[almighty] Clarification around Types
Todd Mancini
tmancini at redhat.com
Thu Jan 26 14:24:54 UTC 2017
Can teams be hierarchical? No. The purpose of teams is to have a 'flat
list' that maps to the hierarchy of areas. This is a tremendous help in
setting up a reasonable UI with reasonable navigation. When navigation is
based upon hierarchies, it becomes very difficult and unruly quickly.Teams
are a great way to collapse the complex hierarchy down to a flat list. It
also permits navigation which does not map well to the hierarchy. (e.g. if
the hierarchy is mapped to functional areas -- in our case, things like
Platform, Planner, Deploy, etc. -- then what do you do for people that
interact with all of those areas? Cross-cutting concerns like Security and
Documentation. Teams let you define a Security team which maps to the
security areas of each of those functional areas -- essentially providing a
pivot on the original hierarchy.)
Are releases a type of iteration? Yes and no. Iterations are paths. There
is no pre-defined meaning on the elements of those paths. So while we may
set up something such as:
Carnation
Sprint 124
Sprint 125
Sprint 126
Sprint 127
Daffodil
Sprint 128
Sprint 129
...
the saas has no idea that Carnation and Daffodil are 'releases'. We, as
this team, are simply using this convention and all agree that those are
releases. That's why I answered 'yes and no'. Yes, inasmuch as you can use
iteration paths to represent releases. No, inasmuch as the system does not
have a predefined concept of a 'release'.
This model lets you do things such as: map Experience E73 to Daffodil, and
map the child Features of E73 to the various sprints. (e.g. Feature F1221
is a child of E73 and has an assigned iteration of 'Daffodil/Sprint 129')
On Thu, Jan 26, 2017 at 8:48 AM, Leonard Dimaggio <ldimaggi at redhat.com>
wrote:
> Questions/Comments on the diagram:
>
> Team Hierarchy
>
> - Teams contain 1..n people (each person can be on 1..n teams) and
> provide user perms
> - Each team can be associated with 1..n areas
>
> Organization Hierarchy
>
> - Spaces (formerly called "projects"), each space contains one backlog
> - Areas - hierachical, expressed in paths
> - Iterations (can also have child iterations)
>
> Planning Hierarchy
>
> - Epics
> - Stories
> - Tasks
>
> Execution Hierarchy
>
> - Release (can also have child releases)
> - Iterations (can also have child iterations)
> - Sprints
>
> Open Questions:
>
> - Can teams be hierarchical?
> - Are releases a type of iteration?
>
> Thx!
>
>
>
> On Mon, Jan 23, 2017 at 4:39 PM, Adam Jolicoeur <ajolicoe at redhat.com>
> wrote:
>
>> Hi Everyone,
>>
>> As part of this conversation, I thought it would make sense to share the
>> Areas diagram that UX has created, through discussions with Todd and
>> Michael V. Additional information can be found on GitHub:
>> https://github.com/fabric8io/fabric8-ux/issues/101.
>>
>> -Adam
>>
>> On Jan 23, 2017, 4:06 PM -0500, Michael Virgil <mvirgil at redhat.com>,
>> wrote:
>>
>> The intent is as Konrad describes.
>>
>> Based on the current behavior, there are some issues that can be
>> addressed in the design/implementation to make this clearer. Would also
>> need to ensure that the context and providing what can be created within
>> that context from the backend. UX team is currently working on refining the
>> UX/interaction to clarify this user experience - in this area.
>>
>> In general; the process template selected when creating the Space defines
>> the process hierarchy (planning work Item relationships, among other
>> things).
>>
>> The intent:
>>
>> - The quick create, should be just that, quickly create goal based
>> work items at a level that I've determined by my setting my context
>> - The context; where I am working within the hierarchy; Theme,
>> Initiative, Epic, Story, Task... Vision, Scenario, Value Prop, Experience,
>> Feature, Task... or whatever
>> - User Experience needs to be clean, clear and obvious to the user
>>
>> I should be able to set my context and then...
>>
>> - Create a bunch of tasks for a story...
>> - Create a bunch of stories to refine the goal of an Epic...
>> - Create a bunch of stories in the backlog...
>> - Create a bunch of initiatives while brainstorming...
>> - etc.
>>
>> It would be easy to envision a common set of filters/views/tree
>> structures that would be applied to all these hierarchies. The look and
>> feel would be the same and comfortable for parent/child relationships -
>> defined in the process template. Should not need any custom code in the
>> backend or FE to accommodate these process templates. If we focus on
>> Agile/Scrum for the first implantation using the templates, I would assume
>> adding Waterfall, Red Hat or other process would not be a major feat...
>>
>> I would also anticipate that the process template would allow a team over
>> time, not out of the gate, to define the policies, rules and/or constraints
>> of the process. This could include the rules/policies for work items
>> (parent/children) relationships, iterations (milestones), policies for
>> permissions, or even the fields to present on views. I think there is a lot
>> we could configure and potentially allow to be defined within the template.
