I was reading the "Hey That's Illegal!" thread and everyone kept writing "e-mail, e-mail, e-mail".  While I understand what they are trying to do, I do not think a lot of people, including IS/IT, network engineers, programmers, etc... realize how unsecure the internet, along with e-mail is.  

I use PGP (Pretty Good Privacy) whenever I write to buddies, not because I'm breaking the law, but because I don't want a script kiddy, cracker, or alphabet group soup member reading what I have to say.  

If you want somewhat secure e-mails you should really be using PGP.  If you do download it or if you are using it now, here are some thoughts to remember:
[list]
[*]Use the strongest public key you can make.  You are putting the public key out on the net for [b]ANYONE & EVERYONE[/b] to see and use.[/*]
[*]Use a passphrase with different characters, numbers, letters (both upper & lower case), etc. Don't use a simple name or phrase.  A key can be broken, even when it is the strongest key available, by simple passphrases.[/*]
[*][b]DON"T WRITE YOUR PASSPHRASE DOWN!![/b] Don't give it to your wife, if see wants to use PGP have her make her OWN keys. No one should know you passphrase except you.[/*]
[*]Don't use PGP with people you don't know.  If you are afraid you are a "target" of the alphabet soup group, who knows if Jimbob223 that you met on the boards last month is an actual snitch or just a regular joe.  Just because PGP is encrypted in transit doesn't mean that the person on the other end won't use the information to screw you![/*]
[/list]

You can download PGP from several sites on the net, however I suggest using MIT's site because you don't have to register or give them any information on you.

[url]http://web.mit.edu/network/pgp.html[/url]

Hope this helps someone out there.
Edited because the title looked as if I was conspiring to commit a felony.

Does anyone here use PGP?

I have a key and its out on the public servers.  Never really use it.   But if I have to,  it's there.

Did they fix the (minor) security flaw regarding digitally-signed documents?  (Came about after Clinton signed the digital signature law, which mandated a sig system nobody was using.)

BTW, there's also Hushmail.com for those who don't want to screw around for a month figuring out how to use PGP.  The only problem is, Hushmail has never subjected itself to public scrutiny -- you just have to trust that they got things right, and that they aren't reading your mail themselves, and that they aren't a government front. [:)]

rofl,  it did take me a few days to figure out how the hell I was supposed to use PGP correctly.

mtnpatriot, there's no reason not to use PGP for email to strangers.  You're right that there's no guarantee that the person is any more trustworthy because of it, but there's no harm in encrypting email when you don't need to.  As long as you're just as careful with strangers when using PGP as when you're not, it can't hurt to have a secure channel.  If nothing else, if both people encrypt&sign, you have communications between you and the possible fed that you can demonstrate in court that they tried to set you up.  They'd have to make up some pretty elaborate story and provide a sacrificial lamb to claim they weren't involved.  The key doesn't even have to identify them specifically.  Possession of the decrypted email could be demonstrated by a mathematician or cryptographer as near incontrovertible proof that they are the owners of the private key.

71-Hour Achmed, I believe Hushmail did release it's source for peer review.  I haven't looked for an updated copy for 2.0, but I believe I did find a copy of the 1.x source.  This was one of the things they had to do to gain acceptance with the crypto community.  From everything I've heard about them, they're ok.

Thanks, good to know that they've been reviewed.

BTW, [u]Crypto[/u], by Steven Levy, is a very good read.

Good point

I am set up to use PGP.  While it can take a little while to understand, it is a hell of a lot easier to use than S/MIME, which requires that you get one-year certificates from only two issuers.

Unfortunately, no one else I know uses PGP, so I am frustrated in my secret desire to play a spook.

Both E-Sign and the UETA (the UETA replaces federal law if a State adopts it) do not allow an offical standard, and encourge new tech.
E-Sign knocked out some fairly successful state laws because they specifed a security floor.

The guy who developed PGP was charged with exporting a weapon when he provided it on the internet.  The charges were later dropped.  I wonder if he provided backdoors to decrypt files encrypted using his software in exchange for not going to Club Fed.  I read a good book on the topic of encryption I believe that it was titled "The Code Book".

Ok, why is it that no one else brought up key signing?  Having a public key on a keyserver makes it easy for other people to find, but if you want a secure channel, you need to be able to authenticate the key.

Are there enough people on the various gun boards who use PGP that would make it worthwhile to promote people signing each others keys?  I'd be willing to help.  I have a PGP signature from Thawte(a certification authority) and a couple of other people, one of which is signed by PRZ for those who use that metric.

