Louis Suarez-Potts wrote:
In sum, then, how about this:
* We ask CollabNet to make download.openoffice.org point to an outside
server. (It may have to be renamed, say to "downloads.openoffice.org)
If Bouncer already supports User-agent parsing, then perhaps the
solution is already there in Bouncer?
If not, it may be advantageous to put a functional spec together and
offer this up as a project for someone to work on. From everything I've
heard, making downloads easier seems to be one of the biggest priorities.
I haven't seen the underlying source for Bouncer but it seems to me that
a lot of the mirror management functionality that it provides is useful
and probably doesn't need reinventing. It may be a trivial matter of
extending Bouncer to support User-agent parsing. The download page
could parse the User-agent on its own and then dynamically craft the
download URL to be fed to Bouncer. The rest would be done by Bouncer.
I'll take a look at the Bouncer URLs to see if I can glean some useful
info out of them. If the URL query string already contains locale and
platform details, it may be relatively simple to setup a working prototype.
Ed
* We work on the homepage as has been and is being done.
* We being work on portal.openoffice.org secondarily, with the notion
that portal.ooo will be a more or less comprehensive place for endusers
and even developers, but the focus on endusers.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
