[wp-hackers] GSoC 2009 Ideas: Open ID integration with Wordpress & Template Versioning

[Vidyarth E.S]

Hi everyone,

I have 2 ideas I would like to discuss here
*
I) Open ID integration:*
I was thinking about why not integrate and adequately market Open ID (
http://openid.net/) into the Core Wordpress files. That way we don't have to
sign up for each of the wordpress blogs out there, and can instead use our
Open IDs! This would allow the OpenID login box to be automatically included
as part of the theme (currently it requires editing the theme files if you
don't want it under the comment box) as well as greatly expand the number of
blogs that support it. This is also a concern shared by 596 users on the
ideas page in wordpress (http://wordpress.org/extend/ideas/topic.php?id=40)

This I believe would make Wordpress more open and enhance the probability of
"talkback" for most blogs.

Some points to note from my research so far are

   1. There is an OpenID bounty program that pays out $5000 to any project
   that adds OpenID to their core. You'll notice they have WordPress listed as
   one of the projects they hope to take up the challenge.
   http://iwantmyopenid.org/bounty

   2. The issue that is raised by Matt Mullenweg on the wp-hackers list in
   Oct. 2005 <
   http://comox.textdrive.com/pipermail/wp-hackers/2005-October/002969.html>,
   is that

   '.. currently there exists lots of identification methods and none of
   them tend to be universally adopted. Moreover, the next reply tells a way to
   identificate other than just for blogs would be much more interesting, more
   'universal'. So, I doubt too OpenID would ever make in WP Core.is that there
   were competing identity standards.'

   But since then, of the 7 that he lists 5 (OpenID, Ping Identity, Identity
   Commons, Sxip and LID) have collapsed together into one and it's called
   OpenID. He suggests letting them duke it out and they have. Now we have
   OpenID. The other 2 he mentions are Yahoo (now BBAuth) and Passport. Those
   are both proprietary. Recently there is another competing ID mode -  a web
   Jabber identification (http://shearproject.info/jcas/), but I believe it
   has entered the market slightly later and hence is losing out to OpenID.
   Furthermore here is what jaykul says (
   http://wordpress.org/support/topic/70225?replies=7#post-492486) says

   "Regarding Jabber .... I know there's a web/jabber-id project, but I'd
   have to say that I don't think it's a good login identifier. The key to the
   OpenID system is that (similar to CardSpace -- they are essentially
   compatible) you get to choose which infromation you make available to sites
   that you log into using your ID. Some sites can *require* that you
   include certain pieces of information (like your email, or real name, or
   whatever) but you don't implicitly give them access to anything else. Doing
   the same thing with JabberIDs would require quite a bit of modification to
   the Jabber spec (currently your ID/vcard access is basically all or nothing)
   and would require web sites to implement a Jabber login themselves so the
   website could log into Jabber to check your ID in some way..."

   In summary at least, OpenID has emerged as the de facto, decentralized,
   open standard for identity and WordPress gets paid to add it to the core.

   3. Giving readers the ability to register an identity easily on any
   WordPress blog with one set of details, instead of having to create a new
   login and password for every blog. This would then mean that blog owners
   could turn off anonymous commenting, in turn cutting down somewhat on spam.
   But as Otto points out in the above mentioned thread (point no 2) "Most
   comment spam is done by automatic systems using the trackback functionality.
   Requiring users to register to comment just means that most users won't
   comment and reduce the number of real comments"  Although I agree with Otto
   partially, I think it is the number of real "not so serious comments" that
   are usually moderated out that are reduced which I believe is good.

   4. I think to an extent giving the users an option to use open id to
   register in wordpress as a whole might help bringing in more people to
   wordpress.

*II) Template versioning in Theme Editor:*
Also another idea that I am quite interested in is the versioning of  theme
editor template and style files. I believe several people edit (especially
when it is bulk) on other editors (like notepad++ etc) for the fact that the
theme editor lacks several attributes of WYSiWYG editor (changing that also
interests me especially since a lot of users find it troubling, but I think
it might take more time than what is allocated for GSoC). So the only thing
that benefits people is simple versioning of files and a nice interface to
manage those file versions that enables quick restoration of previous
versions coupled with ability to compare versions.

Summarizing here is what I think should be done

   1. Build a versioning system for template files within the theme editor.
   2. UI for showing different versions of a file (only the title like time
   modified etc)
   3. UI for comparing 2 versions of a file.

For the purpose of GSoC points 1 and 2 might be realistically achievable.

What are your thoughts on both of the above ideas, specifically on any
technical complexity or feasibility issue for either or rather simply if
they are good enough ideas or not?

Regards

Vidyarth
Final Year Computer Engineering Student, National University of Singapore