Recommended Posts

FORGET THE DICTIONARY If your password can be found in a dictionary, you might as well not have one. ?The worst passwords are dictionary words or a small number of insertions or changes to words that are in the dictionary,? said Mr. Kocher. Hackers will often test passwords from a dictionary or aggregated from breaches. If your password is not in that set, hackers will typically move on.

NEVER USE THE SAME PASSWORD TWICE People tend to use the same password across multiple sites, a fact hackers regularly exploit. While cracking into someone?s professional profile on LinkedIn might not have dire consequences, hackers will use that password to crack into, say, someone?s e-mail, bank, or brokerage account where more valuable financial and personal data is stored.

COME UP WITH A PASSPHRASE The longer your password, the longer it will take to crack. A password should ideally be 14 characters or more in length if you want to make it uncrackable by an attacker in less than 24 hours. Because longer passwords tend to be harder to remember, consider a passphrase, such as a favorite movie quote, song lyric, or poem, and string together only the first one or two letters of each word in the sentence.

OR JUST JAM ON YOUR KEYBOARD For sensitive accounts, Mr. Grossman says that instead of a passphrase, he will randomly jam on his keyboard, intermittently hitting the Shift and Alt keys, and copy the result into a text file which he stores on an encrypted, password-protected USB drive. ?That way, if someone puts a gun to my head and demands to know my password, I can honestly say I don?t know it.?

STORE YOUR PASSWORDS SECURELY Do not store your passwords in your in-box or on your desktop. If malware infects your computer, you?re toast. Mr. Grossman stores his password file on an encrypted USB drive for which he has a long, complex password that he has memorized. He copies and pastes those passwords into accounts so that, in the event an attacker installs keystroke logging software on his computer, they cannot record the keystrokes to his password. Mr. Kocher takes a more old-fashioned approach: He keeps password hints, not the actual passwords, on a scrap of paper in his wallet. ?I try to keep my most sensitive information off the Internet completely,? Mr. Kocher said.

A PASSWORD MANAGER? MAYBE Password-protection software lets you store all your usernames and passwords in one place. Some programs will even create strong passwords for you and automatically log you in to sites as long as you provide one master password. LastPass, SplashData and AgileBits offer password management software for Windows, Macs and mobile devices. But consider yourself warned: Mr. Kocher said he did not use the software because even with encryption, it still lived on the computer itself. ?If someone steals my computer, I?ve lost my passwords.? Mr. Grossman said he did not trust the software because he didn?t write it. Indeed, at a security conference in Amsterdam earlier this year, hackers demonstrated how easily the cryptography used by many popular mobile password managers could be cracked.

IGNORE SECURITY QUESTIONS There is a limited set of answers to questions like ?What is your favorite color?? and most answers to questions like ?What middle school did you attend?? can be found on the Internet. Hackers use that information to reset your password and take control of your account. Earlier this year, a hacker claimed he was able to crack into Mitt Romney?s Hotmail and Dropbox accounts using the name of his favorite pet. A better approach would be to enter a password hint that has nothing to do with the question itself. For example, if the security question asks for the name of the hospital in which you were born, your answer might be: ?Your favorite song lyric.?

USE DIFFERENT BROWSERS Mr. Grossman makes a point of using different Web browsers for different activities. ?Pick one browser for ?promiscuous? browsing: online forums, news sites, blogs ? anything you don?t consider important,? he said. ?When you?re online banking or checking e-mail, fire up a secondary Web browser, then shut it down.? That way, if your browser catches an infection when you accidentally stumble on an X-rated site, your bank account is not necessarily compromised. As for which browser to use for which activities, a study last year by Accuvant Labs of Web browsers ? including Mozilla Firefox, Google Chrome and Microsoft Internet Explorer ? found that Chrome was the least susceptible to attacks.

SHARE CAUTIOUSLY ?You are your e-mail address and your password,? Mr. Kocher emphasized. Whenever possible, he will not register for online accounts using his real e-mail address. Instead he will use ?throwaway? e-mail addresses, like those offered by 10minutemail.com. Users register and confirm an online account, which self-destructs 10 minutes later. Mr. Grossman said he often warned people to treat anything they typed or shared online as public record.

