Security is a Process, not a Purchase Anchor link

The first thing to remember before changing the software you use or buying new tools is that no tool is going to give you absolute protection from surveillance in all circumstances. Using encryption software will generally make it harder for others to read your communications or go through your computer's files. But attacks on your digital security will always seek out the weakest element of your security practices. When you use a new secure tool, you should think about how using it might affect other ways someone could target you. For example, if you decide to use a secure texting program to talk to a contact because you know that your phone might be compromised, might the fact that you're using this program at all give an adversary a clue that you are talking about private information?

Secondly, remember your threat model. You don't need to buy some expensive encrypted phone system that claims to be “NSA-proof” if your biggest threat is physical surveillance from a private investigator with no access to internet surveillance tools. Alternatively, if you are facing a government that regularly jails dissidents because they use encryption tools, it may make sense to use simpler tricks—like a set of pre-arranged codes—rather than risk leaving evidence that you use encryption software on your laptop.

Given all that, here are some questions you can ask about a tool before downloading, purchasing, or using it.

How Clear are its Creators About its Advantages and Disadvantages? Anchor link

No software or hardware is entirely secure. Creators or sellers who are honest about the limitations of their product will give you a much stronger idea of whether their application is appropriate for you.

Don't trust blanket statements that say that the code is “military-grade” or “NSA-proof”; these mean nothing and give a strong warning that the creators are overconfident or unwilling to consider the possible failings in their product.

Because attackers are always trying to discover new ways to break the security of tools, software and hardware often needs to be updated to fix new vulnerabilities. It can be a serious problem if the creators of a tool are unwilling to do this, either because they fear bad publicity, or because they have not built the infrastructure to fix problems.

You can't predict the future, but a good indicator of how toolmakers will behave in the future is their past activity. If the tool's website lists previous issues and links to regular updates and information—like specifically how long it has been since the software was last updated—you can be more confident that they will continue to provide this service in the future.

Check for Recalls and Online Criticism Anchor link

Of course, companies selling products and enthusiasts advertising their latest software can be misled, be misleading, or even outright lie. A product that was originally secure might be discovered to have terrible flaws in the future. Make sure you stay well-informed on the latest news about the tools that you use.

Products Mentioned in This Guide Anchor link

We try to ensure that the software and hardware we mention in this guide complies with the criteria we've listed above: we have made a good faith effort to only list products that have a solid grounding in what we currently know about digital security, are generally transparent about their operation (and their failings), have defenses against the possibility that the creators themselves will be compromised, and are currently maintained, with a large and technically-knowledgeable user base. We believe that they have, at the time of writing, the eye of a wide audience who is examining them for flaws, and would raise concerns to the public quickly. Please understand that we do not have the resources to examine or make independent assurances about their security, we are not endorsing these products and cannot guarantee complete security.

Registering for a Social Media Site Anchor link

• Do you want to use your real name? Some social media sites have so-called "real name policies," but these have become more lax over time. If you do not want to use your real name when registering for a social media site, do not.
• When you register, don't provide more information than is necessary. If you are concerned with hiding your identity, use a separate email address. Be aware that your IP address may be logged at registration.
• Choose a strong password and, if possible, enable two-factor authentication. Check out our guide to enabling two-factor authentication here.
• Beware of password recovery questions whose answers can be mined from your social media details. For example: “What city were you born in?” or “What is the name of your pet?” You may want to choose password recovery answers that are false. One good way to remember the answers to password recovery questions, should you choose to use false answers for added security, is to note your chosen answers in a password safe.

Change Your Privacy Settings Anchor link

Specifically, change the default settings. For example, do you want to share your posts with the public, or only with a specific group of people? Should people be able to find you using your email address or phone number? Do you want your location shared automatically?

• How to change your Facebook privacy settings

Remember, privacy settings are subject to change. Sometimes, these privacy settings get stronger and more granular; sometimes not. Be sure to pay attention to these changes closely to see if any information that was once private will be shared, or if any additional settings will allow you to take more control of your privacy.

