Comments on the National Identity Management Systems Act (2021)

Dr. Ronnie Yearwood and Niel Harper recently collaborated to provide expert comments on the National Identity Management System Act (2021) just passed by the Government of Barbados. Given that this piece of legislation was quickly passed with no opportunities for public debate or feedback, we felt it necessary to articulate and ventilate some of our key concerns with the statute in its current form.


Disability and Accessibility

  • In line with the obligations under the United Nations Convention on the Rights of Persons with Disabilities, there are no provisions in the Act for mandatory accessibility features in the digital ID and related services. As such, persons with disabilities may be excluded.

Oversight and Liability

  • There is no mention of a supervisory and oversight body that ensures the digital ID system is used for its intended purposes (to prevent abuse and misuse), to audit and certify the digital ID provider and third-party trust services, to address complaints, and ultimately provide redress.
  • There is no mention of the liability to be assumed by the government or trust services providers to ensure due diligence, transparency and accountability of their operations and services related to the digital ID. The digital ID service provider (Government) and trust services providers should be liable for damage caused to any natural or legal person due to failure to implement robust privacy and security controls or otherwise disadvantage individuals via the delivery of the digital ID system.

Breach notification

  • The Act does not speak to data breach notification and the relationship between this statute and the Data Protection Act (2019) which is critically important.  Furthermore, the Office of the Data Commissioner does not have the staffing or capabilities to oversee the various activities related to large scale data collection and processing.

Comprehensive digital ID ecosystem

  • The Act does not comprehensively cover electronic signatures, electronic seals, time stamps, electronic documents, and website authentication. The legal effect of the above needs to be clearly defined to avoid confusion. Existing practices, standards and legislation exist that can be built upon to address these matters which are integral to a functional digital ID system. Without those features, the Government will essentially be replacing the existing physical ID cards and not truly realizing the value of a digital ID ecosystem that delivers identity, authentication and trust services.


  • The Act does not speak to an interoperability framework that guarantees the digital ID system is built using open standards and can be seamlessly integrated into national and cross-border digital identity ecosystems.


Discrimination and equality before the law

Section 5 (9) “A person who is a visitor shall not be eligible for registration in the National Register unless that person is a person to whom subsection (1) applies.

(Section 5(1) covers persons, for example born in Barbados or citizens of Barbados who “shall be registered in the National Register.”)

  • The point is that a person who is a visitor to Barbados shall not be eligible for registration in the National Register unless section 5(1) applies.
  • Is it that only Barbadians and persons resident in Barbados must register to gain access to public services (see section 5(10)) regarding the fact that if you are not registered under the Act you cannot get a national registration number, cannot be added to the electoral register to vote, cannot obtain a permit to drive, or qualify to access any goods or services requiring presentation of the ID?
  • This looks somewhat discriminatory because the same requirement does not seem to be placed on foreigners for any access to services. I have not seen a reason for this proposed by the government.

(Also see section 12(1) reads: “A person who is issued an identification card may be required to produce his identification card (c) for the purpose of voting in an election in Barbados; (d) for the purpose of accessing goods or services provided by the Government or the private sector… and that identification card shall be prima facie evidence of the identity of the person shown on the identification card…”)

Voter’s rights, registration and identification

Section 5(10)(d) “A person who is not registered under this Act shall not qualify to be added to the register of electors or the revised register of electors prepared under the Representation of the People Act, Cap. 12

Section 34(1) An identification card authorised under section 25 of the Representation of the People Act, Cap. 12 or under the Statistics Act, Cap. 192 shall remain valid for a period of 12 months from the date of the commencement of this Act.

  • Therefore, section 34(1) provides that an ID card under the Representation of the People Act shall only remain valid for 12 months from the commencement of the new ID law. When has the Act been commenced?

Section 12(1)(c) “A person who is issued an identification card may be required to produce his identification card for the purpose of voting in an election in Barbados.”

  • This needs clarification as there should be more than one valid piece of identification to enable voters’ rights.

Section 12(2) “Where a person is unable to produce his identification card for the purposes mentioned in subsection (1), the person authorised to require such identification shall, unless another form of identification is authorised by law, defer consideration of the person or refuse access until such time as the relevant identification card is produced.”

