Hi,
I read this in Stephen's summary from the FOSSP thread and I wanted to discuss it:
== Role Process == Jóhann does not like the term PRD and feels that the term as defined[4] doesn't really make sense for a community project. Moreover, he feels it doesn't make sense when applied to the role projects. He envisions that each role we harden and tie into our deployment API should be developed as a project of its own, following a different (non-PRD-driven) process. He suggested that Stage-Gate[5] might be a useful approach for this.
Why doesn't a PRD make sense for a community project? Why, specifically is having a PRD at odds with the product having various roles?
Stage-Gate doesn't appear to be a substitute for creating a PRD... it looks like a standard software development process, to be honest. And you could say we're following it right now - Stage 0 / Discovery was the Fedora.next proposal and iterations, Stage 1 / Scoping is the ongoing working group discussions... But please correct me if I'm wrong - I never heard of Stage-Gate before.
The main alternative to using PRDs that I'm aware of is the agile model, which I don't think we're following. But even in agile you have to put requirements together in the form of user stories [1]...
That being said: isn't it a "Pandora's Box" to start geeking out on software development methodologies? Is this really the best use of our time? We need requirements, no matter what.
Please, can we just come up with a requirements document?
~m
On 11/19/2013 04:44 AM, Máirín Duffy wrote:
Hi,
I read this in Stephen's summary from the FOSSP thread and I wanted to discuss it:
== Role Process == Jóhann does not like the term PRD and feels that the term as defined[4] doesn't really make sense for a community project. Moreover, he feels it doesn't make sense when applied to the role projects. He envisions that each role we harden and tie into our deployment API should be developed as a project of its own, following a different (non-PRD-driven) process. He suggested that Stage-Gate[5] might be a useful approach for this.
Why doesn't a PRD make sense for a community project? Why, specifically is having a PRD at odds with the product having various roles?
I think this is getting misunderstood.
I've never said it should not be used.
I have disliked where it's and how it's trying to be used as in applied to the "Server Community" as opposed to the "Server Role" and Server
If you looked at each application in one of the server roles you see a "links to PRD?" behind that application.
Stage-Gate doesn't appear to be a substitute for creating a PRD...
It is not however it is what can be applied to our "Server Communities"
PRD's ( or requirements documents ) are part of various stages.
it looks like a standard software development process, to be honest.
Last time I checked we are in the business of delivering software.
And you could say we're following it right now - Stage 0 / Discovery was the Fedora.next proposal and iterations, Stage 1 / Scoping is the ongoing working group discussions... But please correct me if I'm wrong
No you are correct stage gate is just a phase driven approach as in once you complete stage one you go to stage to etc.
The main alternative to using PRDs that I'm aware of is the agile model, which I don't think we're following. But even in agile you have to put requirements together in the form of user stories [1]...
That being said: isn't it a "Pandora's Box" to start geeking out on software development methodologies? Is this really the best use of our time? We need requirements, no matter what.
Please, can we just come up with a requirements document?
For the first stage gate was just something that I intended Stephen to look into not share with the rest of the world before I could form a proposal and it being shot down without me even having one but given that I had expect our discussion might lead to more confusion I started writing this late last ( and into the ) night [1] which hopefully clears things up a bit.
Again requirement documents are *part of* the stage gate process.
JBG
1. https://fedoraproject.org/wiki/User:Johannbg/FOSSP#Stage_Gate.27s_Process
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 11/19/2013 02:50 AM, "Jóhann B. Guðmundsson" wrote:
On 11/19/2013 04:44 AM, Máirín Duffy wrote:
Hi,
I read this in Stephen's summary from the FOSSP thread and I wanted to discuss it:
== Role Process == Jóhann does not like the term PRD and feels that the term as defined[4] doesn't really make sense for a community project. Moreover, he feels it doesn't make sense when applied to the role projects. He envisions that each role we harden and tie into our deployment API should be developed as a project of its own, following a different (non-PRD-driven) process. He suggested that Stage-Gate[5] might be a useful approach for this.
Why doesn't a PRD make sense for a community project? Why, specifically is having a PRD at odds with the product having various roles?
I think this is getting misunderstood.
I've never said it should not be used.
I have disliked where it's and how it's trying to be used as in applied to the "Server Community" as opposed to the "Server Role" and Server
To reiterate (because I think I failed to make the point properly in my late-night email yesterday), no none is disagreeing that we don't need a requirements and goal document. As I understand it, Jóhann feels that we probably need to do this differently for both the Platform and the Roles. I am in general agreement on this, though my stance was that for the January deliverable, it's probably acceptable for the Platform PRD to include the line "A process will be established for promoting services into featured roles".
If you looked at each application in one of the server roles you see a "links to PRD?" behind that application.
Stage-Gate doesn't appear to be a substitute for creating a PRD...
It is not however it is what can be applied to our "Server Communities"
PRD's ( or requirements documents ) are part of various stages.
it looks like a standard software development process, to be honest.
Last time I checked we are in the business of delivering software.
And you could say we're following it right now - Stage 0 / Discovery was the Fedora.next proposal and iterations, Stage 1 / Scoping is the ongoing working group discussions... But please correct me if I'm wrong
No you are correct stage gate is just a phase driven approach as in once you complete stage one you go to stage to etc.
In my attempts to clarify, I think I used mud instead of spring water...
The main alternative to using PRDs that I'm aware of is the agile model, which I don't think we're following. But even in agile you have to put requirements together in the form of user stories [1]...
That being said: isn't it a "Pandora's Box" to start geeking out on software development methodologies? Is this really the best use of our time? We need requirements, no matter what.
Please, can we just come up with a requirements document?
For the first stage gate was just something that I intended Stephen to look into not share with the rest of the world before I could form a proposal and it being shot down without me even having one but given that I had expect our discussion might lead to more confusion I started writing this late last ( and into the ) night [1] which hopefully clears things up a bit.
Again requirement documents are *part of* the stage gate process.
Apologies, Jóhann. I probably did jump the gun on bringing that up. I did note in my summary though that I was talking about that being the process for promoting roles to "featured" status and NOT the process for the Server Platform itself. I think perhaps that got overlooked.
JBG
https://fedoraproject.org/wiki/User:Johannbg/FOSSP#Stage_Gate.27s_Process
Thanks
for putting that together. I'll take a look today.
On 11/19/2013 02:50 AM, "Jóhann B. Guðmundsson" wrote:
I think this is getting misunderstood.
I've never said it should not be used.
I have disliked where it's and how it's trying to be used as in applied to the "Server Community" as opposed to the "Server Role" and Server
Okay. I think this is a straw man then and we're in agreement, because I don't see where anybody is saying or has said the 'Server Community' itself needs a PRD. A community != product, so having a PRD for it makes no sense.
For the first stage gate was just something that I intended Stephen to look into not share with the rest of the world before I could form a proposal and it being shot down without me even having one but given that I had expect our discussion might lead to more confusion I started writing this late last ( and into the ) night [1] which hopefully clears things up a bit.
Okay, that definitely does. Thank you for putting that together.
~m
server@lists.fedoraproject.org