What is threat modeling and where do I start? Anchor link

Threat modeling helps you identify threats to the things you value and determine from whom you need to protect them. When building a threat model, answer these five questions:

1. What do I want to protect?
2. Who do I want to protect it from?
3. How bad are the consequences if I fail?
4. How likely is it that I will need to protect it?
5. How much trouble am I willing to go through to try to prevent potential consequences?

Let’s take a closer look at each of these questions.

What do I want to protect?

An “asset” is something you value and want to protect. In the context of digital security, an asset is usually some kind of information. For example, your emails, contact lists, instant messages, location, and files are all possible assets. Your devices may also be assets.

Make a list of your assets: data that you keep, where it’s kept, who has access to it, and what stops others from accessing it.

Who do I want to protect it from?

To answer this question, it’s important to identify who might want to target you or your information. A person or entity that poses a threat to your assets is an “adversary.” Examples of potential adversaries are your boss, your former partner, your business competition, your government, or a hacker on a public network.

Make a list of your adversaries, or those who might want to get ahold of your assets. Your list may include individuals, a government agency, or corporations.

Depending on who your adversaries are, under some circumstances this list might be something you want to destroy after you’re done threat modeling.

How bad are the consequences if I fail?

There are many ways that an adversary can threaten your data. For example, an adversary can read your private communications as they pass through the network, or they can delete or corrupt your data.

The motives of adversaries differ widely, as do their attacks. A government trying to prevent the spread of a video showing police violence may be content to simply delete or reduce the availability of that video. In contrast, a political opponent may wish to gain access to secret content and publish that content without you knowing.

Threat modeling involves understanding how bad the consequences could be if an adversary successfully attacks one of your assets. To determine this, you should consider the capability of your adversary. For example, your mobile phone provider has access to all your phone records and thus has the capability to use that data against you. A hacker on an open Wi-Fi network can access your unencrypted communications. Your government might have stronger capabilities.

Write down what your adversary might want to do with your private data.

How likely is it that I will need to protect it?

Risk is the likelihood that a particular threat against a particular asset will actually occur. It goes hand-in-hand with capability. While your mobile phone provider has the capability to access all of your data, the risk of them posting your private data online to harm your reputation is low.

It is important to distinguish between threats and risks. While a threat is a bad thing that can happen, risk is the likelihood that the threat will occur. For instance, there is a threat that your building might collapse, but the risk of this happening is far greater in San Francisco (where earthquakes are common) than in Stockholm (where they are not).

Conducting a risk analysis is both a personal and a subjective process; not everyone has the same priorities or views threats in the same way. Many people find certain threats unacceptable no matter what the risk, because the mere presence of the threat at any likelihood is not worth the cost. In other cases, people disregard high risks because they don't view the threat as a problem.

Write down which threats you are going to take seriously, and which may be too rare or too harmless (or too difficult to combat) to worry about.

How much trouble am I willing to go through to try to prevent potential consequences?

Answering this question requires conducting the risk analysis. Not everyone has the same priorities or views threats in the same way.

For example, an attorney representing a client in a national security case would probably be willing to go to greater lengths to protect communications about that case, such as using encrypted email, than a mother who regularly emails her daughter funny cat videos.

Write down what options you have available to you to help mitigate your unique threats. Note if you have any financial constraints, technical constraints, or social constraints.

How Does End-to-End Encryption Work? Anchor link

When two people want to communicate securely (for example, Akiko and Boris) they must each generate crypto keys. Before Akiko sends a message to Boris she encrypts it to Boris's key so that only Boris can decrypt it. Then she sends the already-encrypted message across the Internet. If anyone is eavesdropping on Akiko and Boris—even if they have access to the service that Akiko is using to send this message (such as her email account)—they will only see the encrypted data and will be unable read the message. When Boris receives it, he must use his key to decrypt it into a readable message.

