 Hi So Okay, I'll introduce the handheld microphone throws me off This is Roy. He's Okay, yeah Yes, so I'll take you to this left list I'm Roy. I work with Wunderkraut. I work in Belgium and the Netherlands. I Am a Drupal UX designer. I contribute to core my big break was the views UI so you can blame the three columns, etc there on me Some stuff for the Drupal 8 has been the star guide and we've redesigned the content creation page in the modules page So I specialize in interaction design things and I like to apply that to Drupal core and country The last line can probably safely be ignored I'm Jeff Eaton. I'm with Lullabot. I'm a digital strategist there I did a lot of like core hacking and spawning off of lots of miscellaneous Drupal contrib modules in the pre Drupal 7 core days. I mostly have lived in contrib since then If anyone remembers the small core hashtag, I'm to blame a lot of my time is spent doing architecture work and content strategy for Large like publishing and media clients now But I also end up talking to lots of friends and small orgs that want to they decide They want to use Drupal because it's all full of cool things, but it has a learning curve Roughly equivalent to a barbed wire fence So that's what we're going to be talking about a little bit sort of obliquely This is Sort of what I like to think of as the tool that people want when they get Drupal It does all kinds of cool stuff. It can lift heavy things. It's fun to drive And this is what they get Now it's cool. There's no mistaking that well That actually might be going a little far what they actually get is this They still have to put it together themselves Well, actually what they really get is this Because it doesn't even actually come with instructions. It's just assumed that you're an enterprising person You you you can figure it's got little bumps and they click together go make yourself a bulldozer You'll enjoy it. It can be any kind of bulldozer you want. It could be a dragon. Whatever This For those of us who are full-time Hard-core we like to go build stuff we get hired to build stuff people This it just looks messy not terrifying or daunting we go Oh, well that needs to get sorted maybe the you know the colored pieces go over there And you know the windows go over there, but the underlying problem that New arrivals to the Drupal world Don't really get any guidance about how this giant bin full of Lego bricks of you know Content types fields views build modes display modes Feeds blocks menus breadcrumbs all of them are Bigly familiar to anybody who's worked with web stuff or has talked about working with a web publishing system But figuring out how all of these different building blocks get combined into an actual coherent hole is something we don't give anyone much actual We don't really demonstrate that much has anyone here ever installed Drupal and used the standard install profile. I Said used the standard install Okay, Larry is going to be our outlier for the afternoon Just core you installed Drupal core you used the standard install profile and it was excellent Yeah, it's basically the world's most complicated version of WordPress out of the box which is sad because that Grossly under sells what Drupal is capable of I think it's like everybody in the room is probably familiar with you The fact that you can build big complicated things with Drupal But there was a really really interesting post a while ago By someone named Stevie who basically ranted vigorously about Google's platform This was during the days when you know Google plus was very very new and oh there was all kinds of debate about this wild new platform that they were creating and One of the things that he said was that simply making something a platform You can do a lot of stuff with does not guarantee success That is not like there the population of people who go aha a new kit with which I can build crazy things excellent Those may be our people, but that's not a very big slice of the population There has to be some sort of killer app some sort of thing that people engage with and say oh This is what's on the other side of figuring out how to click all these pieces together. This is something that's useful I could actually do something with this that killer app in the case of Facebook that he was talking about it was Photographs and the wall of your friends updates, you know Facebook has this huge rich platform But still roughly 45% of people's time is spent clicking next on a photo album Because that's what grabs people and the fact that that was built on top of the Facebook platform Was a killer app that gave them something to build on and extend from Rather than just saying ta-da big social graph. You should totally do something with this so that brings us to the philosophical question of Whether or not we should do something like that with Drupal I'm gonna say yes, and just move on There are any objections to that feel free to register them on Twitter But there actually really has never been too much Disagreement about the idea that Drupal should be in some way useful out of the box Like there's never been a whole lot of support for the idea that it should just be utterly stripped down to bare metal and Do nothing and I hear a but Well, yes, perhaps there is a but The problem is is once you start actually asking, okay, let's make a really cool engaging thing that Drupal could just do out of the box There's an entire world full of people who use Drupal and an entire Universe full of potential use cases and how do you possibly choose because for a long time? We've said no no no the standard