No, he didn't.  Phil Zimmerman is a solid guy.  Also, that was dropped a few years ago, and it wasn't until the 7.x branch of PGP that NAI stopped releasing the source code.  Source is available for all versions prior to 7.x.  There are some groups who are extra paranoid who declare that 6.5.8 is the last trustworthy version.  Other tin foil hat conspiracy theorists believe 2.6.2/2.6.3i are the last trustworthy versions.  Those who wear the ultimate tin foil crown claim 2.3a is the only trustworthy version.  However, there's also GnuPG for those who want all the benefits of newer features but still only trust open source software.

Qwijibo

Thanks for the clarification.

I setup PGP, but none of my friends or family use it.  Unless MS puts it as the default in Outlook, it won't catch on with the masses.

One thing I just learned... When Symantec bought the rights for PGP, they were required to put a "back door" for the NSA.

You may remember when anything above 40 bit encryption was illeagle to export, then one day we could export 128 bit encryption. Well that was day that the encryption software folks worked out with the government to put a master key in their software.

This information was obtain from a reliable source. But no details were given, but just "don't use it if you really want to keep things private".

I believe that the there are still certain versions (freeware, GNU, maintained in Europe) that don't have master keys in them.

Yes, I do have a tin foil hat. Ever since my Alumininum foil hat melted.

I found that book I refered to earlier its "The Code Book: The Science of Secrecy from Ancient Egypt to Quantum Cryptography"  by Simon Singh.  If you are interested in this subject it's a good read.  

I also found a German Navy Enigma sim that can be downloaded.  I don't remeber the site but it was fun to play with.

Guzzler, can you please point to some kind of proof to back up your assertion, please?  The last I knew, PGP was open source, which means anyone and everyone can download the sources, inspect them for any "funny business" and then compile the application.  If PGP had been weakened somehow, it would have been widely publicized by now.  You can't hide something like a back door in open source.

As far as using PGP in everyday e-mails, why bother?  Do you really think your missives are so controversial and secretive that the government is going to bother to read them?  Please!  If you are involved in a group that is plotting something so dangerous and secretive, then your biggest threat would be from an insider leaking information to the authorities.  It won't be from the interception of your e-mail.

The best policy is to treat every e-mail just like a post card--that is, automatically assume it is a non-secure comm channel and don't send something you would be uncomfortable having other, unintended people read.  

Is there anything to substantiate a claim like this other than the "fact" that the NSA has had alien supercomputers since Roswell that are not subject to the laws of physics, so they're infinitely fast and have infinite storage capacity?

I guess it's just coincidence that people have been held in contempt of court until they either gave up their keys or the court determined that they would never give them up, no matter how long they were imprisoned without trial.  Seems like a lot of work when it would only take them an hour to decrypt it.



I don't recall Symantec every owning PGP.  Network Associates has had it for a while now, even though it keeps changes from PGP, Inc. to NAI to PGP Security, it hasn't been bought and sold several times.  Unless I missed something, NAI got the commercial rights to it shortly after after ViaCrypt disappeared, and took on the job of maintaining the code which was being housed at MIT.

It seems somewhat suspicious that an Additional Decryption Key was added into PGP, either when the software didn't support that feature, or when it would have been obvious that there were extra decryption keys available.  All the time people spent figuring out how to put an unsigned ADK into someone else's key, they broke the keys and messages down into hex and marked the different segments, but somehow missed a secret NSA ADK?  There are a lot of myths out there about different versions being compromised, and that's why some people will only use really old versions.  None of the claims ever seem to be substantiated with examples, even though plenty of people have been able to give examples of ways to exploit bugs in PGP, create fake keys with matching fingerprints (but different key sizes), insert ADK's in other people's keys, etc.

Why is it that these claims don't ever seem to have sources, evidence, or never even seem to rate mention by the ultra paranoid crypto weenies who can jump in an instant from the possibility of a subliminal channel in public key cryptography to an absolute proof that the NSA is monitoring everything everyone ever does with a computer 24x7?

trickshot, as of 7.0, NAI is no longer releasing the source.   In an earlier reply, I gave the breakdown of "trusted" versions based on paranoia levels.

Though, my personal feeling on the matter is that if Zimmerman is still vouching for the people who are running the PGP division and says there have been no back doors put in, I tend to believe him.  Also, if one was put in, as soon as someone proved it, NAI would suffer a horrible death.  Betraying your customers like that is something a company like Microsoft can get away with, but NAI has no hold on their customers other than providing good products.