End-to-end encryption involves some effort, but it's the only way that users can verify the security of their communications without having to trust the platform that they're both using. Some services, such as Skype, have claimed to offer end-to-end encryption when it appears that they actually don't. For end-to-end encryption to be secure, users must be able to verify that the crypto key they're encrypting messages to belongs to the people they believe they do. If communications software doesn't have this ability built-in, then any encryption that it might be using can be intercepted by the service provider itself, for instance if a government compels it to.

You can read Freedom of the Press Foundation's whitepaper, Encryption Works for detailed instructions on using end-to-end encryption to protect instant messages and email. Be sure to check out the following SSD modules as well:

• An Introduction to Public Key Cryptography and PGP
• How to: Use OTR for Windows
• How to: Use OTR for Mac

Text Messages Anchor link

Standard text (SMS) messages do not offer end-to-end encryption. If you want to send encrypted messages on your phone, consider using encrypted instant messaging software instead of text messages.

Some end-to-end encrypted instant messaging services use their own protocol. So, for instance, users of Signal on Android and iOS can chat securely with others who use those programs. ChatSecure is a mobile app that encrypts conversations with OTR on any network that uses XMPP, which means you can choose from a range of independent instant messaging services.

Email Anchor link

Most email providers give you a way of accessing your email using a web browser, such as Firefox or Chrome. Of these providers, most of them provide support for HTTPS, or transport-layer encryption. You can tell that your email provider supports HTTPS if you log in to your webmail and the URL at the top of your browser begins with the letters HTTPS instead of HTTP (for example: https://mail.google.com).

If your email provider supports HTTPS, but does not do so by default, try replacing HTTP with HTTPS in the URL and refresh the page. If you’d like to make sure that you are always using HTTPS on sites where it is available, download the HTTPS Everywhere browser add-on for Firefox or Chrome.

Some webmail providers that use HTTPS by default include:

• Gmail
• Riseup
• Yahoo

Some webmail providers that give you the option of choosing to use HTTPS by default by selecting it in your settings. The most popular service that still does this is Hotmail.

What does transport-layer encryption do and why might you need it? HTTPS, also referred to as SSL or TLS, encrypts your communications so that it cannot be read by other people on your network. This can include the other people using the same Wi-Fi in an airport or at a café, the other people at your office or school, the administrators at your ISP, malicious hackers, governments, or law enforcement officials. Communications sent over your web browser, including the web pages that you visit and the content of your emails, blog posts, and messages, using HTTP rather than HTTPS are trivial for an attacker to intercept and read.

HTTPS is the most basic level of encryption for your web browsing that we recommend for everybody. It is as basic as putting on your seat belt when you drive.

But there are some things that HTTPS does not do. When you send email using HTTPS, your email provider still gets an unencrypted copy of your communication. Governments and law enforcement may be able to access this data with a warrant. In the United States, most email providers have a policy that says they will tell you when you have received a government request for your user data as long as they are legally allowed to do so, but these policies are strictly voluntary, and in many cases providers are legally prevented from informing their users of requests for data. Some email providers, such as Google, Yahoo, and Microsoft, publish transparency reports, detailing the number of government requests for user data they receive, which countries make the requests, and how often the company has complied by turning over data.

If your threat model includes a government or law enforcement, or you have some other reason for wanting to make sure that your email provider is not able to turn over the contents of your email communications to a third party, you may want to consider using end-to-end encryption for your email communications.

PGP (or Pretty Good Privacy) is the standard for end-to-end encryption of your email. Used correctly, it offers very strong protections for your communications. For detailed instructions on how to install and use PGP encryption for your email, see:

• How to: Use PGP for Mac OS X
• How to: Use PGP for Windows
• How to: Use PGP for Linux

Creating Strong Passwords Using Dice Anchor link

There are a few passwords that you should memorize and that need to be particularly strong. These include:

• passwords for your device
• passwords for encryption (like full-disk encryption)
• the master password, or “passphrase,” for your password manager
• your email password

One of many difficulties when people choose passwords themselves is that people aren't very good at making random, unpredictable choices. An effective way of creating a strong and memorable password is to use dice and a word list to randomly choose words. Together, these words form your “passphrase.” A "passphrase" is a type of password that is longer for added security. For disk encryption and your password manager, we recommend selecting a minimum of six words.

Why use a minimum of six words? Why use dice to pick words in a phrase randomly? The longer and more random the password, the harder it is for both computers and humans to guess. To find out why you need such a long, hard-to-guess password, here’s a video explainer.

Try making a passphrase using one of EFF's word lists.

If your computer or device gets compromised and spyware is installed, the spyware can watch you type your master password and could steal the contents of the password manager. So it's still very important to keep your computer and other devices clean of malware when using a password manager.

Syncing Your Passwords Across Multiple Devices Anchor link

Many password managers allow you to access your passwords across devices through a password-synchronizing feature. This means when you sync your password file on one device, it will update it on all of your devices.

Password managers can store your passwords “in the cloud,” meaning encrypted on a remote server. When you need your passwords, these managers will retrieve and decrypt the passwords for you automatically. Password managers that use their own servers to store or help synchronize your passwords are more convenient, but are slightly more vulnerable to attack. If your passwords are stored both on your computer and in the cloud, an attacker does not need to take over your computer to find out your passwords. (They will need to break your password manager’s passphrase though.)

If this is concerning, don't sync your passwords to the cloud and instead opt to store them on just your devices.

Keep a backup of your password database just in case. Having a backup is useful if you lose your password database in a crash, or if your device is taken away from you. Password managers usually have a way to make a backup file, or you can use your regular backup program.

Sometimes, You Will Need to Disclose Your Password Anchor link

Laws about revealing passwords differ from place to place. In some jurisdictions you may be able to legally challenge a demand for your password while in others, local laws allow the government to demand disclosure — and even imprison you on the suspicion that you may know a password or key. Threats of physical harm can be used to force someone to give up their password. Or you may find yourself in a situation, such as travelling across a border, where the authorities can delay you or seize your devices if you refuse to give up a password or unlock your device.

We have a separate guide to crossing the U.S. border that gives advice on how to deal with requests for access to devices while travelling to or from the United States. In other situations, you should think about how someone might force you or others to give up your passwords, and what the consequences would be.

Basic techniques Anchor link

Circumvention tools usually work by diverting your web traffic so it avoids the machines that do the blocking or filtering. A service that redirects your Internet connection past these blocks is sometimes called a proxy.

HTTPS is the secure version of the HTTP protocol you use to access websites. Sometimes a censor will only block the insecure (HTTP) version of a site. That means you can access the blocked site simply by entering the version of the web address that starts with HTTPS.

This is useful if the censorship you are fighting blocks individual web pages based on their contents. HTTPS stops censors from reading your web traffic, so they cannot tell what keywords are being sent, or which individual web page you are visiting.

Censors can still see the domain names of all websites you visit. So, for example, if you visit “eff.org/https-everywhere” censors can see that you are on “eff.org” but not that you are on the “https-everywhere” page.

If you suspect this type of simple blocking, try entering https:// before the domain in place of http:

Try installing EFF’s HTTPS Everywhere extension to automatically turn on HTTPS where possible.

Another way that you may be able to circumvent basic censorship techniques is by trying an alternate domain name or URL. For example, instead of visiting http://twitter.com, you might try the mobile version of the site at http://m.twitter.com. Censors that block websites or web pages work from a blacklist of banned websites, so anything that is not on that blacklist will get through. They might not know of all different versions of a particular website's name—especially if the administrators of the site know it is blocked and register more than one domain.

Encrypted proxies Anchor link

Numerous proxy tools utilize encryption to provide an additional layer of security on top of the ability to bypass filtering. The connection is encrypted so others cannot see what you are visiting. While encrypted proxies are generally more secure than plain web-based proxies, the tool provider may have information about you. They might have your name and email address in their records, for instance. That means that these tools do not provide full anonymity.

