The Role of Identity Management in the GDPR

Unless you have been living in a darkened room for a long time, you will know the countdown for the EU's General Data Protection Regulation is dramatically coming to a head.  May 2018 is when the regulation really takes hold, and organisations are fast in the act on putting plans, processes and personnel in place, in order to comply.

Whilst many organisations are looking at employing a Data Privacy Officer (DPO), reading through all the legalese and developing data analytics and tagging processes, many need to embrace and understand the requirements with how their consumer identity and access management platform can and should be used in this new regulatory setting.

My intention in this blog, isn't to list every single article and what they mean - there are plenty of other sites that can help with that.  I want to really highlight, some of the more identity related components of the GDPR and what needs to be done.

Personal Data

On the the personal data front, more and more organisations are collecting more data, more frequently than ever before.  Some data is explicit, like when you enter your first name, last name and date of birth when you register for a service for example, through to the more subtle - location, history and preference details amongst others. The GDPR focuses on making sure personal data is processed legally and data is only kept for as long as necessary - with a full end user interface that has the ability to make sure their data is up to date and accurate.

It goes with out saying, that this personal data needs to have the necessary security, confidentiality, integrity and availability constraints applied to it.  This will require the necessary least privileged administrative controls and data persistence security, such as the necessary hashing or encryption.

Lawful Processing

Ah the word law! That must be the legal team. Or the newly appointed DPO. That can't be a security, identity or technology issue.  Partially correct. But the lawful processing, also has a significant requirement surrounding the capture and management of consent.  So what is this explicit consent? The data owner - that's Joe Blogs whose data has been snaffled - needs to be fully aware of the data that has been captured, why it is captured and who has access.

The service provider, also needs to explicitly capture consent - not an implicit "the end user needs to opt out", but more the end user needs to "opt-in" for their data to be used and processed.  This will require a transparent user driven consent system, with sharing and more importantly, timely revocation of access.  Protocols such as User Managed Access may come in useful here.

Individuals Right to be Informed

The lawful processing aspect, flows neatly into the entire area of the end user being informed.  The end user needs to be in a position to make informed decisions, around data sharing, service registration, data revocation and more.  The use of 10 page terms and conditions thrust down the end user's screen at service startup, are over.

Non-tech language is now a must, with clear explanations of why data has been captured and which 3rd parties - if any - have access to the data.  This again flows into the consent model - with the data owner being able to make consent decisions, they need simple to understand information.  So registration flows will now need to be much more progressive - only collecting data when it is needed, with a clear explanation of why the data is needed and what processing will be done with it.  20 attribute registration forms are dead.

Individuals Right to Rectification, Export and Erasure

Certainly some new requirements here - if you are a service provider, can you allow your end users to clearly see what data you have captured about them, and also provide that data in a simple to use end user dashboard where they can make changes and keep it up to date?  What about the ability for the data owner to export that data in a machine readable and standard format such as CSV or JSON?

Right to erasure is also interesting - do you know where your end user data resides?  Which systems, what attributes, what correlations or translations have taken place?  Could you issue a de-provisioning request to either delete, clean or anonymize that data? If not you may need to investigate why and what can be done to remediate that.


Conclusion

The GDPR is big.  It contains over 90 articles, containing lots of legalese and fine grained print.  Don't just assume the legal team or the newly appointed DPO will cover your company's ass.  Full platform data analytics tagging will be needed, along with a modern consumer identity and access management design pattern.  End user dashboards, registration journeys and consent frameworks will need updating.

The interesting aspect, is that privacy is now becoming a competitive differentiator.  The GDPR should not just be seen as an internal compliance exercise.  It could actually be a launch pad for building closer more trusted relationships with your end user community.

The Role of Identity Management in the GDPR

Unless you have been living in a darkened room for a long time, you will know the countdown for the EU's General Data Protection Regulation is dramatically coming to a head.  May 2018 is when the regulation really takes hold, and organisations are fast in the act on putting plans, processes and personnel in place, in order to comply.

Whilst many organisations are looking at employing a Data Privacy Officer (DPO), reading through all the legalese and developing data analytics and tagging processes, many need to embrace and understand the requirements with how their consumer identity and access management platform can and should be used in this new regulatory setting.

My intention in this blog, isn't to list every single article and what they mean - there are plenty of other sites that can help with that.  I want to really highlight, some of the more identity related components of the GDPR and what needs to be done.

