20:00:55 #startmeeting Server Working Group Weekly Meeting (2016-08-30) 20:00:55 Meeting started Tue Aug 30 20:00:55 2016 UTC. The chair is sgallagh. Information about MeetBot at http://wiki.debian.org/MeetBot. 20:00:55 Useful Commands: #action #agreed #halp #info #idea #link #topic. 20:00:55 The meeting name has been set to 'server_working_group_weekly_meeting_(2016-08-30)' 20:00:55 #meetingname ServerSIG 20:00:55 The meeting name has been set to 'serversig' 20:00:55 #chair nirik adamw mhayden jds2001 mjwolf sgallagh dperpeet smooge vvaldez 20:00:55 Current chairs: adamw dperpeet jds2001 mhayden mjwolf nirik sgallagh smooge vvaldez 20:00:55 #topic Roll Call 20:00:56 .hello sgallagh 20:00:57 sgallagh: sgallagh 'Stephen Gallagher' 20:01:04 morning. 20:01:07 * mhayden woots 20:01:14 .hello langdon 20:01:15 langdon: langdon 'Langdon White' 20:01:28 .hello vvaldez 20:01:28 vvaldez: vvaldez 'Vinny Valdez' 20:01:29 nirik: Where are you that it's still morning? 20:01:50 it's always morning on IRC. :) 20:02:02 lol 20:02:38 s/morning/Monday/ 20:04:00 mourning? something 20:04:47 .hello jstanley 20:04:48 jds2001: jstanley 'Jon Stanley' 20:05:08 * mhayden tips his hat to jds2001 20:05:11 morning/afternoon/evening, as the case may be :D 20:05:12 dperpeet let me know he's on PTO today 20:05:40 This time isn't ideal for him (being 10pm his TZ), but he can probably make it sometimes. 20:05:53 Otherwise, he'll stay involved in the channel and by email 20:07:07 I'm going to be somewhat disappointed if we don't have quorum today. 20:07:59 sgallagh: how many acks did we have for this time from the chairs? 20:08:16 Eight out of nine chairs 20:08:17 sgallagh: .hello sorry to join late 20:09:05 Oh, I miscounted. We have six, which is enough to go forward. 20:09:52 #topic Agenda 20:09:52 #info Agenda Item: Server SIG PRD 20:10:23 The primary item on the list is kicking off the PRD refresh process. 20:10:35 Do we have other items of interest that we need to cover today? 20:11:00 we could talk about the cloud image, but thats not urgent by any means 20:11:26 #info Agenda Item: Server Edition cloud image 20:11:38 nirik: Sure, if we have time in the hour, we can start discussing that. 20:11:43 Any other topics? 20:12:21 #topic Server SIG PRD 20:12:46 my apologies. I had this down for friday 20:13:12 smooge: Welcome! 20:13:41 Last week, I sent out an email to kick off a new round of brainstorming for the PRD. 20:13:50 #link https://lists.fedoraproject.org/archives/list/server@lists.fedoraproject.org/thread/R7X5SSMR3MRELNR7AZ4M5ZR4YNTSN3N4/ 20:14:22 when i read about that model, i want to eat cereal 20:14:35 mhayden: It does indeed come from that company, yes. 20:14:42 well i cant find much of anytihing on it. 20:14:57 willing to try anything though :D 20:15:15 * mizmo here (sory im late) 20:15:18 So, I suppose I didn't give enough original information on it, but we're basically going to use it only for the structure and constraint it provides to the planning process. 20:15:24 Hi mizmo :) 20:16:22 sgallagh: similar to https://fedoraproject.org/wiki/Objectives/Fedora_Modularization,_Prototype_Phase ? 20:16:22 smooge and I discussed the use of the model today in #fedora-server and I sent the log of it to the mailing list: 20:16:24 #link https://lists.fedoraproject.org/archives/list/server@lists.fedoraproject.org/thread/S7U4CG76ALAJGZ4D2BUVAN42SK53INHF/ 20:16:50 mhayden: exactly like that; they used the Kellogg Logic Model to produce that infographic there 20:17:09 mhayden yes 20:17:26 * langdon typing very slowly today :( 20:17:30 .hello adamwill 20:17:31 adamw: adamwill 'Adam Williamson' 20:17:37 sorry, i thought i saw a cancellation announcement... 20:17:38 In this model, we plan from right to left and execute from left to right. 20:17:48 adamw: No, I was moving the time 20:17:50 oh theres a prettier version of the model template that my intern made 20:17:51 ah. 20:17:51 * langdon also digging for the link to Matt's talk on it at flock 20:19:05 * vvaldez liked the example in the chat log 20:19:15 So my goal for today was to start with the righmost column and see if we can get on board with our Mission and Vision (such as whether we're happy with the existing version or need to rethink it) 20:19:27 what's our current mission 20:19:30 whats our curent vision 20:19:32 vvaldez: The exercise one or the star trek one? :) 20:19:39 #link https://taiga.fedorainfracloud.org/project/sgallagh-fedora-server-planning-flock-2016/kanban 20:19:46 mission statement is here -> https://fedoraproject.org/wiki/Server 20:19:57 sgallagh: the exercise, especially the part "I can't *directly* make my employees healthy, but I can 20:19:57 provide a path to it." 20:19:58 "Fedora Server is a common base platform with 'featured application stacks' built on top of it. We commit to produce, test, and distribute these application stacks." 