The simplest form of an encrypted web proxy is one that starts with “https”— this will use the encryption usually provided by secure websites. However, be cautious—the owners of these proxies can see the data you send to and from other secure websites. Ultrasurf and Psiphon are examples of these tools.

Tor Anchor link

Tor is open-source software designed to give you anonymity on the web. Tor Browser is a web browser built on top of the Tor anonymity network. Because of how Tor routes your web browsing traffic, it also allows you to circumvent censorship. (See our How to: Use Tor guides for Linux, macOS and Windows).

When you first start the Tor Browser, you can choose an option specifying that you are on a network that is censored:

Tor will not only bypass almost all national censorship, but, if properly configured, can also protect your identity from an adversary listening in on your country’s networks. It can, however, be slow and difficult to use.

To learn how to use Tor on a desktop machine, click here for Linux, here for macOS, or here for Windows, but please be sure to tap “Configure” instead of “Connect” in the window displayed above.

Here are some other iOS features you should think about using if you're dealing with private data: Anchor link

• iTunes has an option to backup your device onto your computer. If you choose the “Encrypt backup” option on the Summary tab of your device in iTunes, iTunes will backup more confidential information (such as Wifi passwords and email passwords), but will encrypt it all before saving it onto your computer. Be sure to keep the password you use here safe: restoring from backups is a rare event, but extra painful if you cannot remember the password to unlock the backup in an emergency.

• If you back up to Apple's iCloud, you should use a long passphrase to protect the data, and keep that passphrase safe. While Apple encrypts most data in its backups, it may be possible for the company to obtain access for law enforcement purposes since Apple also controls the keys used for iCloud encryption.

• If you turn on data protection as described above, you will also be able to delete your data on your device securely and quickly. In the Passcode settings, you can set your device to wipe all its data after ten failed attempts to guess your passphrase.

• According to Apple’s old Law Enforcement Guide, “Apple can extract certain categories of active data from passcode locked iOS devices. Specifically, the user generated active files on an iOS device that are contained in Apple’s native apps and for which the data is not encrypted using the passcode (“user generated active files”), can be extracted and provided to law enforcement on external media. Apple can perform this data extraction process on iOS devices running iOS 4 or more recent versions of iOS. Please note the only categories of user generated active files that can be provided to law enforcement, pursuant to a valid search warrant, are: SMS, photos, videos, contacts, audio recording, and call history. Apple cannot provide: email, calendar entries, or any third-party App data.”

The above information applies only to iOS devices running versions of iOS prior to 8.0.

• Now, Apple states that "On devices running iOS 8 and later versions, your personal data is placed under the protection of your passcode. For all devices running iOS 8 and later versions, Apple will not perform iOS data extractions in response to government search warrants because the files to be extracted are protected by an encryption key that is tied to the user’s passcode, which Apple does not possess."

REMEMBER: While Apple will be unable to extract data directly off a phone, if the device is set to sync with iCloud, or backup to a computer, much of the same data will indeed be accessible to law enforcement. Under most circumstances, iOS encryption is only effective when a device has been fully powered down (or freshly-rebooted, without being unlocked). Some attackers might be able to take valuable data from your device's memory when it's turned on. (They might even be able to take the data when it has just been turned off). Keep this in mind and, if possible, try to make sure your device is powered off (or rebooted and not unlocked) if you believe it's likely to be seized or stolen.

• If you are concerned about your device getting lost or stolen, you can also set up your Apple device so that it can be erased remotely, using the “Find My iPhone” feature. Note that this will allow Apple to remotely request the location of your device at any time. You should balance the benefit of deleting data if you lose control of your device, with the risk of revealing your own position. (Mobile phones transmit this information to telephone companies as a matter of course; WiFi devices like iPads and the iPod Touch do not.)