install is literally for everyone But that makes it so generic that what you get is the standard install the world's most complicated copy of wordpress with a feed aggregator built in You know, it's a function. It's full of functionality But there isn't any particularly compelling story about what you're going to do with standard install other than Turn off some of the stuff and go build your actual website About two years ago. Did anyone hear? There's been there's been Debate and heated vigorous Discussion around the idea of making Drupal's install profile out of the box do really cool things and serve a particular use case But the argument is always boiled down to which one of those use cases it should be, you know It will it be a personal portfolio site Well, we try to show off some sort of like small business website or something like that One of the big problems is is you had to actually be able to build it with core in order to ship it with Drupal Which was no small feat in and of itself And then there were some crazy people who argued that you should actually just strip everything out and leave it bare Metal with a 404 page and that would be Drupal's default state. No one listened. It's probably for the best But again, where are they now? It's called symphony now But about two years or so ago a lot of conversations started rolling I'm biased because I started some of the conversations About something called snowman. It's a slightly ambiguous name. It was originally named tsunami The idea of you know, it would just it would be that you some sort of Install profile that could like you could start a movement with it or whatever Unfortunately, roughly 45 minutes after that was announced at Drupal con a tsunami killed several hundred thousand people Changing the name really felt like the best move But snowman works to it's still water related roughly But the idea was that rather than try to have some sort of Gigantic huge all-encompassing community discussion about what the most Fundamentally perfect representation of Drupal this would be and how do we make that into install profile or whatever? a couple of us just wrote a description of what the what a use case could be that would cater to Drupal's Drupal cores out of box strengths and Actually be an interesting kind of website to have and to use a use case that we could actually Explain to someone and someone might actually care about At the time the idea was that basically the audience for this would be a small group of people a small team You know friends, whatever who are collaborating together on some kind of project And they wanted to tell the world about that project and they also wanted to invite and encourage other people to come in and join Them in working on it Coincidentally this sounds not unlike trying to run a small open-source project or something like that We may collectively have a little bit of experience doing that so it's also useful because it's a kind of Story about what this example install profile this site that it would make what it would be used for It's a story that I think we can probably kind of Understand it doesn't take tons of work to get in the mind of some people working on a project that they want to Evangelize and get other contributors for and there was briefly a great deal of excitement about this You know there were buffs and there were meetings and there was you know issue queues and stuff like that It didn't last a whole long time for a very important reason. We still had to build the damn thing with core and In 2011 when this stuff was really starting to get rolling There were a couple of significant roadblocks we encountered one you couldn't list things Turns out if you want to actually you know think about a particular use case a particular kind of site You very quickly run into things like oh, and we could totally make a page where the news lives, I guess we can't Okay back to the front page then you know those kinds of just intractable problems in Drupal core meant that either You were gonna give up on it ever shipping with core which sort of you know missed the whole point of what we were trying to do or You ended up writing a bunch of custom code and jamming it into an install install profile and basically cheating Anyways, you know you were jamming a new custom module into the install profile anyways So this was a real problem. The other one was there was no way to do relationships between content types And you know there were there have been Drupal modules out there like node reference and node relationship and stuff like that That could do this say Have you know an employee in an office related to each other or an album and an artist related to each other Those kinds of things are really sort of you know a meat and potatoes feature of a lot of websites But there was no way to do that. So there was a whole in a whole other swaths of functionality that were totally Totally out of our reach No way to do dates Which meant that any kind of event calendars were out of the question no way to do links Which means that directories of things were going to be really annoying to build You know this this whole actually building it with core thing was really starting to kill the buzz Then we also were thinking about the fact that if we actually made something useful all of the tools for Content administration that ship with core are pretty generic and vanilla too So if you start thinking about the use case for a site where that people are actually using Kind of want to start customizing this stuff too and then also The act of creating an install profile that