20:20:07 Please sign in there with your FAS accounts; we'll use that as our whiteboard 20:20:40 Don't think the video is up yet 20:20:42 On that board, we have the existing mission and vision (slightly abridged, because it was very wordy) 20:20:53 langdon: Matthew's didn't get recorded properly, I'm told 20:21:02 I don't know the details 20:21:13 Ohh bummer 20:21:14 sgallagh, how do i join the server team on taiga 20:21:58 mizmo: Excellent question. I was assuming you would be able to request membership and I'd approve it 20:22:07 sgallagh, im logged in but dont see a way to do that 20:22:09 But this is my first time using taiga, so I apologize if things go wrong. 20:22:21 nirik: Do you know how this works? 20:22:23 I didn't have to do anything.. It just showed up for me 20:22:30 not off hand. 20:22:33 langdon, youre not listed as a team member 20:22:34 mizmo: Can you at least see the page? 20:22:36 I think you have to invite people... 20:22:38 You can't request 20:22:40 but not sure how 20:22:44 sgallagh, i can see it but i cant do anything - read only 20:23:06 On the admin tab.. You can add users.. But you can also just make all logged in have rights 20:23:11 ah, I see it now 20:23:18 sgallagh, you have to go to admin / members / + new member 20:23:22 under the gear in the lower left 20:23:37 And you have to use their fas email address 20:24:05 langdon: How do I just give everybody rights (for today, at least)? 20:24:06 i.e. jstanley@fedoraproject.org ? 20:24:20 or the email address assoicated in FAS? 20:24:27 Change the group rights to all the things 20:24:32 it has to be FAS email address which becomes tricky 20:24:37 .hello duffy 20:24:41 mizmo: duffy 'Máirín Duffy' 20:25:07 Jds2001 no.. The one on .fasinfo 20:25:22 Same as reported by hello 20:25:58 mizmo, nirik: Did you get invitations? 20:26:10 If not, I'm just going to make the edits myself today and we can fix that later 20:26:18 I don't want to burn the whole meeting on this. 20:26:23 nope... ;( 20:26:38 sgallagh, not yet 20:26:43 If you have an account it should just start giving you perms.. 20:26:45 /me sighs 20:27:02 sgallagh, let's fix it after the meeting 20:27:07 Yes 20:27:09 we can #info any changes and update kanban later 20:27:31 OK, so let's start with the Vision Statement 20:27:35 anyhow, the first item in the list is quite broad 20:27:45 #info Current Vision Statement: Fedora Server is the preferred [community] platform for system administrators and developers seeking to deploy applications and services that use the latest technology on a stable foundation with effective resource utilization. 20:27:45 should we be able to measure the impacts? 20:27:53 jds2001: No 20:28:07 Impacts are long-term "this is how we want to change the world" statements 20:28:13 (I think) 20:28:34 right, but if you can't measure it, how do you know if you've achieved it? Or if you're way off? 20:28:34 At Flock, we proposed the more abridged version: "Admins default to choosing Fedora Server as a deployment platform for apps and services" 20:28:45 #info Flock abridged Vision Statement: Admins default to choosing Fedora Server as a deployment platform for apps and services 20:29:13 jds2001: If I understand the model right, the Outcomes are the things that should be measurable. But I'm also quite new at this 20:29:39 That said, a vision is somewhat aspirational 20:29:50 I'll take suggestions on how to rephrase it, though. 20:30:30 can you paste the taiga link again? i lost the backscroll (sorry) 20:30:39 https://taiga.fedorainfracloud.org/project/sgallagh-fedora-server-planning-flock-2016/kanban 20:30:42 https://taiga.fedorainfracloud.org/project/sgallagh-fedora-server-planning-flock-2016/kanban 20:30:48 thanks! 20:31:12 so a vision statement is meant to inspire / invoke a higher purpose 20:31:40 i wonder if admins choosing fedora server is too self referential 20:31:44 you know what i mean? 20:31:46 like how does it make their lives better 20:31:48 how does it make the world a better place 20:31:56 you want to hit those kind of points with a vision statement 20:31:57 Right, but I also want to stress that we need to focus on Fedora Server's vision, not the Fedora Project as a whole, either. 20:32:04 Though obviously we want those to align 20:32:10 so for example 20:33:06 "Deploying applications and servers on top of a free software platform is easy for even junior admins to do using Fedora Server." 