Board ticket #177 so far (Fedora Plasma proposal and related general questions)
Matthew Miller
mattdm at fedoraproject.org
Tue Apr 1 19:37:09 UTC 2014
As an outcome of last week's FESCo meeting, I filed a ticket with the Fedora
Project Board to discuss the Fedora Plasma proposal and related general
questions. Because the board trac instance is private, important discussion
was happening out of view. We've agreed to put it _into_ view -- going
forward, comments on this ticket will be CC'd to this list. But we also
didn't want to lose the conversation so far, so I'm posting it here (in
entirety).
---------------------------------------------------------------------------
At the FESCo meeting today, we agreed on two items for the board based on
the Fedora Plasma proposal put forth by the KDE SIG (See board ticket
https://fedorahosted.org/fesco/ticket/1265 and advisory-board list post
https://lists.fedoraproject.org/pipermail/advisory-board/2014-March/012449.html).
First, we're suggesting that the board consider Fedora Plasma as a
possibility for Fedora 22. (FESCo will establish milestones based on that,
and a KDE spin — possibly under another name — will remain release-blocking
for F21 and have a place on the download page.) Are you okay with a Fedora
Plasma Product in F22?
Second, what non-technical criteria should constrain whether we even see fit
to bring more product proposals to the Board in the future?
---------------------------------------------------------------------------
comment:1 Changed 6 days ago by mattdm
FESCo IRC meeting log if you want to dig through it :)
http://meetbot.fedoraproject.org/fedora-meeting/2014-03-26/fesco.2014-03-26-18.00.log.html#l-233
---------------------------------------------------------------------------
comment:2 Changed 6 days ago by jwboyer
For the sake of sanity, I would suggest we as a Board evaluate the questions
in the reverse order they are listed here. If we're going to approve or
reject another Product for F22, it would be good to do so using the
non-technical criteria we establish, not just doing it one off and then
trying to figure those criteria out.
---------------------------------------------------------------------------
comment:3 Changed 26 hours ago by inode0
Cc kevin, notting, toshio, ausil, pjones, tmraz, mitr, sgallagh added;
fesco@… removed
---------------------------------------------------------------------------
comment:4 follow-ups: ↓ 10 ↓ 11 Changed 25 hours ago by jwboyer
Taking a quick crack as this. We need to figure out what exactly we want
from Fedora.next and a Products based approach, and also what we don't want.
Fedora.next is being created to redefine exactly what Fedora "is", and to
help spread it to areas where we have not had a lot of market adoption.
Things like cloud and server are clearly examples of the latter. The
Workstation product is an example of the former. So we're going for both
clarity/focus, and new initiatives. We also want to be able to scale our
processes and resources to meet these goals and future initiatives.
With that in mind, we can perhaps focus on some of the things we don't want.
I think we don't want to make things confusing for users. We don't want to
push into markets that are poorly suited to Fedora. We don't want to
completely eliminate or exclude any existing set of users, but we can't be
everything to everyone. We need to be careful of our ability to scale, which
will hopefully improve but may not be immediately ready to support numerous
items.
In my opinion, the basis for the non-technical criteria for product
promotion are fairly simple.
New Products should not overlap or conflict with an existing Product
Products that conflict/overlap can be evaluated as replacements for
existing products
Addresses a new, relevant, and broad usecase or userbase that Fedora is
not currently serving
The usecase should be something the Board sees as being a long term
investment
The Product should be coherent with all of Fedora's foundations
I'll elaborate on these a bit.
We don't want conflicting or overlapping Products as that introduces
confusion and wastes resources. From a marketing perspective, we should be
presenting clear choices for people in terms of what Product solves which
usecase/userbase. If we want to reassess a current Product and replace it
with a new proposal, that's fine but we shouldn't present conflicting ones.
New proposals should solve a need that Fedora isn't currently solving.
Examples (not necessarily good or feasible at this time) could be mobile OS,
supercomputing, media center, gaming consoles. If there is an existing
product that could meet these needs without completely reworking itself,
then we should look at adapting that instead of creating a new Product.
Products are long term investments. They require significantly more process
and resources than simple spins. They are intended to be the "face of
Fedora" and should neither be rushed or short-lived. We wouldn't want to
create a Product for every brief trend in computing that shows up. E.g.
Fedora Netbook might fall into the "use a spin, not a product" category.
I think the last bullet is fairly self-explanatory.
Anyway, those are my initial thoughts on the broader question. Would love
some further feedback.
