdot. At least if you put in a https, it redirects, so they have it set up; they just don't use it. You would think that a technology site would be up on current technology.