?At some point, you will get hacked ? it?s only a matter of time,? warned Mr. Grossman. ?If that?s unacceptable to you, don?t put it online.?

more

A good one I've found is to use a word from another language, and eastern european, asian etc... something that can be spelled out on an english keyboard, then using numbers to replace certain letters... but alas, xendrome does also make a good point

  • Like 1

The problem is that most people have too many accounts and hard to remember all the passwords. They choose the easy way is to have the same password for pretty much all accounts. The way I have all passwords different for each account, but it's still easy to remember all.

Just create one complex password (mix lower case, upper case, numbers, special character ... etc...). Then you can add the last two letters (or 3 up to you) based on the account. Pick one logic, so you won't forget.

Eg. my password is sAmpL3pa55.

So if I have account with Hotmail, and I pick the first and last letter, so my password now is sAmpL3pa55hl

For newegg, I have sAmpL3pa55ng

... and so on.

This is the easy logic to have different password for each account you own, and you still have the strong passwords.

?That way, if someone puts a gun to my head and demands to know my password, I can honestly say I don?t know it.?

I think that in this situation it would be probably better to give them the password and have the chance to live. You might end up dead anyway, but at least it's a chance.

This:

password_strength.png

If you REALLY want to test your password strength (all client-side javascript), try this one, it's amazing at analyzing password strength: https://dl.dropbox.com/u/209/zxcvbn/test/index.html

  • Like 2

OR JUST JAM ON YOUR KEYBOARD For sensitive accounts, Mr. Grossman says that instead of a passphrase, he will randomly jam on his keyboard, intermittently hitting the Shift and Alt keys, and copy the result into a text file which he stores on an encrypted, password-protected USB drive. ?That way, if someone puts a gun to my head and demands to know my password, I can honestly say I don?t know it.?

That's a great way to get yourself killed; assuming you're ever in a situation like that. Nobody in this day and age has an excuse to make a password based off a word that can be found in a dictionary are something such as "12345." People that do this honestly deserve to be hacked.

password_strength.png

This is the most widely repeated advice on passwords, and it's completely wrong. "correct horse battery staple" is about as secure as "xkcd" because - guess what - crackers use this newfangled thing called a dictionary.

The best password advice I've seen to-date is this https://www.grc.com/haystack.htm

The problem is that most people have too many accounts and hard to remember all the passwords. They choose the easy way is to have the same password for pretty much all accounts. The way I have all passwords different for each account, but it's still easy to remember all.

Just create one complex password (mix lower case, upper case, numbers, special character ... etc...). Then you can add the last two letters (or 3 up to you) based on the account. Pick one logic, so you won't forget.

Eg. my password is sAmpL3pa55.

So if I have account with Hotmail, and I pick the first and last letter, so my password now is sAmpL3pa55hl

For newegg, I have sAmpL3pa55ng

... and so on.

This is the easy logic to have different password for each account you own, and you still have the strong passwords.

the problem with that is that attackers actually take this into account as well, and if you look at the recent high profile cracks a lot of users do append the site's name (or a derivative thereof) onto a "general" password and it's no better because the pattern is trivial to figure out. now, part of it could be mitigated if everyone actually used sane password storage practices, but that seems to be quite a rarity.

  • Like 3

Not really. They still have to figure out the complex password that you create. My logic is the easy way to have diff passwords for each account. I don't recommend people to create a simple one and attach the site name after like 123newegg.

the problem with that is that attackers actually take this into account as well, and if you look at the recent high profile cracks a lot of users do append the site's name (or a derivative thereof) onto a "general" password and it's no better because the pattern is trivial to figure out. now, part of it could be mitigated if everyone actually used sane password storage practices, but that seems to be quite a rarity.

I think people are being a little paranoid here. There are over 6 billion people in the world. There is security in just the shear number of people who use computers. No one cares about an individual person unless your really important. As long as you make your passwords reasonably hard to guess there should be nothing to worry about.

I think people are being a little paranoid here. There are over 6 billion people in the world. There is security in just the shear number of people who use computers. No one cares about an individual person unless your really important. As long as you make your passwords reasonably hard to guess there should be nothing to worry about.

sure it's unlikely that someone would individually target you, but when a huge corp's databases leak and you have a weak/non-unique password, you bet the attackers will take advantage of that. they won't care who you are, but they will care about your bank account.

