Brand brainstorm

From FedoraProject

Revision as of 05:15, 21 March 2010 by Mchua (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

These were results of an actionability brainstorm on branding from the 2nd day of Marketing FAD 2010.

Note.png
This list needs to be gone through and put into a ticket queue.
If you would like to help, please let the marketing list know! We'll show you how.

Contents

What can we measure?

Awareness

Do people know that your brand exists? What sorts of terms do they associate with your brand? How strong are those associations?

What are you associated with? Do you come to mind? What do you want to be - a community brand, a technology brand, an open source brand, a Unix brand?

Measurable by asking people "think of technology brands" (or whatever you want associations with) and seeing how many of them mention Fedora.

Associations

  • open source
  • technology
  • community
  • linux
  • innovation
  • people friendly
  • free software
  • freedom
  • user friendly
  • learning environments
  • friendly community
  • Red Hat
  • beta
  • leading edge
  • rapidly changing
  • bleeding edge
  • secure
  • developer ease of use
  • open minded
  • leading
  • focused on developers
  • well documented
  • widely distributed
  • peer review

How strong?

  • ask fedora contributors (or users) what 5 adjectives they'd use to describe the community, and (separately) the product
  • ask people who haven't committed to any particular distro what words and qualities they associate with each - what is the difference between Fedora, SUSE, Ubuntu, Gentoo, etc?
  • ask students at college to view some material and ask their perception <- this requiere a filter for those that already have some idea about fedora.
  • ask people at conferences, etc. via ambassadors.
  • ask people to complain
  • Social media. ---> twitter, identica, review sites, etc. Do complaints go up and down at certain peak periods? <--- hard to monitor...
  • Mine the heck out of http://fedoraproject.org/awstats/ and make recommendations based on the results.
  • Deploy a dead-simple web app that simulates the "association" task above and asks people to just enter words that they associate with Fedora.
  • comments in web articles regarding fedora and linux at large
  • look at co-tweeted/dented hashtags - #fedora goes with #what?
  • do word frequency analysis on articles written by other people on Fedora.
  • ask non linux/fedora users to test and give us feedback
  • film new users/contributors using Design Team's usability camera setup
  • do interviews with users / contributors (podcasts)
  • put a whiteboard in the elevator and put the question on it "what do you think of fedora - write ONE word here" <--- "what is the one word that comes to mind" - want to avoid people saying, "i think fedora is great!"
  • survey handouts at install fests
  • ask users/contributors to other distros/projects what they think about Fedora, why they made the other choice (why not #foo)
  • look at referral sites from http://fedoraproject.org/awstats/

Consideration

DEFINITION: Would someone consider using your brand? just ask to try

  • which problem do you have with your Operating system right now and how can fedora maybe solve them?
  • Survey: "would you consider using Fedora in the following situations?" "Why? Why not?"
  • "what options would consider for doing X?" where X is a task that you might accomplish with Fedora, or with something else - see if they mention Fedora in there.
  • If you need some apps to help with college, buy or learn to use Fedora?
  • Have you ever considered using fedora?/have you considered using fedora in the last month? <-measure change
  • People to ask: Friends. Family. Developer groups (e.g. LUG). Students. People interested in Freedom. Co-Workers. Non-platform-specific open source audiences.
  • Make limesurvey work.
  • Reach out to distro-agnostic upstreams and downstreams and ask for people willing to give us 20 minutes of their time. Especially to upstreams not already heavily involved in Fedora. (are we asking questions or using questions to lead into a sales pitch on why they should come here??)
  • Set up a "survey station" at Events booths - have people fill out survey while waiting to ask questions or get help
    • Ask them to fill it out at the same time as giving them swag to encourage participation
  • In #fedora, when people ask for help, ask them for a moment of their time to answer survey questions, and then talk about their answers with them. And determine how we increase bandwidth and number of ops to handle it, or find an alternate channel with some note in /topic?
  • add survey link in channel topic
  • freemedia requests

Trial

Definition: People who actually take the first step and try it out (compare to "Consideration," which is people who are willing to try Fedora but have not yet done so).

