# SAM

## Sprint 1 - Review & Retrospect

* When: November 15, 2019 from 1 pm to 3 pm.
* Where: Meeting room 202, E.J. Pratt Library

## Process - Reflection

#### Decisions that turned out well

##### Good decision 1:
**Decision**: Splitting the team into three subteams:
* The back-end team: Designs and develops the the database and the API.
* The front-end team: Designs and develops the react component and static webpages.
* The algorithms/research team: Designs and develops the SAMpy package which includes all the algorithms needed to produce matchings

**Success criteria**:
* Being able to work on various features concurrently and independently
* Providing opportunity for teammember to focues on the skills that they want to develop and sharpen

**Artifacts**:
* SAMpy package link: [SAMpy](https://github.com/csc301-fall-2019/team-project-sam/tree/research/components/research/sampy)
* Research branch: [research](https://github.com/csc301-fall-2019/team-project-sam/tree/research)
* Front-end branch: [frontendsetup](https://github.com/csc301-fall-2019/team-project-sam/tree/frontendsetup)
* Develop/n_37 branch: [n_37](https://github.com/csc301-fall-2019/team-project-sam/tree/n_37)

##### Good decision 2:
**Decision**: Channels with separate responsibilities on Slack
**Success criteria**:
* Algorithms: Channel for the algorithms/research team. This channel was used to discuss mostly where we were and what our expectations of the other team members were regarding the matching, statistics and diff tool algorithms.
* Logistics: This channel was used to discuss where/when meetings were supposed to happen, and if there were any changes in this schedule.
* Product: Discussing all interactions with end user and evaluating them.
* Deliverable-cramming: While working on deliverable one many team had many fruitful but intense discussions and in order to keep other channels neat following channel was created.
* Tech-stack-issues: Discussing various aspects of technologies which were used and providing help and useful resources for better understanding of utilized technologies.
* Tutorial: This channel was used to discuss matters regarding the information presented during the tutorials.
* Urgent: As the name suggests, this channel was used for discussions that had high priority in our todo list, and that required the attention of multiple team members.

##### Good decision 3:
**Decision**: Meeting every Friday for at least 2 hours
**Success criteria**:
* Resolving major questions
* Staying connected as a team and keeping team spirit up
* Constantly reminding of each individual's responsibilites within team

##### Good decision 4:
**Decision**: Using GitHub projects as our Scrum board:
* Organized in four columns: Backlog/to do, In Progress, Review, Done
* Create milestones
* For each major task in a milestone, create a task and put an Epic tag on it and link it to a the corresponding milestone
* Create issues for subtasks, assign them to one or more team members, link them to specific milestones and put a label on them

**Success criteria**:
* Ease the organization of each team
* Be able to follow the progression of everyone
* Know on which task everyone is working done
* Visualize progression of each team

**Artifacts**:
* GitHub project link (for Sprint 1): https://github.com/csc301-fall-2019/team-project-sam/projects/2
* GitHub project link (for D2 - Documentation):
https://github.com/csc301-fall-2019/team-project-sam/projects/3
#### Decisions that did not turn out as well as we hoped

##### Bad decision 1:
**Decision**: Grey working alone on the front-end
**Failure criteria**:
* Difficulty handling workload and learning new tools
* Resulted in communication and integration issues between technologies
* Difficult to establish what's needed in the front end without knowing the structure of the backend
* Having more overlap between teams could result in more efficient development

##### Bad decision 2:
**Decision**: Splitting the subteams work on non overlapping slots during the reading week
**Failure criteria**:
* Not achieving expected goals set for the reading week
* Resulted in communication issues and, hence, being less productive

#### Planned changes

List any process-related changes you are planning to make (if there are any)

* Ordered from most to least important.
* Explain why you are making a change.

**Change**: Increase inter-team communication by establishing what is expected and needed of other teams
**Reason**: Progress for some teams was stilted by not having enough information about the progress of other teams. By keeping closer tabs on each other and understanding team dependencies, each team will be able to make faster progress.