actually set up anything non-trivial in Drupal 7 Required tons of custom code you are basically hand writing all of the code to go in there and build out new content types Add individual fields set all the default settings for the site the more functional you made it You know the you were signing up for hundreds and hundreds and hundreds of lines of custom install code living in the install profile and That was also not cool You know it making an install profile in D7 was possible But dicey and it required deep understanding of certain APIs. So basically It was very not cool, and it got us back to I guess we could just sort of have wordpress with forums turned on and an Aggregator for some reason, you know, it very quickly looked like Heavy amounts of core development work were the only way that you know We're the only way that this could happen and it wasn't like you know People are going to put views in core or something like that. So back to the drawing board No way We were sad we had many sadds for a while, but then it turned out Not too long ago we sort of stepped back and a couple of us who'd been involved in these discussions started talking amongst ourselves and saying, you know We've got views in core That's a pretty big one. We've got entity relationships in core So we can make those content models that connect things to each other. We've got a date field We've got a link field there's views integration for those we could actually do a List of upcoming events that doesn't require people to do bizarre and unholy hacks with the date posted field on a new piece Of content you could just make this stuff now Then views bulk operations a tool that lets you make little customized admin screens and just you know Say, oh check check check check unpublish these and things like that That's in core Config management the the config management initiative CMI that went into core and it's dramatically Simplified the actual process of building install profiles Because most of the grunt work of say defining a content type or defining fields or defining a view say Get stored in config files. They can now just be moved over and into install profile directory And it becomes something that that install profile can initialize without using a bunch of custom code So that problem is at the very least reduced dramatically Suddenly we've also got abilities like the ability to do custom entity display modes like well, I've got a I've got a Staff bio I could actually make a bunch of different ways that this could appear throughout the site You know, I could have the page where that person's bio lives I could have a different something else for the staff photo gallery, you know There's a much richer vocabulary sitting right in Drupal core For building this kind of stuff of the front-end and back-end themes are responsive now We've got fieldable blocks. So you could actually use a picture in a sidebar block. It's a knob I know it's like, ah, it's 2013. It's amazing Got the editorial like the list keeps going on and on down there now There are There are other debates about whether or not this is making core bigger and heavier and stuff like that but unarguably the Vocabulary that ships with Drupal core at this point is for the first time in history rich enough to describe and implement the functionality of a Lot of different kinds of interesting and useful websites. This is like the dawn of a new era and Once everyone actually manages to like get themselves over the shipping Drupal 8 hurdle and Sobers up and comes back from the long and well-deserved vacations and starts working on Drupal 9 You know The ripple effects of this stuff being baked in rather than something that people have to go out and discover After learning basic Drupal vocabulary and then they go and they find the contrib module repository And then they download stuff and then they break everything and then Because this is cool, but this has made the pile of Legos only bigger bigger So the takeaway today is a kind of shocking and suddenly unexpected We could actually just build that use case that we were talking about an install profile for a team of people That is doing an open-source project and wants to tell the world about it We could just build that like today You would probably break because Drupal 8 is in alpha, but that's alpha breakiness not we had to do crazy things breakiness so Stepping back two years into the wayback machine and This is also where you can start feeling feeling very free to just like throw things or object or Pause and say I think that's a terrible idea and we will attempt to rebut you in a gracious and friendly fashion The core principles that we were talking about at that point For an install profile that could ship with core that would do something useful The first one was before any other things, you know before like giving people an example of how to use this API or demoing Drupal or whatever Make that use case and actually serve the needs of that use case that we described like as if we actually were trying to ship an Open-source product for small teams Trying to tell the world about their project or whatever don't just treat it as like the the north wind database of Drupal Purely useful for saying oh well you need to do X go look at this example in the snowman profile That's useful But it's a secondary concern the idea of showing off Drupal capabilities and saying oh well Let's walk through this feature demo of Drupal in the form of this install profile That's not the goal because that also leads to massive feature creep You get like the Frankenstein monster install