# Definitions and Preface
In this document, “Craig” refers to the conglomeration of software and hardware interfaced through bots called “Craig” and “Giarc”, and the web sites “craig.chat” and “craig.horse”, which collectively provide the service of recording and processing audio on Discord (opens new window) and/or from a browser; “the Craig webapp” refers to the web site “web.craig.chat” and its associated server technology; “you” refers to a user who initiates recordings with Craig; “other speakers” refers to individuals recorded by Craig who did not initiate the recording; and “Craig's maintainers” refers to volunteers who maintain Craig. As Craig is a piece of software, it is not a legal entity capable of ownership or entering into a contract; however, like all software, it is capable of some autonomous action, and so is referred to as an actor as necessary herein.
The nature of recording multiple individuals creates two dimensions of privacy: That you do not use other speakers' audio without their permission, and that Craig's maintainers and Craig do not use any audio without permission. Resolving any privacy concerns between yourself and other speakers is your concern; when you initiate a recording with Craig, you are acting on their behalf in allowing Craig and, if necessary, Craig's maintainers, access to their audio. Although for technological reasons it is of course Craig who is recording you and other speakers, the privacy concerns between you and other speakers are identical to those if you were recording them yourself, as it is you who is granted access to the recordings.
# Collected Data and Rationale
Craig records all audio data transmitted to it—that is, sent to it via Discord, the Craig webapp, or any other technology—as well as the usernames, discriminators, user identifiers, user avatars and other identifying information for you and all other speakers, as well as the name and identifiers of the Discord server and all relevant communication channels where it is used. This information is needed for the following reasons:
- The collection and dissemination of audio data is, of course, Craig's principal purpose!
- For your convenience, Craig labels each track with the username and discriminator of the speaker recorded.
- Craig must keep track of which channels are being recorded and which users are responsible for those recordings.
- Craig's maintainers periodically look for trends in Craig's usage in order to make sure it is adequate for its major users.
- Long-term trends in Craig's usage inform decisions such as server provisioning.
- Logs of particular behaviors which have invoked bugs in Craig are necessary to fix such bugs and prevent them from arising in the future.
Craig does not have access to any individual's real name, address, financial information or similar, unless they've chosen to, for example, use their real name as their Discord username or otherwise provide it directly to Craig. Like most web sites, all Craig web services log the IP addresses of all users, for the purpose of identifying and preventing malicious use. Craig does not collect any information on any person who does not speak in a recorded channel, use any Craig commands, or participate in a recording through the Craig webapp.
# Third Party Data
If you link Craig with any other service, Craig will collect the information necessary to interact with that service. In the case of Google Drive, Microsoft OneDrive, Dropbox and Patreon, this includes your email address, your real name, and file metadata. This information is not retained if you unlink these services. The exception to this is Patreon information, which is gathered from Patreon in order to deliver benefits to patrons when their Discord account is linked via Patreon. Patreon information is removed when the user is no longer a patron of Craig. Craig's maintainers will not share, disclose, or sell information from these services to any third party. Information gathered from Google and Microsoft (by linking these services to your Discord account) are used to upload recordings to their respective cloud storage solutions.
Data collected by these services are handled in accordance to their respective privacy policies. (Patreon (opens new window), Google (opens new window), Microsoft (opens new window), Dropbox (opens new window))
# Your Rights
Excluding the provisions set forth in this section, collected data may be kept indefinitely. Audio data and information pertaining to other speakers is normally kept only until the relevant recordings expire, but it is sometimes necessary to retain recordings and, with them, this information.
By using Craig, you and any other speakers are granting Craig's maintainers an exclusive, non-transferrable license to use your and their data for the purposes outlined above. This does not grant Craig's maintainers or anyone else the right to transfer it intentionally to any other person. When Craig's maintainers retain recordings for testing or debugging purposes, to the degree that is possible, they ask you for permission before doing so. Under no circumstances will Craig's maintainers intentionally transfer your or other speakers' data to any other person without explicit permission from you, but it is assumed that you can grant permission for any other speakers in a recording you initiated.
The copyright status of your audio is not affected by recording it with Craig; generally speaking, each speaker has sole copyright over their own creative work. It is your responsibility to assure that you respect the ownership of any audio recorded with Craig.
Anonymized data and non-identifiable metadata, such as the number of active recordings (but not their identities), the total length of all recordings, and the size of largest recordings by various metrics, are regularly published.
Everyone has the right to know the details and extent of their data collected by Craig, or to have that data expunged. This right supercedes all of the above. To request your data or to request that it be expunged, please contact Craig's maintainers at [email protected]. Please note that since your email address is not collected by Craig, you will need to provide sufficient evidence that the person making the request is the same as the person whose data is being requested. A simple way to do this is to put a codeword in the email, and send that same codeword to one of Craig's maintainers via Discord.
When possible, only the person initiating a recording is given access to it; if Craig cannot give private access, it responds in a public channel. It is your responsibility to assure that no one is granted access they should not have. Anyone with the appropriate URL may download a recording, and it is the responsibility of the person who initiated that recording to protect the secrecy of that URL as appropriate. Craig's maintainers will never transfer the download URL to any person other than you and other speakers in the recording without your permission.
Craig's maintainers have security procedures and measures to protect user data. However, no one can perfectly guarantee the security of any Internet-accessible system. They cannot be held liable for access to your private information through illegal intrusion, even if it can be construed as a consequence of their negligence.
It is immoral, and in many jurisdictions illegal, to record anyone without their permission or knowledge, and Craig is designed to make it clear when recordings are occurring. To use the proper legal terminology: Don't be a dick.