This Plone add-on adds concepts from the EU's General Data Protection Regulations to Plone configuration, which makes it easier to create Plone sites that respect the privacy rights of indivuals.
- ZCML based declaration of uses of data
- User-facing privacy management form
- Integration with core Plone features
The following core Plone overrides are included:
- The sendto_form now validates a to email address against people who have opted-out. The legal basis chosen by default here is legitimate_interest.
- The analytics viewlet also relies on legitimate interest, on the basis that it assumes the tracking is unobtrusive and that this will be allowed by the upcoming changes to the ePrivacy regulations. If the tracking is obtrusive or the site owner doesn't want to make this assumption it should be overridden to use consent.
Users can define a new data processing reason as configuration. For example, an add-on that provides for embedding media might cause users to be tracked. The ZCML would be modified to include:
<configure
xmlns="http://namespaces.zope.org/zope"
xmlns:gdpr="http://namespaces.plone.org/gdpr"
xmlns:i18n="http://xml.zope.org/namespaces/i18n"
i18n_domain="collective.privacy">
<gdpr:data_use_category
name="show_example_media_embed"
title="Embedded media from example.com"
description="We use example.com to embed media into the site. example.com monitors
the usage patterns of users to provide enhanced analytics to site owners."
legal_basis="consent"
identifier="collective.privacy.identifiers.CookieIdentifier"
storage="collective.privacy.storage.CookieStorage"
cookies="media_cookie_*,other_cookie"
/>
</configure>
This would add a new item to the privacy controls that relies on consent to proccess data. This means that by default the permission is denied until an end user gives permission.
Note that the i18n domain of your configuration must be collective.privacy if you want to translate titles and descriptions of your new data processing reasons.
You can then guard your uses of the data, for example:
<div tal:condition="python: context.portal_privacy.processingIsAllowed('show_example_media_embed')">
...
</div>
GDPR provides for six legal bases for processing, all of which are supported by collective.privacy.
They are:
Processing is disallowed by default, users can opt-in. There are rules as to what makes consent valid, which must be followed.
Processing is allowed and users cannot object.
Processing is allowed and users cannot object.
Processing is allowed and users cannot object.
Processing is allowed by default, but users may object. This is only suitable for certain specific types of processing.
Processing is allowed by default, but users may object.
It is necessary to tell one user from another when managing their preferences. In some cases different identifiers are more useful than others. For example, when sending email we want to key users on the email address, but using cookies should be managed by the browser, regardless of the user's logged in state.
The way of choosing which is used is called a identifier. The following are available:
This identifier should be used in cases where the storage is cookie based. It allows the current user to be identifier, but not other arbitrary users.
This identifier should be used when the user needs to be identified by email address. It can optionally use the email address of a logged in user to identify the current request, but in general it cannot identify the current user.
The identifier is a UUID derived from the email address using a one-way function, not the email itself.
This identifier should be used to identify a connection. It can be used to identify the current user or other arbitrary users. It is less reliable than the CookieIdentifier as users IP addresses can change.
The identifier is a UUID derived from the IP address using a one-way function, not the IP itself.
This identifier can only be used to identify logged-in users. It can identify any users who are registered on the site, but not anonymous visitors. As such, it's appropriate for data processing that only occurs for registered users.
The identifier is a UUID derived from the user name using a one-way function, not the username itself.
The storage determines how the user's preferences are persisted. There are three storages available:
This storage uses a cookie called 'dataprotection' on the user's browser. Consent is not required to set this cookie as it is set to comply with legal obligations and cannot be used to track the user.
This storage uses BTrees inside the portal_privacy tool to store the time the user consented or objected. It is currently the only storage that allows for the preferences of users to be queried outside of a request they have initiated.
This is a stub storage to be used with legal bases such as vital_interest where the user has no option to object to processing.
This attribute lists the cookies that should be deleted if the user objects to the use of the corresponding data processing. It is optional and may contain wildcard (*).
This product has been translated into
- French
- Dutch
Install collective.privacy by adding it to your buildout:
[buildout]
...
eggs =
collective.privacy
and then running bin/buildout
If you use this product combined with Varnish ensure that your Varnish config does not remove cookies for requests where caching should be ignored
Example of config:
if (req.http.Cache-Control ~ "(private|no-cache|no-store)" || req.http.Pragma == "no-cache") {
return (pass);
}
Thanks to Jazkarta ( http://jazkarta.com/ ) and YES! Magazine ( http://www.yesmagazine.org/ ) for each sponsoring some of the development costs of this add-on.
The irony that these are both US companies is not lost on us.
- Issue Tracker: https://github.com/collective/collective.privacy/issues
- Source Code: https://github.com/collective/collective.privacy
If you are having issues, please let us know.
The project is licensed under the GPLv2.
N.B., the GPL states:
THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
This add-on has not received any contributors from lawyers and should not be interpreted as legal advice.