profile Who's only purpose is to show off all of the checkboxes and all of the features and stuff like that That doesn't necessarily help anyone so we can show off Drupal capabilities, but that's a second You know priority and then the third priority is like Teaching site builders how to build things showing them an example of how you might assemble things they could absolutely Learn from a pre-assembled website that already has a bunch of Drupal stuff clicked together and built but that isn't the Inherent primary purpose of it. That's a happy accident It's you know probably be less complicated than logging on to a client site and clicking around to learn how Drupal is built Because it would be a simpler site But again, there's that priority list of you know What things we consider first and then finally it wouldn't be intended as a starting point for custom site builds This wouldn't be like the Zen Yeah, it would not be the Zen install profile. He would say well, I want to build a site and it's kind of community-ish So clearly I'll start with snowman No I mean that that again not the idea because that raises all kinds of questions about like you know What crazy use cases could we imagine this being extended for? Way too complicated Serves that use case it would probably if someone really used it They would probably eventually start adding things to it But like it's the goal isn't to be as you said Zen so Going back to that with those principles with all of these new cool Lego bricks We've got in Drupal 8 a small group collaborating on a project They want to tell the world about they want to convince others to join in with them Say Alice and Bob are making a new CMS Thank you So what So the first question is and this is sort of the process that we go through with client sites too, you know What do they need their site to do? What job are they hiring their website to perform for them that this would need to perform and Like what kind of discrete things would it need to do and this is For those of us who are used to thinking in terms of like, okay, what are the feature checklists? We'll take this off. We'll take this off We'll drop this module in this can be a bit of a shift to thinking in this way We're not thinking about features. We're thinking about Jobs that this website will be asked to perform that may they'll they'll take the form of different features or content types But that it will get there later So the first one again with this new unnamed CMS is to share some sort of vision Like why are we doing this? What is this about? What is its purpose? You know, what's the plan stuff like that, you know any open-source project that you know people want to you want people to start getting involved There's got to be some sort of Mission and vision that gets conveyed The other one is building trust, you know, one of the big challenges for an open-source project It's you know ramping up is getting people to say yes, okay, this looks legit Looks like you know people are having some success with this. I'll try it out So that's one of those like jobs the website is supposed to perform getting people to take that sort of step of trust of even saying I'm gonna spend my next two hours tinkering around with this. I'm Spending my time on this so I need I need some degree of trust and the other one is encouraging participation You know it the part of the job of this website is getting other people to say yes I will start investing my time. What is that next step? You know, how can I how can I get involved? So these are like three core jobs that a website for our hypothetical people needs to perform It honestly was hypothetical So I'll pause this seems a very quiet. Thank you for thank you for bringing some levity Like are there any other kinds of Fundamental jobs that a website for a small open-source community would be responsible for This isn't rhetorical Events Releasing it like keeping people up to date like ongoing engage maintaining ongoing engagement or something like that Yeah, all the features of github Yeah Probably not gonna happen with d8 core, but we'll see about d9 Any other thoughts? Oh, no, we're just gonna ship it No, but I mean, I think that's a good idea You know the assumption that this would probably have to evolve over time as they figured out more needs or whatever And again, like that's actually a very good idea because this is help This is like providing training wheels and getting in What you're suggesting is that we might consider and what's next so yeah out into contrary bar Yeah, like what what are the next steps for the people who are running this website now that they've gotten and it's installed Documentation for the project book module click done Oh That's a good like the actual like two-way dialogue with people or I like it Twitter is technically not a solid feedback mechanism for most projects Although I hear it's how you get Dave Reed to fix no I'm hearing case studies and case studies could be a way to build trust Yeah, a showcase of people who have actually used it. It's yeah, okay a team page Yeah, who is actually making this thing? Okay, I Gotta say I'm feeling very happy about some of the answers here because I do actually have So, you know under the the core vision like you know, there's some basic like about information like what is this thing? You know, it's got to be more than just a logo and a repo. You know, what is it? You know, what's our plan? What do we you know? What are we hoping to achieve where are we going with it? Frequently asked questions because you literally are not allowed to put a website on the