Dates
- 1:00pm on Monday, February 9, until 11:30 on Friday February 13
- Jump to: Monday | Tuesday | Wednesday | Thursday | Friday
Goals
Proposer | Proposed Goals (max 2) |
---|---|
KTL |
|
JK |
(K-T, I swear I didn't copy you, we just are thinking the same) |
XW |
|
KSK |
|
JFB |
|
FE | 1. Migrate to Github! Ship it! |
Logistics
Location
- SLAC, 2575 Sand Hill Road, Menlo Park, CA. Maps and directions
- Main meeting room: building 40, conference room #195 ("Sycamore"). Capacity: 48 people
- all coffee breaks and lunches will be delivered to Sycamore
Attendees
Attendee list (must be logged in)
Meeting rooms for parallel sessions
Each room can accommodate 20-25, unless noted otherwise. Buildings 040 and 052 are about 3 min walking distance apart.
- Monday:
- 052-108 "American River". 3pm-5pm
- 052-206 "Truckee River". 3pm-5pm
- Tuesday
- 040-195 "Cypress". 9am-5pm
- 052-103 A&B "Mad River". 9am-5pm
- 052-108 "American River". 11am-5pm
- Wednesday
- 040-195 "Cypress". 9am-5pm
- 052-108 "American River". 11am-2:30pm
- 052-203 "Merced River". 12-5pm
- Thursday
- 040-195 "Cypress". 9am-5pm
- 052-108 "American River". 9am-5pm
- Friday
- 040-195 "Cypress". 9am-1pm
- 052-108 "American River". 9am-1pm
Hotels
- Stanford Guest House, ask for "LSST Data Management" group. Room rate: standard queen $124, double queen $144. Space is very limited, so reserve early. Group rate is valid until Jan 9.
- Atherton Park Inn & Suites in Redwood City. Room rate: $147. No group code, rate is not guaranteed, the sooner you reserve the better. Link to the reservation page.
Agenda
The idea of this agenda is to provide as much productive working time as possible. There is nothing like face to face time and one of the most valuable things about a meeting like this is the ability to meet in both large and small groups. Interactions between pairs or triplets of teams are facilitated without the overhead of additional travel. These meetings can take place in the Strategy track. In many cases, some members of a team are needed in a particular discussion/meeting, but the rest of the team can contribute elsewhere. The hope is that people not needed directly by the strategy track can still get something done by joining the groups in the hack sessions. Keeping hack sessions on track will be difficult, but by planning ahead and having the CAMs actively involved, it should be possible to get a lot done.
Plenary track – all attendees are expected to attend
Hack/Strategy track – these are held in parallel so that team members not immediately needed in the strategy track can work in the hack sessions.
Planning track – This is planned in the evening. The intent is that the schedule can be a bit fluid, by default dinner will be ordered in while planning. If there is free time, planning can be moved, on a team by team basis.
Each day would follow the following logical structure:
Developer Track | Strategy & Management Track | |||||
---|---|---|---|---|---|---|
Plenary Talks / Demos | ||||||
Reports from yesterday, plans for today | ||||||
Lunch | ||||||
Work in smaller groups, in parallel. Agenda TBD by participants in "unconference" format. | Talks and Discussions | |||||
... | ||||||
Dinner | ||||||
Summer 2015 Sprint Planning |
The evening schedule is daunting, but we can't short change the planning and it is hard to hold in parallel with any other work since it benefits from input from the entire team.