  • This does not appear to make sense because what other forms of ID are relevant if every other ID is being subsumed by the digital ID based on the Government’s argument that the digital ID is to prevent fraud and bring efficiency in one form of ID.
  • In the alternative if we consider that other forms of ID can be considered as noted in the law, it is not stated what these forms of ID are and section 5(10) it states that a person not registered under the law shall not, “qualify to access any service or goods which require the presentation of an identification card to obtain such services or goods.” So, in effect the digital ID becomes the only way to access public services.
  • Overall, sections 12 (1) and 12 (2) links the provision of universal public services to the digital ID and denial of such services for failure to produce the digital ID. Section 18 also links the ID to access of public services.
  • If the point of the digital ID is to confirm a person is who they are to access the service, then how it is rational to exclude other forms of ID, while someone is denied access to vital public services, (which we assume to be health, welfare, education as the Act does not define or specify these services).
  • Section 12 does not define these goods and services.
  • Is section 12(d) proportionate? Does it achieve what Government claims the law should address, i.e., identity theft? However, what identity theft occurs if someone presents themselves for universal public services such as medical treatment or other related public services?
  • Can the Government realistically transition the entire population off the existing ID cards in 12 months? What contingencies are in place if this is not achieved? Most importantly, this appears to suggest that after 12 months the only valid form of identification to vote will be the digital ID card. Can this lead to voter disenfranchisement and violation of the individual’s constitutional rights? Should not other forms of identification such as a verified passport or driver’s license be satisfactory to allow an individual to vote?

Driver’s licence

5(10)(e) “A person who is not registered under this Act shall not (e) qualify to obtain a driver’s permit or licence”

  • Is this disproportionate, and could harm chances for employment as well as affect mobility?  What is the justification?

Access to Goods and services

5(10)(f) “A person who is not registered under this Act shall not qualify to access any service or goods which require the presentation of an identification card to obtain such services or goods.”

  • This is exclusionary in nature and also does not take into consideration that other forms of identification are valid.
  • Also, there is no definition of goods and services.

Fingerprinting – Optional, as a “refusal”

Section 7(5) Where a person refuses to consent to submitting fingerprints, that refusal shall be indicated in the applicable field set out for that purpose in the National Register

(a) by affixing his name, signature or mark; or

(b) where a person is unable to affix his name, signature or mark the registering officer shall indicate on the Certificate of Registration that the person is unable to sign,

and the person shall not be subject to any penalty, fine or term of imprisonment.

  • Fingerprinting is framed as a “refusal” which is indicated in the Register as some sort of mark against a person for not wanting to have their fingerprints taken, and in that the individual is not subject to penalty. However, how is it an individual would be subject to penalty for something that is optional?

Privacy & Security

Section 19(1) “The Commission shall ensure the security and confidentiality of the records of a person registered.”

Section 19(2) Commission shall take measures including security safeguards to ensure that the information in the possession or control of the Commission, including information recorded in the National Register or embedded in the chip of the identification card or the national identity credential, is secured and protected against any loss, unauthorised access or use or unauthorised disclosure thereof.

  • These sections need more robustness. Evaluation of security and privacy controls should be based on international standards. It should not be left up to the Commission to determine the adequacy and effectiveness of the controls around security and privacy. Standards such as ISO 15408, ISO 27001, ISO 27701, and others are worthwhile mentioning.

Data disclosure

Section 20(1)

A Commissioner, an officer or an employee of the Commission or an expert retained to assist the Commission shall comply with the provisions of the Data Protection Act, 2019 (Act 2019-29) and shall keep confidential all information coming to his knowledge during the performance of his functions under this Act or any enactment which relates to the private affairs of a person except insofar as disclosure is necessary for

(a) the administration of this Act or any regulations made thereunder;

(b) compliance with the provisions of any enactment; or

(c) compliance with an order of a court.

  • Data can be disclosed not only to a Commissioner but to an officer or an employee of the Commission overseeing the register, or an expert retained by the Commission. An individual’s information can be disclosed to a third party. This is in the administration and compliance with the law, and compliance with a court order. But how does an individual challenge such disclosure or have a right to be heard on the disclosure? Is the individual made aware? What about issues of privacy and procedural fairness?
  • What about obtaining informed consent from data subjects if their information is being shared with a third-party? This is in contradiction with the Data Protection Act and other global data protection laws such as GDPR (EU), CCPA (California), PIPA (Canada), and others.

