Module 2: Legislative and Policy options to Promote e-Commerce and Expand Internet Use

III. E-commerce legislation

Providing for Rules that Will Give Functional Equivalence to Electronic Signatures and Documents

A. From a Legal Perspective, What Should SMEs Be Concerned about If They Plan to Engage in e-Commerce?

The growth of e-commerce depends on the legal enforceability of electronic contracts and electronic signatures. Without such legal clarity, people and companies will and should be very hesitant to engage in electronic commerce.

Without this assurance, SMEs will not be able to fully assess the risks of doing e-business: such as whether there is a likelihood of the transaction being able to be successfully completed, whether it can be challenged, and whether the recipient will have legal recourses in such circumstances, irrespective of the location of the parties.

B. What Sort of Legal/Regulatory Regime Should Be in Place to Provide This Assurance?

The legal regime that will allow e-commerce to grow to its potential, therefore, must first and foremost be one that gives electronic transactions, contracts and signatures the same validity and enforceability as traditional transactions, contracts and signatures.

For this to happen, there must first be a mechanism that will (a) reliably and securely prove the origin, receipt and integrity of information, (b) identify the parties involved, and (c) associate those parties with the contents of the communication.

Without such a mechanism, it would be very difficult to rely on electronic contracts and electronic signatures, especially in cases where the parties have not met or do not really know each other – which would likely be the case in a globalized economy.

In other words, electronic documents and signatures must be functionally equivalent to traditional documents and signatures.

To illustrate, if Person A sends an electronic document over the Internet to Person B, ideally, Person B should be assured of the following:

1. Data Origin Authentication. First, that the message really did come from Person A. That is, Person B must have some assurance that the message has in fact come from its purported sender, Person A.

2. Message Integrity. Second, that the message received by Person B is the exact message that Person A sent. Person B should be able to verify that the message has not been intentionally or accidentally altered during transmission.

3. Non-Repudiation. Finally, that Person A cannot later deny that he or she did in fact send the message. No one else should have been able to send the message but Person A, and Person B should be able to prove it unequivocally.

Whatever the methods or technologies used to achieve these objectives, the crucial factor is trust. The issue then, for electronic commerce, is how to build trust and confidence in electronic transactions on open networks between parties who may have no pre-existing relationship.

C. What Are Electronic and Digital Signatures?

“Electronic signature” generally refers to any distinctive mark, characteristic and/or sound in electronic form, representing the identity of a person and attached to or logically associated with the electronic data message or electronic document or any methodology or procedures employed or adopted by a person and executed or adopted by such person with the intention of authenticating or approving an electronic data message or electronic document” (Philippines Republic Act 8792, or the Electronic Commerce Act of 2000).

There are many methods for creating an electronic signature. These methods range from simple ones – such as typing a name at the bottom of an email message – to more complex and secure ones – for example, biometric technologies, such as fingerprints or retinal scans.

There are yet more types of authentication methods, such as magnetic strip cards with personal identification numbers (PIN), user names and passwords, public key cryptography, writing tablets with electronic pens and even smart cards that generate a unique access code every few seconds. As technology advances, the list of viable electronic signature alternatives is sure to grow.

D. Are Electronic Signatures the Same as Digital Signatures?

Note that technically speaking, electronic signatures are different from digital signatures. A “digital signature” does not refer to the image of a signature in any way. Unlike both an “electronic signature” which is simply any form of mark intended to be a signature, and a “digitized signature” which refers to an electronic image of a signature, a “digital signature” is actually a term of art that refers to scrambling data in order to provide security and authentication.

Digital signatures are created and verified using cryptography, the branch of applied mathematics that concerns itself with transforming messages into seemingly unintelligible form and then back into its original form.

E. How Do Digital Signatures Work?

Military communications have relied on encryption for thousands of years. In fact, Alexander the Great communicated with his generals by sending messages in which each letter was shifted a certain number of positions. This was a form of “secret key encryption,” i.e., anyone who knew the secret code (or key) would be able to send and receive messages securely.

Today, commercially available encryption software creates encryption so strong that it is all but impossible to break the code and ascertain the original message without the use of the authorized software.

To be secure, a secret-key coding system requires some method of distributing the secret key to intended users, without it falling into the hands of other parties.

Now, before going further into the basics of cryptographic and digital signature technology, a few terms need to be defined first.

“Encryption” is simply the process by which information is scrambled by use of a code.

A “hash function” is a process that creates a relatively small number that represents a much larger amount of electronic data. For instance, if I had a ten-page word processing document on my computer, I could use special hashing software to derive a particular number associated with that document.

If even one comma is changed on the document, the resulting hash number from the changed document would be completely different. This number is called the “message digest.”

Digital signatures use a “one-way hash function,” i.e., there is no way to reverse engineer or derive the content of the message based on the resulting message digest. When a digest is sent along with the message, the recipient can check to see if the message had been tampered with by using the same hashing software to make her own digest of the message and then checking to see if the two numbers match.

1. Public Key Cryptography

By its nature, the Internet is poorly suited for a secret-key system because it is an “open” network in which a message may make several “stops” or “hops” before arriving at its final destination.

Figure 1. Message hops from origin to destination on the Internet

This creates a serious risk that a third party could intercept a secret key at some point along its routing, which would allow the third party to read messages, or even send encoded messages purporting to be from the authorized holder of the key.

Physically delivering a secret key to every user through a secure channel, on the other hand, would be slow, expensive and unwieldy. It would effectively rule out one-time transactions between people and firms that have not previously exchanged secret keys.

Public key cryptography eliminates the need for users to share a secret key, which makes it ideally suited for communications over open networks such as the Internet.

In a public key system, each user has a type of software that generates two related keys, a public key and a private key. The fundamental characteristic of this pair of keys is that only this particular public key can decrypt a message encrypted with its corresponding private key and vice versa. Similarly, only this particular private key can decrypt a message encrypted with its corresponding public key.

The process is illustrated in the two diagrams below.

Figure 2. Sending a digitally signed message

Let us assume that (1) an SME owner has a business proposal that he wants to digitally sign and send to you. He would then (2) run his message through one of several standard algorithms known as a “hash function” that performs a series of mathematical operations on the original message. The hash function produces a number called a (3) “message digest” which can be thought of as a fingerprint of the message, because any change in the message, no matter how slight, will cause the hash function to produce a completely different message digest. (4) Using his private key, the SME owner then encrypts the message digest. The message digest encrypted with the SME owner’s private key forms the actual (5) “digital signature” for the message. Both the digital signature and the actual message are then sent to you.

Figure 3. Verifying a digitally signed message

Upon receipt of the message, your computer and software would then perform two separate operations to verify the SME owner’s identity and to determine if the message had been altered in transit.

To verify his identity, your system would (1) take the digital signature and (2) use the SME owner’s public key to decrypt the digital signature, which would then (3) produce the message digest. If the operation is successful, you would then know for a fact that the SME owner (who alone has access to his private key) must have sent the message.

In order to ensure that the message had not been altered, your system would (4) run his message through the (5) same hash function that was used, which would then (6) yield a message digest of the message. You would then be able (7) to compare the two message digests, and if they are identical, confirm that the message has remained unaltered in transit.

Generally then, users of this system would keep their private key very safe (perhaps password-protected, or even embedded in a smartcard) but they would make their public key freely available by sending it to all potential recipients of messages or posting it to an Internet public key directory.

In this way, the private key holder (in the example, the SME owner) can send a message to anyone on the Internet, and, if his public key decrypts the message, the recipient knows it must have come from the private key holder.

Conversely, anyone on the Internet who wants to send the private key holder a message can encrypt the message with his public key and send the message with the knowledge that only the private key holder can read the encrypted text.

Note that all the above processes would actually happen automatically and in the background, and you would only be made aware if something is wrong with the verification and the process reveals that there is reason to doubt the integrity of the message or sender – as in the case of credit card verification processes.

2. Public Key Infrastructures and Certification Authorities

The process of public key cryptography described above can work well between parties who know each other.

But what happens in transactions between parties who have never met each other before?

In the example above, how would you know for certain that the SME owner, and not someone else posing as him, did in fact send the message?

In general terms, how can one bind the identity of a particular party to a particular public key? This need has been widely perceived in the marketplace, and several companies are stepping into the so-called “trusted third party” business. Such a company is known as a “certification authority” (CA).