>>
>> Just some things we need to think about.
>>
>> For now, lets just get a simple end-2-end use case for the parent/child
>> hierarchy working using a process template and we can extend it over time.
>>
>> Michael
>>
>> On Mon, Jan 23, 2017 at 8:36 AM, Aslak Knutsen <aslak at redhat.com> wrote:
>>
>>> If that's the same QuickAdd bar, then I think it should visually be
>>> moved inside the Open WI draw to show the context things are being added
>>> and the 'type selector' be filtered same as 'allowed links'.
>>>
>>> -aslak-
>>>
>>> On Mon, Jan 23, 2017 at 2:23 PM, Konrad Kleine <kkleine at redhat.com>
>>> wrote:
>>>
>>>> Aslak,
>>>>
>>>> yes we need something superior, but if the quick add bar simply
>>>> performs two actions rather than one, we're safe:
>>>>
>>>> 1. Add work item (userstory)
>>>> 2. Link to currently "open" work item (epic)
>>>>
>>>> That implies three things (at least)
>>>>
>>>> 1. we can have some form of navigation that makes allows for filtering
>>>> user stories based on a concrete link type (e.g. child of)
>>>> 2. the quick add bar knows about which link type can be used (maybe the
>>>> one which was used to filter in the first place?)
>>>> 3. the quick add bar adds the new work item (userstory)
>>>> 4. the quick add bar creates the link between the epic (currently
>>>> filtered on) and the newly created user story
>>>>
>>>>
>>>>
>>>> On Mon, Jan 23, 2017 at 2:08 PM, Aslak Knutsen <aslak at redhat.com>
>>>> wrote:
>>>>
>>>>> That workflow is unrelated to 'Epics'.
>>>>>
>>>>> It's what we do in every planning sessions on some level.
>>>>>
>>>>> Create/Read a userstory, add 10 tasks.
>>>>>
>>>>> It's a feature for, I create/have a parent, now create n children.
>>>>>
>>>>> We need a better approach than;
>>>>>
>>>>> * Create parent
>>>>> * Create child 1
>>>>> * Create child 2
>>>>> * Create child n
>>>>> ...
>>>>> * Open parent
>>>>> * Search and Link child 1
>>>>> * Search and Link child 2
>>>>> * Search and Link child n
>>>>> ...
>>>>>
>>>>>
>>>>> -aslak-
>>>>>
>>>>> On Mon, Jan 23, 2017 at 1:58 PM, Konrad Kleine <kkleine at redhat.com>
>>>>> wrote:
>>>>>
>>>>>> If I were a user of the system I'd probably go in and
>>>>>>
>>>>>>
>>>>>> 1. create an epic from the work item type dialog.
>>>>>> 2. go to the quick add and add 10 features
>>>>>>
>>>>>> IMHO we shouldn't first build for custom tailored UI solutions that
>>>>>> fit a work flow specific to epics. I'm not saying we shouldn't do this at
>>>>>> all but if we have the tools in place to do operate generically, then this
>>>>>> is good start, no?
>>>>>>
>>>>>> On Mon, Jan 23, 2017 at 1:51 PM, Todd Mancini <tmancini at redhat.com>
>>>>>> wrote:
>>>>>>
>>>>>>> Can someone describe to me the experience of this common usage
>>>>>>> pattern: a user wants to create a new Epic and ten related Features. What's
>>>>>>> that workflow for that?
>>>>>>>
>>>>>>> On Mon, Jan 23, 2017 at 7:39 AM, Konrad Kleine <kkleine at redhat.com>
>>>>>>> wrote:
>>>>>>>
>>>>>>>> We currently have the a little icon next to the left most field
>>>>>>>> once you've clicked the quick add button. I suggest to make this a toggle
>>>>>>>> just like we used to have it for the work item symbol in the top left
>>>>>>>> corner of the details page: https://cloud.githubuserconten
>>>>>>>> t.com/assets/193408/21686784/d717829e-d366-11e6-9641-b61cece
>>>>>>>> 54c99.png. Currently this is just a symbol and no longer a menu.
>>>>>>>>
>>>>>>>> On Mon, Jan 23, 2017 at 1:26 PM, Aslak Knutsen <aslak at redhat.com>
>>>>>>>> wrote:
>>>>>>>>
>>>>>>>>> Then what happens to QuickAdd when there are multiple types shown
>>>>>>>>> in a list?
>>>>>>>>>
>>>>>>>>> On Mon, Jan 23, 2017 at 1:21 PM, Konrad Kleine <kkleine at redhat.com
>>>>>>>>> > wrote:
>>>>>>>>>
>>>>>>>>>> Hey Aslak,
>>>>>>>>>>
>>>>>>>>>> > The QuickAdd feature is currently hard coded to add items of
>>>>>>>>>> type 'UserStory'. With no fixed types and no option to change the type this
>>>>>>>>>> looks wrong.