20:33:10 not saying we use that but 20:33:17 it sort of focuses on the users needs / goals first 20:33:23 how we make their life better 20:33:33 could even push that angle more eg 20:33:53 I'm just curious since I don't have the history, but why "admins" vs "sysadmins" or "operators" or something else? 20:34:20 "Fedora Server alleviates the headaches and stress of deploying new applications and services, making it easy for admins of any experience level to do quickly and easily on an open source platform" 20:34:49 it should be sort of a 'hell yeah' kind of statement. do we want to do this? HELL YEAH 20:34:51 vvaldez, mainly to save typing 20:34:53 Maybe we should take "Fedora Server" out of the statement entirely and focus on the user's perspective. 20:34:58 +1 20:35:01 smooge: fair enough, thanks 20:35:11 "Fedora Server" could come into it by the title "Fedora Server Vision Statement" 20:35:19 "User Y doesn't need advanced degrees to do Y"... 20:36:02 We envision a future where even part-time / reluctant admins can smoothly and easily deploy the apps and services they need to run their organizations / get the job done / cure cancer" 20:36:08 id also want to capture somewhere that if User Y *does* have advanced degrees, we're not hiding the knobs from him. 20:36:13 yes 20:36:19 well, if we want 'preferred' or whatever that means we want a lot of people using it right? 20:36:24 mizmo: That. But wordsmithed :) 20:36:27 so here, the points we got so far 20:36:41 - even jr / reluctant / inexperienced admins can do it 20:36:54 "Even mhayden can do it" <-- lofty goal 20:36:55 "low friction to entry" 20:36:56 - it has a technical depth with knobs as needed 20:36:57 saying it's easy or whatever seems like a side track for vision... if it's easy more people will use it, but... 20:37:18 - preferred / more people use it 20:37:23 Yeah, I think maybe that's focusing on an Outcome rather than an Impact 20:37:28 nirik, not necessarily, eg VHS winning over betamax is the classic example 20:37:50 theres a lot of popular in terms of userbase software that have a shitty / difficult to use interface 20:38:11 Well, in that case I think the Impact was "Bring home video to the masses" and the Outcome was "VHS defeats competing services" 20:38:16 i had gotten the impression we want more junior folks to be able to grok it to pull them into the fold but correct me if i'm wrong 20:38:23 sure, I am not saying it shouldn't be there, but it's a detail. The vision should be high level... 20:38:30 do we want to talk about cutting edge technology or anything of that nature? 20:38:30 +1 nirik 20:38:37 s/cutting/leading :) 20:38:44 mizmo: Right, but that's maybe focusing on an implementation detail 20:38:46 - leading edge tech 20:38:53 What we want is to grow our community 20:38:57 sgallagh, its not implementation detail: it's your audience 20:39:09 right - grow your community - to whom? who is going to be in it 20:39:10 How we do that is by making it easier to join. 20:39:27 fair 20:39:33 but still, need to focus on the who 20:39:41 oh also 20:39:43 - software freedom (right?) 20:39:49 +1 20:39:53 +1 20:39:54 Comment on the "easy".. Part of that is scale.. Admins are expected to manage *many* more machines these days than they ever used to.. I wonder if you want to capture that part of" easy" as well as "junior" 20:40:10 do people want it cutting edge or do they want it easy to install and foolproof 20:40:14 - helping admins scale 20:40:18 " Fedora server is the ubiquitous choice for open source software deployments" 20:40:18 "Administrators of all skill levels can deploy services that meets their needs" 20:40:19 langdon: something like "at scale"? 20:40:38 * nirik isn't good at this stuff. ;) but oh well. 20:40:46 vvaldez: yeah 20:40:54 follows standards, cutting edge, free 20:41:10 so administrators of all skill levels can deploy services that meet their needs today cant they? it just is a shitty experience with a lot of pain and tedium no 20:41:11 "Administrators of all skill levels can deploy services that meets their needs using open standards and platforms" 20:41:22 sgallagh: i like where you're going 20:41:23 maybe another way to think about this 20:41:25 what sucks with today 20:41:29 mizmo: I'd argue that no, they cannot. 20:41:30 what sucks that we want to fix 20:41:32 Everything is always too fast and too slow.. You want to allow the freedom to choose by use case 20:41:34 "deploy services on a modern platform" perhaps? 20:41:35 At least, not of all skill levels 20:41:46 sgallagh, at the lower end pay someone else to do it 20:41:57 sgallagh, thus in the cloud 20:42:23 i guess thats an elephant in the room? 20:42:33 /me notes that this level of debate is exactly why we start at Impact: if we don't know where we're going, we can't get there. 20:42:34 why are people deploying their own apps and services on a platform when they could use some cloud service 20:42:38 langdon: what do you think, we need more "containers" in the vision, no? :) 20:42:46 do we want to affect that? 20:42:46 Ha 20:43:04 mizmo: That's a good observation. 