On Wed, Apr 17, 2013 at 10:05 AM, William G. Thompson, Jr. <[email protected]
> wrote:

> On Tue, Apr 16, 2013 at 11:15 AM, Daniel Fisher <[email protected]> wrote:
> > On Tue, Apr 16, 2013 at 9:32 AM, William G. Thompson, Jr. <
> [email protected]>
> > wrote:
> >>
> >> Team,
> >>
> >> Unicon has already implemented MFA in CAS for a few clients and we
> >> were just awarded a contract to build a modular extension to the CAS
> >> Server to support multi-factor, and levels of assurance along the
> >> lines of the vision articulated in
> >> https://wiki.jasig.org/display/CAS/Level+Of+Assurance+-+Head+document
> >>
> >> This project requires that the resulting CAS server customizations be
> >> released as free and open source software for adoption with and in CAS
> >> (i.e. as cas-addons or the Jasig distribution).  We intend to engage
> >> with the community throughout the implementation.
> >>
> >> This project provides a real world deployment scenario to build upon
> >> the authN API work that Marvin has completed thus far.  So I think
> >> this is good news for CAS 4.1/5.0 in that we can be assured that the
> >> work to date on authN API, MFA, and LOA will yield deployable code
> >> even though it didn't ship with 4.0.
> >>
> >> In terms of "big features" for 4.0, I think we've made substantial
> >> progress along a number of paths that lead to a more supple,
> >> extensible and elegant CAS server that is a joy to install, configure
> >> and run[1].
> >>
> >> * Externalizing SAML as an optional support
> >> * Upgrades to all dependancies
> >> * OAuth support improvements
> >> * Externalized Services Management Webapp
> >> * A bunch of security related improvements including encrypting
> >> clearpass cache by default
> >> * Many other improvements.
> >
> >
> > I don't see anything here that merits a major release. What's the problem
> > with pushing the release date so the authN API features can be included
> in
> > 4.0?
>
> Hi Dan,
>
> The current dev cycle started last summer and we've already pushed the
> release date several months.  In that time, additional scope has
> continually been added.  At some point in time the project has to trim
> scope and cut a release.
>
> I'm sure there would be no objection to sliding the date a few days or
> even a week or so.
>
>
I can certainly appreciate the value of adhering to a schedule. If this was
a 3.x cut no one would be complaining, but it's a major release and the
features aren't indicative of one. So what's the rush? Less than 12 months
is a short dev cycle for a major release anyway. Lets push it back and give
people a reason to want 4.0.

--Daniel Fisher

-- 
You are currently subscribed to [email protected] as: 
[email protected]
To unsubscribe, change settings or access archives, see 
http://www.ja-sig.org/wiki/display/JSG/cas-dev

Reply via email to