As shown in this video https://www.youtube.com/watch?v=EshNTl23liY there is a new confidentiality classification drop-down currently labelled "Intellectual Property". There are two problems with this: * the classifications in the drop-down do not actually relate to copyrights or anything else that might be (poorly) termed "Intellectual Property"; * the term itself - see [1] for a good summary, but essentially, legal (and supposedly limited) monopolies on reproduction of intellectual creations do not equal property, however much that continued characterisation might (and does) benefit those with a vested interest in propagating that legally misrepresentative perception. The dropdown would be much better- and more accurately-named "Confidentiality". [1] https://en.wikipedia.org/w/index.php?title=Intellectual_property&oldid=730213735#The_term_.22intellectual_property.22
Let's ask to ux-team their opinion about this one.
For all who hate youtube videos like me: the concerns are about the classification toolbar (View > Toolbars > Classification), where "Intellectual Property" is a label along with National Security and Export Control with the options to flag as Non-Business, General Business, Confidential, Internal Only. "Confidentiality" sounds good to me (non native speaker). But from the UX POV it's bikeshedding. Questions are rather: * How to add more options? (department XY, lab Foo, person Bar) * How to adopt to country/organization specific situations? (4th options like read/write, comment, read only, share instead of Confidentiality, Nat. Sec. etc) * Why show a toolbar to just set a flag? (we would find alternatives to reactivate the message panel) * How to assign colors with flags? (red is am error, and classifying as Internal might not always be wanted as such) * Any relation between the classes? (when I want set the first options only, the others have an arbitrary value) + The toolbar is non-standard, misses transparency first of all, but is not usable when placed at the vertical side. + Labels have no accelerator meaning the toolbar is not accessible for handicapped people. (No youtube video explains the use case, purpose, target group etc. And a review needs these info. So I might be wrong on some points.)
(In reply to Heiko Tietze from comment #2) Agreed. When I first encountered this toolbar, it was in relation to a bug reported against the background colour not being graduated in the same way as the other toolbars for OSX. I literally had no idea of what this was, or what it was supposed to achieve. At present, I fail to understand what useful purpose it might have, especially seeing as the tag definitions do not appear to be editable nor their content modifiable, or capable of being localised (I hope I am wrong with this last one, else what a mess the toolbar will be for non US-English users). I assume that there is a spec somewhere that defines the rationale behind it ? Apologies to all for digressing from the bug report, but I'm trying to understand how this came into being and what use-case it serves.
Setting to new with regard to the use of the word "Intellectual Property", a better definition would indeed appear to be "Confidentiality"
The "Intellectual Property" string comes from <https://www.tscp.org/wp-content/uploads/2013/08/TSCP_BAILSv1.pdf>, Appendix B (page 20). If you don't like that string, please complain to the creators of the TSCP standard. The string in question is in sfx2/source/view/view.src, STR_CLASSIFIED_INTELLECTUAL_PROPERTY, but I would find it very confusing to rename it, when it was a string so far taken literally from the standard.
Hi Miklos. This toolbar and classification system is designed for organisations following the TSCP document classification framework. TSCP is a defence industry initiative and as such this classification system is geared towards that use, not general business document classification. As your blog says, the work in LibreOffice was even "made possible" (I assume funded?) by the Dutch Ministry of Defence: http://vmiklos.hu/blog/classification-toolbar.html There is nothing necessarily wrong with the work being done for that purpose of course. But it should be very clear in the interface that this is not a general-purpose document classification system (e.g. for general business use). Since it is specific to TSCP standards, it should be labelled as such. According to the above blog post, the toolbar is disabled by default, and enabled via: View → Toolbars → Classification Instead the toolbar should then be termed "TSCP classification" or "TSCP BAILS classification" or similar. That should also be extended to the toolbar itself, which has enough room (looking at http://vmiklos.hu/blog/classification-toolbar-multicat.html) for the addition of e.g. "TSCP classification" on the left of the toolbar. However, this wouldn't be a necessity as long as the toolbar is not enabled by default, since it should be clear to anyone who has manually enabled the toolbar what its purpose is. But as mentioned, without properly describing what the toolbar is for anyhwere in the interface, the initial issue remains - these descriptions are not accurate or helpful in isolation. A general user, looking to categorise for example general business documents, who enables a seemingly fit for purpose "Classification" toolbar, would be presented with what seem to be a badly-named and confusing set of three dropdowns, each with identical options. They will not have any idea that they are BAILS classifications for TSCP-compliant documents!