20:43:18 Just because you can buy it in the cloud doesn't mean it is easy.. Choosing providers, config, etc.. 20:43:37 im just thinking about my non profit use case.... i just need something to work, i'm buying it in the cloud, but the cloud provider only does 80% of what we need 20:43:52 your workloads need to be optimized for clouds, too 20:43:59 Yeah.. I was thinking more like transparent deployment.. 20:44:00 failure-tolerant and such 20:44:09 so right now its not worth it bc its too much of a pain in the ass to roll my own deployment 20:44:27 You just click the "target" button.. That machine, that cloud, that laptop, etc 20:44:28 mhayden: I think mizmo is talking more about basic infra like email and storage, both of which are readily available by players like Google 20:44:29 the end users' orgs do not care about cloud vs this vs that, they just want the biz prob solved 20:44:39 sgallagh: got it 20:44:39 So do we want them to self-deploy, and if so, how? 20:44:44 sgallagh, not even just that, eg we have an app that a custom CRM kind of thing for non profits 20:44:58 its a niche product 20:45:27 well what apps / services are we talking about 20:45:28 I think it is the ease of use focus. people to want to spend hours on forums trying to figure out how to configure/setup something up 20:46:36 mjwolf especially when they have to do it repeatedly for the same thing every few months.. Because one forgets when you do it infrequently 20:46:41 i always think about the app catalog the cpanel mojo apps plugin thing offers, pretty much a smorgasbord of FLOSS apps from mediawiki to pligg to various CRMs and DBs 20:46:55 but theres also custom apps that the org wrote they want to deploy 20:46:55 And.. Strongly agree with mizmo about biz problem 20:47:01 So let me ask a dangerous question: Is the entry-level admin/market something we actually want to address? Is that a solved problem that we should ignore and refocus on something else? 20:47:30 sgallagh: good question, how is it solved today though? 20:47:35 sgallagh, i dont think it's a solved problem 20:47:41 (being said persona) 20:48:12 I like using the term "Default choice" which could apply to entry or experienced admins 20:48:15 mizmo: Sure, but you're also a highly-technical person with a background in related technologies. I don't know that you are necessarily representative of all small businesses. 20:48:40 which would require more community buzz/involvement to become the known default 20:48:52 jds2001: I meant in terms of the mom-and-pop small business, cloud providers like Google basically have that locked up. 20:49:09 sgallagh: in terms of email/calendaring/that sort of thing yeah 20:49:15 Fedora Project's mission wants to see that powered by FOSS, but I'm not sure that's our place to accomplish. 20:49:15 sgallagh, the pieces i represent are the ongoing support (regardless of expertise or not) - we can only choose solutions that have a support ecosystem and that is a very small list of options that mostly suck 20:49:22 sgallagh, we have needs beyond email / calendaring 20:49:29 /me nods 20:49:41 (Right now, I'm kind of playing Devil's Advocate) 20:49:57 so I'm setting up a workstation for my daughter (I know not server but). I had to configure multiple repositories, google a ton. I wanted to load codecs, get skype running etc 20:49:58 What I guess I'm asking is whether our Vision should be focused on scale, as previously mentioned. 20:50:14 sgallagh, also the LACK OF TIME. two kids and a job, and this is extra. i think a lot of small biz have NO TIME for this shit. they want to focus on.... producing frobbles and beezsnitches, or whaever their core biz is, and this type of stuff they dont care about its not their passion 20:50:23 Instead of trying to capture the nickel-and-dime users, should we attempt to go after the people hosting the cloud providers? 20:50:25 so for server I think you need an easy way to see what is out there and have it work and install without doing a bunch on configuring googling etc 20:50:37 mjwolf: a lot of that is legal, and specific to workstations 20:50:47 (Yes, I'm aware that doing too good a job of that might cause issues with our principal funding source) 20:50:50 but feel your pain at the same time :D 20:50:58 sgallagh, well the advantage of aiming at the small biz end is you get the ubiquity if a lot of small orgs use you 20:51:01 * langdon is waiting for his beezsnitch to be delivered.. 20:51:21 langdon: ??? 20:51:26 sgallagh, but they're ppl who dont care about the tech 20:51:31 * langdon joking 20:51:42 sgallagh, the advantage of doing the at scale stuff is its people who do have more of an interest in the tech for techs sake 20:51:43 * vvaldez prefers frobbles 20:51:44 /me missed the reference 20:51:57 sgallagh, 16:50:14 timestamp 20:52:05 i was channeling seuss sorry 20:52:28 Oh, I missed one line of your replies, hence the confusion. 