Data security: Difference between revisions

From FreeMind
Jump to navigationJump to search
No edit summary
No edit summary
Line 5: Line 5:
* Unauthorized access to data or data theft is limited by '''[[encryption]]''' feature. Thus, the user can select a particular branch of the mind map as encrypted, password-protected. There is no encryption by default. This function relies on the encryption services provided by the Java platform. What extent of true security this achieves is unclear; we are aware of no security audit in relation to this FreeMind feature. Having an encrypted section seems out of scope of the core FreeMind feature set, and there is a risk that a user creates a false sense of security achieved by using this function. FreeMind is not designed as a password management software and it appears advisable to avoid using FreeMind for the purpose. In case of more serious doubt, this internal encryption feature would be disabled and users would be asked to use e.g. zip to encrypt the whole mind map as, say, PersonalConsiderations.mm.zip. Then, the user would have the encryption method and its strength under strict control, using encryption tool that is specifically designed to address that task.
* Unauthorized access to data or data theft is limited by '''[[encryption]]''' feature. Thus, the user can select a particular branch of the mind map as encrypted, password-protected. There is no encryption by default. This function relies on the encryption services provided by the Java platform. What extent of true security this achieves is unclear; we are aware of no security audit in relation to this FreeMind feature. Having an encrypted section seems out of scope of the core FreeMind feature set, and there is a risk that a user creates a false sense of security achieved by using this function. FreeMind is not designed as a password management software and it appears advisable to avoid using FreeMind for the purpose. In case of more serious doubt, this internal encryption feature would be disabled and users would be asked to use e.g. zip to encrypt the whole mind map as, say, PersonalConsiderations.mm.zip. Then, the user would have the encryption method and its strength under strict control, using encryption tool that is specifically designed to address that task.
* Data loss risk is mitigated by FreeMind saving '''automatic backups''' into a certain folder, per [[Automatic backup]]. However, this same feature could make data theft easier (needs clarification). In any case, it is a recommended practice to set up a regular backup procedure for any important files, not just FreeMind files. What FreeMind provides here is a crutch for people not disciplined enough to setup a proper regular backup process.
* Data loss risk is mitigated by FreeMind saving '''automatic backups''' into a certain folder, per [[Automatic backup]]. However, this same feature could make data theft easier (needs clarification). In any case, it is a recommended practice to set up a regular backup procedure for any important files, not just FreeMind files. What FreeMind provides here is a crutch for people not disciplined enough to setup a proper regular backup process.
* Since FreeMind mind maps are plain text files, any spyware aware of FreeMind file format can take away information from FreeMind mind maps. In this regard, FreeMind mind maps fare no worse or better than plain text files, so cherished by many a Unix user.
* Since FreeMind mind maps are plain '''text files''', any '''spyware''' aware of FreeMind file format can take away information from FreeMind mind maps. In this regard, FreeMind mind maps fare no worse or better than plain text files, so cherished by many a Unix user.
* FreeMind 0.8.0 introduced '''undo''' and after a disagreement in the team, the version was released with undo that cannot be disabled. This is a violation of Mill's harm principle, treating FreeMind users as kindergardeners who need tutelage. And it is not clear what kind of security risk this mandatory use of undo creates on its own, especially given that the storage for undo purposes is in XML format, ready in a form that can be transmitted over the network.
* FreeMind 0.8.0 introduced '''undo''' and after a disagreement in the team, the version was released with undo that cannot be disabled. This is a violation of Mill's harm principle, treating FreeMind users as kindergardeners who need tutelage. And it is not clear what kind of security risk this mandatory use of undo creates on its own, especially given that the storage for undo purposes is in XML format, ready in a form that can be transmitted over the network.
* Since FreeMind 0.8.1, FreeMind uses a '''plugin architecture'''. In some software contexts, plugin architecture has been found to be a security concern. The latest releases without plugins is FreeMind 0.7.1 and FreeMind 0.7.1-XT. See also [[Plugin]].
* Since FreeMind 0.8.1, FreeMind uses a '''plugin architecture'''. In some software contexts, plugin architecture has been found to be a security concern. The latest releases without plugins is FreeMind 0.7.1 and FreeMind 0.7.1-XT. See also [[Plugin]].

Revision as of 05:25, 28 March 2023

Data security:

Security of the data is of concern. One is the risk of data loss, the other one is of data theft. The following is written not by a security expect; in case of doubt, it is advisable to contact a security professional to perform an analysis or audit.

  • Unauthorized access to data or data theft is limited by encryption feature. Thus, the user can select a particular branch of the mind map as encrypted, password-protected. There is no encryption by default. This function relies on the encryption services provided by the Java platform. What extent of true security this achieves is unclear; we are aware of no security audit in relation to this FreeMind feature. Having an encrypted section seems out of scope of the core FreeMind feature set, and there is a risk that a user creates a false sense of security achieved by using this function. FreeMind is not designed as a password management software and it appears advisable to avoid using FreeMind for the purpose. In case of more serious doubt, this internal encryption feature would be disabled and users would be asked to use e.g. zip to encrypt the whole mind map as, say, PersonalConsiderations.mm.zip. Then, the user would have the encryption method and its strength under strict control, using encryption tool that is specifically designed to address that task.
  • Data loss risk is mitigated by FreeMind saving automatic backups into a certain folder, per Automatic backup. However, this same feature could make data theft easier (needs clarification). In any case, it is a recommended practice to set up a regular backup procedure for any important files, not just FreeMind files. What FreeMind provides here is a crutch for people not disciplined enough to setup a proper regular backup process.
  • Since FreeMind mind maps are plain text files, any spyware aware of FreeMind file format can take away information from FreeMind mind maps. In this regard, FreeMind mind maps fare no worse or better than plain text files, so cherished by many a Unix user.
  • FreeMind 0.8.0 introduced undo and after a disagreement in the team, the version was released with undo that cannot be disabled. This is a violation of Mill's harm principle, treating FreeMind users as kindergardeners who need tutelage. And it is not clear what kind of security risk this mandatory use of undo creates on its own, especially given that the storage for undo purposes is in XML format, ready in a form that can be transmitted over the network.
  • Since FreeMind 0.8.1, FreeMind uses a plugin architecture. In some software contexts, plugin architecture has been found to be a security concern. The latest releases without plugins is FreeMind 0.7.1 and FreeMind 0.7.1-XT. See also Plugin.
  • Since FreeMind is apparently used by many users and organizations, one might by inclined to think that enough eyeballs double checked that there is no genuine security issue. However, we know the story about spinach and iron, and we are reminded of the quote that people would sooner die than they would start to think, and indeed, this is what most of them do.
  • If an organization actually did publish a security audit, it would be preferable to post it here, for traceability.
  • As per Scripting#Security, security level for scripting can be customized. The most secure option is to disable scripting altogether.
  • Bundling JRE with FreeMind is a concern: since such FreeMind installs a dedicated Java JRE locally as a separate copy, it will not receive any automatic security updated. Requires a solid analysis as to whether there is a genuine security risk.

See also