Monday Feb. 9 | ||||
---|---|---|---|---|
Time | Plenary Track (Sycamore) | Hack Track (American River) | Strategy Track (Truckee River) | Plan Track (Sycamore) |
1:00PM | Introductions | |||
1:30PM | ||||
2:00PM | Long-term Schedule | |||
2:30PM | Break (refreshments) | |||
3:00PM |
| SUI developers hangout | Package Reorganization Discussion |
|
3:30PM | ||||
4:00PM | ||||
4:30PM | International Networking | Calibration Review | ||
5:00PM | Summer 2015 Planning (all teams in parallel) Dinner (order in) | |||
5:30PM | ||||
6:00PM | ||||
6:30PM | ||||
7:00PM |
Tuesday Feb. 10 | ||||
---|---|---|---|---|
Time | Plenary Track (Sycamore) | Hack Track (American River) | Strategy Track (Cypress) | Plan Track (Sycamore) |
9:00AM | Code, Communications Tools Refresher | |||
10:00AM | ||||
10:30AM | W2015 Retrospective (all teams) | |||
11:00AM | ||||
11:30AM | Lunch (provided) | |||
12:00PM | ||||
12:30PM | ||||
1:00PM | Github move / Package Reorganization Implementation | SLAC/IPAC Team Coordination Meeting (link) |
| |
1:30PM | ||||
2:00PM | ||||
2:30PM | ||||
3:00PM | Break (refreshments) (Level 3 discussion Gregory Dubois-Felsmann) | |||
3:30PM | Hackathon: TBD (all teams) in Sycamore | Architecture Deep Dive I in Truckee River | ||
4:00PM | ||||
4:30PM | ||||
5:00PM | ||||
5:30PM | ||||
6:00PM | Social Dinner (off campus, Buca di Beppo, 643 Emerson Street, Palo Alto, CA 94301. MAP, driving directions) | |||
6:30PM | ||||
7:00PM | ||||
7:30PM | ||||
8:00PM |
Wednesday Feb. 11 | ||||
---|---|---|---|---|
Time | Plenary Track (Sycamore) | Hack Track (American River) | Strategy Track (Cypress) | Plan Track (Sycamore) |
9:00AM |
| SUI developers hangout | ||
9:45AM | Camera Debriefing - CCS and Visualization | |||
10:30AM | W2015 Demo 1 (all teams) Tool for Camera Team: New Measurement Framework: Perry Gee Firefly Push & qserv interface : Trey Roby lead: Mario Juric | |||
11:00AM | ||||
11:30AM | Lunch (provided) | |||
12:00PM | ||||
12:30PM | ||||
1:00PM | Tools Improvement | Data Loading/Ingest (link) | ||
1:30PM | ||||
2:00PM | User Workspace slides | |||
2:30PM | ||||
3:00PM | Break (refreshments) | |||
3:30PM | Hackathon: TBD (all teams) ??? | Package reorganization (not renaming!!!) | ||
4:00PM | ||||
4:30PM | ||||
5:00PM | ||||
5:30PM | ||||
6:00PM | Summer 2015 Planning (all teams in parallel) Dinner (order in) CAM Session: Loading/sequencing Spreadsheet Bugs with SP | |||
6:30PM | ||||
7:00PM | ||||
7:30PM | ||||
8:00PM |
Thursday Feb. 12 | ||||
---|---|---|---|---|
Time | Plenary Track (Sycamore) | Hack Track (American River) | Strategy Track (Cypress) | Plan Track (Sycamore) |
9:00AM | SUI developers hangout | CAM Training: EV Training, Tools - KL W15 Results to Date, Closeout - KL Kanban vs. Scrum, Example - FE Agile issues - JK
(S15 Epics - CAMs 1:1 w/KL) (Key Performance Metrics - JDS)
| ||
10:00AM | SUI developers hangout | |||
10:30AM | W2015 Demo 2 All teams meas_astrom replacement: afw::table changes: Jim Bosch Firefly/qserv - Trey Roby Firefly in iPython notebook- | Summit-Base networking (with Camera group) | ||
11:00AM | ||||
11:30AM | Lunch (provided) How to modify 1-line of code demonstration SCONS, eups tutorial |
| ||
12:00PM | ||||
12:30PM | ||||
1:00PM | Visualization Requirements Design |
| DM Sizing Model | |
1:30PM | ||||
2:00PM | ||||
2:30PM | ||||
3:00PM | Break (refreshments) | |||
3:30PM | Base Center Design | Github / Repo Rename: Pedal-to-the-Metal Hackathon (all teams) | Integrated System Database Requirements Design |
|
4:00PM | ||||
4:30PM | Camera Debrief Part II - Chris Stubbs on sensor characteristics (link) | Base Center Design (continued) | ||
5:00PM | ||||
5:30PM |
|
|
| |
6:00PM | Summer 2015 Planning (all teams in parallel) Dinner (order in) | |||
6:30PM | ||||
7:00PM | ||||
7:30PM | ||||
8:00PM |
Friday, February 13 | ||||
---|---|---|---|---|
Time | Plenary Track (Sycamore) | Hack Track (American River) | Strategy Track (Cypress) | Plan Track (Sycamore) |
9:00AM | Copyrights/Headers | |||
10:00AM | Summary and Wrap Up | |||
10:30AM | ||||
11:00AM | Alert Generation, Brokering/Distribution | Hackathon: TBD |
| |
11:30AM |
|
Suggestions for this meeting
Suggestions about agenda or anything else you can think of related to this meeting goes here, the DMLT will keep revisiting them.
Item | Desired Outcome | Duration (estimator: estimate) | Needed Attendees | Prerequisite Work |
---|---|---|---|---|
Introductions | Everyone knows everyone else | KTL: 1 hr | Everyone | Prepare one slide per person? |
Summer2015 planning | Epics, story points, dependencies, people, order. Ensure we address LDM-240 Key Progress Metrics. | KTL: 4 hr | JK, MJ, CAMs, science leads, everyone invited | Groom backlogs, generate potential epics, estimate lengths, tentative assignees, sequence/dependencies |
Long-term schedule | Everyone knows the long-term schedule for LSST, and the key dates for DM deliverables to other LSST systems and to ComCam and other commissioning activities | gpdf: 0.5 hr | JK, MJ, CAMs, science leads, everyone invited | Refresh visual materials with project milestones; extract summaries of inter-system dependency dates as slides |
Baseline architecture deep-dive | Complete dataflow descriptions, process descriptions, component designs with explicit TBDs | KTL: 8 hr | KTL, MJ, RHL, GPDF, DP, science leads, everyone invited | Finish writing up design documents |
Hackathon: low-priority issues | Clear out small but useful issues that keep getting delayed | KTL: 4 hr | Identify appropriate issues, owners, participants | |
Coding/communication tools training/refresher | Educate all developers on best practices | KTL: 1 hr | KTL, new developers, everyone invited | Ask for areas of current confusion |
Tools improvement | List of suggested changes | KTL: 1 hr (in parallel with previous) | FE, existing developers | |
Reorg: Discussions on big API changes including package reorganization | Decisions, identification of potential problems and implementation plan | KTL: 1.5 hr | Pkg reorg: JimB, MJ, KTL, FE, everyone invited | Identify issues, owners, participants |
Hackathon: Repo renaming / re-organisation | Implement low-hanging fruit from the Reorg: session above. Potentially move repos to GitHub as they are done. | 2.5 hr | KTL, FE, JimB, all stack repo co-ordinators | Naming scheme, GitHub CCB approval, outcome of KTL Reorg meeting above, README template |
Social event (dinner) | Everyone knows everyone else better | Everyone | Find and reserve venue | |
Review all places calibration touches design | Explicit TBDs, needed information from calibration, needed data for calibration | KTL: 2 hr (in addition to deep-dive) | RHL, MJ, GPDF, KTL, everyone invited | Publish draft calibration plan update |
User workspace requirements and design | Design sketch | KTL: 2 hr | XW, TR, DC, JacekB, MJ, GPDF, KTL, everyone invited | |
Debug visualization requirements and design breakout | Design sketch | XW, TR, RHL, JimB, GPDF, KSK? | ||
Briefing from Camera personnel | Understanding of current Camera activity and DM dependencies | KTL: 1 hr | GPDF, PatH/ StuartM/ TonyJ/ AaronR/ AndyR? | Develop outline with Camera team, determine presenter(s) |
Leadership Team sidebar | Decide on process issues and major items | KTL: 2 hr | DMLT, CAMs?, science leads? | Identify issues and proposed resolutions |
Integrated system database requirements and design | Design sketch | KTL: 2 hr | KSK, Sims person?, JacekB, GPDF, MJ?, KTL? | |
Data loading/ingest – current and future requirements and design | Design sketch | KTL: 2 hr | JacekB, JimB, SMM, AS, JohnS | |
Dominique Boutigny presentation about CFHT | Understanding of issues and successes | KTL: 1 hr | DB, Everyone? | |
Hackathon: finish up W2015 loose ends | W2015 ready for QA prior to release | KTL: 4 hr | Identify issues, owners, developers | |