Using Signal Anchor link

In order to use Signal, the person that you are calling must have Signal installed. If you try to call or send a message to someone using the Signal app and they do not have any of the aforementioned apps installed, the app will ask if you would like to invite them via SMS, but it will not allow you to complete your call or send a message to them from inside the app.

Signal provides you with a list of other Signal users in your contacts. To do this, data representing the phone numbers in your contact list is uploaded to the Signal servers, although this data is deleted almost immediately.

How to Send an Encrypted Message Anchor link

Note that Open Whisper Systems, the makers of Signal, use other companies' infrastructure to send its users alerts when they receive a new message. It uses Google on Android, and Apple on iPhone. That means information about who is receiving messages and when they were received may leak to these companies.

To get started, tap the compose icon in the upper-right corner of the screen.

You will see a list of all the registered Signal users in your contacts.

When you tap a contact, you'll be brought to the text-messaging screen for your contact. From this screen, you can send end-to-end encrypted text, picture, or video messages.

How to Initiate an Encrypted Call

To initiate an encrypted call to a contact, select that contact and then tap on the phone icon.

At this point, Signal may ask for permission to access the microphone. Tap "OK."

Once a call is established, your call is encrypted.

How to Initiate an Encrypted Video Call

To make an encrypted video call, simply call someone as described above:

and tap the video camera icon. You may have to allow Signal to access video from your camera. This shares your video with your friend (your friend may have to do the same):

How to Start an Encrypted Group Chat

You can send an encrypted group message by tapping the compose icon in the upper-right corner of the screen (the square with a pencil pointing to the center), and then tapping the icon in the same place with three figures.

On the following screen, you'll be able to name the group and add participants to it. After adding participants, you can tap on the "+" icon in the upper right corner of the screen.

This will initiate the group chat.

If you wish to change the group name, or add or remove participants, this can be done from the group chat screen by tapping the overflow icon (the three dots in the upper-right corner of the screen) and selecting “Edit group.”

How to Verify your Contacts

At this point, you can verify the authenticity of the person you are talking with, to ensure that their encryption key wasn't tampered with or replaced with the key of someone else when your application downloaded it (a process called key verification). Verifying is a process that takes place when you are physically in the presence of the person you are talking with.

First, open the screen where you are able to message your contact, as described above. From this screen, tap the name of your contact at the top of the screen.

From the following screen, tap "Verify Safety Numbers."

You will now be brought to a screen which displays a QR code and a list of 'safety numbers.' This code will be unique for every different contact you are conversing with. Have your contact navigate to the corresponding screen for their conversation with you, so that they have a QR code displayed on their screen as well.

Back on your device, tap "Scan Code." At this point, Signal may ask for permission to access the camera. Tap "OK."

Now you will be able to use the camera to scan the QR code that is displayed on your contact's screen. Align your camera to the QR code:

Hopefully, your camera will scan the barcode and show a "Safety Numbers Verified!" dialogue, like this:

This indicates that you have verified your contact successfully. If instead your screen looks like this, something has gone wrong:

You may want to avoid discussing sensitive topics until you have verified keys with that person.

Note for power users: The screen displaying your QR code also has an icon to share your safety number in the top-right corner. In-person verification is the preferred method, but you may have already authenticated your contact using another secure application, such as PGP. Since you've already verified your contact, you can safely use the trust established in that application to verify numbers within Signal, without having to be physically in the presence of your contact. In this case you can share your safety number with that application by tapping the "share" icon, and send your contact your safety number.

Disappearing Messages

Signal has a feature called “disappearing messages” which ensures that messages will be removed from your device and the device of your contact some chosen amount of time after they are seen. To enable "disappearing messages" for a conversation, open the screen where you are able to message your contact. From this screen, tap the name of the contact at the top of the screen, then tap the slider next to "Disappearing Messages."

A slider will appear that allows you to choose how quickly messages will disappear:

After you select an option, you can tap the "<" icon on the top-left corner of the screen, and you should see information in the conversation indicating that “disappearing messages” have been enabled.

You can now send messages with the assurance that they will be removed after the chosen amount of time.