Personal Data

On the the personal data front, more and more organisations are collecting more data, more frequently than ever before.  Some data is explicit, like when you enter your first name, last name and date of birth when you register for a service for example, through to the more subtle - location, history and preference details amongst others. The GDPR focuses on making sure personal data is processed legally and data is only kept for as long as necessary - with a full end user interface that has the ability to make sure their data is up to date and accurate.

It goes with out saying, that this personal data needs to have the necessary security, confidentiality, integrity and availability constraints applied to it.  This will require the necessary least privileged administrative controls and data persistence security, such as the necessary hashing or encryption.

Lawful Processing

Ah the word law! That must be the legal team. Or the newly appointed DPO. That can't be a security, identity or technology issue.  Partially correct. But the lawful processing, also has a significant requirement surrounding the capture and management of consent.  So what is this explicit consent? The data owner - that's Joe Blogs whose data has been snaffled - needs to be fully aware of the data that has been captured, why it is captured and who has access.

The service provider, also needs to explicitly capture consent - not an implicit "the end user needs to opt out", but more the end user needs to "opt-in" for their data to be used and processed.  This will require a transparent user driven consent system, with sharing and more importantly, timely revocation of access.  Protocols such as User Managed Access may come in useful here.

Individuals Right to be Informed

The lawful processing aspect, flows neatly into the entire area of the end user being informed.  The end user needs to be in a position to make informed decisions, around data sharing, service registration, data revocation and more.  The use of 10 page terms and conditions thrust down the end user's screen at service startup, are over.

Non-tech language is now a must, with clear explanations of why data has been captured and which 3rd parties - if any - have access to the data.  This again flows into the consent model - with the data owner being able to make consent decisions, they need simple to understand information.  So registration flows will now need to be much more progressive - only collecting data when it is needed, with a clear explanation of why the data is needed and what processing will be done with it.  20 attribute registration forms are dead.

Individuals Right to Rectification, Export and Erasure

Certainly some new requirements here - if you are a service provider, can you allow your end users to clearly see what data you have captured about them, and also provide that data in a simple to use end user dashboard where they can make changes and keep it up to date?  What about the ability for the data owner to export that data in a machine readable and standard format such as CSV or JSON?

Right to erasure is also interesting - do you know where your end user data resides?  Which systems, what attributes, what correlations or translations have taken place?  Could you issue a de-provisioning request to either delete, clean or anonymize that data? If not you may need to investigate why and what can be done to remediate that.


Conclusion

The GDPR is big.  It contains over 90 articles, containing lots of legalese and fine grained print.  Don't just assume the legal team or the newly appointed DPO will cover your company's ass.  Full platform data analytics tagging will be needed, along with a modern consumer identity and access management design pattern.  End user dashboards, registration journeys and consent frameworks will need updating.

The interesting aspect, is that privacy is now becoming a competitive differentiator.  The GDPR should not just be seen as an internal compliance exercise.  It could actually be a launch pad for building closer more trusted relationships with your end user community.

Rencontrez ForgeRock à SIdO Lyon, les 7 et 8 Avril

Salon Internet des ObjetsJe serai présent avec notre équipe au SIdO, l’événement 100% dédié à l’Internet des Objets qui aura lieu à Lyon les 7 et 8 Avril 2015.

Outre notre présence dans l’espace coworking pendant les 2 jours, Lasse Andresen, CTO de ForgeRock, animera un workshop avec ARM et Schneider sur la place de l’Identité dans l’Internet des Objets, le Mercredi 8 à 13h30.

N’hésitez pas à venir nous rendre visite dans l’espace coworking.


Filed under: General, InFrench Tagged: conference, ForgeRock, france, identity, internet-of-things, iot, Lyon, privacy, security

IoT World Forum Review: Interop, Data & Security

This week saw the 2 day Internet of Things World Forum conference take place in London. There is clearly a general consensus, that the IoT market is a multi-trillion dollar opportunity, through the implementation of items such as consumer wearables, embedded predictive failure components and data collecting sensors.



The rapid rise in connected devices and IoT ecosystems, is seemingly beingdriven by several key factors, includingfalling cost of both connectivity anddata storage. These lowering barriers to entry, coupled with more developer friendly ecosystems and open platforms, is helping to fulfil new revenue generating business opportunities in multiple verticals including manufacturing and healthcare.

Matt Hatton from Machina Research started off discussing the progression from local standalone projects (Intranets of Things), through to more internal or enterprise focused deployments (Subnets of Things).  David Keene from Google, extended this further, to say the progression will reach the concept of Web of Things, where accessibility and 'findability' will be key to managing and accessing data.

It was clear that data aggregation and analytics will be a major component in any successful IoT infrastructure, whether that is focusing on consumer enhancements, such as the Jaguar connected car project as described by Leon Hurst, through to smart health care, either in the form of Fitbits, or more advanced medical instrumentation.

API's and machine processing were certainly referenced more than once.  The new more connected web, will provide interaction touch points that only machines can understand, coupled with better data aggregation, distributed data storage and centralised querying. API's of course need protection too, either via gateways or via token management integration for standards such as OAuth2.

One aspect that was conspicuous in it's absence, was that of data privacy, and identity and access management.  The IoT landscape is creating vast amounts of data at stream like speeds.  The concept of little data (small devices in isolation) to big data (aggregated in cloud services) requires strong levels of authentication and authorization, at both the device, service and end user level.  The ability to share and transparently know where data is being accessed will be a key concern in the consumer and health care spaces.

Dave Wagstaff from BSquare, brought up the interesting concept, that many organisations are now subtly moving away from a product based business model, to a software and services based approach. With the the increased capability of devices, organisations now can perform much more in the way of remote monitoring, predictive failure and so on, where the end user really is just paying an insurance or subscription for their physical thing.

Bernd Heinrichs from Cisco followed a similar pattern, where he described the German view of Industry v4.0 (or 4.1...) where innovative production concepts are helping to reduce energy, increase uptime and generate better component output.

From a new market opportunity perspective, Francois Menuier from Morgan Stanley, observed that 6% of all consumers now own a wearable, with 59% of them using that wearable daily. In addition many wearable owners, argued that this was an additional purchase and not one to replace existing technology, solidifying the view that new market initiatives are available in the IoT world. However many consumer wearables generate huge amounts of deeply personal data that needs to be protected and shared securely.

Jon Carter from Deutsch Telekom went through the 7 steps for a successful IoT implementation, which ended with the two main points of applying a minimum viable product concept to design and also leverage secure and open platform.

Dr Shane Rooney from the GSMA focused his thoughts on security within the mobile network operator network, including the concept of device to device and device to service authentication, as well the the need for greater focus on data privacy.

Overall an interesting couple of days. Whilst most manufacturers and platforms are focused on interoperability and data management, identity and access management has a strong and critical role in allowing 3rd party data sharing and interactions to take place. It will be interesting to see if the 2015 and 2016 start to introduce these concepts by default.





Would You Sell Your Privacy for Service Improvement?

When you put the question so bluntly, most people would probably say no.  But in reality this is the common situation many users face when signing up to cloud services, applications and retail sites.

Think of the following common scenario:  you want to get a quote for car insurance / car valuation / current house price or similar.  You will probably be faced with several click through forms where you fill in the necessary product information.  But, and there's always a but, you then need to fill in some personal contact information as a minimum before you are provided with the information you're looking for.  A sort of exchange of data for data.  Just so happens yours is personal.  In addition, you may also need to sign away how that personal data is going to be used.  Perhaps marketing emails or letters via the service provider themselves, or perhaps by a 'trusted' third party.  A final, more subtle exchange of data, is that the service provider now clearly knows you are looking for new insurance / moving house / selling your car.  That is quite a powerful personal context to each of those scenarios.

Is The Exchange of Information Worth It?

This is obviously a subjective question as the information you exchange will have a different value to each data owner.  You could argue that personal contact information is pretty much public domain anyway.  They have been Yellow Page equivalents and directory enquiries facilities for decades.

Finding someones personal or work email address is also pretty trivial these days, simply as they are so commonly used to sign up to so many services.  Many also will take the view, that if someone does send you a marketing or spam email, your ISP filters will simply place it into a junk folder and no harm is done.

If in return, you receive a free white paper, document, quote, temporary access to a new service, or signup to a freemium product perhaps giving away some personal contact information is a good deal?

However, what happens to the data that is harvested?  Where does it end up?  Contact information is one thing, but adding in additional details such as your personal circumstances, how many people live in your property if it's for an insurance quote for example, or perhaps releasing your mobile number, could result in future impacts on your privacy.

Is There An Impact On Privacy?

