Skip to content

Rhaptos Software Development

Personal tools
You are here: Home » Developer Blog » Manpreet's Blog » Roles and their impact on author browse & search

Roles and their impact on author browse & search Roles and their impact on author browse & search

Document Actions
Submitted by manpreet. on 2007-05-14 17:20.
While looking at the author browse subcategories (Content By, Other courses Using, Derived Copies, Also Maintains) - the following issues came to light:
  • Author browse currently lumps Editors and Translators in with Authors without giving any clue as to whether the content is authored work, edited work or translated work by that person
  • The subcategories contain two role based views i.e. "Content By" (Author, Editor,Translator) and "Also Maintains" (Maintainer). The other two subcategories - "Other courses containing" and "Derived Copies" - are more of a "content lifecycle" type where they bring to light works based on this authors work.
  • Should all roles be supported in the browse interface? Currently these are: Author, Maintainer, Copyright Holder, Editor and Translator. Upcoming: Illustrator. There might be others over time.
  • In the current browse system - Maintainers are not in the author browse. But you can get to the browse list of a maintainer's content from their member profile. This list then hangs in no-man's land because the member name is not listed in the browse author pane 2. See "Kevin Duh" as an example.
  • Member profile only gives information about authorship and maintainership, none of the other roles or subcategories.
  • Do all roles have content lifecycle? e.g. we know an author's content can be used in a course by anyone, as well as copies can be derived and modified. We keep track of this information and display it in the two lifecycle subcategories mentioned above. Does the same apply to edited works? Translated works? We know that maintainers lose maintainership once a copy is derived. Is a derived copy that is modified in any way different from a derived copy that is essentially the same but translated?
  • Current Search treats Author, Editor and translator as the same. Maintainer and Copyright holders are not searchable.
After discussion with Kathi and learning the historical perspective from Max, Jenn, Ross and Kyle, here are some solutions:
  • Member profile should include everything i.e. Author of, Translator of, Editor of, Maintainer of, Copyright holder of and Content based on works by.
  • Search should include everything - all roles. Search should also display related author profile, member lists, endorsements etc. when the match is on one of the roles.
  • Advance Search should allow the selection of role to search.
  • Author browse should support "All Content | Authored |Edited |Translated |Illustrated" as the subcategories i.e. only role based subcategories. If it is a singular case i.e. there is only authored content or translated content - then only that category should display. The other subcategories should only appear if there is any content under them. "All Content" should appear in the case of 2 or more subcategories.
  • Authors, Editors, Translators, and Illustrators should show up in the author listing in Pane 2.
  • To support the lifecycle based subcategories - Pane 3 should have an additional link "View Content Based on this person's work" next to the "view profile" link. This link will take readers to a new page type which provides a listing of content based on the author's work - including the ability to see "Other courses using content by" and "Derived copies". This new page will also be linked to from the member profile.
Kathi - let me know if I forgot anything from our discussion.
Everybody else - please comment if you see any missing interactions/dependencies. Or even if you have any positive or negative thoughts on this new direction.

Re: Roles and their impact on author browse & search

Posted by maxwell at 2007-05-21 17:47
I don't necessarily see Kevin's type case as being out in no-man's land simply because they aren't linked to from the browse interface, since the "edited content page" is still linked to from that person's profile, which is linked to from the content object, which I would guess is the more likely path of a user seeking out info on a particular cnx person anyway. I'm not sure that making a full loop back to/from /content/ is sine qua non.

That having been said, I do like your idea of keeping pane 3 as having links by allroles, role1, role2, etc., and moving the "life cycle" content out to another space. I think it would clear up some current confusion and give more room (on the separate page) to explain exactly what those other life cycle things are. I'm curious as to why maintainers and licensors are excluded from panes 2 and 3, though.

Also, regarding whether things other than authorship do/should carry on in a work's life cycle, I think that authorship should probably be the only role to carry through, since that's the only role that counts in terms of carrying through, AFAIK, in terms of the attribution license.

Re: Re: Roles and their impact on author browse & search

Posted by kef at 2007-05-22 16:33
The no-mans land was referring to the fact that the maintainer is not displayed on the results page even when coming from the "Maintains 67 modules" so it isn't crystal clear why each of the results is showing up.

I think we should have browse over members and search should definitely search everything.

On the derived copy issue -- I wonder if this really ought to be following the copyright holder and the author.

Re: Re: Re: Roles and their impact on author browse & search

Posted by maxwell at 2007-05-23 11:59
If you came to that page from a link reading "Maintainer of: _67 modules_" and there is also text at the top of that page reading "Content maintained by MaintainerName", then I still don't see why there would be any confusion. I mean, it might be *nice* to have that listed as a detail on the actual result, but then would we also need to list maintainership in other content listings (e.g. in the listing of content in Spanish) for consistency?

I don't see why the Copyright Holder needs to be shown as having content derived from them since they might have had nothing to do with the creation/originality of the work, in the same way that at the top of derived content we don't write "Based on: _Module X_ by _Author Y_, copyright _Author Z_.

Re: Roles and their impact on author browse & search

Posted by jenn at 2007-05-22 17:12
I think I agree with everything!

Didn't quite follow this, though: "Search should also display related author profile, member lists, endorsements etc. when the match is on one of the roles."

Do you mean that types of information other than content should show up in the search results -- i.e., return an author line with profile info, or an endorser line with the endorser's blurb, inline with the courses and modules? Frightening thought. Or are you just talking about extra snippets of metadata showing up so they can be highlighted as having contributed to the matching?

Re: Re: Roles and their impact on author browse & search

Posted by manpreet at 2007-05-23 10:00
The search results should have an additional category of information - what I am calling "Content Related" results which include member lists, profiles, lenses etc. The intent is not to display these inline with the actual content results but some other visual solution where they show up as related but do not interfere with the actual content results. Max & Sebastian are working on this as we speak.

Re: Roles and their impact on author browse & search

Posted by reedstrm at 2007-05-24 10:21
I talked this over w/ Manpreet, and also agree with everything (Which is good, since I'm implementing it :-) ) The logical separation of direct (contributing) roles from lifecycle roles is good. Note that _none_ of the roles carry though a derived copy, by default. The issue of rights holders of parent works is one of attribution: we currently attribute the author, but if the author is not identical to the rights holder, what's the right thing? I'm not sure.
Developer Blog
« November 2008 »
Su Mo Tu We Th Fr Sa
            1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30            
2008-11-10
13:39-13:39 Suggestion for live site slowness reports
Categories:
Content (55)
Copyright (0)
Deep Code (3)
Development (203)
Markup (22)
Metadata (1)
Printing (7)
Style (9)
Testing (2)
Usability (6)