Germany Builds Encrypted, Identity-Confirmed Email

jfruhlinger writes "Looking to solve the problems of spam, phishing, and unconfirmed email identities, Germany is betting very, very big. The country will pass a law this month creating 'De-mail,' a service in which all messages will be encrypted and digitally signed so they cannot be intercepted or modified in transit. Businesses and individuals wanting to send or receive De-mail messages will have to prove their real-world identity and associate that with a new De-mail address from a government-approved service provider. The service will be enabled by a new law that the government expects will be in force by the end of this month. It will allow service providers to charge for sending messages if they wish. The service is voluntary, but will it give the government too much control?"

Comments

No end-to-end encryption though

[Anonymous Coward]

As far as I've read, they decrypt messages in the middle "to check the messages for viruses".

Re:No end-to-end encryption though

[moonbender]

Yup. Sounds like a bad joke right? A new messaging standard, incompatible with everything else, that doesn't even do end-to-end encryption! It's pathetic. It purports to solve problems that are already pretty much solved -- spam, reliable delivery -- while not solving all the difficult ones and introducing new dangers for the customers, like missing a "registered email". Oh, and you'll be charged per mail! The worst outcome would be if people ended up using it, but at this point I'm guessing it'll be a huge dud; some government entities will support it, as will a few corporations, but that's it.

Re:No end-to-end encryption though

[Anonymous Coward]

As a native German, I can confirm this. Encryption is only used for Client Server communication.

There are further flaws in the concept. For example, our government lately decided that de-mail addresses do not have to be visually distinguishable from other mail addresses (i.e. de-mail addresses do not share a common tld, nor do the tlds have to contain something like "de-mail"). Instead, they came up with the idea that email client vendors could implement a mechanism for telling users whether an email address is a de-mail address..

Re:No end-to-end encryption though

[somersault]

Spam has not been solved, just covered up. It is a pointless waste of incoming bandwidth and server power (if you do your own filtering). This would do nothing to stop spam either, it doesn't matter if you know the identity of the sender if the sender's machine is a zombie. There will always be more idiots with compromised machines.

Re:No end-to-end encryption though

[rotide]

Until any email to _any_ government agency (applications for services, jobs, taxes, etc, etc, etc) _requires_ you use this service..
Until any company wishing to do business with the government is _forced_ to use this service to keep their contract..

There are ways to make sure it's not a "dud", if they are willing to make the laws, and it sounds like they are.

Re:No end-to-end encryption though

[divisionbyzero]

YThe worst outcome would be if people ended up using it, but at this point I'm guessing it'll be a huge dud; some government entities will support it, as will a few corporations, but that's it.

I don't think they will be so lucky. I'd bet the government will require it for some communication and account access. Over time it will become more inconvenient to have multiple email accounts and people will just default to using de-mail.

Re:No end-to-end encryption though

[bemymonkey]

Hmmm, I haven't gotten much info about this, but IIRC it's not really about replacing or upgrading E-Mail, but rather about replacing snail-mail entirely. Documents with signatures and so on can now be sent as e-mail instead of in quaint old envelopes...

Re:No end-to-end encryption though

[moonbender]

Documents with signatures can already be sent as e-mail!

Re:No end-to-end encryption though

[bemymonkey]

Really? Here in Germany they're not always accepted on the other end. This would allow people to verify that the signed document actually came from the person who supposedly sent it...

Re:No end-to-end encryption though

[moonbender]

Really? Here in Germany they're not always accepted on the other end. This would allow people to verify that the signed document actually came from the person who supposedly sent it...

I didn't say that anybody accepts (or sends out) signed emails. I said it's already possible to sign emails, so there's no reason to come up with an alternate infrastructure. Instead of spending X to get government services and a few companies to use de-mail, they could have spent Y << X to get government services and a few companies to install GPG. Of course that'd result in widely deployed public cryptography -- including strong end-to-end encryption -- something that must not be.

Re:No end-to-end encryption though

[bemymonkey]

Is end-to-end encryption absolutely necessary for sender verification?

Re:No end-to-end encryption though

[JSlope]

My experience shows that people are not interested in easy to use end-to-end encryption.

Re:No end-to-end encryption though

[moonbender]

No. But in pretty much any decently engineered crypto setup, if you can do signing you can also do (end-to-end) encryption. Which is why they didn't create a decently engineered crypto setup.

Re:No end-to-end encryption though

[dgatwood]

...it doesn't matter if you know the identity of the sender if the sender's machine is a zombie.

Depends on how it is designed. If the system required that each server in the chain be a trusted server that signs the message with a valid SSL certificate, then the spammers would have to either buy a cert for each individual zombie (too expensive to be profitable) or tie them all to a single domain name and cert that could then be trivially blocked (either by revoking the cert or by blocking mail from that domain).

Either way, the high cost and low effectiveness of such a scheme would almost completely eliminate any financial benefit to sending spam, while having minimal effect on legitimate mail servers. This does, however, require that every step in the chain require authentication in some form.

The requirements for a secure mail system without spam are actually fairly simple:

• All mail hosts must have an SSL certificate.
• All mail users must have an SSL certificate.
• All supported certificate providers must use a CRL or OCSP.
• All supported certificate providers must provide an abuse email address for reporting spam, and must take action promptly.

That's it. It's not magic. Spam is relatively easy to combat. It does, however, require a completely redesigned mail system. Unfortunately, it sounds like this one isn't that system.

Re:No end-to-end encryption though

[bemymonkey]

Then why the fuss? Sure, end-to-end encryption would be nice, but the main purpose (sender verification) is fulfilled, as far as I can tell...

Re:No end-to-end encryption though

[Candid88]

"It purports to solve problems that are already pretty much solved -- spam, reliable delivery -- while not solving all the difficult ones and introducing new dangers for the customers"

A strange conclusion. I don't see how spam has been "pretty much solved" at all. Current anti-spam techniques are far from ideal and phishing is an extremely serious, still-emerging, problem. Also, making wild predictions on a technologies uptake upon initial announcement is a complete guessing game. If you could know for sure you'd be far too busy making millions and living it up on a yacht in the Caribbean to be commenting on Slashdot.

Sure, encryption and backwards compatibility would be good, making the system far from perfect, but for the vast majority of email use, this system would - from initial appearances - help combat the issues that are without doubt the most serious current and emerging issues affecting email.

Re:No end-to-end encryption though

[plover]

This is not a problem of encryption or SSL. Zombies can simply bypass all security measures by emulating the end user.

Think of a zombie that opens up your copy of Outlook Express, fake-clicks "Create new email message", types something about penis enlargement, types in a hundred addresses, then fake-clicks "Send". As far as the entire chain of email is concerned, the email came from somersault@example.com. You (and by you I mean your computer acting on your behalf) sent the spam, so De-mail could block you until you clean up your act.

If your machine is infested, having your email client require you to type a secret password is pretty much useless, too. The malware could sniff your secret password keystrokes and the zombie would simply play them back whenever it wanted to send more spam.

So if the idea is "If De-mail detects spam coming from one of the authenticated users, it will refuse to permit that user to send more email until they can demonstrate they are not going to send more spam." It'll be up to the end user to correct the problem before being allowed to rejoin the collective, which is a good thing.

Re:No end-to-end encryption though

[plover]

YThe worst outcome would be if people ended up using it, but at this point I'm guessing it'll be a huge dud; some government entities will support it, as will a few corporations, but that's it.

I don't think they will be so lucky. I'd bet the government will require it for some communication and account access. Over time it will become more inconvenient to have multiple email accounts and people will just default to using de-mail.

I can see the commercial sector driving adoption on its own. As a business, I might ask all business to be transacted through De-mail to ensure legitimacy of contracts and payment. Or as an insurer, I might offer reduced rates of coverage to business transactions that take place over De-mail, as I would trust them to have less chance of being fraudulent.

Re:No end-to-end encryption though

[moonbender]

They're spending a lot of money implementing a new technology to accomplish something that old technology would have done cheaper and better, and they're enforcing uptake of the new, inferior technology by legislative means, at the same time obstructing the uptake of the better alternatives.

Re:No end-to-end encryption though

[infotechvina]

does it work with all viral mail via internet? phuong phap hoc tieng anh hieu qua nhat

Re:No end-to-end encryption though

[dgatwood]

Zombies can simply bypass all security measures by emulating the end user.

Not really. An end user's ISP typically has throttling in place such that if the user tries to send millions of emails out in a day, they A. will not go through, and B. will result in the user's account getting disabled rather quickly. If spammers are not able to run their own SMTP servers on zombie machines, spam ceases to be profitable, as it requires being able to send out huge volumes of email in a short period of time, and continuing for a long period of time.

With even the most basic host signing requirements, you could stick a fork in the spam. It's done.

Re:No end-to-end encryption though

[Dan541]

In other words it's not actually email but a government controlled message system.We should ask the people of Egypt, Libya, China ect if this is a good idea.

Re:No end-to-end encryption though

[plover]

Changing the subject doesn't invalidate my previous point. Your previous comment was talking about spam being thwarted by SSL, but that's what zombies easily bypass. Each zombie could easily send out 100 emails a day and not trigger "suspicion" flags at the ISP level. With a hundred thousand zombies, that's ten million spams that the security software would never catch. And I'd bet that a competent botherder could probably quote each major ISP's spam threshold from memory, so if Comcast's throttle is 1,000 emails per day he probably limits his Comcast zombies to about 900 spams per day.

The point of my previous post is that certificates are not a cure-all, because they're meaningless on a compromised box. Once it's owned, it can't be considered trustworthy. And most home users couldn't tell you whether or not their box is working right or not, because they can't even tell the difference between useful software and the shovelware that stores like Best Buy poop onto the hard disks before delivering the equipment. There's almost no chance they will detect malware zombie code.

Re:No end-to-end encryption though

[dgatwood]

Each zombie could easily send out 100 emails a day and not trigger "suspicion" flags at the ISP level.

The fact that spam bots can masquerade as the user is largely irrelevant. The reason we have spam is ultimately that there's no good way to verify that a message was sent by a given sender.

The only reason spam is possible with authenticated mail clients is that the ISPs require all outgoing mail to go through their servers, and thus the ISPs are forced to not do comparisons between the ISP's mail drop username and the "From" field in the message. Were it not for this flaw, any botnet would shrink as soon as somebody replied to the spam with a message that says, "Dude, your machine is sending out spam. Run a virus scan."

The only reason that such a technique is ineffective today is that the routing and other header fields are free-form text that can be modified en route, and trying to track down who had a given dynamic IP at a given moment is generally not worth the ISP's time anyway unless there's a subpoena attached to the request. As such, it's not readily possible to start out with an email message as received and definitively and programmatically pin it on a specific account at an ISP.

However, if the email system required each server in the path to sign the message, that would no longer be the case. Routing information would be verifiable and would be canonical in form, thus allowing trivial verification of the entire path all the way back to the initial ISP. Because the system would be inherently verifiable, there would be no reason for ISPs to try to route third-party mail drop clients through their SMTP servers, and thus the authentication to the mail drop server would be the From address.

Further, because the signatures of each server all the way back to the original server could be checked for authenticity, the ISP would be absolutely certain that the user (or at least the user's computer) actually sent the message. Thus, eliminating botnets would be as simple as adding a button in the mail client that says "report as spam" that automatically bounces it to the sender's ISP's abuse address. The ISPs could then collect these reports, and if a threshold is exceeded, the user's account would get disabled. No maximum number of messages per day, no value that the botnet authors have to stay below, just a "this user sent ten messages marked as spam" threshold that (after someone looked to see if the messages really did seem to be spam) would result in the user losing access. Such low volume and poor continuity would make spamming infeasible.

That's why forcing all of the mail servers to have legitimate certificates is the key to eliminating the spam problem. It allows the normal abuse management process to work again like it did back when all the servers were at least ostensibly trusted, and it allows for trivial automation of the verification process, which is key to detecting and eradicating affected machines in a timely manner.

Re:No end-to-end encryption though

[climb_no_fear]

Where did you see that a German national ID is required? Maybe I overlooked it in the article. I saw German resident but nothing about a German ID card. This is not a minor detail since 10% of German residents are not German and therefore do not have a German ID. Does this mean I can't receive these mails? Could make fiing my taxes interesting...

out of thin air?

[StripedCow]

So why didn't we read about this on slashdot before? Or did I miss something?

Re:out of thin air?

[ludwigf]

Wikipedia: "The project was announced in 2008"

Google: couldn't find a coverage of de-mail on /. before

Living in Germany I've heard about it several times before.

Re:out of thin air?

[Anonymous Coward]

Slashdot is powered by your submissions, so send in your scoop.

Lot of peoples complain about slashdot, but not much are doing anything about it.

Re:out of thin air?

[stealth_finger]

Living in Germany I've heard about it several times before.

I used to work at DHL and they never shut up about it.

Re:out of thin air?

[tomhudson]

Because the editors choose the shittiest submissions. (I sent a few too.)

You sent in a few of the shittiest submissions?

No wonder you're posting A.C.

Re:out of thin air?

[muuh-gnu]

DHL, i.e. "Deutsche Post" isnt participating in De-Mail at all. Since the basic purpose of De-Mail was to obsolete a large part of legally binding snail mail, and Deutsche Post realized they would be hit the hardest by this, they developed their own competitive service called "Deutsche Post ePostBrief", which works exactly the same as De-Mail, but of course isnt compatible with De-Mail, so you cant interchange legally binding emails between providers. Deutsche Post is kinda alone in their camp, since basically everybody else (ISPs, Email-Providers) is in the De-Mail camp.

What both of course have in common is that there is no end-to-end encryption, so now you have not only to trust your lawyer/bank/doctor for confidential stuff, but now you also have to trust the carrier. Oh, and, in order to not hurt their snail mail business, every "Deutsche Post ePostBrief" will cost EUR 0,55, exactly as much as a snail mail.

Re:out of thin air?

[smurfsurf]

http://service.deutschepost.de/faq/wie-steht-die-deutsche-post-zum-geplanten-de-mail-gesetz

Natürlich unterstützen wir die De-Mail-Initiative des Bundes und werden – sobald das Gesetz in Kraft ist – eine Akkreditierung als De-Mail-Anbieter beantragen.
Schon jetzt erfüllt der E-POSTBRIEF die hierfür erforderlichen Standards, soweit sie nach dem derzeitigen Gesetzesentwurf absehbar sind.

(Loose translation: We support the de-mail intiative and will apply for an accredition as soon as the law is enacted. e-postbrief already meets all criteria of the upcoming law.)

They were stalling the legislation, but they have no chance but to participate as they could not stop it.

Re:out of thin air?

[Gofyerself]

If you were a real slash dotter, you would be on your mothers basement couch 24/7 waiting to make the next first post.

Re:out of thin air?

[countertrolling]

Well, you know how submissions work. For better chances, always make sure it links to an advertising partner. If it's not on one of the mainstream *tech* rags, then chances are Slashdot won't have it. That's where its aggregator does most of its searching, and is why Slashdot is always showing up late to the party. And to see all the false rumors on the front page lately is kind of a sad display of editorial *quality*

No, thank you.

[Mortiss]

I can encrypt on my own and Gmail already does a fine job removing spam. I don't need a Government oversight and much less a possibility of paying per message for this "privilege".

Re:No, thank you.

[Lennie]

It is not even encrypted. Just a the mailservers use encryption for the transport and the system is seperate from the normal internet mailservers.

My guess is, it is SMTP-authentication over SSL/TLS for sending mail so they know exactly who send it (atleast which e-mail client).

Re:No, thank you.

[Kvasio]

nobody prohibits you from using your gmail account, this is just that when dealing with state offices (e.g. tax office, land registry, local authorities, voting), their registered email would be useful.

Re:No, thank you.

[TheRaven64]

Then why not use existing standards? We already have S/MIME, which allows a digital signature to be used to sign and encrypt mail. Simply pass a law saying that emails with S/MIME encryption and a certificate signed by the government's CA are viewed as legally binding. Then, anyone can continue to use existing clients, can continue to use existing servers, and can just get a certificate signed by the government if they want to opt in to this.

Re:No, thank you.

[Lennie]

Yes, I'm sure that is interresting. But why not use DNSSEC, SSL/TLS-certificates, SSL/TLS Certificate Authorities and DKIM which already solve all these problems.
1. SSL/TLS-certificates are created by the Certificate Authorities
2. SSL/TLS encryption for communication between mailservers
3. SSL/TLS encryption with authentication for delivery from the user to the mailserver
4. DKIM signing of the e-mail on the mailserver to verify that the mail came from the user
5. DNSSEC to publish the DKIM key
6. DNSSEC to verify the domain
7. All we need is an interface in the e-mail client (or maybe partly on the mailbox server where the e-mail is delivered to) which checks that the the above is valid and the domain has an HTTPS EV-like certificate (green bar in the browser) to prevent phishing with similair looking domains.

I think simpler is possible too, it is more a list of technology which can already be applied. :-)

Re:No, thank you.

[b4upoo]

You may be able to encrypt beyond the government's ability to decrypt But how can you handle a court forcing you to reveal the contents? Worse yet I would not be so certain that simply using encryption may in itself be enough to attract one of our current star chamber types of discovery.
                              Freedom of speech is not lost at the moment it is squelched. The freedom to speak dies the moment you understand that your speech might be inspected or that you may be required to reveal your information. Simply knowing that the freedom can vanish at the will of authorities kills speech.
                              Think about book publishers forced to pay lawyers to go through a book page by page before publication. You know how it works. You spin a story about superman and some kid puts a bed sheet on his shoulders and jumps of his roof. You get sued. So you are chilled. You must not even publish harmless fantasy stories. If you draw a cartoon you must fear that some idiot will think the cartoon girl is under age and off to prison you go. What freedom?

Re:No, thank you.

[Thundersnatch]

You may be able to encrypt beyond the government's ability to decrypt But how can you handle a court forcing you to reveal the contents?

IANAL, but at least in the USA, the fifth amendment protects you against self-incrimination. I do not think you can be compelled to divulge an encryption key if doing so would provide any evidence you committed a crime. Any decent lawyer and/or the ACLU's could probably prevail with this argument in court.

The trick, of course, is that the prosecution will typically give another party who has access to the encrypted data immunity from prosecution, so the 5th amendment does not apply. Then that party can be compelled to decrypt the messages. This makes getting at encrypted emails easy if the prosecution is only after one of the sender or recipient.

From what I can gather, in civil cases, so long as decryption would not provide evidence of a crime on your behalf, you could be compelled by the court to decrypt the data, even if it harms your arguments in the civil case. But if you're willing to risk perjury, you can always testify you forgot the key and likely get away with it.

Every mistake in the book

[Anonymous Coward]

They put a price on every email.

The system will not provide end-to-end encryption: Mail will only be encrypted to and from the mail service providers.

While the accounts are free, individual mails will cost money.

Mail delivered to these accounts will count as delivered to the recipient, so any respite associated with the delivery starts running. Don't read your email regularly - miss deadlines.

Did I mention that mails cost money?

I have recommended to everyone who has asked me to stay away from this system if at all possible. Don't even get an account.

Re:Every mistake in the book

[maweki]

"Mail delivered to these accounts will count as delivered to the recipient"

Yeah, well, not true. Actually, you start counting three days after the technical delivery. Check your mail reguarily and gain three days. Check your analog mailbox not often enough (people in my house check their boxes once a week) and you will miss deadlines as well.

Re:Every mistake in the book

[Thad Zurich]

Sender-pays may well be the ultimate spam defense (but see comment about botnet operators...) That bit about delivery=receipt has to be reworked; it's not comparable to snail-mail return receipts, which have to be signed by the recipient. Might make more sense for business-to-business, since the other real value is in non-repudiation.

Re:Every mistake in the book

[rmstar]

Mail delivered to these accounts will count as delivered to the recipient, so any respite associated with the delivery starts running. Don't read your email regularly - miss deadlines.

How is this different from mail delivered to your snailmail box? "I wasn't at home" has not been a particularly good excuse for a very long time.

The lack of end-to-end encryption is another matter entirely, and a rather obvious strategy to ensure that the government can eavesdrop. So much is clear.

Re:Every mistake in the book

[stealth_finger]

Mail delivered to these accounts will count as delivered to the recipient, so any respite associated with the delivery starts running. Don't read your email regularly - miss deadlines.

Surely not, even regular email has read reports, so unless it does that automatically and it counts that. Any company will tell you with regular post proof of postage isn't proof of receipt.

Re:Every mistake in the book

[Splab]

Yeah, I hardly ever check the mailbox, once a week tops - unless I'm expecting something.

Re:Every mistake in the book

[crtreece]

Anything sent via snailmail that is expected to be time sensitive and/or legally binding would require a signature, it would not just be left in the mailbox.

Or it would be sent via FedEx or UPS, again requiring a signature.

Not so sensitive items, bills and such, don't require a signature, but you're still on the hook. Mail carrier left the door to the mailbox open, and your mortgage payment invoice got blown down the road? You are still on the hook for the payment.

Re:Every mistake in the book

[maxwell demon]

To me this is why I'll never get an account. It's mandatory by law you check your account regularly because delivery by the provider equals reception by you. Sick in a hospital? On vacation? Good news! Once you'll return you're life will be fucked because you missed a few legal messages to which you had to respond yesterday.

How is this different from legal messages arriving in your physical mailbox when you are away (in hospital/on vacation)?

Re:Every mistake in the book

[maxwell demon]

It will make things harder for the botnet operators as well because, unlike now, the infection will only remain undetected until you receive your next bill from your de-mail provider. And after that, people will try to remove that bot ASAP. So as soon as you use a bot to send de-mail, that node will soon be lost for the botnet.