Again this could be a subjective answer.  There have been many discussions in the past 36 months regarding government surveillance of both individuals and government officials.  I am not an advocate for or against surveillance, but was there a physical impact on the individual during the snooping?  Note the word during.  Many people talked about the invasion of privacy and human rights, but that was only after they knew they had been observed.  The reaction was generally a retrospective one, not an active one.  That is not to say it's less valid, but the context needs to be applied.  The same could be said regarding commercial use of personal data. It's all good, until it isn't.  Whilst no one uses your personal data maliciously, is there a problem to address?  This can probably be classified in the same file as the infamous 'unknown unknowns' approach to threat intelligence by the US government.

eCommerce, Digitization and 'Sticky' Customers

Many organizations have no interest in using personal data maliciously.  The increased digitization of previously physical services and the increased use of online retail, has lead many organizations trying to get a better picture of their consumers, customers and potential customers.  Note the subtle difference between a consumer (who is actively using a service perhaps without registration eg Google), a customer (paying for a service or good) and potential customer.  All have different characteristics from a marketing and customer servicing perspective.  A organisation wants to get the individual perhaps down a consumer --> customer --> repeat/upselling customer route as quickly as possibly, but with a stickiness towards the latter part of the journey - ie when they're a customer keep hold of them.

The information exchange at the beginning of that cycle is key to helping organisations follow that flow.  From a individuals perspective, there needs to be a strong service improvement or cost saving aspect in order to sacrifice some of the data that is being asked for.

By Simon Moffatt



Would You Sell Your Privacy for Service Improvement?

When you put the question so bluntly, most people would probably say no.  But in reality this is the common situation many users face when signing up to cloud services, applications and retail sites.

Think of the following common scenario:  you want to get a quote for car insurance / car valuation / current house price or similar.  You will probably be faced with several click through forms where you fill in the necessary product information.  But, and there's always a but, you then need to fill in some personal contact information as a minimum before you are provided with the information you're looking for.  A sort of exchange of data for data.  Just so happens yours is personal.  In addition, you may also need to sign away how that personal data is going to be used.  Perhaps marketing emails or letters via the service provider themselves, or perhaps by a 'trusted' third party.  A final, more subtle exchange of data, is that the service provider now clearly knows you are looking for new insurance / moving house / selling your car.  That is quite a powerful personal context to each of those scenarios.

Is The Exchange of Information Worth It?

This is obviously a subjective question as the information you exchange will have a different value to each data owner.  You could argue that personal contact information is pretty much public domain anyway.  They have been Yellow Page equivalents and directory enquiries facilities for decades.

Finding someones personal or work email address is also pretty trivial these days, simply as they are so commonly used to sign up to so many services.  Many also will take the view, that if someone does send you a marketing or spam email, your ISP filters will simply place it into a junk folder and no harm is done.

If in return, you receive a free white paper, document, quote, temporary access to a new service, or signup to a freemium product perhaps giving away some personal contact information is a good deal?

However, what happens to the data that is harvested?  Where does it end up?  Contact information is one thing, but adding in additional details such as your personal circumstances, how many people live in your property if it's for an insurance quote for example, or perhaps releasing your mobile number, could result in future impacts on your privacy.

Is There An Impact On Privacy?

Again this could be a subjective answer.  There have been many discussions in the past 36 months regarding government surveillance of both individuals and government officials.  I am not an advocate for or against surveillance, but was there a physical impact on the individual during the snooping?  Note the word during.  Many people talked about the invasion of privacy and human rights, but that was only after they knew they had been observed.  The reaction was generally a retrospective one, not an active one.  That is not to say it's less valid, but the context needs to be applied.  The same could be said regarding commercial use of personal data. It's all good, until it isn't.  Whilst no one uses your personal data maliciously, is there a problem to address?  This can probably be classified in the same file as the infamous 'unknown unknowns' approach to threat intelligence by the US government.

eCommerce, Digitization and 'Sticky' Customers

Many organizations have no interest in using personal data maliciously.  The increased digitization of previously physical services and the increased use of online retail, has lead many organizations trying to get a better picture of their consumers, customers and potential customers.  Note the subtle difference between a consumer (who is actively using a service perhaps without registration eg Google), a customer (paying for a service or good) and potential customer.  All have different characteristics from a marketing and customer servicing perspective.  A organisation wants to get the individual perhaps down a consumer --> customer --> repeat/upselling customer route as quickly as possibly, but with a stickiness towards the latter part of the journey - ie when they're a customer keep hold of them.

The information exchange at the beginning of that cycle is key to helping organisations follow that flow.  From a individuals perspective, there needs to be a strong service improvement or cost saving aspect in order to sacrifice some of the data that is being asked for.

By Simon Moffatt