Worth re-reading Alex's comment #3 in light of my last reply as well - demonstrates what the UX problem will be for most users very well.
https://cgit.freedesktop.org/libreoffice/help/tree/source/text/swriter/classificationbar.xhp spells out TSCP explicitly. I have no problems with renaming the toolbar is that helps, the UI strings are here: $ grep Classification officecfg/registry/data/org/openoffice/Office/UI/*WindowState.xcu officecfg/registry/data/org/openoffice/Office/UI/CalcWindowState.xcu: <value xml:lang="en-US">Classification</value> officecfg/registry/data/org/openoffice/Office/UI/ImpressWindowState.xcu: <value xml:lang="en-US">Classification</value> officecfg/registry/data/org/openoffice/Office/UI/WriterWindowState.xcu: <value xml:lang="en-US">Classification</value> Though be aware that this will create confusion -- we're past the 5.2 string freeze, so probably 5.2 will have the Classification name, and then 5.3+ will have a different name. Feel free to submit a gerrit patch that changes the names.
Ah, sorry, I commented too fast. :-) Renaming from "Classification" to "TSCP Classification" sounds good, users who already saw "Classification" would still find it reasonably easily. I'll do that.
Thanks :)
Miklos Vajna committed a patch related to this issue. It has been pushed to "master": http://cgit.freedesktop.org/libreoffice/core/commit/?id=adbaad53624bd88c58cd1ee2ecfb7207c9ab59ee tdf#101154 Classification -> TSCP Classification for the toolbar name It will be available in 5.3.0. The patch should be included in the daily builds available at http://dev-builds.libreoffice.org/daily/ in the next 24-48 hours. More information about daily builds can be found at: http://wiki.documentfoundation.org/Testing_Daily_Builds Affected users are encouraged to test the fix and report feedback.
This is a string change, so not possible to backport to libreoffice-5-2 in general. If there is an explicit ACK from l10n that this exception is OK, technically no problem to backport there.
How would I go about getting an ack from l10n?
@Sophie: Can we backport this string change to 5.2? http://cgit.freedesktop.org/libreoffice/core/commit/?id=adbaad53624bd88c58cd1ee2ecfb7207c9ab59ee
(In reply to freedesktop from comment #13) > How would I go about getting an ack from l10n? @Sophie: Do we have / Can we introduce a flag like "needsl10nEval" that makes an issue appear on your mailing list or get your attention somehow else?
Why is it so urgent to break the string freeze in the 5.2 series for something that could not even be considered a big issue? You realize it’s putting a big burden in translators? Why isn’t it okay for this to be fixed in 5.3 and beyond only?
sorry, but this needs to be reverted. https://wiki.documentfoundation.org/TSCP-classification explains why. It is TSCP based, but good for ~any organization. So TSCP confuses.
https://gerrit.libreoffice.org/66176
This should absolutely *not* be reverted, for the reasons already given above. Reverting this change re-introduces all of the above problems. Please instead propose a *change* which addresses your concern ("TSCP confuses" - which is odd because it is concise, accurate and eminently searchable), and doesn't re-break all the above.
And can you please explain what the impetus for this proposed revert is 3 years later? Who has a problem with "TSCP classification" exactly? is there some user testing that has indicated that the status quo is a problem?
(In reply to DN from comment #19) > This should absolutely *not* be reverted, for the reasons already given > above. please read what I've written and the wiki. Thanks for understanding.
(In reply to Cor Nouws from comment #21) > please read what I've written and the wiki. Thanks for understanding. and the Help is clear too on this aspect.
> please read what I've written and the wiki. Thanks for understanding. What you've written on the wiki re. the name is: "The implementation in LibreOffice allows working with classification in probably any organization and environment. Therefore using the term "Classification" as such, is fine." You wrote that 6 days ago. It does not explain how a toolbar with a dropdown containing the label "intellectual property" and the options "Non-Business", "General Business", "Confidential" and "Internal Only" (and dialogues repeating those options), constitutes a general-purpose "classification" feature. It is *at best* a *confidentiality* classification feature. There are *plenty* of other classification metrics aside from confidentiality. What *is* confusing, is labelling a specific type of classification toolbar simply "classification". Naming it "confidentiality" would be *better*, but this doesn't explain at all why this toolbar has such propagandised terminology as "Intellectual Property" as a label. Again, I refer you to https://en.wikipedia.org/wiki/Intellectual_property#The_term_%22intellectual_property%22 for a thorough examination of this. "TSCP classification" at least clearly explains the choice of terminology, namely that this is from an external organisation and classification process. Why have you suddenly after three years declared that the current wording "confuses"? Where is the evidence for this?
(In reply to DN from comment #23) > Why have you suddenly after three years declared that the current wording > "confuses"? Ah, I understand the following is not clear: From LibreOffice 6.0 on the process of applying classification in LibreOffice is much more mature and flexible than what was offered in the initial version. And apart from that: also in older versions can one adapt the content of the policy file to the needs. See the wiki.