This is the most widely repeated advice on passwords, and it's completely wrong. "correct horse battery staple" is about as secure as "xkcd" because - guess what - crackers use this newfangled thing called a dictionary.

The math is easy. Can you explain why he's wrong?

Assume I use diceware and assume I give you the dictionary that I used to generate passwords (7776 words). Assume I tell you my password is at most 6 words long. Calculate the key space taking into account what you know:

7776^6 = 2.2E23

Compare to a 12 character "random, but easily typed character" password: a-zA-Z0-9 and all of the typical symbols: !@#$[];',. etc. Let's just call it 80 characters.

Sigma(n=1,12) 80^n = 7.0E22

So 6 random words form a dictionary that the attacker knows is an order of magnitude larger search space than 12 random characters.

My comparison assumes the 'best case' for random passwords: brute force search of the entire key space. I also assumed the worst case for diceware passwords (the attacker knows exactly which words are valid in my password, that I used only lower case letters to type them, that it's exactly 6 words long - not 4, not 7) and still diceware is better than 12 random digits by a large amount. Bumping it to 16 random characters vs 6 random words does not erase the advantage diceware ware if you allow me a minor change like "maybe I don't use spaces" or "maybe I capitalize some words".

The XKCD comic restricted the comparison space - he assumed the attacker knew the strategies in both cases and tuned his algorithm accordingly. He was also considering the common advice to start with a random word and modify it some way - that ends up in a much smaller amount of entropy than a purely random password. I tried to correct for these short comings in my example just to show that his advice still holds.

In his example and looking at his concerns (how hard is it to generate and memorize a strong password) things favour the random words approach even more. If the attacker doesn't have information about what passwords should look like and they resort to brute forcing the entire a-z0-9+symbols search space then the longer password will be stronger - that tends to favour diceware for the reason he highlighted.

The best password advice I've seen to-date is this https://www.grc.com/haystack.htm

Using your recommended site to evaluate passwords:

First I used diceware to make a 6 random word (The minimum recommended length) password:

  • Password: cash party island beset waxen coil
  • Search Space: 1.65E60
  • Massive Cracking Array Scenario: 5.23 trillion trillion trillion centuries

Note that the advantage calculated here is much higher than in my example because here he's assuming the attacker only knows that he has to search a-z+spaces, not that he can restrict his key space to combinations of a specific list of 7776 words.

Using keychain to generate a 12 character random password:

  • Password: zXn6(iy77&:r
  • Search space: 5.23E23
  • Massive Cracking Array Scenario: 1.74 centuries

Assuming compute speed doubles ever year and that 1.74 centuries starts looking pretty damn small. If you're sending 'sexy pictures' with a 12 character password to a mistress now - they'll be pretty easy to crack (1 month) in 10 years when your wife is looking to divorce for a history of cheating. What are the odds those files end up laying about on a gmail account waiting for a sopena?

In order to reach the same "durability" as I had with diceware I had to use a 30 character random character password. That seems to demonstrate exactly the point Randal was making: a few random words is just as strong and infinitely easier to memorize than random passwords or using a common strategy of mangling an uncommon word in predictable ways.

The math is easy. Can you explain why he's wrong?

Assume I use diceware and assume I give you the dictionary that I used to generate passwords (7776 words). Assume I tell you my password is at most 6 words long. Calculate the key space taking into account what you know:

7776^6 = 2.2E23

Compare to a 12 character "random, but easily typed character" password: a-zA-Z0-9 and all of the typical symbols: !@#$[];',. etc. Let's just call it 80 characters.

Sigma(n=1,12) 80^n = 7.0E22

So 6 random words form a dictionary that the attacker knows is an order of magnitude larger search space than 12 random characters.

My comparison assumes the 'best case' for random passwords: brute force search of the entire key space. I also assumed the worst case for diceware passwords (the attacker knows exactly which words are valid in my password, that I used only lower case letters to type them, that it's exactly 6 words long - not 4, not 7) and still diceware is better than 12 random digits by a large amount. Bumping it to 16 random characters vs 6 random words does not erase the advantage diceware ware if you allow me a minor change like "maybe I don't use spaces" or "maybe I capitalize some words".

