Re: [Treebase-devel] treebase.domain.name

[William P. <wil...@ya...>]

On Jan 13, 2010, at 12:47 PM, Hilmar Lapp wrote:

> Isn't there a difference between the domain name of a resource, and  
> the PURL domain it may or may not use? We don't want to serve TB off  
> of the PURL domain and masquerading as such, right? And why is the  
> domain treebase.org not stable as a basis?

Don't know if I quite understand the nuance here between a stable treebase.org and using purls. But my $0.01 thought is that using purls for all phylows URIs is something we, at one point, decided to do. 

By contrast, all non-phylows URLs can use whatever domain TreeBASE is running off of (and for the deployment build, treebase.org would point to this). 

So, for example, our dev-build would be built using the global phylows-property parameter "http://treebasedb-dev.nescent.org:8180/treebase-web/", while the deployment build would be built using the global parameter "http://purl.org/phylo/treebase/". That way we can easily test new changes to TreeBASE on the dev server without all of its /phylows/ links rerouting us to the deployment build. 

I suppose we could view the access-code-phylows as a "special" /phylows/ string that doesn't need domain name stability. So in that case we would not make use of the global phylows-property parameter, and would instead use whatever the domain name happens to be. 

bp