I posted this on [https://lists.fedoraproject.org/pipermail/council-discuss/2014-December/013066.html council-discuss] a few weeks ago and there was a little discussion. Because this is really a continuation of something we've previously agreed to, I think this is straightforward. Since objectives are non-trivial, we should go for the "full +3 with no -1s" consensus, but also, since this has technical components which impact F22 planning -- which is starting now! -- I'd like to have the final decision as soon as possible.
= Objective: Fedora Flavors, Second Phase. =
Overview:
Expected Impact:
Timeframe:
Aspects:
Coordinate Working Groups' development of updated PRDs and changes and features for each release.
Work with FESCo and Fedora Program Manager to develop process whereby flavor-specific Changes are handled primarily at the WG level.
Work with Outreach (marketing, ambassadors, etc.) to identify and plan representation at new conferences specific to the various target audiences.
Plan, coordinate, and schedule release engineering and infrastructure changes in advance of the F22 and F23 alpha releases.
Lay groundwork for possible different release cycles and lifespans.
Tooling and infrastructure for spins and remixes to increase differentiation. # Note: perhaps this is big enough to be an independent objective of its own, along with better promotion for spins.
Work with Council and community to develop concrete process for expansion (or possible contraction) of Fedora flavors as identified needs change, working from the product definition previously approved by the Board
Metrics:
PRDs updated. Changes filed, changes accepted, changes completed.
Conference reports; user data from those conferences.
User and contributor surveys, and other user and contributor measures
Usability is increased as measured by user testing specific to each target group
(My only regret here is that we are not up to Phase Four, for full-on Fedora alliteration.)
Additionally, I would like to nominate Stephen Gallagher for the role of Objective Lead. Stephen brought the "three products" idea to Fedora.next, was instrumental in its realization in the F21 release, and, also, tells me that he will have significant time to dedicate to this over the next year. (Please vote on this nomination as a separate +1/-1 along with any +1 votes to the proposal overall.)
I am
+1 to objective +1 to Stephen Gallagher as objective lead
I'm generally for this, but I'm confused where the sudden switch from Products to Flavors came from. Could we stick with Products since that is what we started with and what the media just finally jumped on? Switching now seems odd.
See https://lists.fedoraproject.org/pipermail/marketing/2014-November/016818.html -- this is at the request of product management in Red Hat, as they were getting feedback that it was genuinely confusing their customer messaging. I wish that we'd gotten that feedback earlier, but here it is now. It's been expressed to me as quite important from that side. Of course we can discuss earlier, but it seems to me like we have bigger things to worry about than what we call things.
(Of course, with that said, I'm up for a round of finding a better word than "flavors" if we want to go for it. I feel like the product/spin distinction was relatively clear, and we've kind of lost that here.)
Did you only send that to the marketing list? I don't recall seeing it on devel (where we talk about products all the time), nor on any of the Product WG lists. Given that we've been saying Product everywhere for over a year, it seem like a pretty big oversight to not do a very wide announcement and then just act like people are going to magically have heard about this request and will naturally just use the new phrase.
I'm not a huge fan of flavor, but I don't have anything immediately better. I just think it's too nuanced (and weak) to really convey how these things are supposed to be decidedly different in many aspects, particularly when you consider they have their own governance structures and decision making committees. I'll think about it and see if I can come up with another suggestion. We should probably come up with a new acronym for PRDs as well.
It did just go to marketing, although I also talked to docs, translation, and ambassadors. RH is primarily concerned with user-facing messaging and press rather than with devel discussion. I apologize that I didn't also make sure everyone here was aware.
Maybe we can come up with something new which captures the intentionality we are aiming for and better inherently explains the difference from spins as alternative offerings, without confusing the sales and support message that "product" gets into.
Replying to [comment:5 mattdm]:
I think I heard about something the devops people called silos once ;).
Absolutely
I wouldn't use "silos". That has a bit of a negative connotation for IT. (People in silos don't talk each other.)
Replying to [comment:7 bruno]:
That was my entire point. The communication around the flavor wording was done in silo style. Anyway, so much for sarcasm. Time to be productive.
Just piping up to say that I will accept the appointment to Objective Lead if it is offered to me.
As for the Flavors rename, I think we probably should have a separate ticket and public conversation about it. I agree with Josh that the transition did not happen in a transparent manner (and I personally dislike "Flavors" for a lot of reasons, most importantly because it does not translate well outside of English).
Btw. "flavour" could be pretty hard to translate to other languages than English and it should be considered when we come with another term. Product is pretty much international term now with pretty clear meaning. Variant maybe but there could be clash with RH product marketing again.
The Changes process was designed from the beginning to be more SIG centric (before we had WGs) but in the end, implementation was more FESCo based. It shouldn't be that hard to extend it to WGs/flavours (and I already added a few fields to cover that). Ultimate plan is to create Changes web app that could make it even easier (a few folks volunteered but in the end it's not that easy app as it sounds, also infra guys were interested... not F22 thing but could be done within F22/F23 releases).
For spins vs products/flavours, we had an interesting debate a few days ago with websites. One idea was to have "Fedora Lab" for spins. But in that case, we would need clear path to propagate spins into real products as it would be strange to have "forever lab" projects and with current strict rules set by former Board... I think Fedora should be inclusive meritocratic project, if you invest your energy, you should have the same handling as others even it would lead to less targeted/aimed deliverables.
