jpr's blog

[phpwiki]
silbia's been working on getting the dependancies in place for
trac and subversion. in helping resolve some of the issues it seems trac
uses the one true templating system called clearsilver. after having
looked at the drupal templating options (thankfully nicely seperated via
the theme engine) where folks hype phptemplate and template toolkit. also
working with sympa where template toolkit is used. it seems that every
template tool vendor has the perfect solution. the one that you need to
use. guess that's a sure sign that the best way hasn't been discovered
yet.

finally got my sympa work under cvs. that just took too long. nothing
wrong with the process, it's just tedious to do this after the fact. much
better to get it under source control to start with than to try to back
port it later. :(

now I can work on merging this up to the 5.1 code base. I'd like to get
this with the nice new look.

probably need to get the wayf thing working first though. can't seem to
get the picture in my head on how to do this. i need a wayf for the vocore
and a wayf for the federations. the vo core wayf should be able to direct
the user to their prefered home institution regardles of what

the title says it all. i finally got off my but and submitted the patch
to the sympa developers to support seperation of identity and email during
authentication in support of shibboleth and myvocs. many thanks to the
sympa team in helping develop this patch.

[phpwiki]
been looking into the updates that would be needed to support folksonomy in drupal, known as free tagging. i like the idea of letting the end user determine the vocabulary for entries. I looked at the [drupal support for tagging|http://drupal.org/node/20936]. it really needs it to allow end-user control over vocab. from the description of the features, though, it seems like it needs some work to add features. i'd like to get the [wordpress features jim described|https://mywebspace.wisc.edu/jamesphelps/web/WordPress/wordpress_folksonomy.html] because in order to use tags in a meaningful way, you really need to have a list of popular tags to choose from or reference.

[phpwiki]
this idea has been rolling around in my head for a while and the thoughts
on tagging seem to make it an even more powerful option.

bayesian filters for spam seem to be a good way to train a filter to
answer a simple yes/no question about a given piece of content. "yes this
content belongs in catagory x" or "no this content does not belong in
catagory x". Normally the catagory is "inbox". After a while of training
the filter should get pretty good at identifying terminology patterns that
associate something with catagory x or not. if you think of each person's
inbox as a distinct catagory, then there are obvious millions of distinct

[phpwiki]
Looks like tagging support has been added to drupal 4.6 (and provides more
motivation to upgrade the @lab site). This is a logical extension to
durpal's already strong taxonomy support. The major enhancement is that
it makes term creation an end-user feature rather than restricting it to
admins.

http://drupal.org/node/20936

[phpwiki]
had thought earlier today that an alt solution to doing the mod
for sympa to get the netid map to email is to just have an external code
that does the user registration and then have sympa do an ldap lookup
against this db to get the netid map to email. conversation with jason
suggested one could avoid the ldap lookup and just pass the email to
sympa. this is true if the sympa is a consumer app, but when it is the
vocore then it needs to be the one acting as the AA backend, so it can't
consume it's own attributes (unless the AA queried across two dbs or the
reg front-end populated the user_table in sympa completely, which maybe

Hi Tom,

Sorry to have taken so long to respond. We had a network issue locally
that took my attention away from this for a few days.

Here's a first pass at some annotation. I wanted to keep the diagram
light and mainly to identify the components in the solution so I left
out the flow. I'll try to clarify below. Let me know if I need to take
another pass.

It will probably be best to identify the goal and major components first
and then discuss the flow between components. The basic idea is to
build the foundation of a consistent system environment that can be
leveraged by participating applications, much like what we're accustomed

the oasis docs are hard to wrap your mind around. the two that sesm most
referenced in the early part of the shib architectural overview are
samlcore (which defines the message structures) and samlbind (which
describes how to transport saml in protocol). samlbind identifies a
mapping of saml to a protocol as a binding and defines the rules for
packing and unpacking into that protocol as the saml profile. maybe i've
just been reading to much of this today and it's all bleeding together.

[phpwiki]
an issue to deal with when picking an application as the vocore is that
you'll need to map its abastractions and terms into the general system
environment. for example, we're using a mailing list as the vo core which
speaks of lists and subscribers. this is a general groups and members
concept. unless web interfaces are redefined to use a vo-specific naming
convention then users need to map these concepts themselves.

an advantage of using a mlist as the vocore app is that end users can
self regsiter for group membership and there is a concept of administrator
approval for membership. another aspect is that this instantly gives