The XKCD comic restricted the comparison space - he assumed the attacker knew the strategies in both cases and tuned his algorithm accordingly. He was also considering the common advice to start with a random word and modify it some way - that ends up in a much smaller amount of entropy than a purely random password. I tried to correct for these short comings in my example just to show that his advice still holds.

In his example and looking at his concerns (how hard is it to generate and memorize a strong password) things favour the random words approach even more. If the attacker doesn't have information about what passwords should look like and they resort to brute forcing the entire a-z0-9+symbols search space then the longer password will be stronger - that tends to favour diceware for the reason he highlighted.

Using your recommended site to evaluate passwords:

First I used diceware to make a 6 random word (The minimum recommended length) password:

  • Password: cash party island beset waxen coil
  • Search Space: 1.65E60
  • Massive Cracking Array Scenario: 5.23 trillion trillion trillion centuries

Note that the advantage calculated here is much higher than in my example because here he's assuming the attacker only knows that he has to search a-z+spaces, not that he can restrict his key space to combinations of a specific list of 7776 words.

Using keychain to generate a 12 character random password:

  • Password: zXn6(iy77&:r
  • Search space: 5.23E23
  • Massive Cracking Array Scenario: 1.74 centuries

Assuming compute speed doubles ever year and that 1.74 centuries starts looking pretty damn small. If you're sending 'sexy pictures' with a 12 character password to a mistress now - they'll be pretty easy to crack (1 month) in 10 years when your wife is looking to divorce for a history of cheating. What are the odds those files end up laying about on a gmail account waiting for a sopena?

In order to reach the same "durability" as I had with diceware I had to use a 30 character random character password. That seems to demonstrate exactly the point Randal was making: a few random words is just as strong and infinitely easier to memorize than random passwords or using a common strategy of mangling an uncommon word in predictable ways.

you're right, it's not completely the same, but the fact that it's using real words from a dictionary means it's not all that strong either. essentially the difference is between a 4 character password where each character can be one of ~70 choices and a 4 character password where each character can be one of ~10,000 choices (arbitrary example), while yes, it is stronger, it still takes a sane amount of time to crack. an order of magnitute, as you have calculated, is not really that much stronger in terms of passwords. The mistake that Randall made is exactly the one that you pointed out in the haystack calculator that I linked - it doesn't take into account dictionary attacks. Steve Gibson's method, on the other hand, is not vulnerable to a dictionary attack (of course, it might have other weaknesses of its own).

essentially the difference is between a 4 character password where each character can be one of ~70 choices and a 4 character password where each character can be one of ~10,000 choices (arbitrary example), while yes, it is stronger, it still takes a sane amount of time to crack.

What are you talking about - it's like you didn't even read the post.

an order of magnitute, as you have calculated, is not really that much stronger in terms of passwords.

It's the difference between a year and a decade. An order of magnitude is the difference between minimum wage and 1%. Between failing a math class and having the top score.

A 'six word' password is 30,000,000,000,000,000,000,000,000,000x times stronger than a 12 random character password when it comes to resisting brute force attacks 10x stronger when I give you the dictionary I used, the number of words, I used, and the combination method (spaces,all lower case, etc).

If you're not seeing how this is true the you're literally struggling with the concept that 23 > 22.

The mistake that Randall made is exactly the one that you pointed out in the haystack calculator that I linked - it doesn't take into account dictionary attacks.

I pointed out that issue, then did the math for you to show that even if you correct for that mistake you still don't eliminate the advantage of 6-word passwords.

Once again: show the math. Given your difficultly with inequality I can see why your hesitant to trundle into the lofty world of exponents but I believe in you! You can do it if you try!

Steve Gibson's method, on the other hand, is not vulnerable to a dictionary attack (of course, it might have other weaknesses of its own).

Diceware passwords aren't vulnerable to dictionary attacks in any meaningful sense either. I get the feeling you don't actually know what that term means or how it actually works in practice.