>>>>>>>>>> > Should the Process Definition have some option to define a
>>>>>>>>>> 'default' Type that should be used here?
>>>>>>>>>> > Should the Quick Add have an option to select which type it
>>>>>>>>>> should quick add? Optionally, should the Quick Add remember the users
>>>>>>>>>> previously used type? Or
>>>>>>>>>> > possible types based on some form of role, e.g. PM's would most
>>>>>>>>>> likely create Experiences, SM UserStories and Devs Tasks?
>>>>>>>>>>
>>>>>>>>>> As far as I understood UI and UX, they want to make the quick add
>>>>>>>>>> bar context sensitive. So when you're viewing a list of user stories, the
>>>>>>>>>> quick add will just add user stories. But if you're viewing a list of
>>>>>>>>>> epics, the quick add shall add epics. I have no clue about the technical
>>>>>>>>>> realization yet but that's at least what I heard from talking to different
>>>>>>>>>> people. It seems the most convenient to me instead of having some process
>>>>>>>>>> defining what you can quick add.
>>>>>>>>>> The only problem I see is how to get the "context", e.g. how to
>>>>>>>>>> know that the list of work items is currently just showing features, epics
>>>>>>>>>> or user types. Shall this be done with filters or maybe implicit filters?
>>>>>>>>>>
>>>>>>>>>> - Konrad
>>>>>>>>>>
>>>>>>>>>> On Mon, Jan 23, 2017 at 12:08 PM, Aslak Knutsen <aslak at redhat.com
>>>>>>>>>> > wrote:
>>>>>>>>>>
>>>>>>>>>>> A few points to clear up around the generic type system.
>>>>>>>>>>>
>>>>>>>>>>> 'Epic'
>>>>>>>>>>>
>>>>>>>>>>> In the UX designs, Epic is special, e.g. it shows up as summary
>>>>>>>>>>> points in the same draw as Iterations and can be created directly from the
>>>>>>>>>>> 'right-click' menu: https://redhat.invisiona
>>>>>>>>>>> pp.com/share/QU9U8D8GF#/screens/212141138
>>>>>>>>>>>
>>>>>>>>>>> What is the generic term/feature for this? Is the intent to be
>>>>>>>>>>> able to define as part of the Process Template which Type should be the
>>>>>>>>>>> 'special/top' type? What is the equivalent type in the PDD style,
>>>>>>>>>>> Vision/Experience?
>>>>>>>>>>>
>>>>>>>>>>> Part of https://github.com/fabric8io/fabric8-planner/issues/657
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> 'WorkItem Links hierarchy'
>>>>>>>>>>>
>>>>>>>>>>> Same as a Feature can be a Child of an Experience, a UserStory
>>>>>>>>>>> is a Child of an Epic. What are the Link Forward and Reverse names for this
>>>>>>>>>>> relationship?
>>>>>>>>>>>
>>>>>>>>>>> Relates to? Belongs to? Child of? Part of?
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> 'Quick Add'
>>>>>>>>>>>
>>>>>>>>>>> The QuickAdd feature is currently hard coded to add items of
>>>>>>>>>>> type 'UserStory'. With no fixed types and no option to change the type this
>>>>>>>>>>> looks wrong.
>>>>>>>>>>>
>>>>>>>>>>> Should the Process Definition have some option to define a
>>>>>>>>>>> 'default' Type that should be used here?
>>>>>>>>>>> Should the Quick Add have an option to select which type it
>>>>>>>>>>> should quick add? Optionally, should the Quick Add remember the users
>>>>>>>>>>> previously used type? Or possible types based on some form of role, e.g.
>>>>>>>>>>> PM's would most likely create Experiences, SM UserStories and Devs Tasks?
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> PM's input appreciated.
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> -aslak-
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> _______________________________________________
>>>>>>>>>>> almighty-public mailing list
>>>>>>>>>>> almighty-public at redhat.com
>>>>>>>>>>> https://www.redhat.com/mailman/listinfo/almighty-public
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>> _______________________________________________
>> almighty-public mailing list
>> almighty-public at redhat.com
>> https://www.redhat.com/mailman/listinfo/almighty-public
>>
>>
>> _______________________________________________
>> almighty-public mailing list
>> almighty-public at redhat.com
>> https://www.redhat.com/mailman/listinfo/almighty-public
>>
>>
>
>
> --
> Len DiMaggio (ldimaggi at redhat.com)
> JBoss by Red Hat
> 314 Littleton Road
> Westford, MA 01886 USA
> tel: 978.392.3179 <(978)%20392-3179>
> cell: 781.472.9912 <(781)%20472-9912>
> http://www.redhat.com
> http://community.jboss.org/people/ldimaggio
>
>
>
> _______________________________________________
> almighty-public mailing list
> almighty-public at redhat.com
> https://www.redhat.com/mailman/listinfo/almighty-public
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/almighty-public/attachments/20170126/512f1e95/attachment.htm>
More information about the almighty-public
mailing list