Thanks for all the smart comments in the thread so far, and big thanks to everyone who's worked on this proposal. GWT is a pretty amazing reflection of the unique value Wikimedia can provide to the cultural and educational sector already, so I am really happy this is getting additional attention.

A couple of points:

1) I agree with Gergo's point quoted above, and within the context of the current proposal, would recommend budgeting for at least a 20 hour/week developer supporting Dan with code review and integration, ideally someone with prior MW experience. This cannot entirely eliminate integration effort on the WMF side of things, but will ensure that the people who have the greatest interest in seeing the project through to completion are set up for success in doing so. With or without this, let's really carefully negotiate what exactly everyone's commitments are so we avoid a repeat of phase 1.

2) I say "within the context of the current proposal", because I would ask you to give serious consideration to the following question: Does GLAMWikiToolset need to existing within MediaWiki? When it was first developed, we didn't have OAuth (so a tool outside MediaWiki couldn't perform user actions), and our APIs were less mature. Today we have many examples of external tools that are doing amazing things. Magnus' tools have made tens of millions of edits to Wikidata. The Wiki Edu Foundation has created wizard.wikiedu.org and  dashboard.wikiedu.org for managing student assignments and courses. 

I know we have GWT and so it seems natural to just fix bugs and improve it. But consider the long term development velocity. GWT is used by a very small subset of Wikimedia users, it's not "core site functionality" and does, as far as I can tell (I may be missing something), not benefit dramatically from deep integration. You pay a lot of cost for this integration without necessarily getting a lot of "bang for the buck".

I would wager that if you started over with a new external tool, applying all the lessons learned so far and spending extra effort on UX, you could pretty quickly catch up with current functionality and then would move at a faster velocity from there. Consider where we want to be in 2016, 2018, 2020 -- is the strategy of maintaining a deeply integrated MediaWiki extension for this really sustainable or desirable? I think it's at least worth seriously considering the alternatives.

Erik