20:52:30 i think theres a lot of things we want for the vision and we have to connect the dots 20:52:32 Scale does not mean hosting providers in my mind.. It means managing the things on the providers.. But that may be more the domain of cockpit 20:52:33 langdon, your order of beezsnithces and frobbles is back ordered due to a problem with our ordering system. no one knows how to fix. we will let you know when we find someone who understands this Login: thing 20:52:33 I only saw langdon's response 20:52:45 so yeah, i think that if you go for SMB, then the enterprisy stuff comes after 20:53:04 Smooge arggh 20:53:34 My sniggertbiscuit won't work! 20:53:45 eg ubiquity, ease of use, software freedom, leading edge technology, technical depth / knobs, wide audience (experience), scale, 20:54:05 OK, so we're approaching the top of the hour. Is everyone going to turn into a pumpkin, or do we have a chance to keep hashing this out? 20:54:14 * mizmo has time (unless you want me to go :) ) 20:54:25 How about home servers? A la workstation.. Target developers or tech people who manage servers for their personal or semi-formal stuff 20:54:45 "Fedora Server is the defining technology solution of the 21st century"... 20:54:47 langdon: that overlaps with the SMB target, no? 20:54:49 langdon, what are those peeps deploying 20:54:50 (I'll show myself out_ 20:54:56 langdon: id think that a lot of the concerns are the same 20:55:02 agree, there a ton of other use cases like home NAS servers and such, DNS/DHCP for home etc that could fall into 20:55:13 Jds2001 yes I should think.. But limits near term scope 20:55:21 sgallagh: I have some time 20:55:48 i have a bit of time, but I've been putting off a conversation with someeone for this meeting :D 20:55:55 told them i'd be avaialble around 5 :) 20:55:59 What about this: "The overwhelming majority of businesses (large and small) run at least some Fedora Servers" 20:56:02 Mizmo I think if you polled the room we could easily find the answer cause I bet all of you have at least one :) 20:56:10 (as a wild, long-term goal) 20:56:19 langdon, well for us but what about the people we want 20:56:22 what do we want people deploying 20:56:27 what do they want to depoy 20:56:29 deploy 20:56:35 sgallagh: what problem does that solve? 20:56:43 sgallagh: we're back to needing the user perspective 20:56:51 sgallagh, then you're basically just envisioning ubquity. which is fine. but how does that help the user 20:56:53 seems timid. ;) we should want the world! 20:56:53 sgallagh, that helps us 20:57:04 just bc everyone uses us doesnt mean we solve their problems 20:57:11 jds2001 +1 20:57:16 True. (See Microsoft for example...) 20:57:20 I expect they are very similar to the developer profiles workstation wants.. And/or members of the server wg or Fedora devel ML 20:57:20 exactly 20:57:37 Anyone should easily be able to obtain a stack of software that they want to deploy, and install, configure, and deploy that software easily, regardless of entry-level or super experienced administrator with one billion machines. Everyone appreciates ease of use. Everyone appreciates curated recipes for doing such things. Regardless of "home server," cloud, traditional server, in containers, etc. 20:57:52 I'm worried we're getting into "be all things for all people" territory again, though 20:58:09 rbergeron++ 20:58:09 sgallagh: Karma for rbergero changed to 1 (for the f24 release cycle): https://badges.fedoraproject.org/tags/cookie/any 20:58:11 rbergeron, +1 20:58:11 Because you can spend time making that recipe, or you can spend time doing other stuff. Or you can at least have something to start from if you want to do crazier stuff. 20:58:34 rbergeron, i like how you're stating it, "Anybody should..." 20:59:07 rbergeron: coming out of nowhere with the game-winning play 20:59:23 rbergeron is like that :D 20:59:26 And people appreciate being able to collaboratively work on what those solutions look like, and they appreciate knowing that they ahve been tested, and work well. 20:59:31 i would say the vision should basically be, worry about the shit you're doing, not the intricacies of deployment. you do your thing we do ours 20:59:38 focus on your beeznitches 20:59:58 And you can put a wrapper and a clicky thing around those, or not, or whatever, but: recipes are nice. 21:00:21 coudl the vision include, a recipe for anything anybody could want to deploy? 21:00:29 OK, so: "Anyone should easily be able to obtain, configure and deploy software that enables their business needs using crowdsourced and tested recipes." ? 21:00:46 crowdsourced? 21:00:52 we'd be curating them, no? 21:01:04 Actually, rearranging a couple words: "Anyone should be able to easily obtain, configure and deploy software that enables their business needs using crowdsourced and tested recipes." 