[wp-hackers] Potential Solution for Absolute vs. Root Relative URLs?

[Mike Schinkel]

Hi all:

Okay, I may be stepping into it (I really hope not) but I'm wondering if I can refocus the absolute/relative URL discussion?  

In the other thread I think Jonathan Bailey hit the nail on the head; this debate is primary premised on the requirements of each use-case, and each person is arguing the approach that they believe best resolves their respective use-cases.  Said another way, we have lots of blind men here and everyone is trying to describe an elephant. (Jonathan unsubscribed from the list so I've cc'd him given that I'm trying to address his point; I hope he appreciates my cc-ing him.)

I'd like to propose something but first I'd like to document my perceptions?   There seem to be several factions represented here:

1.) Those who passionately care about the **End-user Blogger** use-case.  This (it seems to me) is the primary use-case for WordPress and that which most of the core WordPress team care passionately about. Anything that benefits the end-user blogger use-case is viewed as "good" while things that don't benefit them or worse that complicate things for them are "bad." This use-case is benefitted greatly by inter-changable themes and the broad array of plugins, all of which end-users can add to their site.

2.) **Professional small shops** who build and deploys sites for small business clients.  This use-case is mostly synergistic with use-case #1 because most of these can use the exact code that benefits end-user bloggers. Often these client sites have a large blogging component too and a site developer can easily assemble a great site using available plugins. And for occasional site deployment it's easy enough in most cases to spend the "just 5 minutes" it takes to move things around.

3.) **Enterprise-level/Larger Agency** use for large-scale sites, especially ones that are more CMS oriented than blog oriented.  This group's needs are often in conflict with the end-user blogger use-case. Things that make management easier make it more difficult for the end-user, add unnecessary overhead to a site, etc.  So it is the people using WordPress for this use-case that get very frustrated because use-case #1 always gets priority in the world of WordPress.

Is this a fair assessment?

If yes, I think that while we won't solve the larger conflict today we might actually be able to come up with a truce for the absolute vs. root relative URL standoff.  First and foremost I believe that WordPress will continue absolute URLs, period.  It's apparently what fits the blogging use-case best, it's what the core WordPress team believes in and it's what most of the more influential members of the WordPress community believe in. Given that, chances it will change are nil.

OTOH for use-case #3 there are those who would really, really like relative URLs (and please even for those who like Otto disagree that it is needed please just agree to disagree.)

So here's the proposal:

1.) Is there any reason WordPress can't add a constant like ROOT_RELATIVE_URL that if defined would allow root-relative URLs, ASSUMING that the people advocating here create the patch, get the code accepted by the core team, and then do lots and lots of testing?

2.) If #1 is unacceptable and there are hooks that could be added to make it easier to implement root relative URLs, is there any reason why the core team would have a problem adding those hooks?  I know there is now a root relative plugin but its implementation may be much more complex than if changes were made to the core (I'm only assuming since I've not really studied it much.)  Will the core team add requested hooks to address this use-case?

Anyway, thanks in advance for considering what hopefully could be an amicable and final solution to what has become a debate that flares up repeatedly on the hackers forum. :-)

-Mike
http://about.me/mikeschinkel