---------------------------------------------------------------------------
comment:5 Changed 24 hours ago by mjg59
My initial reaction was that adopting further products would inevitably
weaken the strong brand that Fedora.next allows us to attach to our
deliverables, but I'm now starting to think that I'm thinking about this the
wrong way.
What do we mean when we say "Fedora"? There's the media we give out, and
there's the community that produces the contents of that media. These don't
have to be the same thing. It seems reasonable to wrap the current set of
three products under the "Fedora" banner, but to permit the community to
work within the project to create their own branding and associated
products. Rather than thinking of Plasma as a product, consider the creation
of a Fedora KDE brand that could produce a range of Plasma-derived products
(for instance, a Plasma desktop aimed at educational or scientific
environments and a Plasma Active image aimed at mobile devices) without
having to worry about whether they overlapped with the "stock" Fedora
products. This would avoid the situation where arguments about overlapping
use cases and cohesive marketing overshadow the real work that people are
willing to put into improving the whole Fedora project because doing so
gives them an opportunity to ship code that they're interested in.
In this fantastical future, it would be up to individual brands to decide
their own marketing focus and, as such, how to arrange their products in a
way that made sense for their brand. I'd imagine that Fedora itself would
probably keep pretty close to the existing three products model, and perhaps
oversight of those would remain with our current governance model.
---------------------------------------------------------------------------
comment:6 Changed 24 hours ago by jwboyer
I don't immediately see how your brand proposal and my thoughts on Product
promotion conflict at all. I also don't see how it is more than the existing
Spin mechanisms we have in place with perhaps a bit more marketing thought
put behind them.
Creating a brand around a Spin seems like it wouldn't conflict with the main
Products that Fedora is promoting on it's primary sites. However, if such a
spin/brand were to become wildly useful and better suited for Fedora
overall, we'd still want to have some process for evaluating which of those
should possibly be a main Product (perhaps by replacing an existing
Product).
If I've missed something, please elaborate?
---------------------------------------------------------------------------
comment:7 follow-up: ↓ 8 Changed 23 hours ago by mjg59
Right now spins are kind of pushed off as an afterthought. They're mostly
lumped together under a single tab on the download site. There's no link
between related spins - Fedora KDE appears under "Desktops" , Fedora
Scientific-KDE under "Spins".
It's not surprising that people feel like spins are an afterthought, and in
that case it's natural for people to feel that they need to be more than a
spin to seem like a real part of the product. Providing a more structured
way for communities to brand and promote their spins would remove much of
that stigma.
And, right now, that's the only reason we're being asked about new products.
Nobody has, as yet, suggested a new product that fits nicely into the Fedora
brand. I'm not convinced that they will. It may be that worrying about
criteria for product eligibility is trying to solve the wrong problem, and
we should instead ensure that the various Fedora communities are empowered
to promote themselves more meaningfully.
---------------------------------------------------------------------------
comment:8 in reply to: ↑ 7 Changed 22 hours ago by jwboyer
Replying to mjg59:
Right now spins are kind of pushed off as an afterthought. They're
mostly lumped together under a single tab on the download site. There's no
link between related spins - Fedora KDE appears under "Desktops" , Fedora
Scientific-KDE under "Spins".
Yes.
It's not surprising that people feel like spins are an afterthought, and
in that case it's natural for people to feel that they need to be more than
a spin to seem like a real part of the product. Providing a more structured
way for communities to brand and promote their spins would remove much of
that stigma.
Yes.
And, right now, that's the only reason we're being asked about new
products. Nobody has, as yet, suggested a new product that fits nicely into
the Fedora brand. I'm not convinced that they will. It may be that worrying
about criteria for product eligibility is trying to solve the wrong problem,
and we should instead ensure that the various Fedora communities are
empowered to promote themselves more meaningfully.
So, yes to the last part. As to the "solve the wrong problem" part, I'm not
convinced an answer of "you asked us to come up with Product criteria so we
instead did this other thing instead" is really the complete approach here.
I'm not saying branding isn't a good idea (in fact I believe that was one of
the things the Board said we needed to do after approving the Products in
the previous ticket), but it doesn't answer the question we've been asked.
More specifically, we've been asked to evaluate the Plasma Product proposal
for F22. If we're going to say yes or no to it, we probably need reasons
why. We probably need those reasons even if we say "hey, you should make
your own Brand instead" because they didn't ask for more Branding. They
asked to be made a Product.
Since we seem to be in strenuous agreement on the Branding aspects, could
you comment on the proposed criteria?
---------------------------------------------------------------------------
comment:9 Changed 21 hours ago by mjg59
I think your proposed criteria are sensible, and that the Plasma proposal
fails to meet them. The reason why I'm interested in this other discussion
is that I think it's more helpful to return with proposed alternative ways
forward rather than merely answering a question in a way that may discourage
future contributions.
---------------------------------------------------------------------------
comment:10 in reply to: ↑ 4 Changed 21 hours ago by inode0
Replying to jwboyer:
In my opinion, the basis for the non-technical criteria for product
promotion are fairly simple.
New Products should not overlap or conflict with an existing Product
Products that conflict/overlap can be evaluated as replacements
for existing products
Addresses a new, relevant, and broad usecase or userbase that Fedora
is not currently serving
The usecase should be something the Board sees as being a long term
investment
The Product should be coherent with all of Fedora's foundations
I think points 2-4 are fine, I don't agree with point 1 though. Every
product overlaps in some way with other products and this is just a rathole
to try to define how much and what sort of overlap is unacceptable. Anything
that egregiously fails on point 1 will almost surely also fail on point 2
anyway.
I can imagine being alright with a product having substantial overlap if
there is a clearly different use case or target audience in mind that isn't
being served by the existing product.
---------------------------------------------------------------------------
comment:11 in reply to: ↑ 4 Changed 14 hours ago by gholms
Replying to jwboyer:
In my opinion, the basis for the non-technical criteria for product
promotion are fairly simple.
New Products should not overlap or conflict with an existing Product
Products that conflict/overlap can be evaluated as replacements
for existing products
Addresses a new, relevant, and broad usecase or userbase that Fedora
is not currently serving
The usecase should be something the Board sees as being a long term
investment
The Product should be coherent with all of Fedora's foundations
Strike the first one from the list and I'll completely agree. As the saying
goes, "Every program attempts to expand until it can read mail." A certain
level of redundancy in the tools in play is both inevitable and a
distraction from the redundancy that does make a difference: that of use
cases. If this was instead a "Mobile" product, for instance, its
distinctiveness from the "Desktop" product would be quite clear even if both
of those products were capable of running regular desktops.
---------------------------------------------------------------------------
comment:12 follow-up: ↓ 13 Changed 5 hours ago by pjones
Can we phrase the overlap criterion in terms of goals and audience rather
than features and maybe even use cases? It seems like it's written with the
intent of implying "we shouldn't have two different cloud Products" and
maybe being read closer to "we shouldn't have two Products with
ZOMG-amazing-mail-client".
---------------------------------------------------------------------------
comment:13 in reply to: ↑ 12 ; follow-up: ↓ 14 Changed 5 hours ago by
jwboyer
Replying to pjones:
Can we phrase the overlap criterion in terms of goals and audience
rather than features and maybe even use cases? It seems like it's written
with the intent of implying "we shouldn't have two different cloud Products"
and maybe being read closer to "we shouldn't have two Products with
ZOMG-amazing-mail-client".
... maybe? Something more like:
New Products should not aim to solve the same goals or reach the same
users as existing Products
That rephrase does map more closely to what I was intending. Given everyone
seems to want to drop that criteria entirely, I wonder if that makes it more
palatable?
---------------------------------------------------------------------------
comment:14 in reply to: ↑ 13 ; follow-up: ↓ 15 Changed 4 hours ago by inode0
Replying to jwboyer:
... maybe? Something more like:
New Products should not aim to solve the same goals or reach the
same users as existing Products
That rephrase does map more closely to what I was intending. Given
everyone seems to want to drop that criteria entirely, I wonder if that
makes it more palatable?
That is actually what I thought it really meant but I don't see it as saying
anything that isn't still captured in the second point
Addresses a new, relevant, and broad usecase or userbase that Fedora is
not currently serving
What is the difference?
---------------------------------------------------------------------------
comment:15 in reply to: ↑ 14 Changed 4 hours ago by jwboyer
Replying to inode0:
Replying to jwboyer:
... maybe? Something more like:
New Products should not aim to solve the same goals or reach the
same users as existing Products
That rephrase does map more closely to what I was intending. Given
everyone seems to want to drop that criteria entirely, I wonder if that
makes it more palatable?
That is actually what I thought it really meant but I don't see it as
saying anything that isn't still captured in the second point
Addresses a new, relevant, and broad usecase or userbase that Fedora
is not currently serving
What is the difference?
Heh, good point. Apparently I REALLY wanted to get that point across or
something.
Peter, does the second bullet cover what you were thinking of as well?
---------------------------------------------------------------------------
comment:16 Changed 4 hours ago by pjones
I guess so, yeah.
--
Matthew Miller -- Fedora Project -- <mattdm at fedoraproject.org>
"Tepid change for the somewhat better!"
More information about the advisory-board
mailing list