21:01:07 we want a curated product, not a choose-your-own-adventure product, right? 21:01:10 jds2001: That's not mutexed 21:01:17 could just say "pre-existing recipes" to emphasize you dont gotta write it yourself every time 21:01:22 The solutions can be crowdsourced but blessed 21:01:30 cuz crowdsourced vs curated vs etc is implementation detail 21:01:31 I like curated 21:01:39 the main point is it's there, i dont have to do it, i can take it and use it and save time 21:01:43 right? 21:01:59 Mizmo and give feedback /contribute 21:02:12 langdon++ 21:02:12 mizmo: Karma for langdon changed to 7 (for the f24 release cycle): https://badges.fedoraproject.org/tags/cookie/any 21:02:15 mizmo: and optimally, i know it's been tested or blessed by someone or something 21:02:18 rbergeron++ 21:02:18 mizmo: Karma for rbergero changed to 2 (for the f24 release cycle): https://badges.fedoraproject.org/tags/cookie/any 21:02:24 mizmo: +1 21:02:28 OK, another try: : "Anyone should be able to easily obtain, configure and deploy software that enables their business needs using readily-available recipes." 21:02:49 how about dropping business. ;) 21:02:52 do we wanna say readily-available and trustworthy recipes? 21:03:04 nirik++ could be biz, orgs, or personal use 21:03:04 mizmo: Karma for kevin changed to 23 (for the f24 release cycle): https://badges.fedoraproject.org/tags/cookie/any 21:03:06 I like tested/cursted/something to indicate quality 21:03:07 just drop biz 21:03:15 Trust is good too 21:03:29 Eh, I was using "business need" in the strict definition, but I'm good with dropping it. 21:03:39 just say needs 21:03:39 * langdon still typing slowly 21:03:43 Yet another try: : "Anyone should be able to easily obtain, configure and deploy software that enables their needs using readily-available and trustworthy recipes." 21:03:48 I feel the same, something that indicated it's known to work 21:03:58 that works 21:04:02 enables their needs sounds funny like its making me need more 21:04:15 right in the needs! 21:04:16 you meet needs 21:04:17 how about "solving their problems"? 21:04:17 Should we scope "software" to server-y stuff? 21:04:17 mizmo: The first one's always free... 21:04:24 you enable greatness 21:04:26 lol 21:04:29 "solving their non-emotional problems" 21:04:36 hahaha 21:04:44 i get very emotional about servers 21:04:47 * mizmo wipes away a tear 21:04:56 * mizmo is verklempt 21:05:00 * nirik packages a zippy the pinhead recipe. 21:05:09 Yet another try: : "Anyone should be able to easily obtain, configure and deploy software that addresses their needs using readily-available and trustworthy recipes." 21:05:18 address! yes! 21:05:23 yes! 21:05:25 sgallagh++ 21:05:26 vvaldez: Karma for sgallagh changed to 8 (for the f24 release cycle): https://badges.fedoraproject.org/tags/cookie/any 21:05:28 s/software/services ??? 21:05:38 Micro! 21:05:38 that makes it more sever-y 21:05:41 can we say anybody instead of anyone? im weird and nitpicky but anybody sounds less formal to me 21:05:57 like langdon suggested too, is software the right term 21:06:05 do we want to say apps/services? something else? 21:06:21 bits? 21:06:23 do we want to throw in something about software freedom 21:06:25 I liked services or "software services" 21:06:26 solutions? 21:06:48 do we want to say anything about scale 21:06:50 mjwolf: I think "solutions" is a little too buzzword-y for this 21:06:55 Solutions is soooo corporate-y :) 21:07:02 agree 21:07:14 Scale is droppable imo 21:07:17 I'm for "software services" 21:07:27 how about freedom 21:07:32 Yeah, I think "at scale" is covered under "meeting needs" 21:07:35 Services *should* imply scale 21:07:44 mizmo: That's already covered in the Fedora Vision 21:07:45 do we say freedom is covered by the overarching fedora proejct vision and leave it at that 21:07:46 yeh 21:07:51 lol great minds think alike 21:07:53 readily-available, open source and trustworthy... 21:08:03 mizmo: Apparently ours do as well 21:08:10 sgallagh, bwahaha 21:08:11 open source is sorta implied :) 21:08:25 sgallagh: i might easily obtain and configure software that addresses their needs using blah blah blah... and deploy to the location / image type (??) of their choice 21:08:26 nirik, ah yeah langdon made the point about how the recipes could be openly worked on right, feedback, mods whatever 21:08:33 I would leave oss out, personally, just cause of the whole 3rd party repo discussion going on 21:08:36 though that might still fall under "easily ... deploy" 21:08:44 * mhayden unfortunately must depart and sit in traffic on the freeway 21:08:47 rbergeron: Sorry, I'm not sure I understood... yeah? 21:08:54 rbergeron, if its addressing their needs its deployed where they want it i think 21:09:00 i think! 21:09:08 mhayden: you dont know what traffic is :D 21:09:13 mhayden, enjoy :( i have that dance coming up soon 21:09:20 sgallagh: you should be able to take a recipe and put it into an ami, install it directly on bare metal, stick it in a docker container 21:09:31 "Anyone should be able to easily obtain, configure and deploy software services that address their needs using readily-available and trustworthy recipes." 21:09:55 rbergeron: but those could be classified as their "needs", right? 21:10:00 ....and if ami, stick that crap in AWS, etc, if docker stick it in your k8s cluster, whatever. 21:10:04 I think I'd actually leave open-source out of this statement. Open source is arguably an implementation detail for how we're going to accomplish that 21:10:05 jds2001: sure 21:10:24 jds2001: at some point the whole thing could ust be "fedora server: serving your neeeeeeeds" :) 21:10:28 i guess a plurality of deployment targets means flexible 21:10:35 rbergeron++ 21:10:35 geppetto: Karma for rbergero changed to 3 (for the f24 release cycle): https://badges.fedoraproject.org/tags/cookie/any 21:10:36 do we want to highlight flexibility? 21:10:41 Fedora server: to server all your server needs. 21:10:53 rbergeron: If we get to the point where Fedora Server is serving me a T-bone steak on command, I'll retire a happy man. 21:10:56 Also, full. 21:10:56 at first i read that, to sever all your server needs 21:11:11 * nirik chuckles 21:11:13 mizmo: that's what amazon lambda is for! serverless! yo! 21:11:34 heh 21:11:37 One plus side to rbergeron 's point is marketing... If you say it we can make hay with the all the targets 21:11:44 all your needs are servered 21:11:57 I'll also note that the details about how we get there are probably better placed within our *Mission* rather than our Vision 21:13:08 "Anybody should be able to easily obtain, configure and deploy software services that address their needs using readily-available and trustworthy recipes." 21:13:24 mizmo: I honestly disagree with you on "Anybody", but not enough 21:13:31 why??? 21:13:33 to argue about it if everyone else prefers it. 21:13:49 Personal pet-peeve. I think it sounds *too* informal. 21:14:02 i think a vision statement should be less formal cuz its the inspirational piece 21:14:07 the mission should be more formal / technical 21:14:08 imho 21:14:21 I don't know about formal/informal. they both are pronouns 21:14:28 how about: 21:14:35 i suppose anybody is discriminatory since it requires you have a body 21:14:36 "Everyone is able to... " 21:14:48 nah, nevermind 21:14:51 * nirik is tired 21:14:59 I gots to go.. 21:15:03 mizmo: Also true: we might well have AI consumers down the road ;-) 21:15:14 /me mumbles something about the singularity 21:15:18 i dont want to serve the robot overlords 21:15:28 I for one salute our new robot overlords. 21:15:38 "I'm from Skynet and I'm here to help. 21:15:39 anyway.. I think we are in the weeds 21:15:41 "Any admin"? 21:15:42 yep 21:15:43 close enough 21:15:49 we got vision? 21:15:55 are you inspired by that statement? 21:15:57 "Anyone should be able to easily obtain, configure and deploy software services that address their needs using readily-available and trustworthy recipes." 21:16:11 mizmo: +1 21:16:13 is this a rallying cry? 21:16:34 +1 21:16:38 is "recipes" too implementation-y 21:16:42 i guess its fine 21:16:45 +1 21:16:46 it is the right meaning 21:16:52 mizmo: No, generic term. 21:16:53 mizmo: implementation would be "playbook" :) 21:17:06 i was going with Anyone should be able to easily obtain, configure and deploy software services that address their needs 21:17:10 jds2001, yeh but puppet uses 'recipes' right 21:17:13 everything after that is implementation 21:17:18 jds2001: Or lmishell script! ... what, it could happen. 21:17:32 sgallagh: :) 21:17:34 mizmo: It's a sufficiently generic term, IMHO 21:17:38 smooge, well, the part i dont think is implementation is the part where i can choose from stuff someone already made available, i dont have to do it myself, and it's centralized 21:17:39 chef uses recipes. puppet uses strings 21:17:46 what about "formulas"? 