But then, normal mail won't go away anyway. So spam will continue to exist on normal mail, but de-mail will likely be mostly spam-free (mostly, because after all some companies happily pay for "snail spam" to be delivered to your physical mail box).

Re:Every mistake in the book

[mxs]

Mail delivered to these accounts will count as delivered to the recipient, so any respite associated with the delivery starts running. Don't read your email regularly - miss deadlines.

How is this different from mail delivered to your snailmail box? "I wasn't at home" has not been a particularly good excuse for a very long time.

Actually that is a very, very good excuse when you require proof of delivery/acceptance -- since those are usually signed-for. Recipient not there to sign ? No proof of personal delivery. The difference with DE-Mail is that messages count delivered when they hit your service provider, no matter whether you read your account or not. This can have far-reaching consequences under German law.

The lack of end-to-end encryption is another matter entirely, and a rather obvious strategy to ensure that the government can eavesdrop. So much is clear.

Yes, and the lies and bullshit they spew when defending this are even more so. Too bad too few people will get the message -- or care.

Basically the whole things boils down to a giant waste of money and resources for everybody. Well, everybody not implementing such a system and getting paid for it.

Re:Every mistake in the book

[smurfsurf]

> How is this different from legal messages arriving in your physical mailbox when you are away (in hospital/on vacation)?

You claim to the sender (and have to prove if he disagrees) that you were not able to retrieve the letter for that reason and any deadline has to be restarted. The legal term is restitutio in integrum (although it seems the US uses restitutio in integrum only for demages?).

Re:Every mistake in the book

[maxwell demon]

Any why should that not work with de-mail?

Re:Every mistake in the book

[countertrolling]

Yeah, something like this would sell much better in the states, where they are dumb enough to pay to receive calls.

Re:Every mistake in the book

[rmstar]

Recipient not there to sign ? No proof of personal delivery.

IIRC, this is not true. In particular the kind of mails that involve legal proceedings can be considered as delivered even if you weren't there. It sometimes is even written explicitly on top that for legal purposes you were there personally. German laws are strange.

Re:Every mistake in the book

[mxs]

Recipient not there to sign ? No proof of personal delivery.

IIRC, this is not true. In particular the kind of mails that involve legal proceedings can be considered as delivered even if you weren't there. It sometimes is even written explicitly on top that for legal purposes you were there personally. German laws are strange.

There was a reason I put in "personal" there ;-) You can get products like proof of delivery (Einschreiben Einwurf) which do not prove you personally received it, but which do prove that the letter was delivered to the address given. Then there is a product with proof of PERSONAL delivery (Einschreiben eigenhändig), which proves the recipient has personally received the piece of mail (but which also requires the recipient to sign for it of their own volition).

You are however correct that under German law, the actually important part is when a piece of mail enters the domain of the recipient -- i.e. even regular postal mail will start legally binding deadlines if it just enters your mailbox -- no matter whether you open it or not, or whether you are home or not. Question is how you prove this.

If you want to prove to a court that your letter from that day was delivered, the best option is to use a personal courier who observes you putting what you want to deliver into the envelope and personally delivers it to the recipient, noting when and how he did so -- to be used as a witness later on. Otherwise a recipient could simply claim he received an empty envelope, the letter was put into the wrong mailbox, or funny games like that.
With the DE-Mail proposal, such "defenses" no longer work, and the act of delivery is a LOT easier and cheaper to prove for the sender. Legally speaking, the recipient will have limited his/her options considerably.

Dibs on the nickname!

[ClayJar]

From the sound of it, it'll almost inevitably end up costing money. With that in mind and by the powers vested in me by absolutely nobody in particular, I hereby dub it "feemail".

(One *could* say that it is supposed to be a kinder, more respectable alternative to the rough-and-tumble wild west of existing (e)mail, but then there are those who think it's just a prettier version that will inevitably cost a bunch of money.)

Re:Dibs on the nickname!

[OeLeWaPpErKe]

Don't you think it's probably more meant as a kinder, faster alternative to confirmed (ie. legally valid) snailmail delivery, in addition a way for that long-awaited legally valid electronic signature ?

Because that's what Europe's various governments have been trying to create for a long while now. Belgium, Holland, France. The good news (for you at least) ? They all failed miserably. Somehow I doubt this will work better.

They are busting it again.

[Anonymous Coward]

Typical mix of greedy corporations in bed with clueless *and* greedy lawmakers.

I bet you:

* Mails will live unencrypted at provider's server (check!)

* Users won't have any control on their keys and identities (check)

* There will be a central place to map identities to Real Life users (check)

Darn. And OpenPGP is out there for years. Sad. But hey, with OpenPGP the Deutsche Telekom and other parasites won't be able to leech on "consumers", right?

Re:They are busting it again.

[binarylarry]

Poor RMS.

Re:They are busting it again.

[HungryHobo]

as far as I can see everything this service provides has been done better for free elsewhere.
nothing novel.
but it'll probably be pushed hard by the german government.
and if it works even poorly then other governments will follow their lead because since what happened in egypt, tunisia and libya governments the world over are suddenly terrified of the net.

Re:They are busting it again.

[OeLeWaPpErKe]

The problem is (some) users are idiots, easily tricked into revealing their encryption keys, and OpenPGP is no protection on a virus-infested pc (of course palladium, or "trusted execution" could solve that).

Cooperation

[Kwesadilo]

This sounds like completely run-of-the-mill encrypted email that you also have to pay per message and identify yourself for. The one significant advantage that I can see is that you might be able to convince other people to actually use it.

Looks like my Aunt was right...

[fortfive]

...when she sent me an forward claiming the government was going to start charging for email!

d-mail ?

[advance-software]

Isn't that going backwards ?

Shouldn't the next one be f-mail ?

Why use de-mail when gpg exists?

[bl8n8r]

Why would I volunteer to use a government sponsored program that I may get charged for when I can just use Enigmail in Thunderbird, or gpg the message otherwise?

Second problem: "It will allow service providers to charge for sending messages".

Major fail. It sounded almost good until I read that.

Re:Why use de-mail when gpg exists?

[betterunixthanunix]

I imagine people will use it for the same reason people use Hushmail: ignorance.

Re:Why use de-mail when gpg exists?

[peragrin]

well that and ease of use.

Setting up openPGP is a pain for someone who has never had to deal with it before, and not mention then you have to have the other end using the same encryption.

while it makes sense, people aren't very smart about these techie things and really don't want to think about it.

I don't encrypt my email simply because 99.9999999999999% of end users don't know what it is or how to decrypt it, or even which tools to decrypt it with.

Re:Why use de-mail when gpg exists?

[Velex]

This is the fault of email client developers. I haven't used KMail in quite some time (I've since switched to a GTK/XFCE desktop so Claws-Mail is the client of choice these days), but when I had a KDE 3.x desktop, I remember that I was struck by how seamless KMail made GnuPG, even S/MIME. If all email clients made GnuPG as seamless as KMail, you'd see more use of encryption.

Really, encryption need not be difficult, not much more difficult than typing https or getting redirected to https when you just type foobar.com. It's simple. Your email client should generate a key or detect if you already have one in your OSes security system (GnuPG, etc). Then it should advertise that you have a key by attaching the public half to your messages. When another client sees a public key, it should cache it, and wa-lah! Now that client can send encrypted emails back with no problem.

I'm sure I'm over-simplifying, but the number 1 reason nobody encrypts their emails is because of this: look at who your popular email programs are. There's Outlook, Yahoo, GMail, and HotMail. Do a single one of those support OpenPGP out of the box? Absolutely not. In fact, the only one of those that even supports S/MIME is Outlook, and its support is a pain-in-the-ass at best!

Re:Why use de-mail when gpg exists?

[mxs]

Why would I volunteer to use a government sponsored program that I may get charged for when I can just use Enigmail in Thunderbird, or gpg the message otherwise?

Second problem: "It will allow service providers to charge for sending messages".

Major fail. It sounded almost good until I read that.

As a sender, you get to deliver stuff to DE-Mail addresses and they count as legally delivered. This is going to be very good to have for collection agencies or governmental agencies. Senders also get to save a bit compared to paper delivery while legally on the same footing. Senders also get proof of identity for the recipient. Senders get to spout bullshit about using the latest and most secure email standard ever.

Recipients get shafted, in more ways than one.

Re:Why use de-mail when gpg exists?

[Dr. Evil]

What's wrong with Hushmail?

It provides SSL encryption on servers protected by Canadian laws, including Canadian privacy laws. While they respect U.S. court orders, there's no reason to believe that such orders could be executed in secret outside an investigation of a crime recognized by both Canada and the U.S.

PGP is stronger, but a people aren't using it, so practical applications are limited.

Re:Why use de-mail when gpg exists?

[peragrin]

So you have to install an add-on, and then the recpients have to install an add on, and both sides then need to exchange keys.

so you have 4 steps before you can even consider sending encrypt mail. Someone needs to simplify the process( while not actually removing steps)

Re:Why use de-mail when gpg exists?

[Chaonici]

I completely agree. Lack of widespread email encryption is likely the fault of webmail developers and Microsoft (with regards to Outlook) not supporting the encryption in their interfaces. And of course they wouldn't- it's completely contrary to the interests of a large corporation to give its customers privacy.

On the other hand, I use Evolution for my email, and it supports GPG out of the box. When writing a message it's a simple matter of checking a box in a menu at the top to encrypt it; two boxes for signing. (Of course, I have to have a key pair first, and while it's trivial for me to generate one, the average user would have a bit more difficulty with it.) Unfortunately, as aforementioned, I never have need for it, because none of my contacts use encryption.

I think instant messaging has the same problem: lack of encryption by the client. Pidgin has a few plugins that provide end-to-end encryption, like OTR, but most of the more popular clients (like Windows Live Messenger) don't support it at all. (And the Empathy developers actively refuse to add message encryption because they believe that IM encryption should be done on the protocol level, not by the client.) Oh well, at least we have XMPP...

Re:Why use de-mail when gpg exists?

[Chaonici]

From a security-conscious standpoint, the fact that Hushmail has the capability to read their users' emails is a concern. Never mind that they only said they'd do it if the government told them to (which should be no real comfort at all). As we've been saying to the FBI recently, any backdoor at all could potentially make the entire system worthless because there's no way to guarantee control of who uses it.

PGP, on the other hand, has no central authority that can give up your communications. No need to trust a third party.

Re:Why use de-mail when gpg exists?

[betterunixthanunix]

What's wrong with Hushmail?

This:

protected by ... laws

I might as well not even bother with encryption if I am going to turn to "laws" to protect me. Hushmail is snake oil cryptography, which is what I said when it was first described to me years before the DEA bust.

U.S. court orders

Court orders should not result in plaintext being produced by a third party, regardless of why the orders were issued.

PGP is stronger, but a people aren't using it,

Then people should be educated, not given snake oil.

Re:Why use de-mail when gpg exists?

[Dr. Evil]

"I might as well not even bother with encryption if I am going to turn to "laws" to protect me. Hushmail is snake oil cryptography, "

I disagree here. While it's true that you can't expect any service provider to protect you more than the laws permit, if you choose those laws, the situation and the country very carefully, you can ensure that the service provider has more to lose than you do.

The idea that a company is going to break the law arbitrarily with your data is paranoia. Your landlord could bug your apartment, your phonecompany could tap your phone, your girlfriend could be bugging you. Your ISP could be reading your email. It's all the same. It's not unprecedented, but the punishment is severe. You need to weigh the motive against the punishment.

If I'm in Baltimore, and I need to send my x-rays to my doctor in Los Angeles, I can be reasonably confident that the Canadian company is not going to violate Canadian privacy laws to harvest and sell the data to healthcare or insurance providers in the U.S.

If I'm sending instructions to somebody smuggling people over the Mexican border though, I can expect that the U.S. legal system could approach a Canadian court to have any existing records revealed.

So if you're trying to send your medical records to your doctor, Hushmail might be a good option. If you're trying to profit from smuggling illegal aliens, then you might want to consider something with no legal escrow.

Re:Why use de-mail when gpg exists?

[AtillaTheMagyar]

Email is different from physical home security and to compare the two I think is a bit of a red haring. That DEA case referenced where Hushmail hacked their user to get the password to decrypt their private key and stored messages shows a fundamental weakness in their system's design. I would never leave my private key on someone's server, even if it's encrypted. It's just too tempting for a government agent to strong-arm the provider into doing exactly what hushmail did. Court order? Sure, they complied with the law but here we're talking about the 'ability' of a third-party to decrypt messages. I have one key, you have another. I should be able to send to you without anyone else being able to decrypt it. Luckily, there are other systems out there like GPG which people can feel safer with. There are even some companies trying to automate everything like TrulyMail so non-technical users can also get things up and running. Is there a perfect solution? Not yet, but I see things getting better with time.

Re:Why use de-mail when gpg exists?

[Dan541]

I don't encrypt email for the same reason. It's secure enough for it's purpose. Certainly allot more secure than regular postal mail sitting in a letter box next to the road. Email encryption really needs to become integrated as a standard within the clients. I think encryption as a whole will become much more widespread over time; especially with today's governments.

Re:OpenPGP

[Anonymous Coward]

If you want encrypted mails then yes. If you want to do a legally binding offer or request or or or, then you cannot use OpenPGP, because there are no rules who does what with the keys. (You could create a contract with someone saying that mails signed with a specific OpenPGP key are your mails, but good luck on getting anyone to do so). With something like this, once you sign it, this key is your key. Everything signed with it is as if you had said it in public or written it with a (classical) signed paper.

legal binding

[sourcerror]

The point is that mails sent through De-mail have legal binding, so you can use as proof at court.

Re:legal binding

[Anonymous Coward]

Unfortunately that also means you cannot leave your account unchecked for more than a day or two because you might miss a legal deadline. It's your duty by law to regularly check it. Sounds like a hoot.

Re:legal binding

[Anonymous Coward]

E-mail with a qualified signature is already legally binding in all of the EU. A more realistic approach would be to set up a state (or preferably EU) run CA, that would be inserted in the CA root lists of all common OSes. As an addition, they should ensure that OS email clients would automatically discard unsigned email coming from the government.

This would actually have a chance to succeed.

Czech govt. already did

[jmak]

And it's been a failure, for a number of reasons:

- it cost a fortune to deploy
- one message costs an equivalent of about 1 USD, which means no one uses it except for communicating with the government
- it relies on a proprietary (although free as beer) rather obscure application for Windows, fortunately a non-profit foundation later developed a cross-platform library for accessing the mailbox
- once you register into the system, any official letter you get is automatically considered delivered, so you cannot deny receiving it, that's why any sane lawyer will discourage from getting such an account ever unless you are obligated to

Obviously, because so much money already burnt, the mailbox system is here to stay.

For certain uses I see great value

[germ!nation]

If it allows banks, utilities and other real world important billing and information emails to be able to be considered trustworthy then I can see a lot of value.

Obligatory

[moonbender]

Your post^Whuge government engineering proposal advocates a

( ) technical (x) legislative (x) market-based ( ) vigilante

approach to fighting spam. Your idea will not work. Here is why it won't work. (One or more of the following may apply to your particular idea, and it may have other flaws which used to vary from state to state before a bad federal law was passed.)

( ) Spammers can easily use it to harvest email addresses
(x) Mailing lists and other legitimate email uses would be affected
( ) No one will be able to find the guy or collect the money
( ) It is defenseless against brute force attacks
( ) It will stop spam for two weeks and then we'll be stuck with it
(x) Users of email will not put up with it
( ) Microsoft will not put up with it
( ) The police will not put up with it
( ) Requires too much cooperation from spammers
(x) Requires immediate total cooperation from everybody at once
( ) Many email users cannot afford to lose business or alienate potential employers
( ) Spammers don't care about invalid addresses in their lists
( ) Anyone could anonymously destroy anyone else's career or business

Specifically, your plan fails to account for

( ) Laws expressly prohibiting it
( ) Lack of centrally controlling authority for email
( ) Open relays in foreign countries
( ) Ease of searching tiny alphanumeric address space of all email addresses
( ) Asshats
( ) Jurisdictional problems
( ) Unpopularity of weird new taxes
( ) Public reluctance to accept weird new forms of money
(x) Huge existing software investment in SMTP
(x) Susceptibility of protocols other than SMTP to attack
( ) Willingness of users to install OS patches received by email
(x) Armies of worm riddled broadband-connected Windows boxes
( ) Eternal arms race involved in all filtering approaches
( ) Extreme profitability of spam
( ) Joe jobs and/or identity theft
( ) Technically illiterate politicians
( ) Extreme stupidity on the part of people who do business with spammers
( ) Dishonesty on the part of spammers themselves
( ) Bandwidth costs that are unaffected by client filtering
( ) Outlook

and the following philosophical objections may also apply:

(x) Ideas similar to yours are easy to come up with, yet none have ever been shown practical
( ) Any scheme based on opt-out is unacceptable
( ) SMTP headers should not be the subject of legislation
( ) Blacklists suck
( ) Whitelists suck
( ) We should be able to talk about Viagra without being censored
( ) Countermeasures should not involve wire fraud or credit card fraud
(x) Countermeasures should not involve sabotage of public networks
(x) Countermeasures must work if phased in gradually
(x) Sending email should be free
(x) Why should we have to trust you and your servers?
( ) Incompatiblity with open source or open source licenses
( ) Feel-good measures do nothing to solve the problem
( ) Temporary/one-time email addresses are cumbersome
(x) I don't want the government reading my email
( ) Killing them that way is not slow and painful enough

Furthermore, this is what I think about you:

( ) Sorry dude, but I don't think it would work.
(x) This is a stupid idea, and you're a stupid person for suggesting it.
( ) Nice try, assh0le! I'm going to find out where you live and burn your house down!

Re:Obligatory

[Yvanhoe]

Yeah, funny, but done correctly it would be a system parallel to the regular emails, that would be used to send official mails like taxes declaration or agreement of a contract. The governement would not have to be able to read the content of the email. I think this is not about fighting spam, but fighting scams.

Ultimately, the main problem I see with this is that many people will have trouble with keyloggers and rootkits, but having a centralized governement sponsored identity checker for crypto messages is a good thing to have.

Re:Obligatory

[moonbender]

Cryptographically signing emails has been possible for decades. The government could have lead by example by simply doing that on a wide scale, encouraging businesses to do the same. For instance, after buying stuff online, you unfailingly get an invoice per mail, something I think businesses are pretty much required to do (if they don't snailmail it, of course); why not just require them to sign it for it to be a valid invoice. Of course, signing and encrypting go hand in hand, and LEO and the interior intelligence service are scared out of their wits of public key crypto gaining wide usage. Hence this train wreck.

Re:Obligatory

[sjames]

They could also just sign people's keys for them and require sufficient proof of ID at the time to make it official.

Re:No End-To-End Encryption

[lennier1]

Once the encryption on the end can be faked so someone else will end up with the costs and even have the cops knock down their door?

Re:OpenPGP

[Alain Williams]

This is the way to go, it is what I use when I want to send encrypted email. There are some big problems with PGP/GPG where government could help, these are:

• not enough people use it. A government push would speed adoption, if government departments use it then others will follow -- that is probably all that they need to do.
• helping with key management and verification. I would be happy to pay a small charge (say £10 one off) to have my key verified against passport, ...

Once they have done that then the normal commercial forces would kick in: some people would pay for s/ware that works, others would use FLOSS; it doesn't really matter -- it is the standard that is important.

Mail signing -- encryption is a completly different problem from spam prevention, we must not conflate the two.

Re:OpenPGP

[Lennie]

Combine it with DKIM and DNSSEC and your are done.

if it's anything like Deutsche Post's E-Postbrief

[itsme1234]

... they better forget it.
It costs from 55 eurocents to send one "email" (to multiple euros if you want confirmation, even if there is no snail-mail/paper involved). The interface is arcane with no 3rd party integration, of course there's no end-to-end encryption (and the "mails" are way less legally protected than normal post) and there are some really nasty conditions attached:
- you have to check your mail EVERY WORKING DAY (that includes Saturdays, not that it matters)
- you can't delegate this "check mail" duty to anybody (note that there isn't anything wrong in letting your wife/neighbour/etc in charge of your physical mailbox if you trust them).

Not for me

[houghi]

There is a reason I do not want my online profile linked to my real life person. Or at least as little as possible.

It is also the reason I did not participate in a GPG signing, as I would then have to identify myself with my real life name. Thanks but no thanks. (Could be that other signings are different. No idea.)

If it needs be, I can drop my online alias and create a new one. e.g. if in 20 years people want to kill me because of something I said that is acceptable now. My boss looking for whatever information he thinks he wants, he won't find anything that wasn't screened by me (if he finds the right person, because others with the same name and similar profiles exist and they are in WAY better shape then I am. One even runs marathons.)

So again, thanks but no thanks.

Re:All is good as long as...

[i-linux123]

It's all fun and games until someone that doesn't know about the system tries to send you an email. I like the idea of having real names registered to email addresses, but certificates already do this.

Email should cost one penny per message

[cjonslashdot]

Charge one penny per sent message. That is all we need to do to stop spam. So simple.

If anyone wants security, there is S/MIME, widely available and widely supported.

Re:Email should cost one penny per message

[mxs]

It's beautiful how you came up with that simple idea all of your own, and so elegant ! Implementation is not something to worry about, that's for the people who don't have ideas, they can do that easy work. Go plebs, implement !

I deduct points for not mentioning CompuServe and it not having any spam. I mean come on, that was so easy to reference !

Re:Email should cost one penny per message

[Stormy Dragon]

1 penny where?

If the sender's e-mail server is charging the penny, how does the recipient's server verify that the penny has actually been collected? If it means only accepting e-mail from servers at known ISP's you're going to break most business e-mail servers. Also, it's essentially just a white list, so why not just implement a white list and forget about the money.

If the recipient's e-mail server is charging the penny, how do you verify who sent the e-mail so you know who to charge? Also, even if you do get rid of spam, you just created a new replacement fraud. The spammers infect a million computers and get them each to send one e-mail to random addresses at the spammer's e-mail server. Viola, the spammer gets to collect $10,000.00 How many people are going to notice their e-mail bill is off by a couple of pennies that month?

This is setting aside that the financial system isn't really prepared to handle billions of one penny transactions every day. You can aggregate, I suppose, but who verifies all the e-mail servers are doing their bookkeeping properly?

Re:Email should cost one penny per message

[mjwalshe]

yep i used to work for telecom gold (dialcom) on the billing side (I wrote the core of the x.400 billing system) and you don't want to go back to that era 20p a mail plus tiered data charges on top of that.

Re:Email should cost one penny per message

[Tom]

That problem has been solved 20 years ago. Some nifty crypto does the trick. There are, in fact, plenty of decentralized electronic currency implementations around. Their problem is that nobody uses them.

Re:Email should cost one penny per message

[cjonslashdot]

I did not say that it was my idea. And you are right: implementing it - getting ISPs and users to accept this - is the hard part!

Re:Email should cost one penny per message

[cjonslashdot]

You are right, that implementing this would require embedding a financial transaction protocol within the TCP infrastructure. Still, that could be used for other purposes and might be quite useful.

Re:Email should cost one penny per message

[VortexCortex]

Charge one penny per sent message. That is all we need to do to stop spam. So simple.

If anyone wants security, there is S/MIME, widely available and widely supported.

Ooh look -- My new email business is to batch zip all incomming / outgoing emails, send / recieve the batches, and unpack them at the other end.

Peering agreements between mail-batch-zip providers will allow all email to traverse freely once again.

If you have any questions, click the "reply" link and fill out the form. You may be charged 1 cent to hit "submit" (unless you are already a customer of Slashdot's mail-batch-zip service).

Re:Email should cost one penny per message

[Jason Pollock]

I agree with you there are better things than charging a penny. We can already see what happens when you do with SMS and VoIP.

For example, people are hacking VoIP lines and then making fraudulent calls to numbers with large termination rates. The guy at the other end gets his cut and disappears.

People are also attacking smartphones and doing similar things - signing up for premium SMS services, etc.

However, we already do have financial systems which are prepared to handle billions of one penny transactions every day - the phone network does this already. It's expensive, but it does already happen.

Re:Email should cost one penny per message

[mxs]

It's the impossible part. It simply is not feasible. As such, the "idea" is a dud and timewaster.

Re:Email should cost one penny per message

[cjonslashdot]

The implication then, is that reflection on what should be is a waste of time.

Good thing that Gandhi did not feel that it would be a waste of time to even contemplate ways to evict the British.

The US political system is very messed up. Any concept for how to truly fix it is inconceivably difficult to implement. Therefore, according to your thinking, let's not even think about it, since it is a waste of time. Let's only think about what is easy to do.

Re:Email should cost one penny per message

[mxs]

The implication then, is that reflection on what should be is a waste of time.

Good thing that Gandhi did not feel that it would be a waste of time to even contemplate ways to evict the British.

The US political system is very messed up. Any concept for how to truly fix it is inconceivably difficult to implement. Therefore, according to your thinking, let's not even think about it, since it is a waste of time. Let's only think about what is easy to do.

That's an interesting reading of what I said, albeit entirely untrue. The implication is not that we should not try to better ourselves, the implication is that we should not go the way of knee-jerk thinking that sees a very simple solution to a very hard problem and makes that simple solution be the silver bullet. Making email cost money is a very elegant and simple solution with one caveat -- it does not work. This has nothing to do with Gandhi or not trying to find a good solution to the spam problem (conflating the two as you do is, by the way, pretty disgusting when looked at comparatively). It's about polemically trying to shoehorn a wrong solution down everybody's throat so as to appear to be doing something, anything at all. Even if it is boneheaded and wrong. And that is precisely what is wrong with US politics as well. If the solution to a problem cannot be conveyed in 2 simple sentences single-celled organisms could immediately get the gist of within 30-60 seconds or so, it is not going to be pursued. The idea that hard problems may require complex solutions is lost on politics and is definitely lost on the populace at large.

Re:Email should cost one penny per message

[cjonslashdot]

I am sorry for misinterpreting your intention. I actually thought about it after I posted my response. I realized too late that you were only saying that the "pay for email" solution was un-implementable. And I agree with you completely about the sound bite culture that we live in. I am not quite as sure though that a solution cannot always be reduced to its essential ideas and expressed concisely: I don't know. Perhaps.

Re: next one

[TaoPhoenix]

No, the next one after gmail would be HeMail, pronounced

Ahee-Mayal.

Homestarrunner FTW!

http://www.homestarrunner.com/main8.html
"Email" tab

There's something like that in Italy as well

[opus_magnum]

named PEC: (http://tools.ietf.org/html/draft-gennai-smime-cnipa-pec-08> ) which has the same legal validity as certified mail.
There's also a variant (CEC-PAC) to communicate with government offices only.

How does this prevent zombie spam?

[goombah99]

Spam sent from zombies will be encrypted and signed with the certificate of the zombied computer. so how does this help?

Re:How does this prevent zombie spam?

[machine321]

You do the same thing you do when any certificate gets compromised, revoke the cert.

Not that I think this is a good idea, though.

Re:How does this prevent zombie spam?

[Z00L00K]

After the damage is done and 200+ persons in the address book already have received the spam.

Plus countless other people. A certificate revocation does take some time.

Re:How does this prevent zombie spam?

[kshade]

A DE-Mail costs about 25 Cent.
Yes, the German postal service is in on this.

Re:How does this prevent zombie spam?

[dadioflex]

Bob Crow, head of the Transport Workers union in the UK, was being interviewed by David Mitchell, of Peep Show fame, and suggested that there should be a 1p charge on all emails being sent as a way to cut the UK deficit. Now the idea was treated with mild scorn by Mitchell and everybody laughed at it, but if it means we could pay for a few more nurses and it stops people sending me links to cat videos then I'm broadly in favour.

Re:if it's anything like Deutsche Post's E-Postbri

[OFnow]

Read every day? So when you go on holiday you get into legal or financial trouble? Cute!

Great big pile of sh*t

[garry_g]

Yet another example of either clueless politicians, attempting to do "a good thing" all the while creating on over regulated, technically inferior system, or the clever attempt to get yet another way of snooping on the people while making them "feel good and safe" ...
The good thing at the moment is that it's not mandatory to have or use the POS email service. At the prices currently discussed(55 âcent per email - same as for a regular letter!), I doubt it will find many people who are interested in using it. Though they have said that prices "may" go down ...
And yes, the standard usually means the mail will be decoded by the MITM, to check for spam (yeah right, at .55â a piece?) or virus/malware (whoah - get a worm on your machine, let it send out millions of DE-Mails - get poor in the process - at least then you won't be able to afford any more internet, removing one more botnet machine from the net), then re-encode for the recipient. The standard is supposed to include the option for end-to-end encryption though, but I'm not sure under which circumstances ... Anyway, as the DE-mail is kept on certain provider mailservers, with current law interpretation, any court could order all the mails to a certain person (or from) to be handed over to law enforcement ...

