When a user saves an ODF or Flat ODF document, LibreOffice currently also saves a config-item set of "view settings": <config:config-item-set config:name="ooo:view-settings"> this lets whoever reads the file know which part of the file the file the saving user was last focusing on, and at what zoom level. In most cases, this may be of little consequence. However, in some cases, the saving user may not want to share this information with others. For example, two parties are negotiating a long contract. The last part of the contract one party was looking at may indicate to ther other party something they had on their mind, which they may not want to share. This is one, privacy-related, reason why saving view information without informing the user is problematic. A second reason has to do with the fact, that the saving user may not want a person opening the file to have their view focused on some particular part of it, just because the saving user was last looking at it. They may well prefer that the opening user would see the beginning of the document, as though they had never seen the document before (which, in fact, is the case). A third and final reason is that the view settings are application-specific; and a user may want to save a "universal" ODF file free of application-specific settings.
(In reply to Eyal Rozenberg from comment #0) > A second reason has to do with the fact, that the saving user may not want a > person opening the file to have their view focused on some particular part > of it, just because the saving user was last looking at it. They may well > prefer that the opening user would see the beginning of the document, as > though they had never seen the document before (which, in fact, is the case). Note that this specific concern (only) is moot, because the mechanism of automatic scrolling to the last view is implemented to only work when (1) the last edit author name is saved in the document, and (2) it's the same as the data of the current user. The reasoning is exactly that the last position is only of interest to the author, not to others. So - either when there is no author data (they didn't fill their user name, or stripped that away), or no current user data, or they are different - the scrolling to the last view will not happen.
(In reply to Mike Kaganski from comment #1) Point taken. The two other motivations remain.
File > Properties: [ ] Apply User Data should do the trick. There is also a setting under Tools > Options > Security: Options "[ ] Remove personal information on saving". Otherwise, duplicate of bug 159130 "Add option to remove personal information from documents at runtime"
(In reply to Heiko Tietze from comment #3) > File > Properties: [ ] Apply User Data should do the trick. It is about reading, not writing, IIRC. > There is also a setting under Tools > Options > Security: Options "[ ] Remove > personal information on saving". It may be useful to extend what it controls. Currently, I believe, it doesn't cover this.
(In reply to Heiko Tietze from comment #3) > File > Properties: [ ] Apply User Data should do the trick. Oh, you're saying that if we uncheck that, the view information is not saved? But - view information is not "user data"... Or, if it is considered user data - users are unlikely to understand this applies to view settings. Moreover, the discoverability here is too weak. Users who are not strongly aware such information is saved will not seek it out; and even if they do, they may fail to enter "File properties" for this purpose. > There is also a > setting under Tools > Options > Security: Options "[ ] Remove personal > information on saving". Same points as before: * view information is not what users will consider to be personal data * Low discoverability > Otherwise, duplicate of bug 159130 "Add option to remove personal > information from documents at runtime" That bug is a bit shapeless IMNSHO... I just commented there. What I think I'd like to see is a toggle in the File > Save... dialog, for this. And if we add toggles for a few other kinds of potentially-sensitive information to that dialog, then - either we have an expandable section with toggles for different things; or we have one toggle for all potentially-sensitive data. ---- Also, it might be a good idea to have a discussion about the privacy-and-what-gets-saved issue a bit more broadly than the scope of just a single bug. And it would probably be a good idea to have more voices in that discussion. ESC session? Design meeting with effort to get more participants? Discourse thread? Something else?