21:17:46 mizmo: puppet == module 21:17:54 no recipes is fine 21:17:55 chef uses recipes 21:18:16 mizmo: https://galaxy.ansible.com/list#/roles?page=1&page_size=20&platforms=fedora 21:18:20 because you could have a system that makes blah blah easy to do, but you still need the legwork eg 21:18:28 there's 716 things for fedora 21:18:29 you could have a ipod without the itunes, so you got a great music player but no music 21:18:40 i think having the catalog of music to play is an important component of the vision 21:19:26 mizmo my main reason is that I find statements over 72 characters hard to grok. 21:20:05 mizmo, but that is a failing on my part so I am not going to wordsmith it down to a roix of that length :) 21:20:07 * jds2001 has gtg 21:20:11 smooge: How far over is it? 21:20:40 Anyone should be able to easily obtain, configure and deploy software 21:20:42 ah, about double 21:21:08 I don't think we can be that abbreviated and still get our message across 21:21:21 yeah. it isn't going to fit in a tweet 21:21:23 * rbergeron has to depart as well, i have promised a child to take her shopping for watercolor painting supplies 21:21:25 I agree with mizmo, the availability of the pre-built helpers is key 21:21:26 but 21:21:40 sgallagh: anyone should be able to *confidently* ... blah blah blah 21:21:51 if we pair it down anymore we may end up with: "Anyone should address their needs" 21:21:59 confidently implies trust and belief and etc. 21:21:59 "Anyone should be able to deploy software services that address their needs." 21:22:19 I don't like dropping the recipe section, really. 21:22:25 (still working on it) 21:22:41 I do kind of like the addition of "confidently" though 21:23:11 "Anybody should be able to avoid reimplementing what others are by building on top of a shared collective" 21:23:12 blah blah blah 21:23:32 something about it being not just that they can deploy, but they can be part of something greater 21:23:50 and that it saves everybody work reinventing the wheel in their own silos 21:24:00 hmm 21:24:05 was hoping it'd be more inspirational than my sleep deprived brain can do right now 21:24:08 "Anyone should be able to confidently manage software services using trustworthy recipes" 21:24:09 FYI, I have a hard stop in six minutes. 21:24:18 we close 21:24:20 manage could imply obtain, configure, deploy 21:24:21 Server Goood. Sysadmin Like. Solve Problems 21:24:27 /me snickers 21:24:29 Just go with what we had a while ago 21:24:57 call it draft 1 and we can wordsmith later 21:24:57 * nirik thinks we should let that perkolate and see if we can think up changes next week or on list. 21:25:00 exactly 21:25:05 its hard to be brillant on the spot 21:25:08 though I like the short one vvaldez said as it sticks in my head 21:25:09 brilliant even 21:25:38 ok I think we are good here. percolate 21:25:43 vvaldez: The only problem with that is that it sounds like it could be Ansible's Vision, not strictly Fedora's 21:26:01 sgallagh: that's a great point, I didn't think about that 21:26:15 Whereas the longer version includes deploy as part of it, which strongly implies providing the deployment platform as well 21:26:33 which one 21:26:38 can you copypasta 21:26:53 mizmo: Your last one before we started trying to shorten 21:26:59 vvaldez> "Anyone should be able to confidently manage software services using trustworthy recipes" 21:27:19 "Anyone should be able to easily obtain, configure and deploy software services that address their needs using readily-available and trustworthy recipes." 21:27:52 this vs ansible is a conundrum i dont know what to do about 21:27:57 what is ansibles vision statement do they have one 21:28:14 AUTOMATION FOR EVERYONE 21:28:14 Deploy apps. Manage systems. Crush complexity. 21:28:14 Ansible helps you build a strong foundation for DevOps. 21:28:22 I'm not worried about overlap, I'm just worried about constraining ourselves to solving only the same problem that they do 21:28:26 see thats more of the type of language you want for a vision, thats from their front page 21:28:32 see the for everyone 21:28:47 strong foundation for devops i dont know what that really means but 21:28:51 the first two lines, a+ 21:29:02 I think Ansible will be one of the bigger hammers in our toolbox, but not the only one 21:29:08 what do we add 21:29:28 i fear we're not getting at that 21:29:30 Ansible provides the recipes, but we provide the actual services 21:29:51 why use ansible + us instead of ansible + ( other os ) 21:30:05 right, that's the question I think we need to answer 21:30:08 and updates and testing 21:30:14 nirik++ 21:30:14 sgallagh: Karma for kevin changed to 24 (for the f24 release cycle): https://badges.fedoraproject.org/tags/cookie/any 21:30:38 services is the biggest thing IMO 21:31:54 OK, I need to turn into a pumpkin. 21:32:20 mizmo: Would you be so kind as to throw together a brief summary in an email to the list and we'll try to continue the discussion there? 21:32:33 (Tomorrow is fine, I know you need to go home) 21:33:54 we could post a few of the final versions then vote on them? 21:34:12 Well, I think we've established that there are still questions we don't have answers to 21:34:19 So I think voting is premature. 21:34:25 makes sense 21:34:52 #info Lots of discussion on this topic, will continue on the list 21:35:08 sgallagh, i gotta run but i can try to summarize tomm 21:35:20 mizmo: Thanks very much (and thanks for helping drive this) 21:35:26 #endmeeting