Section 22(1) A person who is aggrieved by an act performed or a decision made by the Commission may make a complaint to the Data Protection Commissioner within 14 days of the act performed or the decision made by the Commission.”

  • The Electoral and Boundaries Commission and Office of the Data Protection Commissioner are both Government entities. Adequate oversight for the digital ID system cannot be achieved with this insufficient segregation of authority. There should be a collegial body in place to oversee this critical function. The body should consist of representatives from the government, private sector, academia, technical community, and civil society, with power vested equally to each member.
  • The complaints provision in section 22(1) does not appear to make sense in connection with section 20(1) on disclosure, if it is not clear that an individual or data subject does not know that their information was disclosed to a Commissioner or a third-party.

Section 23 “A person who is aggrieved by the decision of the Data Protection Commissioner pursuant to section 22 shall appeal to the High Court within 14 days of the decision.”

  • Processes for efficient redress in the High Court are not in place. This should be performed by an independent arbitration body with sufficient authority. The High Court is not the venue for dispute resolution (arbitration), and this should be addressed in the Act in a clear manner.


Section 26(a-i)

  • This appears to be currently addressed in the Computer Misuse Act 2005. Why not just refer back to this statute as opposed to creating offence/sentencing fragmentation?
  • The Act imposes a fine of $100,000 and/or 3 years in jail and the Computer Misuse Act imposes a fine of $50,000 and/or 5 years in jail for basically the same crime.

Sections 27 and 28.  Same as above.

Data minimisation

  • First Sch, number 25 “Other relevant information required by the Commission”

This appears to point to mission or data creep, so that different kinds of information can be added and centralized, but where is the consultation/oversight for this broad category?

Cybersecurity pros are badly in need of MENTORS: And here’s why…

Finding and keeping cyber-talent is a top global concern for public- and private-sector organizations alike. Yet, the prevailing theory among industry analysts is that there is a talent crisis, with ‘experts’ predicting that by 2022 there will be more than 1.8 million unfilled jobs.

The above graphic highlights one of the industry’s most glaring shortcomings: Everyone wants to hire cybersecurity pros, but no one wants to develop, guide, instruct and enhance the career effectiveness of inexperienced/entry-level candidates. It’s a self-destructive, self-refuelling, self-fulfilling prophecy – And it NEEDS to STOP! We simply don’t have an assembly line of top-tier, experienced cyber pros to choose from.

So how do we develop the next generation of cybersecurity leaders? What are some of the individual actions veteran security leaders can take? How do we help those without the finances to obtain expensive security training and certifications? What role does the government have to play?

There are multiple dimensions to the institutionalisation of cyber capacity building. For example, there’s a national response and an enterprise response — and ideally the two should be coordinated (but most often are not).

There are established commercial training and certification programs, which can verify the capabilities of individuals. However, while these certifications can be used to get hired, organizations still have to continuously invest in their employees’ development. This is particular important given how rapidly the threat landscape changes.

From a national perspective, capabilities need to be developed to build trust in the online systems that underpin the digital economy. Part of building trust is creating a workforce of cyber pros to address key threats. Government should create a workforce development program as part of a national cybersecurity strategy, and it should address training at the college, university and professional certification levels.

But in the absence of such actions by corporations or countries, we cybersecurity leaders need to take up the charge. We need to commit to mentoring as many young professionals as we humanly can. It’s not only incumbent upon us to support their career progress, but also to give back to the profession as well as contribute to the overall trust model that underpins the global Internet. Let’s do our part!

6 Tips for Protecting Against Ransomware

The Internet Society has been closely monitoring the ransomware cyber-attacks that have been occurring over the last couple of days. The malware, which has gone by multiple names, including WannaCry, WannaDecryptor, and WannaCrypt, exploits a flaw in Microsoft Windows that was first reportedly discovered by the National Security Agency (NSA). A group of hackers leaked the code for exploiting this vulnerability earlier this year, and a fix or patch was available as far back as March 2017. Since Friday, 200,000 computers in 150 countries have been compromised using this exploit. The numbers are expected to grow exponentially as people settle back into their work routines and regular use of computer systems this week. As part of our continuing work in online trust and security, there are some key takeaways from this incident that we want to leave with our community.