**Change**: Uphold stricter branching/git conventions
**Reason**: Loose adherence to the git workflow caused confusion among some teams, resulting in overwritten work and lost time. The process outlined in our initial README.md would prevent conflicts and facilitate a smoother collaborative workflow.

**Change**: Establish priorities and common goals to share between teams
**Reason**: Different teams had different ideas of what was important towards the end of the sprint, resulting in a disjoint product. Establishing what is most important to finish, and what teams need from each other to achieve it, will make integration at the end easier.

## Product - Review

#### Goals and/or tasks that were met/completed:
* Frontend
* Home page ([Mockup](https://xd.adobe.com/view/dec6ca3b-0f0d-4b00-6ae0-96d9fe2438c5-a8f7/))
* Matching criteria page (Mockup)
* Matching results page (Mockup)
* Individual student Data page (React app)
* All students overview page (React app)
* Add student page (React app)
* Edit student page (React app)
* University overview page (React app)
* Add university page (React app)
* Edit university page (React app)
* Resources and Database interfaces
* Write Django models to implement the resource storage system (database, implicitly).
* Write specification on each resource’s methods using RFC 2119 keywords
* Database architecture:
* Due to lack of technology knowledge initial database architecture was not implementable within restrictions of the stack and, hence, was changed eventually.
* API
* Student Resource Implementation (done, tested)
* University Resource Implementation (done, tested)
* Region Resource Implementation (done, tested)
* Application Resource Implementation (done)
* Vairous glabal resource Implementation (done, tested)
* Matching algorithms ([SAMpy package](https://github.com/csc301-fall-2019/team-project-sam/tree/research/components/research/sampy))
* Gale-Shapely algorithm
* Unit-test Gale-Shapely
* Matching benchmark/statical overview
* Data extraction and testing of Sophie's exel
* Compute score of student in a wished university
* Rank students per university
* Comparison tool/Differences log
* SAMpy diff tools to compare the results of two matchings both from a university and a student perspective.

#### Goals and/or tasks that were planned but not met/completed:

* Frontend
* Login page
* turned out not to be necassary since we did not implement authentification yet.
* Matching history page
* The resource and database models are not ready yet.
* Two matchings comparison page
* The resource and database models are not ready yet.
* Resources and Database interfaces
* Write JSONSchema descriptors for each resource, most rigourous form of documentation.
* We had to significantly change our design of the schema due to implementation criteria of the database (Django ORM)
* API
* Matching parameters Resource Implementation
* Not enought time

* Comparison tool/Differences log:
* history of all previously saved matches and comparison of selected past matches
* Least priority after the Matching resource.

#### How was your product demo?

As it will be explained below, we didn't present our demo to Sophie, our early adopter, because we were not able to deploy a fully integrated app. However, we are proud to have finished the MVP version of SAMpy and have been able to test it with last year's applications to the study abroad program at Ecole Polytechnique. We created a [report](https://github.com/csc301-fall-2019/team-project-sam/blob/research/components/research/reports/mock_test_report.pdf) explaining how SAM outperforms the manual method of Sophie she used last year. We will meet with Sophie next week to discuss about it.

## Meeting Highlights

##### Highlight 1:

As developers, we decided to organize our work around a long term vision of our development needs. This meant considering our long term maintainability and expandability needs, that is, working on a horizontal basis, rather than hard coding a single page to work correctly and neglecting the quality of our code.

The grading timeline for deliverable 2 is incompatible with our organization. Although we have made significant progress towards an MVP and created the foundation for all features' implementations, we have not produced one single feature to a satisfactory level.

##### Highlight 2:
At the beginning of the sprint for D2, Osman and Guillaume were dependent on Loan (backend team) because he knew more about APIs and Django. During the past two weeks, they have thus learned a lot from him and are now able to work on their own.
This will improve the efficiency of the backend team for the next iteration.