  • Capture the people who come into #fedora saying "I'm trying different distros".
  • Contact everyone who *started* but failed to *finish* the CLA process and try to help them or ask them what happened.
  • we have statistics about user downloading updates
  • link smolt (obvious)
  • Fear of the CLI - "what do you think about the CLI?"
  • Fear of not having everything OOB (Music, etc. ) - "what do you think you're going to miss?"
  • Sick of Redmond/Cupertino
  • Find blogs of people who tried Fedora and gave up(And ask them why and also "If we help you fix the issues would you try/use Fedora again") We shoul also look for people bloging its first try.
  • Legal blockers. Not sure if I'm allowed to use, not sure if i'm allowed to participate. Do people complain / have issues here?
  • Installation statistics
  • make a FAS-data/FAS-authed-services data parsing library/API, see http://blog.melchua.com/2010/03/01/fas-scraper/
  • Use start.fedoraproject.org to link to a survey that first-time installers would use, since the Web browser is the first thing many users fire up
  • look at number of FAS accounts who never actually do anything, graph over time
  • Look into your local user groups, Developer groups, etc.
  • look over the shoulder of your co-workers
  • Casual questions with students and employees
  • freemedia sent

Loyalty

Definition: Repeat visits/contribution/usage.

  • look at repeat FUDCon attendance data
  • Look at smolt data to see how many systems check in more than once
  • look at FAS users and see how many reappear from release to release in the various FAS-authenticated channels (are they blogging, on lists, committing code, pushing packages, etc) - again, see http://blog.melchua.com/2010/03/01/fas-scraper/
  • can we obtain data from installations of preupgrade - stats on
    • stickster: I believe that we have a special client string for preupgrade that tells us when people use it (rather than yum) to pull packages. We can't tell whether any specific run is successful. The fact that they ran pre-upgrade in the first place suggests some loyalty to begin with - eg they wanted to continue using Fedora. Whether they are successful or not is a technical problem 'we' (being marketing) can't solve.
  • freemedia stats over time
  • who comes to a fudcon?
  • who comes to FAD
  • who comes to IRC
  • who reads our mailinglists
  • who ever filed a bug
  • who reads our FWN (will be better with that in Fedora Insight - FI)
  • Talk to people at Events (repeat attendees )
  • How can we get all of this data without seeming like big brother? Do we seem "evil" if we have this data and share it with the world? *raises pinky to mouth*
  • Talk to people who have failed at doing something with Fedora, and if fixed would they come back
  • Talk to people who have *stopped* contributing (see gaps in FAS, perhaps?) and have either restarted, or not restarted... and why - and when they look at Fedora again, what does it look like? (Find friends you used to see around, but no longer see around, and try to touch base with them again)
  • Stats from freemedia, persons that have asked more than one version
  • emphasize personal connections with other contributers
    • Follow up with people who have stopped or decreased their contributions(Written above )
  • look at our trac/bugzilla queues - are there names that are frequent but who we don't recognize?
  • look at the forums - are there names that are frequent but who we don't recognize?
  • ask ambassadors who the active Fedora people are in their local LUG - people who may not be online
    • Material to help ambassadors deliver a talk to their local LUG

Looking at our User base...

How can we get insights on this audience?

  • Work with a group that is installing a product. Work with a group that is working with pre-installed machines. Is it easy? Is it hard?

a couple of suggestions:

- filing bugs? / get email, feedback query? Microsoft, Oracle, etc, all of them do this regarding their products/services.
- several methods to get insights, can be done internally with our own staff (several processes available, surveys, questionnairs, pools, closed group debates, etc)
- corporate model (this might get out of hand due to "community stuff", still, a more personalized approach)
- documentate the audience, and start working on a "user profile" stereotype, see it as "consumer profile" study.
- Internally, data mining and data processing tools (this more of a method, but should thought before hand and analysed due to resources).
- Friendly approach (every bad word about fedora removes us 5 potential "customers", train the community to answer friendly requests and provide personal assistance, through "localized" we should be able to cut some gaps here).
- Technology forecast. We are not creating new trends (maybe we are), but the tools and services we offer (even community based) should be consistent with the degree of knowledge from the audience. Dont push people into using IRC or whatever if they have strong background and pro's for foruns). We have to hit them through means they already are aware off and that have some degree of knowledge.
 - Market study should provide all of this. I can provide proprietary market studies if needed related to open source.
 - Lurk other projects and get to know their flaws. I have some user emails (new developers) from the gnome lists that clearly demonstrate their obstacles when starting programming (in this case GTK). Getting to know this before hand, we can learn a lot on how to make a faster and more efficient integration. Keyword: "INTEGRATION", the smoother and faster, the better the results.
 - Strategial thing: Watch Novel's OpenSuSE and keep an eye on it, lots of Fedora packages being generated by their builder. I've runned into this several times in the past. Why people use their stuff to generate packages for our distro? Why do they use their tool? Documentate.
 - 
  • IRC -- pick a 24-hour log and then use it to make a list of each individual problem (and/or keywords - statistics about wich "word" comes up most) that happens (We have weekly logs for these - http://fedora.theglaserfamily.org/ircstats/fedora-weekly.html and yearly too - http://fedora.theglaserfamily.org/ircstats/fedora.html )[Can be made to 24 or 48 hours if you like with some help from scott]
  • IRC -- lurk on #otherdistro and see if anyone says they came from fedora over a 24/hr period. Wonder if $otherdistro logs their irc chans anywhere. Most do. irclogs.ubuntu.com for example (off the top of my head)[Use a logger bot and then grep the logs]
  • Look around at forums like Linuxquestions.org
  • parse the #fedora logs from the #fedora-ops channel (http://fedora.theglaserfamily.org/ircstats/fedora-weekly.html)
  • look at the weekly psig stats from #fedora-ops
  • the [In the News] Tag -> see Kara Schiltz
  • Mailing lists (users@list.fp.o, etc) -- make a list of all the subjects that occur in a specific period of time... or just use the list archive ;-)
  • look at lists/forums for non-Fedora communities-of-interest and search for keywords relating to tasks that one can complete with Fedora (for instance, "deviantart users are talking about photo filtering... they could use GIMP")
  • make a list of such keywords
    • Monitor mailing list subjects content, etc - make EKG work
    • Download all the mailing list archives and grep for "sucks" "broken" "awesome", etc.
    • topic list of what folks are saying on IRC, in support and social
    • ask for participation in surveys
  • Engage marketing schools to have Fedora market research as a class project.
  • make fedora-tour
  • offer migration list for voluntary swtiching and sending followup qustions
  • try to get a subset of voluntary switchers, to tell us wahy they are trying Frees software and why fedora
  • use the user filming setup that Design has to watch people trying out Fedora for the first time, then call for usability designers to help us analyze those films and find appropriate heuristics
  • Talk to LUGs
  • Ask freemedia requesters to take a survey (small price for free media)
  • can we get any data form br-office spin ?
  • talk with users of spins that wouldn't normally be interested in "general Fedora" (niche users) and ask them why they use that spin (FEL is a good example) and "what's Fedora to you?"

What are our existing points of parity and difference, and how do we prove them?

Parity

  • you can browse the web, use office software, edit photos, "general productivity"
  • you can develop other software in Fedora - libraries are in there, IDEs work, etc.
  • you can view the source
  • Gnome versions
  • hackable and customizable - you can patch/look-at any part of your system you want (iow, it's FOSS)
  • you can be a part of it - social aspects, etc.
  • you can have fun - hear music, look at Internet-TV and videos (are we allowed to say that?) play games, etc. chat with friends.
  • people can get the same os's at a conference. Free cd in the hand - can generally obtain one of each at an event, if you want it.
  • you generally have someone near you who's part of the Fedora community. (whether you can find or reach them is a different question.)
  • Most software available in default repos
  • KDE versions
    • Fedora was the first to release KDE 4.x, now more parity

Difference

  • Fedora exceeds proprietary operating systems in terms of ability to instantly ship software on demand, even when the user doesn't know what to demand. (PackageKit autoinstallation of fonts, codecs, langpacks, etc.)
    • PROOF POINT: In another operating system, load a document with a non-installed font and observe what happens.
  • SELinux (strategical positioning, SE Linux in my opinion should be one of our biggest relevant points, it's being threatened currently by Novel's AppArmor, this can be a nice point to demonstrate technical superiority towards Novel and do some damage)
    • Needs better communication of benefits
    • Usability improvements continue to be improved
    • PROOF POINT: do a usability study comparing an old release to a recent release (although... how many of those improvements are, say, GNOME improvements rather than "Fedora"? Maybe we should point out improvements in upstreams, and then which ones were made by members of the Fedora community, or adopted by Fedora first?)
  • doesn't push non-free software at you (this is why I changed to fedora)
    • PROOF POINT: comparison count of non-free software provided by other distros (in their repos) to non-free software (0) provided by Fedora.
  • packagekit <-- what about package kit are we proving? That we have it?
    • PROOF POINT: (if it's the use of the kit, it's self evident, but still worth calling out)
  • Creation of user tools adopted by other distros
    • PROOF POINT smolt, revisor, Live USB Creator
  • Creation of procedures and practices - we're "open" in everything we do (governance, marketing, etc), not just our source code.
    • PROOF POINT: Marketing FAD 2010.
    • PROOF POINT: SOPs for Infrastructure, Marketing teams
    • PROOF POINT: things like Publican templates being reused for documentation for other projects, other projects citing Fedora's practices (The Sugar Labs oversight board explicitly cited/borrowed-from Fedora's trademark policy, as well as its license approval policy, etc.)
  • choice between hundreds of different programs to possibly do the same thing which can be installed using a single command
    • PROOF POINT: NO PROOF --> This is really a parity point.
  • Community-run marketing and materiel production!
    • PROOF POINT: Marketing FAD 2010 and every piece of Fedora-specific artwork and design since Fedora 8
  • our Infrastructure team KICKS ASS in the community way
    • Need a way of saying this that's marketing friendly?
    • PROOF POINT: MirrorManager, all bandwidth used to distribute is donated from the community, Koji, number of infrastructure employees actually on salary vs how many people have root on what servers vs magnitude of services provided - these stats will likely make *any* IT person's jaw drop, but I'm not sure what the stats are
  • better communication upstream helps get things done/solve faster
    • PROOF POINT: Selinux, pulseAudio, kernel, KDE, gnome, call these out in feature profiles/interviews, point out where upstreams are highlighting Fedora and Fedora contributors in *their* fora and material
  • no matter what distro people are using they are using some fedora code
    • PROOF POINT: kernel, NetworkManager, X.org, freedesktop.org, a bunch of GNOME
  • Uses 100% free software to create free software -- we drink our own champagne (or in some cases eat our own dog food ;) <-- messaging fail ;-)
  • by the people, for the people, of the people
    • PROOF POINT: Who makes Fedora? Who uses Fedora?
    • PROOF POINT: Mo's uses this interview, embedding links in as many deliverables we make as possible as to how those deliverables were created and with what tools, pointing out wherever we *can* find non-FOSS tools being used to make Fedora stuff so it is clear that the list is small and rapidly being eliminated (though I actually can't think of things that would go on this list right now...)
  • Art and Design team - also a good example of nontechnical contributions also dogfooding FOSS
  • If you don't like something and have the will and reasons, you can change it
    • RESTATEMENT: We make it possible for our entire project to be forked if we do not satisfy our community at large.
    • PROOF POINT: All the code that runs or services our project is 100% FOSS, and all publicly available. All data other than that we can't expose by law is available as well.
  • This can be useful for my SWOT ;)
  • Suggested approach on the following topics: Place (distribution), Product (Features) > from this we can work out Promotion.

^^ These last two lines aren't really differences we can prove. Should they be removed?

  • Openly airing our disagreements - we expose our ugliness.
    • PROOF POINT: I'm sure we can find a bunch of mailing list threads that prove that we expose our disagreements. ;) But also follow those up and link them to analysis - "it's good we're having this discussion, because..." or "in my company this would happen instead and having this out in the open is SO MUCH BETTER," etc.

How do we find out more about the previous question?

Brainstorming? skipped this

Foundations

For each foundation, 3 questions:

  • Why is this important to us? (the people gathered here - we'll do a second round with Fedora contributors at large, and a third with the User base people who are not yet contributors)
  • What do we already do in Fedora (revealed preferences) that demonstrate-through-action that these things are important?
  • What groups care about this foundation, and how do we reach them?

Freedom

Why is this important

  • Freedom == Choice == Power
  • In my perspective, this can be very exploited in communication. And some general lines can be established for any kind of audience. I find this foundation by definition too much restrictive.
  • Perception from users in several locations through this world might be misleading. Our definition is objective, but freedom is always relative from person to person. One wrong move and this can turn easilly against us (example: the chinese). That's an interesting point. It does start to get close to being a "political" message in some ways. I felt that way talking about Fedora Governance at FOSDEM -- I felt at points like I was giving a small-government political speech.Yeap. If this dependend on me, I would relate to freedom as with "freedom of action" towards the possibilites Fedora offers for "ONE" to develop his work, and would go away from the "political" scene. Either way, Richard Stallman is all politics ;) How can you even say that with a straight face? ESR is all politics

^^^ I never saw this as a pollitical message its always been Freedom as in the 4 Freedoms that RMS created. --> But below, people are writing about how Fedora is set up to allow you to individually take action, etc. Freedom doesn't come from RMS. :)I know I just couldn't think of a better word (sleep dep) Maybe part of the "freedom" messaging issue is the same thing that the libertarians have; the most prominent carriers of the message are usually gun-toting crazies, not necessarily representing the core normal person who is a libertarian, and that sends a poor message to prospective users and/or contributors. "Be like us! you can look like RMS if you use fedora!" is not the message we want :) I'm not pointing that! I'm poiting that "freedom" in the western civilization has one meaning, middle east, asia, africa and so on has another. So advertising freedom here and there as different values involved. Right, it's something we almost take for granted here, and that people are still fighting for in other places. :) The "great firewall" of China ;)

  • Freedom gives us the latitude to drop the ball, because anyone else is free to pick up the pieces, learn, and carry on. BUS/RAPTOR TEST!
  • The building block upon which all our other actions are founded.
  • Freedom is our conscience.
  • Freedom lets us re-spin re-mix -- we impose no burdens on the recipient that are not part of the free software philosophy, other than the step of removing our trademarks (for which we already provide substitutes)
  • Freedom is something we feel obligated to spread to others, because Freedom is -that important.-
  • Freedom lets me speak openly.
  • Allows community to get involved and control the destiny of the project or a subset thereof.
  • Lets me fork. If I have a crazy idea I want to try, nobody's going to stop me - I can do it anyway, and if it works then I can go back and show people, nad I can do all this trying-out out in the open.
  • freedom lets us get in touch with other freedom arenas like CC
    • Proof point: Documents released under CC

What do we already do in Fedora

  • Don't permit any non-free software or patent-encumbered software.
  • We create a licensing policy which is itself free so others can refer to and use it, http://fedoraproject.org/wiki/Licensing
  • Encourage border cases to make themselves free (ie. Zimbra)
  • Make it easy for others to remix and rebrand the distro we create - and encourage them to do so.
  • Encourage translations
  • Encourage people to get involved as contributors, not just users
  • encourage participation and discussion from everyone without descrimination
    • Value contributions on merit
  • Attempt to educate users on potential inhibitions on their freedom and the free alternatives
  • Empower users in third world countries to
  • Encourage the use of free formats/codecs (ie linking to sites with freely licensed music, streams using non-proprietary codecs
  • Make our best-practices usable by other open source groups - making SOPs, etc
  • Everything in Fedora is done on free open source tools that if someone disagrees they can take our code and create their own distro. (koji)
  • This is a sensible point. I'm staying out of this because my concept is way over that and will for sure find enemies.

What groups care about this foundation and how do we reach them?

  • People who have a problem *right now* who are looking for a solution.
  • Many open source ninja developer types.
  • Disaffected Debian community members.
  • Windows Users / Switchers
  • People tired of being forced to pay money in order to solve simple software problems.
  • Disaffected Apple Users who want Open Apps
  • OSI, FSF, EFF, CC
    • FSF, ironically, doesn't consider Fedora to be free. I think this is something we should work on. <-- FSF's and Fedora's definitions differ, but we are continuing to have discussions with them privately.
      • because of one package (linux-firmware)
  • open formats that let you be owner of your own data
    • Proof Point: you can
  • you are in control of your computer
  • Game Programmers who wants to get a large community of players
    • A good place to kick in with the FOSS Drivers and it's path breaking grounds on this field.
  • Open inventions network
  • Designers / Artists -> design spin
  • hackers (the real ones - blackhat vs. whitehat) -> security spin
  • people who don't have a lot of avenues to explore otherwise - eg angst driven teens with overprotective parents and lots of bandwidth, or others who don't have control over other parts of their lives. Inmates. Canadians stuck indoors all winter

the Antihackers (real security progfessionals) --> White Hat types

  • other projects like OLPC
  • Other distros that like to use the stuff we create
  • groups with few resources - NOT because of the gratis part, but because they're usually ideological and like the libre part

Friends

Why is this important

  • Community serves as a force multiplier.
  • Because all of us is smarter (or dumber) than one of us.
  • Because Microsoft has 93,000 employees.Fedora has XXX contributors...
  • we are all friends, arent we ?! - together we are strong. Union, good point.
  • I have someone to relate to.
  • I have someone to ask for help (technical, or otherwise!)
  • I learn things from my friends. They show me things I never would have thought of.
  • Friends == users, friends == contributors
  • Friends = channel of distribution (if not one of the most important). Where we win or loose a battle.
  • Friends = Resource, we all learned something. We all knew Fedora by friends (assumption, I did).
  • Reinforcement of "Community" essence, translates into bonds (assuming self respect is present).
  • Without a community, Fedora is simply a code base maintained by Red Hat engineers, and all input starts to gate on things that are not open-source in nature.
  • community is FUN (most of the time)
  • Being friends means we can disagree and still have Fun at the end of the day.
  • Friends understand that Community does NOT always = FUN.
  • friends hold us accountable for deliverables - we don't want to disappoint our friends.
  • help us grow as we have feed back from differente perspectives
  • Sometimes you wanna go where everybody knows your name...
  • be introduced to new cultures and personalities
  • you can trust friends
  • helps us remember to be nice to noobs
  • A friend recognizes when you have a problem, and gives you a helping hand even if she thinks you don't need it.

What do we already do in Fedora

  • A community-based accountability model. "I didn't do $FOO because my boss told me to. I did $FOO because I promised my friends that I would, and I don't want to let them down."-- Peer Review
  • We communicate in different ways as friends. In person; Fedora Talk. IRC. Email. Forums. Planet. We work to ensure that these lines of communications WORK, are available, etc.
  • we log obsessively, and "if it wasn't written down it didn't happen"
  • encourage friendships with in-person meetings (FAD/FUDcon) and constant conversation on IRC
  • see: Fedora Planet
  • do a word count on mailing lists and IRC channels for the word "community" - and how many times it's used as a "why we should do this thing that is hard" reason. I'm guessing it'll be a lot.
  • diverse backgrounds and experiences
  • FUN we got a lot of fun, at least time to time
  • Provide friendly community support. in a place where it is accessible, easily, and people can FIND where to go, not wonder, "should I be asking questions here, or there?" Nearly free of background "noise". Good point.We don't want people looking for support and finding support in forums for other distros. That winds up being a potential conversion point.Unfortunatly this is not verified if you use Google to find information. Usually people are sent to U$%&$ somehow :( It's strange how we have such great documentation and Google usually digs it in the most weirding places. And Fedora's documentation indexing is great to my knowledge.
  • Support each other in times of need. Even if its just a kind word.

What groups care about this foundation and how do we reach them?

  • By definition - everyone - ref Love/Belonging in Maslows Hierarchy of Needs
  • Free culture hippies.
  • LUG's(relevant point) - I would give a lot of importance to this point. Today's and tomorrow's professionals.
  • phorum-websites (e.g. http://www.fedoraforum.de/ )
  • researchers of communities, actually (sociologists and anthropologists find us interesting)
  • people who want to build a network or a reputation
  • people who want to be able to tap into a group to find answers quickly - but not everyone will care about this (i.e. Mel has MechE friends who ask her about Linux because it's not worth it for them to take 3 months out of their lives to learn how and where to ask those questions themselves, because they have those questions so rarely.)
  • "new media" people find us interesting in the same kind of way reseachers do
  • people looking to pass the bus/raptor test with their own projects
  • people who are looking to learn stuff from other people, to be exposed to different things they didn't know about before
  • People who are interested in collaborating
  • People who have jobs that are mostly solitary activities.
  • People who participate in other social sites. Think reddit, etc. They wind up having positive experiences in other communities that are forum-based, usually with paired-up IRC channels for being "friends," and they want to duplicate that social experience in other things they do / participate in (hobbies, whatever).
  • provides sense of belonging
  • people who didn't know they'd need support in other areas
  • Tool to personal inter-relations. Several packages available.
  • Trolls

Features

Why is this important

  • It is our biggest differentiator.
  • Fedora's primary sponsor (RHT) cares the most about this, and devotes a ton of resources to Fedora for this purpose.
  • People LOVE shiny --> make it new, make it work, and iterate quickly
  • this foundation helps not just fedora but the entire community, most distros get n% of thier packages from us
  • If "freedom" is our conscience, then "features" is the foundation that our engineering work is most grounded in. This is what we DO with our freedom.
  • Bleeding Edge. Differentiation factor.
  • Technical know-how aquired and shared with the community.
  • new software, new drivers -> new contributors, new fans

Cutting edge (not as painfull as bleeding)Thats usually advertised that way on #fedora. That's where I digged that concept, but true, too agressive.

What do we already do in Fedora

  • Relatively egalitarian feature process. "If you can make it work on a schedule, you can get it in. If not, the next release is only 6 months away."
    • How true is this? ^^^^^^VERY - see openchange - completely outside group - missed schedule first time - so not included. Next release it was included.
  • see: features list for each release (and talking points, etc.) for shiny
  • freely let people add features according to our features policy if you want it listed for the actual release - or "on your own" if you don't mind not having that. depends on the feature and its effects on the distro
  • one release every 0,5 years and no legacy. It's dead after 13 months.
  • provide several flavours, production, updates, rawhide/testing, etc.
  • we provide a HOME to develop features -- fedorahosted -- The home has all the infrastructure people need, the friends / community to help you become accustomed - it's not just a big black hole of "oh hai we have source control." We also have the shepherds, so to speak, to help newcomers get features in, and not just have them be subjected to trial by fire. That is a feature in itself.
  • Latest software/kernel/etc attracts folks who may be looking for $foo version $bar
  • Compliance with community standards. (focus rpm for instance)
  • Ease of use
  • have a mechanism for determining quality (well, we're also making that mechanism for determining quality...) of features included From the marketing scope: Quality = Expectation - Satizfaction. > Quality, < No Quality. Thats how people usually perceive quality.

What groups care about this foundation and how do we reach them?

  • Full-time upstream developers.
  • anyone who wants to use their computer to be productive in any way
  • Working envirorment (relatevely) secure and virus free(carefull, there aren't virus but there's other things like root kits) actually there are some viri but they require you to be rootSuggestion: use the normal "out of the box" security diplomacy. This way we always win.
  • developers who want to see what interesting tech is coming up and therefore what they should learn/work-on
  • Developers who want to expose their "feature" or project to a large audience.
  • Developers who want to use a {tool,library} and know that what they are using is very close to what upstream ships (and thus their work is not relying on hacks, i.e. more applicable)

First

Why is this important

  • Innovation / Creation
  • Leader / Leading
  • first comes, first served
  • We are Innovators.
  • When we do things first we ensure they are Free. Sometimes, when others do things first, they do it in an unfree way.
  • we break and fix more software in 6 months than most projects do in a year, meaning we have already broken and fixed something new before anyone else has even put it on a feature list (pulseaudio)
  • To increase the pace of innovation (ie - find bugs faster, deploy new features faster, increase freedom)
  • it set us apart from the rest
  • First to add new features
  • Upstream to many projects
  • Our developers innovate in a lot of downstreams
  • First to adopt new methods of working in the Open Source way
  • Fast moving


What do we already do in Fedora

  • 6 month release cycle
  • Premiere new technologies early (not the same as changing them after debut) ;-)
  • Encourage packagers to bring in new software they're willing to maintain, and bring in the latest version possible from upstream
  • Encourage people to innovate within Fedora - ie, the egalitarian features process
  • heavy collaboration with upstreams - Fedora people actually go there and do development, submit patches, etc. to upstreams to improve *their* projects rather than just carrying patches in Fedora
  • Lead what linux would be in the future
  • other distros picking up on our tools/processes/features/stuff that Fedora contributors put into upstreams (i.e. network manager)
  • Specific examples
    • SELinux
    • libvirt
    • PolicyKit
    • PackageKit and automatic instalation of print drivers, codecs, etc
    • FOSS Drivers / Nouveau
    • KDE 4.0 in Fedora 9
    • KDE 4.4
    • rpm 4.8
  • Newest Kernel every new release; Newest Software (even beta)
  • people like kernel devs, etc. tend to use Fedora (look at our existing userbase, they tend to be the people who stereotypically want CUTTING EDGE NOW)
  • working pulseaudio
  • LOTs of prereleases (alpha, beta, multiple rc) to make the churn even faster
  • We strive to make solutions to problems - we want to be the first to solve things for you.


What groups care about this foundation and how do we reach them?

  • Tinkerers.
  • People who want to show their friends that what they're doing is BETTER, works better, functions better for their needs, etc. Not necessarily show-offs, but perhaps people who are interested in informing. People who are heavily ingrained in social media "stuff" are good examples here - they want to be First to the social media "market" with information about cutting-edge technologies. We want them to be talking about those technologies, and we want them to be discovering those new technologies via Fedora.
  • People who want to be different. Want to be cool want to be the coolest
  • People who want cutting-edge.
  • Developers who want the latest and greatest code available, platform available, etc.
  • Developers who *need* the latest and greatest code available - i.e. they're developing their code to be compatible with $something that is coming up, but don't have the time to stay on top of all the *other* projects' changes and dependencies and want their distro to do it for them.
  • Developers, Developers, Developers
  • Administrators (RH Admins ;-) ).
  • VM Lovers
  • sysadmins who want to preview what they're going to be expected to maintain N years from now, or what things might make their lives easier N years from now, or who want to *make* their lives easier N years from now by making something work in Fedora NOW.
  • Enthusiasts - autodidacts. experienced personel. Those who like challenges
  • Fanboys (feel free to crucify me).
  • People who like transparency.
  • People who like to participate and learn "new" concepts.
  • people who want to be "forced" to learn how their computer works - to be surrounded by the kind of environment that would tend to make them run across those kinds of opportunities to learn esp when something fails to work and you learn to fix it
  • people who want a learning sandbox (learning tool, learning excersice)
  • Early adopters
  • other distros - even ones we aren't upstream for.
  • Software Testers, bug finders, localization

Fail

This is something that is a part of all. Freedom to fail. Friends to help you when you fail. Features may fail. Sometimes being the first to fail is pretty darn nice.

without failure there is no learning


Recording

  • Use Engineering Services queue
    • File a ticket that includes sort-of pseudocode for what you want, a contributor can knock it out in a couple hours
    • Actionable ideas to code:
      • One line web form: "What one word do you think of when you hear 'Fedora'?" -- feeds a local SQLite database. Ambassadors use at a booth. (Low tech version: post-it notes on a board.)
      • Wishlist: one kind of Fedora Searchengine - e.g. fedoogle: search for "gobby" and get 234 items in wiki, 34556 items in faq, 3456 items in forum, 345456 items in mailinglists, 3764 items in FAS Accounts, etc.
        • could be sriptbased with such kind of webform around it.
        • (e.g. look at mels script)
      • Channels we are monitoring: IRC, forums, mailing lists
  • Look at awstats page, pick the top wiki page that's getting hits, and make sure it is as accurate and excellent as possible. And iterate weekly - delving farther down the lists.
  • Have a standard set of questions to ask of people who are:
    • Hearing about Fedora for the first time
    • Trying Fedora for the first time

Wishlist 2: some wiki tracking plugin: we have some wikipages we must update each cycle (e.g. Questions we get to students, Ambassadors, etc.) but maybe we get lost in work. So a small script we need ASAP wich we can implement wich tells us "Ok, here is a new release upcoming in XX weeks. So we need to cleanup the following wikipages / the following documents"...