(In reply to Cor Nouws from comment #24) > Ah, I understand the following is not clear: From LibreOffice 6.0 on the > process of applying classification in LibreOffice is much more mature and > .. / added a note on top of the wiki page
> flexible I'm using 6.0.x, and the only flexibility I can see is the ability to write a custom BAILS-xml file as documented in the wiki? If so, "out of the box", the end user gets TSCP classification. In a corporate environment with a customised deployment of LO with custom classification config., the toolbar naming might be suitable, since the organisation will have customised the classification to be relevant to that organisation. For all other users, writing a custom XML file and pointing LibreOffice to it doesn't fall within "end user" flexibility. In this case, while there is not a UI for editing classification categories: * The toolbar name in the menu (e.g. TSCP classification) should be drawn from a field in the XML, unless there is something preventing this. By default, this should be "TSCP classification", since that's the only available classification without writing XML, which is *not* an end-user task, and "Classification" currently/still does *not* usefully describe this classification system. * If custom XML is deployed, this would then change the toolbar name to something that usefully describes the replacement system. * If and when LO provides a UI for the end user to change the classification categories, it would then be sensible for the toolbar name to be reverted to "Classification", since for the *end user*, this is a general-purpose and flexible categorisation system. The wiki mentions 6.3 having a broader example.xml for classification; maybe they will no longer be TSCP-specific and the revert in master will end up in that version? In that case: * Presumably the categories will be useful to a broad set of users, and won't be specifically about confidentiality and/or TSCP? * This revert won't end up in LO versions before 6.3?
(In reply to DN from comment #26) > I'm using 6.0.x, and the only flexibility I can see is the ability to write > a custom BAILS-xml file as documented in the wiki? Plus adding extra markings to the labels, pre-define extra stuff, plus paragraph classification, plus hardening with signatures, .. > If so, "out of the box", the end user gets TSCP classification. It is not TSCP-classification. The system is TSCP-based. The provided classifications are examples for any type of use (or not). > For all other users, writing a custom XML file and pointing LibreOffice to > it doesn't fall within "end user" flexibility. Your opinion. What does a private user want to do with classification? > * If custom XML is deployed, this would then change the toolbar name to > something that usefully describes the replacement system. The toolbar name can be changed in Expert configuration. Not yet documented, I think (as most expert configuration :\ ) > * This revert won't end up in LO versions before 6.3? No, all work AFAIC for 6.3. There are some bugs that might get fixed and backported to 6.1/6.2. No idea. But the rest all involves strings and string freeze for 6.2 is already here.
(In reply to Cor Nouws from comment #18) > https://gerrit.libreoffice.org/66176 Don't think just "Classification" is a clear term. It misses the security aspect, which was clear with "TSCP Classification".
(In reply to Heiko Tietze from comment #28) > Don't think just "Classification" is a clear term. It misses the security > aspect, which was clear with "TSCP Classification". Are you really sure :) ? What do the letters "TSCP" add about security in understanding the UI? Few people, also in relevant organizations, have an idea what "TSCP" is. One has to look it up to see "secure" in TSCP. And even more: if you do a little study of TSCP, you'll notice that classification, the marking of documents, is only _one_ aspect of TSCP, and as such only _one_ element of working secure. TSCP does a great job in setting out a framework for secure global collaboration, that can be used by more or less ~any organization. For LibreOffice the name "Classification" says it all. And if someone doesn't know what it is, and how to use it, s/he obviously doesn't need it, or will be informed later by colleagues.
(In reply to Cor Nouws from comment #29) > What do the letters "TSCP" add about security in understanding the UI? Not much, of course, but it's at least an information that this classification is not about big/small documents, good/bad writing style, private/business content etc. Nevertheless I agree with the idea to change. How about: "Change Security Level" or "Modify Security Classification"?
(In reply to Heiko Tietze from comment #30) > (In reply to Cor Nouws from comment #29) > > What do the letters "TSCP" add about security in understanding the UI? > > Not much, of course, but it's at least an information that this > classification is not about big/small documents, good/bad writing style, > private/business content etc. Nevertheless I agree with the idea to change. > How about: "Change Security Level" or "Modify Security Classification"? Could also be "Apply privacy level"; "Change IP-level" ?
(In reply to Cor Nouws from comment #31) > Could also be "Apply privacy level"; "Change IP-level" ? The first one absolutely yes. But IP could be read as something else.
(In reply to Heiko Tietze from comment #32) >>> How about: "Change Security Level" or "Modify Security Classification"? >> Could also be "Apply privacy level"; "Change IP-level" ? > The first one absolutely yes. But IP could be read as something else. But serious now. - A tool bar name does not have verbs, I think. - Varying it at random from 'Security' to 'Intellectual property' to 'Privacy' to .. doesn't sound useful too.
>>> How about: "Change Security Level" or "Modify Security Classification"? >> Could also be "Apply privacy level"; "Change IP-level" ? > - A tool bar name does not have verbs, I think. It's straightforward to conform the suggestions to "Security Level", "Security Level", "Privacy Level" and "IP Level". > - Varying it at random from 'Security' to 'Intellectual property' to 'Privacy' to .. doesn't sound useful too. Did someone suggest "varying the name at random"? There is also my original proposal of "Confidentiality", although without the broadened list of categories targeted at 6.3 it's difficult to know what would make an accurate and precise name.
Reset assignee to default now that this is reopened.