PGP can be used successfully for encrypted email, but, uh, what are you doing to secure the box you are sending that mail from anyway?

hack your box, steal your key, monitor your keystrokes, .....

all that and more, available on your friendly neighborhood internet.  

just decide what level of paranoia is appropriate.  your comms are never secure.  adjust accordingly.

scott out
PS. oh yeah, PGP is, like, so 1995.  heh, i couldn't resist.

If I remember correctly there where two versions of the download, one for the US and one for people from other countries.  What is the difference in these two versions?  I assumed that the US version had a backdoor.

The difference in the 2.x series was the use of the RSAREF libraries vs other code.  It was a legal distinction to make it lawful for people to use PGP for personal or commercial use in the US.  Outside the US, it wasn't necessary to worry about conforming with US law.  In later versions, because of the case for exporting PGP, the source was exported in the form of a book, which was scanned and reconverted back to source code.  I think there stopped being two different versions when that was the release procedure, but US sites still had to make sure they weren't exporting it - so US people could download it from US or international sites, outside people could only get it from international sites.

qwijibo, you just answered your own question. There is no more source code for anything above 7.x. This is roughly when the "back door" was put in. And if I remember correctly, roughly when the export ban was lifted.

I honestly don't know the actual nature of the "back door", just that it does exist. I do not claim to know specifics. From what I have been able to gather, is that it still does take time for decryption process to take place. It isn't a 15 minute crunch, more like a couple of days.

PGP is great for stopping the scipt kiddies from reading you e-mail, even with those that have great talent. But the point is, if someone really wants to read your stuff, it can be done. That is all. Nothing is perfect, at least not for long.

The orginal PGP did make the code breakers shit their pants "Oh my god, we can't read this!!" for awhile. But remember the basic rule... If man can make it, man can break it (sooner or later).

One way to slow things down, is don't put your public key out to the public. Pass it just to the people you are sending messages to (preferably sneaker net).

Hey, they finally brought back the [b][u]Cryptonomicon[/u][/b] web site!
[url]http://www.cryptonomicon.com/text.html[/url]

[16 Sep 1999] USA lifts export controls on crypto
[10 Sep 2000] PGP 7.0 released

There was a year and several versions between these two events.  Details on those, and much more can be found under "latest news" at pgpi.org.



Cryptography is a branch of mathematics.  The nice thing about math is that you can prove mathematical equations and theories.  The data format is public information.  All the world's operating systems must be compromised too, since none of them will display additional fields in the message format that would be required to disclose an additional decryption key.



No, it's not perfect.  It relies on people to use it correctly.  Anyone who runs PGP under Windows clearly does not understand the security risks involved.  There are many attacks against PGP, but they're attacks against people, they're not ways to break the crypto.  The reason expert cryptographers can't break it is that they look for mathematical solutions to the problems.  They aren't going to beat you with a rubber hose until you give them the key.  They're not going to make you watch them dismember your family in front of you until you reveal your key.  A determined attacker has many means of getting the key without breaking it.

So I take it you believe the last trustworthy version is 2.3a, even though it's limited to 1024 bit RSA keys, which are still considered secure, but modern technology is catching up.  384 bit keys, the lowest it supports, have been cracked.  To complete the conspiracy theory, you'd have to claim that this was planned from the beginning, and the NSA has always been able to factor 308 digit numbers, which is why 1024 bit keys were considered secure and reasonable, but because computers were noticeably slower at the time, it was really impractical to have 4096 bit keys.  Of course, no conspiracy theory would be complete without claiming that prosecuting Phil was all a PR stunt to take a little known piece of software and thrust it into the mainstream because of its verboten status.  All the money he spent defending himself, and all the efforts he put forth to circumvent the law were all carefully scripted by the government to get people to have faith in the software.  Releasing the source code as a book, with its first amendment protection, and exporting that to europe where it was scanned, was nothing more than a government plan to make it look like he and thousands of other cryptographers were working to help all the little people, when they were all secretly working for the US government.



So you're simultaneously saying that there is a back door, so the NSA can read anything, and at the same time, the NSA has the ability to factor very large numbers.  If they have the factoring capability, they've discovered some amazing mathematical capabilities that the rest of the world doesn't know anything about.  So they have this ability, but still have to pressure companies to put a back door into software?  This has all the traits of a conspiracy theory - no evidence to support it, claims that cannot be verified, logical contradictions, and a decree that the sky is falling.

I tend to believe a bunch of independent cryptographers when it comes to whether or not to have faith in a particular piece of security software.  A fair number of these people are so paranoid that they make the people here with tin foil hats look like hardcore supporters of our government.

The real world result of people who don't know how the software works believing that there are back doors in it anyway, is that they don't bother using it.  So who wins when people send everything in cleartext?