Weekly Planning Meeting, 11/28

[Marc Stiegler <marcs>]

Highlights 11/28:

The big congratulations this past week go to dean and hill. Dean 
got enfilades working with splay trees (part of the congratulations 
go to markm as well, who helped resolve some design issues). 
Now, once the splay trees are working inside the ent, the ent 
will be ready to merge into Alpha--which will be the biggest 
single task to perform to have a Beta of the backend.

Meanwhile hill finished a new version of alpha, which has tables 
(the stuff hugh has been tearing his hair out over) merged in 
(part of the congratulations go to hugh as well).

We're getting close enough to a quantum leap in the quality of 
the backend for it to be exciting, folks.

Congratulations also go to hugh and roger for accelerating the 
current Alpha backend; whereas last week we could only retrieve 
documents with <20 characters without waiting for the Millennium, 
today we can get 200 characters. Now all we need is another 5 
orders of magnitude, and we'll be in good shape :-)

And congratulations go to bill, who has the whole-xanadu-backup 
part of our backup system tested and working smoothly at last; 
heretofore, we were counting on the incremental backups for everything.

Major Meeting: We will have a Capability Review starting at 11AM 
on Wednesday December 13. Basically, we will go through every 
feature of the InfoFactory frontend, item by item, and determine 
whether we know how to support the requirements for that feature 
in every level of our software, layer by layer (jacque, see me 
about how structuring this undertaking so we can record it and 
so we don't miss anything).This will also act as a warmup for 
a design review on the InfoFactory: people will be able to point 
out problems and issues, though we will not resolve anything 
in this meeting. Resolutions will probably be reached by groups 
of no more than 4 people in the aftermath.

Meetings: Monday, hill and markm do sequence number dangle checks. 
Hill, hugh, ravi, etc., discuss PackOBits. There will be a Debating 
Society meeting on sizzling frontends, Wednesday December 6 at 
7PM; marcs will present the InfoFactory design at 6PM that evening 
for interested parties in preparation for that meeting. Marcs, 
all the programmers, discuss Pearl Harbor options on December 
11 at 2PM. Bobp, markm, hugh, discuss third party interfaces. 
markm, michael, log sessions. Ravi, dean, talk about ent interface. 
Roger, hugh, dean, discuss system independent attributes for 
the frontends.


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

A.       dean

x    11/28/89     replace ocrum objects
x    12/05/89     virtual subtrees
x    11/28/89     enfilade with splay trees
     12/05/89     fix pack o bits
     12/05/89     partial orgls
     12/05/89     specify ent test tools
     12/12/89     merge ent into alpha!!
     12/12/89     upgrade regression test support
     12/12/89     figure out how to generate commands to xlint in smalltalk
     12/19/89     controllers for frontends

B.       hill

x    11/28/89     upgrade BE to use tables, alpha 4
     12/05/89     write mail about X++ upgrades
     12/05/89     confirm functioning of garbage collection
     12/05/89     merge ravi stuff into alpha
     12/06/89     turn length into count everywhere
     12/06/89     replace regions with enclosures
     12/13/89     basket & grandmap use garbage collection
     12/13/89     fix exception handling with sensors between comm handlers
     12/13/89     merge ent into alpha!!

C.       hugh

x    11/28/89     build cache for alpha performance
x    11/28/89     replaced iterators
x    11/28/89     merged with ravi
     12/05/89     finish file generator
     12/05/89     strip ropes out of alpha
     12/13/89     update broadcast
     12/13/89     integrate FE with Alpha

D.       markm

x    11/28/89     calc example for stubble
x    11/28/89     help dean repair the ent
     12/05/89     fix frozen bert grab
     12/05/89     help bobp fix up C interfaces
     12/06/89     spin up michael
     12/13/89     slut implementation of gatekeepers
     12/20/89     filtered backfollow
     12/20/89     respond to semantics email
     12/27/89     merge necromancers with dependencies
     12/27/89     merge necromancers with counted pointers
     12/27/89     complete to do list from alpha 3

E.       marcs

     11/29/89     michael's employment agreement
     12/05/89     negotiate with nat on C++ upgrade
     12/05/89     new infofactory documentation
     12/13/89     figure out how to repair email concept
     12/13/89     Capabilities Review
     12/13/89     figure out whether links are first class objects
     12/24/89     rough draft of AMIX/Xanadu frontend comparison
     12/31/89     rough draft AMIX/Xanadu integration plan
     12/31/89     start second round of patenting work

F.       michael

     12/05/89     find new/delete inversion in bomb test
     12/05/89     dean style, michael style regression testing
     12/05/89     refcount experiment

G.       ravi

     12/05/89     merge tables into current fe stuff
     12/12/89     merge tables into current febe stuff
     12/19/89     design extra doc&links protocol to hide orgls&berts
     12/26/89     waldo to data model with link schnert
     01/02/90     proxify D&L classes
     01/02/90     put D&L in smalltalk
     01/17/89     do comm in smalltalk

H.       bobp

     12/05/89     get put on PERT
     12/12/89     write xanadu system overview
     12/12/89     talk to terry winograd

I.       roger

     12/05/89     scrolling windows under C++
     12/05/89     review controllers
     12/13/89     follow a link
     12/19/89     link lists with simple documents in alpha
     12/19/89     traverse to acad, traverse from acad

J.       roland

     12/05/89     disjunction in formic
     12/12/89     eliminate && || in C++
     12/19/89     write grammar for formic
     12/19/89     parser with expression semantics
     12/26/89     xlint check for assign to a parameter variable and assignment to "this"
     12/26/89     automatic documenter

K.       Bill

x    11/28/89     test whole backup
     12/05/89     fix thinLAN

J.       Lars


K.       Joel

     12/05/89     new info packet draft

L.       Deferred Meetings:

     Markm, hill, discuss explicit deletion of orgls & refcount deletion. 
     Markm, hill, discuss exception handling on sensors between comm handlers.
     Dean, roger, michael, markm, discuss sequence numbers of size 4 billion. 
     Dean, markm, discuss intersecting backfollow ideas. 
     joel, marcs, discuss terminology in X++ for developer release 
(particularly necromancers and related tools)
     Ravi, roger, discuss common view.

M.       Unassigned

     merge contact databases
     scenario documents (markm, dean
     integrating ent is a nasty problem
     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