The CA essentially vouches for the identity of a person who subscribes to their service. It issues a certificate that, in effect, guarantees the identity of the person (or subscriber) associated with a given public key. The CA is responsible for undertaking certain measures to ascertain the identity of the person to whom it issues a certificate. The certificate issued by the CA:

• Identifies the CA issuing it;

• Identifies the subscriber;

• Contains the subscriber’s public key; and

• Is digitally signed with the CA’s private key.

The digital certificate may also contain additional information, including a reliance limit, or a reference to the CA’s “certification practice statement” that gives relying parties notices of the level of inquiry conducted by the CA before issuing the certificate.

Thus, if the SME owner wished to use a CA to vouch for his identity on the Internet, he would have to present the CA with a copy of his public key along with sufficient proof of his identity (or else the CA could also issue his private and public keys). Once satisfied with the identity of the SME owner, the CA would issue him a digital certificate.

Going back to our example, and as shown in the diagram below, the SME owner will send you, along with his digital signature, a copy of his digital certificate.

In addition to the steps described above, upon receipt of the message, you may also confirm with the CA identified in the digital certificate that the sender is who he says he is, and that his certificate has not expired or been revoked.

Again, note that all these activities would be happening in the background, and nearly instantaneously, such that you may not even be aware, or might take for granted, that all these verification processes are actually happening.

Figure 4. Certificate authorities, digital certificates and digitally signed messages

F. A Note on the Concept of “Technological Neutrality

One key issue that often arises when governments attempt to make rules related to the Internet is the question of technological neutrality. That is, when crafting laws or rules, governments must be aware that their actions could have an effect on the development of technology itself to the extent that the laws or rules they craft encourage or discourage use of, or investments in, particular technologies.

The subject of electronic authentication presents a good example.

Rule-makers are more confident in the security and reliability of known electronic authentication methods (such as public key infrastructures). This confidence allows them greater room to grant legal benefits and presumptions to the use of those techniques. They are generally less willing to grant the same level of legal benefits to as yet unknown techniques or to technologies that are not as well known or used.

G. Should the Government Formally Endorse the Validity of Digital (or PKI) Signatures?

1. The arguments for formally recognizing the validity of PKI signatures

Advocates of legislation to govern CAs argue that laws and rules are needed to address a number of key issues even before such signatures are widely used. First, digital signatures must be given the same legal force as traditional signatures. Second, CAs need to be licensed, or at least regulated by the State, to ensure that they are technically proficient, financially sound and operationally secure. And finally, legislation is needed to shield CAs from potentially crippling liability if they have complied with the requirements set by law.

Moreover, they believe that pro-active rule-making is preferable to allowing the validity of digital signatures to be determined by the evolution of technical standards or business practices in the market. Legislation would produce uniformity among different jurisdictions, whereby they argue, not only domestically, but internationally as well. With such a legal infrastructure in place, they believe electronic commerce will gain a broader acceptance because parties to online transactions will be able to use digital signatures that are not only secure, but are also legally enforceable.

2. The arguments against CA legislation/rules

Critics of the legislation point out that proscriptive, technology-specific legislation runs the risk of distorting the market, thus preventing the natural evolution of best business practices, technological innovations and competitive pricing. Many observers believe that detailed statutory and regulatory treatment is simply inappropriate in an infant industry undergoing rapid change.

They also point out that the non-regulatory environment of the Internet allowed it to evolve solely in response to advances in technology, the creativity of providers and the needs of users, rather than in conformance with detailed strictures laid down by government bureaucrats. Freed from government mandates, producers and consumers were able to quickly and easily adapt to new technologies and business models. This was the dynamic that caused such an explosive growth in the Internet and it should not be ignored by policymakers when they consider the best ways to promote electronic commerce and/or authentication.

Furthermore, particularly following the events of 9/11, there has been a growing recognition that other means of electronic authentication, including biometrics and voice authentication, may take on equal or even greater importance in the years ahead. In fact, some of these techniques – and particularly, those that are based on biometric features – may prove to be more reliable and less susceptible to compromise than digital signatures.

The danger then is that specific rules for public key infrastructures could in effect be seen as an endorsement of a specific technological option. This legislative or state-sanctioned endorsement could have the unintended effect of stunting the development of other authentication mechanisms.