This topic is now closed to further replies.
  • Posts

    • With the current hardware prices Microsoft should lift the restriction. Then if you have the correct TPM then allow you to use X feature, if you don't have the correct TPM then don't but still actually let you run windows. 11. With a disclaimer during install that X features would be unavailable.
    • It's good for recycling of course. But commence inflation of a second hand RAM bubble and price gouging on DDR 4 inventory in 3... 2... 1...
    • Bypassed Windows 11 shows surprising stability on ancient, completely unsupported hardware by Sayan Sen When Windows 11 was first released, one of the most complained-about issues with the new desktop Microsoft OS was its higher system requirements, which pushed many relatively modern and powerful processors and devices onto the officially unsupported list. Thankfully, they have not been updated again for the base OS, though systems require four times the memory and storage if they want to run AI-powered apps and features. As such, Windows 11 technically runs on 4GB of memory, and there is no imposed restriction on the generation of memory it supports. Speaking of memory, prices are extremely high nowadays for hardware, especially DDR5 and DDR4 kits due to the current silicon shortage, and there are also reports of it affecting DDR2 as well, and it might only be a matter of time before even DDR1 gets affected. Before that could happen, an enthusiast took an ancient DDR1-based system and decided to try out Windows 11 on it to see how well the modern OS would fare on such hardware. The system runs an outdated graphics card interface standard based on AGP, or Advanced Graphics Port, called AGP 3.0 or AGP8x. AGP was essentially succeeded by the modern PCI Express (PCIe) bus standard. The user behind the experiment is retro hardware enthusiast Omores, who built the system around an ASRock ConRoe865PE motherboard based on Intel's i865PE chipset from way back in 2003, around the time when AGP was still in fashion. What made this board special back in the day was its unusual support for newer Core 2 Duo and even Core 2 Quad processors while still retaining older DDR1 memory support and an AGP8X graphics slot, making it an ideal bridge or link between two vastly different generations. Powering the machine was Intel's Core 2 Quad Q6600 alongside 3GB of DDR1 RAM and an ATI Radeon HD 4650 AGP graphics card, one of the final and most capable GPUs released for the aging AGP interface. While installing Windows 11 itself was relatively easy by bypassing Microsoft's hardware checks, getting the graphics card fully functional proved to be some challenge. Microsoft had quietly dropped native AGP support after the earliest releases of Windows 10, meaning newer versions of Windows no longer include the necessary Graphics Address Remapping Table (GART) drivers required for proper AGP acceleration. Without them, AGP graphics cards typically boot up, though with limited functionality, and can often throw a Code 43 error in Device Manager. To work around the limitation, Omores extracted Intel's legacy AGP440 SYS driver from an early Windows 10 release and paired it with a modified INF file so Windows 11 would correctly recognize the chipset. Following this and combined with AMD's final 64-bit Catalyst AGP drivers from 2012, the Radeon HD 4650 was able to operate with full AGP 8X acceleration intact. The result was said to be surprisingly usable for hardware that is over two decades old. Hardware-accelerated H.264 video playback worked correctly and benefited apps like Firefox, while legacy applications and games ran without major graphical issues. The system also successfully completed the 3DMark 2001 benchmark, although performance naturally lagged behind what the same hardware achieves under Windows 7, which is significantly lighter than Windows 11. There was, however, one unavoidable limitation as Microsoft's Windows 11 version 24H2 introduces a mandatory SSE4.2 CPU instruction requirement that cannot be bypassed through installer modifications or registry tweaks. Since no AGP-era processor supports SSE4.2, Windows 11 version 23H2 effectively becomes the final release capable of running on such systems. Regardless, it is still a very cool feat and quite fascinating to see just how stable Windows 11 turned out to be on such unfamiliar hardware. Source: Omores (Patreon) via O_MORES (Reddit)
    • That will only really help other players that are also responsible for creating the problem.
    • Well, it's good to know that they have found a workaround to a problem that they helped create, I guess...
  • Recent Achievements

    • Reacting Well
      NovaEdgeX earned a badge
      Reacting Well
    • Week One Done
      NovaEdgeX earned a badge
      Week One Done
    • One Year In
      BA the Curmudgeon earned a badge
      One Year In
    • Conversation Starter
      rosiecharles earned a badge
      Conversation Starter
    • First Post
      KMilenkoski1202 earned a badge
      First Post
  • Popular Contributors

    1. 1
      +primortal
      538
    2. 2
      +Edouard
      266
    3. 3
      PsYcHoKiLLa
      151
    4. 4
      Steven P.
      98
    5. 5
      macoman
      66
  • Tell a friend

    Love Neowin? Tell a friend!