SLAC team and IPAC team meeting | face to face meeting, discuss any API issues | XW: 2-4 hr | SLAC and IPAC teams, others welcome | |
W2015 Release demo | Demonstrate W2015 (pre-QA) features to Product Owners | KTL: 3 hr | MJ, RHL, KTL | Loose end hackathon |
W2015 retrospective | Understand successes and issues with W2015 development process | KTL: 1.5 hr | Everyone |
|
DM Infrastructure Sizing Model | Determine future evolution of model, plan for refresh, applicability to annual hardware procurement process, DAC sizing/costing for external requests, Cloud-based model | JK: 3 hr | Jeff Kantor, Don Petravik, Mike Freemon, Toni Perez, K-T Lim, Ron Lambert, Jacek Becla, Gregory Dubois-Felsmann | Familiarize with existing model |
Hackathon: Docuthon 101 | "So you want to change 1 line of code" - single page document that has, or has pointers to, all the information needed to contribute to the LSST stack (aka 'Developer Guide') | FE: 1.5 hr | FE, JS, everyone invited | Skeleton document and docuthon process |
Cost Account Manager training and EVMS process (partial overlap with Leadership Team Sidebar above) | If we don't do this before the AHM (likely) we need to refresh those with training and train those who have not had it yet, discuss improvements to the process, etc. This will also cover the DM Risk Register and monthly Risk Reviews. | JK: 3 hr | Jeff Kantor, Kevin Long, John Swinbank, Margaret Gelman, Frossie Economou, Simon Krughoff, Xiuqin Wu | Training delivered, process improvements captured |
Long lunch/breaks/spare slots for unscheduled discussions
Suggestions for next meeting
- Name Tags
- Shorter days (planning at night is too late, people are burned out)
- Need unscheduled chunks of time (e.g. in the PM) planned on the fly along with pre-planned sessions (e.g. in the AM)
- CAM half-day at front and back end
- Better planning in advance by CAMs/DMLT
- Provide information about cross-subsystem interactions through CAMs/PM/PS
- Provide questions to be answered and decisions to be made rather than session titles
- Useful to have Camera interaction here but only for fractions of the audience
- Should have had "welcoming plenary" on overall project status as well as DM
- Need tutorial but only for new people
- Product demos good
- More meals on our own, or at least with user-selectable options
- More time out of doors
- Make use of Camera, T&S, OCS, etc. people, make sure to invite them and give them reasons to come
4 Comments
Gregory Dubois-Felsmann
I'd like to suggest an additional topic. This can be covered in one of the existing sessions or can be called out explicitly in the agenda: briefing the entire (DM) team on the latest version of the high-level schedule for the project, and in particular on the dates when other systems will be expecting various components of DM software and operational systems to be available. Notably this can include the expectations for the commissioning camera.
My guess is that this could be done in 15-20 minutes.
Jeff Kantor
For Gregory's suggested topic, I don't see a session that this fits into exactly, so Gregory, please add this to the list of topics as a new row.
Gregory Dubois-Felsmann
Done
Frossie Economou
Can I suggest that hackathon sessions are scheduled last (afternoon) in the daily program so that people have the option to keep working / order in pizza in if one of them happens to turn into an orgy of collective productivity.
I have edited those session titles to have a Hackathon: prefix to make identification easy.