Firstly, we want to highlight the extremely negative effects which government stockpiling of vulnerabilities and zero day attacks has on the overall security of the Internet. With over 60 countries known to be developing growing arsenals of cyber weapons, and with many of these exploits leaking into the public domain, the potential for widespread damage is a massive cause for concern. The impact is not only economic in terms of financial loss, but social in terms of how it impacts end user trust, and most importantly human in terms of loss of life (especially given that ransomware attacks have been focusing on hospitals). And with critical infrastructure like power plants, dams, and transportation systems being targeted in nation state cyber offensives, the threat to human life increases exponentially.

Secondly, it would appear that some hospitals are easy targets for ransomware attackers. Their systems house data that is critical to patient care and management, and many of these institutions don’t have the IT resources to support critical process areas like vulnerability management, patch management, business continuity management, etc. In general, hospitals are also now adapting to digital realities and a number of them are playing catchup with regards to cyber readiness. However, the aforementioned challenges are not unique to hospitals, and are faced by many small and medium enterprises (SMEs), and in several instances, large corporations. Individual users are also targeted based on their generally poor Internet hygiene or lack of security awareness.

We want to take this opportunity to emphasize the importance of good online security practices when accessing the Internet. So here are 6 basic tips for protecting against ransomware:

1. Employ strong, multi-layered endpoint security – Using endpoint security that can protect web browsing, control outbound traffic, protect system settings, proactively stop phishing attacks and continuously monitor for anomalous system behavior will allow for better protection of servers, laptops, tablets, and mobile devices.

2. Maintain regular backups of your critical data – Backups can help you to protect your data from more than just ransomware. Other risk events such as malware, theft, fire, flood or accidental deletion can all render your data unavailable. Be certain to encrypt your backed-up data so it can be effectively restored. Backups should also be stored at an offsite location isolated from the local network.

3. Do not open unsolicited emails or messages from unknown senders – Many ransomware variants are distributed through phishing attacks or email attachments. Increased mindfulness when handling ‘suspect’ emails can be effective in combating ransomware.

4. Patch your systems regularly – Patching your systems for vulnerabilities reduces the opportunities for hackers to infect you with ransomware. The fact that a patch was available for the WannaCrypt vulnerability since March highlights the somewhat lax attitude by organizations and individuals to keeping their system patches up to date. That being said, patch management is a complex activity and can impact the availability of key systems. Hence, thorough testing must be conducted to avoid unplanned downtime.

5. Disable macros if possible – Many forms of ransomware are distributed in Microsoft Office documents that attempt to trick users into enabling macros. There are a number of tools available that can limit to functionality of macros my preventing them from being enabled on files downloaded from the Internet.

6. Be aware and vigilant – For individuals, don’t assume that only techies need to know about all the recent malware and trends in online attacks. Subscribe to mailing lists that provide information on common vulnerabilities and exposures. In the case of organizations, developing an information security awareness program is an integral part of improving overall security posture.

Finally, we want to touch on the important work being done by the Online Trust Alliance (OTA), the Internet Society’s newest initiative. The OTA’s mission is to enhance online trust, user empowerment and innovation through convening multi-stakeholder initiatives, developing and promoting best practices, ethical privacy practices and data stewardship. With regards to preventing ransomware attacks, OTA has developed a number of industry best practices that address key threat areas such as email authentication and incident response. These are as follows:

Email Authentication:

Domain-based Message Authentication, Reporting & Conformance (DMARC):

Cyber Incident & Breach Response:

Additional OTA best practices, resources and guidance to help enhance online safety, data security, privacy and brand protection can be found here.

The Spam Toolkit developed by the Internet Society also provides some guidance on addressing online threats.

The Internet Society is committed to the enhancement of online trust, and our work along this vein spans multiple areas. Our goal is to continue to provide our individual members, organizational members, chapters, partners, and other constituents with timely and relevant information and resources that equip and empower them to act.

My original blog article was published on the Internet Society website at: