Three productive discussions were had today.
The first was with Brian, who pointed out that I shouldn't have separate packages for amdsgmap and amdsgmap-web, as it makes it annoying with multiple properties files and the like, and that amdsgmap would never be used independently of amdsgmap-web. He also suggested naming it "gridviewer" as it would be easier for people to remember when typing it into the 'go' box of Internet explorer or Firefox or browser-of-choice. Chalk this up to lessons learned from the beautiful but hard-to-remember 'npdsgmaps-web'. I spent some of today moving packages into a web project and checking to make sure maven would understand what I was trying to do. I think it did, I will check more tomorrow.
The second conversation was with Vaughn, and we discussed the AMDS client and how aware it would be of what servers it would need to call (the end result being 'not very aware at all'). It will have basic hooks for setting the URL to call, the level of security to use (for now it will be 'none' and 'globus', and hopefully things like 'siteminder' and 'cabig' and 'coolfuturenetwork' to help extend the trust mesh of multiple CA's brian was talking about in an earlier post) and a plug for querying the AMDS service at given url/bunker.
The third conversation was in my head, where I was drawing out all of the bits between the client and the gridviewer and plotting out data and function flows. Tomorrow, I hope to code a bunch of them in and commit them with a few tests.
Tuesday, April 21, 2009
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment