Weekly Planning Meeting, 10/17

[Marc Stiegler <marcs>]

Highlights 10/17:

The big congratulations this past week go to joel. Joel went 
to the Software Publisher's Association meeting in Canada, to 
be the only person on a Client/Server Architecture panel who 
was not doing SQL. I understand that this panel wound being a 
Great Application Ideas for Xanadu panel, since no one, including 
the other panelists, wanted to talk about SQL. Joel also got 
the guy responsible for developing collaborative software at 
Microsoft interested in Xanadu; we'll see what comes of that.

Congratulations also go to dean, hugh, and ravi. Dean got text 
selection working in controllers. He now has a very pretty demo 
of controllers in action (dean, those link panes were really 
a beautiful sight to my sore eyes), although they're not ready 
for translating yet (and won't be for a while, since dean has 
gone back to finishing the ent, which hill should be ready for 
real soon now). Dean tells me the controllers are at the "30 
minutes per feature" stage, i.e., all the underpinnings are in, 
now there's just a bunch of capabilities to express. 

Hugh completed the waldo to data model schnert without links, 
which was one of the hard pieces of shared frontend code.And 
ravi got waldoes wrapping waldoes, which will make frontend work 
a lot easier.
 
Condolences go to markm, who has spent the entire week trying 
to find the bug in backfollow that ravi stumbled onto. Between 
backfollow not working, and controllers not being quite done, 
garbage collection not being quite done, and the waldo to data 
model still needing link support, I fear we lost a week this 
week, in spite of all the good things we completed.

Finally, a special thanks goes to sue this week, for cleaning 
up the kitchen. The kitchen had fallen victim to the standard 
tragedy of the commons problem: it's in people's interest to 
mess it up, but not to clean it up. What always makes this particularly 
acute, of course, is that no one ever says thanks or appreciates 
the person who shoulders the  unpleasant, completely prestigeless 
responsibility for cleaning. Those of you who like the kitchen 
better now than you did on Monday, you should thank sue too.

Meetings: Ravi, markm, backfollow problem, 3PM Wednesday. Ravi, 
Dean, discuss inclusion model. Hill, roland, discuss formic bug. 
Roland, markm, discuss abstract class generator. Dean, hugh, 
discuss iterators. Markm, marcs, discuss next step in the protocol 
reduction talks. Hill, roland, markm, discuss formic modifications. 
Marcs, markm, discuss the wisdom of a quick & dirty solution 
to descriptors. Dean, look at hugh's enclosures. Roger, hugh, 
dean, discuss system independent attributes for the frontends. 
Michael, hill, meet on hacking formic scripts. Markm, hill, discuss 
explicit deletion of orgls & refcount deletion. Dean, roger, 
michael, markm, discuss sequence numbers of size 4 billion. Markm, 
hill, dean, discuss enclosures & infinity on Wednesday. Roger, 
markm, discuss 88.1 libraries.

If the meetings section gets to be a full page long, it will 
be time to take one day off from coding, get everyone together, 
and have meetings for 24 hours :-)

Anyway, this week's task list is as follows:

A.       dean

x    10/17/89     text selection in controllers
     10/24/89     call peter deutsch
     10/24/89     partial orgls
     10/24/89     translate enclosures
     10/30/89     virtual subtrees
     10/30/89     specify ent test tools
     11/07/89     merge ent into alpha!!
     11/14/89     controllers for frontends

B.       hill

x    10/17/89     checked pointer package
     10/24/89     help markm repair backfollow
     10/24/89     get new formic working
     10/24/89     garbage collector complete
     10/24/89     change formic, no recipe for deferred copy
     10/24/89     rehack make file for "incremental" links
     10/31/89     confirm functioning of garbage collection
     10/31/89     merge ravi stuff into alpha
     11/07/89     merge ent into alpha!!
     11/14/89     turn length into count everywhere
     11/14/89     replace regions with enclosures

C.       hugh

x    10/17/89     build the waldo to data model schnert w/o links
     10/24/89     build the waldo to data model schnert with links
     10/31/89     update broadcast
     10/31/89     integrate FE with Alpha

D.       markm

     10/24/89     fix backfollow
     10/24/89     slut implementation of gatekeepers
     10/31/89     filtered backfollow
     10/31/89     respond to semantics email
     10/31/89     scenario documents with dean
     11/07/89     merge necromancers with dependencies
     11/07/89     merge necromancers with counted pointers
     11/14/89     complete to do list from alpha 3

E.       marcs
    
x    10/17/89     attend Hackers
x    10/17/89     draft of unix review article
     10/24/89     beta agreement with asymetrix
     10/24/89     figure out how to repair email concept
     10/24/89     figure out whether links are first class objects
     11/24/89     rough draft of AMIX/Xanadu frontend comparison
     11/31/89     rough draft AMIX/Xanadu integration plan
     11/31/89     start second round of patenting work

F.       michael

x    10/17/89     shake up sun on maintenance
x    10/17/89     examine bombs for regression failure
     10/24/89     send stroustrup exception stuff
     10/24/89     hand bjarne version of bombs to hill
     10/24/89     refcount experiment

G.       ravi

x    10/17/89     link type documents
x    10/17/89     waldoes wrapping waldoes
     10/24/89     get backfollow working
     10/24/89     headers/body pair waldo for inclusion waldo
     10/31/89     find/replace waldo

H.       bob

     10/24/89     write xanadu system overview
     10/24/89     draft beta application form
     10/31/89     get put on PERT

I.       roger

x    10/17/89     integrate hugh, dean stuff
     10/24/89     get xanadu doc up as composited text
     10/24/89     scrolling windows under C++
     10/24/89     translate schnerts and models to C++
     10/24/89     traverse to acad, traverse from acad
     10/31/89     link lists with simple documents in alpha

J.       roland

x    10/17/89     formic done
     10/24/89     xlint1
     10/31/89     abstract class generator
     10/31/89     parser with expression semantics
     11/14/89     deal with multifile xlint
     11/14/89     automatic documenter

K.       Bill

x    10/17/89     xylogic board fixed
     10/24/89     replace news, so system doesn't swamp when it runs
     10/24/89     fix whole backup system
     10/24/89     talk to kari about pennsylvania

J.       Lars

x    10/31/89     mockup infofactory in supercard

K.       Joel

x    10/17/89     fluor meeting
x    10/17/89     SPA in canada
     10/24/89     assume command of info packet development
     10/24/89     susan sheridan follow up
     10/24/89     cliff gauntlet, toolkits

L.       Deferred Meetings:

     joel, marcs, discuss terminology in X++ for developer release 
(particularly necromancers and related tools)
     Ravi, roger, discuss common view.

M.       Unassigned

     integrating ent is a nasty problem
     tell stroustrup about bombs in 2.0
     How well will canopy rebalancing perform?
     Tumblers as subclass of partiallyOrdered
     tutorial for users on the frontend
     reference manual for users on the frontend
     test scaffolding, code exercisers
     rollin, rollout a la Univac
     subaccounts, permissions on FE
     sys admin, i.e., archiving and partitioning disks


--marcs