I'd suggest 'edition' as something that (AFAIK) isn't used in RH portfolio, and might translate better across geographies as it's occasionally used already in the software world. Fedora 21: Server Edition, and so on.
Replying to [comment:11 notting]:
+1 to editions. But I agree with Stephen - this needs own ticket and own discussion to be transparent (and get as much feedback as possible on translations, meaning etc.).
naming things -> https://fedorahosted.org/council/ticket/19
Okay, so, back from vacation. If we take aside the naming issue and agree that whatever we approve here we can rename to whatever comes out of ticket #19, do we have the votes? I see no objection, but only a formal +1 from me (and informal from Josh). We need at least one more +1, (for both the objective and the proposed lead). Thanks all!
I have a few changes to some of the aspects that might be worthwhile, but nothing that I would consider blocker. I'm +1 for the objective.
I'm also +1 for the proposed lead.
For my suggested changes on the aspects, I'd change:
to
We want to make sure all the groups in engineering are covered on this as they'll all be impacted in one way or another.
I would also suggest removing the tooling for spins from this entirely, as the Objective is about products (I mean flavors/editions/whatever). It can be split out if we think it's important, but it's orthogonal to this Objective. Another possibility is to just lump it into the "rel-eng and infra" bullet.
in response to the prior comment about removing mention of spins... While it is true that Objective is primarily about products...there's a perception (valid or not) that they're devalued/deprecated. Spins are in the least very related, and I like the explicit mention, to help raise their real and perceived value.
Replying to [comment:16 rdieter]:
+1 to objective +1 to Stephen Gallagher as objective lead in response to the prior comment about removing mention of spins... While it is true that Objective is primarily about products...there's a perception (valid or not) that they're devalued/deprecated. Spins are in the least very related, and I like the explicit mention, to help raise their real and perceived value.
If that's the general feeling, then it should be split into it's own Objective entirely. Perhaps an Objective around Fedora customization/consumability. I'm concerned with the amount of work already listed for this Objective the lead is either going to be overloaded or the spins item is going to be lowest priority and not actually get worked anyway.
that's an option, but I personally consider the overlap here high enough and scope narrow enough... to not warrant splitting it out.
That said, if the lead (stephen or whoever) agrees with your feelings that there's already too much work and/or that it won't get the priority it needs, then I would reconsider.
Replying to [comment:15 jwboyer]:
Work with Engineering representative to plan, coordinate, and schedule release engineering and infrastructure changes in advance of the F22 and F23 alpha releases.
nod
Replying to [comment:18 rdieter]:
Stephen, what's your take on the scope here and the amount that you're interested in this particular aspect?
I agree with Matthew's original comment about Spins: "perhaps this is big enough to be an independent objective of its own, along with better promotion for spins."
I think there's definitely value in putting together a long-term strategy for improvements to the Spins and Remix process (including such things as working on a koji-and-bodhi-in-a-box sort of solution that could be reused by CentOS, Scientific Linux and popular third-party repositories).
Given the other set of tasks on the Fedora.next Objective, I think it's reasonable to suggest that this effort should have its own shepherd, most likely someone more closely involved with the relevant projects. It's definitely something we want to do, but I'm pretty sure I wouldn't be the right person to direct it successfully.
There's one big place, where these two initiatives overlaps and it's process to promote spin/whatever we call it to product and compared to old what old Board agreed on, I think it should as easy as possible but with strict requirement on commitment (and vice versa - de-promoting product if these commitment requirements are not fulfilled). But even I'd like to see closer relations between products and spins as described above, I'm not against having it as two initiatives that coordinate effort.
Given sgallagh's feelings, I withdraw my objection to jwboyer's suggestion to remove mention of spins in this objective proposal.
Okay cool. I'll put this on the wiki as a final draft, with that section struck.
Sorry for the delay; I've put this on the wiki at https://fedoraproject.org/wiki/Objectives/Fedora_Editions,_Phase_2
Any objections to this as a final draft? If not, I'm going to close this as accepted in a week.
Replying to [comment:24 mattdm]:
Sorry for the delay; I've put this on the wiki at https://fedoraproject.org/wiki/Objectives/Fedora_Editions,_Phase_2 Any objections to this as a final draft? If not, I'm going to close this as accepted in a week.
On #fedora-websites, we touched spins again as there are plans for modern websites for spins, ARM and Docker images (and other non primary deliverables). And it would be nice to get an answer were too. With F21, we invested a lot of time to establish products/editions, we have very nice modern web (thank you design and websites), so I do hope spins should be part of F22 phase. As it was already mentioned, there's some overlap but also it can make phase 2 too big. So, should I create a new ticket for spins objective? Or my feeling is that there's some change in support how we want to treat spins recently, that it could be squeezed back to this objective?
Don't shoot me for re-opening it but seems like at least websites would like to work on it for F22. And it would be nice to have goals set. It's not only about spins but all other offerings we have and do not fit to getfedora.
On spins: I think that if there's interest in and a champion for doing something better for and with spins, with scope beyond the websites, it ''could'' make its own objective. But not everything needs to be a large-scope objective -- if the websites team wants to work on it, we can support that just as it is (without necessarily fitting it back into this). Make sense?
Login to comment on this ticket.