Problem is the typical chicken and egg dilemma - too few people use public key crypto, because they don't know (or care) about it, so the ones who would use it don't have any recipients to send to, so less people use it ...
Guess everybody should start using a footer with a link to a web page that explains for computer dummies how to set up and operate GPG/PGP and forget all about this crap government control attempt ...

Re:CAcert.org

[Noughmad]

... the amount of peers who you have met in the real world ...

And how would that help Slashdotters?

I had better opinion about Slashdot readers.

[molibden]

Did anyone try to think before start complaining about "clueless politician"?

1. End-to-end encryption. As far as I can see the system does not provides one and does not attempt to do so. And this is right. End-to-end encryption is between me and my recipient and nobody else has anything to do with it. All middle message relay agents can do whatever they want with my encrypted message, as long as they will deliver it finally to the recipient intact. I don't care. People where using end-to-end encryption on mail message for thousandths of years over much less sophisticated transfer agents with great success.

2. Cost per message. First of all I never see a statement that Germany established "e-mail tax" so all messages _must_ cost something. It _may_ cost something. I do not know any law that prohibits Google, Yahoo and Microsoft from collecting money for emails that they transfer. In fact I'm paying right now to Google and Yahoo for e-mail services and considering number of e-mails I've sent per month the cost is much grater 5c/message. Did you guys have a clue that to relay your messages cost money? Service providers have to pay for computers, electricity, network bandwidth, heating/cooling, physical security, customers support etc. All this cost them money. Why they cannot collect fair price for the services that they provide? I'd prefer to pay fair price for the service that I need/value rather than use it for free and watch all this advertising on the sides of the screen.

I like the idea to establish network of trusted MTA - it will be positive thing. It will not solve all problems, but at least it will help with some.

Re:OpenPGP

[jgrahn]

This is the way to go, it is what I use when I want to send encrypted email. There are some big problems with PGP/GPG where government could help, these are:

• not enough people use it. A government push would speed adoption, if government departments use it then others will follow -- that is probably all that they need to do.
• helping with key management and verification. I would be happy to pay a small charge (say £10 one off) to have my key verified against passport, ...

Once they have done that then the normal commercial forces would kick in: some people would pay for s/ware that works, others would use FLOSS; it doesn't really matter -- it is the standard that is important.

Right on. All I'd have to do is to trust the German key (they could publish the fingerprint in Frankfurter Allgemeine Zeitung or something) and I could communicate with anyone in .de.

And that is why I resent the "OMG I would never trust a system where the government is involved!" comments here. Handing out public identities for people is precisely what governments *are for*. Without the government, we are clearly stuck where we are today: with unsigned and unencrypted mail.

Re:Want to send email in Germany?

[Z00L00K]

Starts to be the same crap everywhere - not only Germany. Look at the "bastion of freedom" (The United States) again and see how it really is.

Feels like the world of Max Headroom is going to be a paradise utopia soon rather than a dystopia.

Soon we will have blipverts... And stuff like AdBlock Plus will be illegal.

difficult from a legal standpoint

[Casandro]

Legally a mail in that system has arrived at your place, even if you cannot get it because you are on vacation or your computer/internet broke down. That's a big legal problem obviously.

Re:difficult from a legal standpoint

[AtillaTheMagyar]

I think a 'reasonable' test would be applied. If I send you a notice of something happening in a month, then it is reasonable that you can act in time, unless you are out of the country or have some other reasonable excuse. If I send you a notice of something that happens in five minutes, then I am clearly being unreasonable. Let us not forget that in addition to logic rules of 'If X then Y' there are also real people who determine when someone has violated an agreement/responsibility.

Re:difficult from a legal standpoint

[Casandro]

You are forgetting that this is a German project. There is no reason in German law.

S/MIME & PGP

[Bert64]

There are already standards for authenticating the sender of mail and encrypting the contents of those mails, it would be far better to encourage use of these existing standards rather that come up with something completely new and incompatible with everything else.

back to the 80's

[mjwalshe]

sounds like DBP has manged to looby for a return to the 70's and 80's with the ptt running the countrys email system

That's the problem isn't it?

[HotNeedleOfInquiry]

I'd love to have widely adopted secure end-to-end non-reputable email, but I think it will be a cold day in hell before *any* government will support a standard that doesn't permit them to read the email at will.

Re:That's the problem isn't it?

[JesseMcDonald]

I'd love to have widely adopted secure end-to-end non-reputable email...

We already have non-reputable email; most of it is known as "spam". I believe you meant non-repudiable.

Better solution where technology exists, policy ez

[isopropanol]

X.509/PKI user certificates. Have whatever department is responsible for passports issue certs for citizens, and whatever department is responsible for other legal entities (Corporations, societies, etc). As a bonus it also works for HTTPS.

Re:Better solution where technology exists, policy

[Rhinobird]

I always thought it would be neat to take a thumb drive of public keys, and a photo ID and have the post office sign them. Maybe a yearly fee to have the USPS host the public keys on the internet.

Yes

[Arancaytar]

This is a completely retarded idea. It was thought up by people who think email works like the postal service. What it does great is accumulate control and bureaucracy where it is not needed; what it does badly is any kind of security.

If the federal government of Germany wanted to actually effectively help people secure their online communication, they would certify actual end-to-end encryption and electronic signature programs for official use, and provide some kind of root CA (or the PGP equivalent). Instead, we will have an incompatible reinvented email implementation that will, based on the German government's track record with electronic passports, be buggy, riddled with critical vulnerabilities and badly supported on non-Windows systems, if it will even be accessible without the web at all.

Re:Yes

[hesaigo999ca]

The idea that this idea is retarded, is retarded.
Simply put, when you add a pay per use campaign on emails, not only does it give you more of a paper trail, it allows the people sending out spam to know they are infected, and after a few times of being charged by their ISP for this extra email, of which I agree should be a cap so as not to offend too much those owners, maybe say a cap of 25$ a month for all the emails their infected pcs put out.....then they could be made to know that this will occur every month unless they clean their machines, this will promote a much safer and secure internet.

The problem is people don't care right now, because they are running pirated windows, and do not bother getting their updates...so their machines are compromised and are not aware....even if they are legit with windows, they could still be infected, and not know it, cuz between u and me, most AVs today suck hairy balls.

I got to tell you , I am super glad they are coming out with this, because it is a first step at least in the right direction, someone else will come out with their version, and so on, and so on....and then we will have options....then we will find a good one of these spam killers...and then all will be good on the internet to play as you wish, spammers will have dried up and died, and the industry will know much less bandwith usage because of less spam, who knows, maybe we will all start to see discounts on the over all bandwidth charges as the ISPs need to spend less on bandwidth and spam filters....

stay away from it !

[datadefender]

I had signed up for an account just to play with it. Then I read the T&C's. Once I did though, I instantly deleted my account. Any email send to you is treated like a registered letter. They require you to check your mailbox every 24h (maybe it was more - cannot recall). So you could really miss deadlines. Its not only that nobody needs this (we already have S/Mime and PGP/GPG) - it can actually be harmful to you to have an account. Therefore: under no circumstances use DE-Mail - don't even get an account - and if you have one - cancel it right away.

finally

[hesaigo999ca]

after 10 years of posting about this, the germans come out with it, its about bloody time!, now we will see a sharp decline in spam emails....just you wait and see. Siting past posts does nothing for my karma, but if you want to see some of them, just check some rants and raves from my past about email spamming.