internet Without a frequently asked questions page The internet police will come for you The team, you know a team page, you know bios who's working on it case studies Yeah News coverage, you know, if it gets written about and people on slash dot get angry about it or something like that clearly You know, you've got to mention that and tell everyone to go there and defend the project and you know That's how it works Under encouraging participation the basic, you know, how to get involved page point to the github project whatever Open forums where people can share their dissatisfaction or they're encouraging stories about how it's changed their life working with the project, you know Those kinds of things or you know post spam as they do Group events I think you know that was mentioned a couple of times of like, you know Oh, we're we're gonna be having a sprint or there's gonna be a screencast or the new You know the the next release is gonna be dropping So if you've got any ideas for what you want in because our new CMS is he's so incredibly light-weight But just if somebody tosses out an idea for a feature, we'll just squeeze it right in before it ships So we'll just you know put up a you know on Friday. We're rolling a new version get your features in I'm just making stuff up right now So That yeah, that was Drupal 3 yeah It's just everyone decided it was time to make a new round number and that became a new version of Drupal. Yeah So what was interesting, you know, we've gone through some of the same process some of the same brainstorming process Way back in the day when snowman was being kicked around initially and Preparing for this and you know it during sleepless hours with markdown files making notes and stuff like that This kind of set of functionality although there were there are a couple of good ones that were also mentioned It could probably be added to this Documentation yeah, but who wants to write documentation? Yeah, maybe documentation is good for build trust Yeah The interesting thing is is we haven't really other than community forums talked about specific features yet We're talking about tasks that the website is supposed to perform for our people our Bob and Alice so This is again other one of those fun community participation times Given the tools that we know of in Drupal 8 right now the stuff that we just talked about is now like you know in hand Does this kind of stuff look doable? Any any any initial thoughts content types I'm tricking you into actually beginning the implementation work right now Yes That's why it's only for registered users Yeah, who would have why why would a spam bot ever register for a site? That's a good point though. This is something that we'll have to think about That's one of those next steps. Yeah Yeah, ooh, so like you know some Okay, so I'm not sure whether that yeah I'm not sure which of those sections that would go in but the idea of like what's the selling What's the case for how this software is good like a? Showcase Yeah, oh, yeah, yeah, like I mean and well now now we're just into content writing guidelines, you know Products page. Wow. Who's who's ready to monetize now? so I I am fairly confident that barring the uncomfortable spam question This is something that could probably be knocked out in d8 in a relatively short period of time There would be tweaking to be done and stuff like that But I don't think it would fundamentally tax a site builders experience, you know, you know a knowledgeable Drupal site builder if They were comfortable installing an alpha version of Drupal and occasionally getting Ajax errors Which brings us to The sort of next steps and this is the call to whoever is Interested and either thinks we're on the wrong path or has a different idea or or wants to actually build this What comes next? Building it in Drupal 8 is a Very important next step. I mean we we would want to start like Breaking this down into some specific content types, but there's not too many that are implied, you know under that under that diagram We would have to figure out how to agree that it's actually done because that's one of the problems when you get a community of people Or even a small group of people deciding to add new functionality to a website It will just keep accruing functionality until the sun burns out But I think that's also doable the next step is actually capturing the install profile again with with the improvements that have gone in in the Cmi work that isn't as difficult as it was, but there's still some coding one big Remaining technical hurdle for this entire concept of shipping some sort of usable thing in core is that some kind of Not dummy content, but pre-populated content at least in some areas whether it's you know Blocks that you know populate with you know are populated with certain messages or you know a first post pages Just so that the whole thing isn't just an empty husk when it's first installed, you know, those are important because people are much Many people are much more comfortable editing something existing than the white sheet of paper and we're giving a white sheet of paper up till now and we're Making something that is trying to be that has that demo content. I know it's a technical Thing to figure out, but it's pretty crucial. Yeah, like we're filling this vision Yeah, that's one of the things that core doesn't currently do, you know, there isn't a clean easy mechanism for like Just saying here's six nodes that this install profile should just have once it's installed It's doable, but that's another one of those places where some custom code would probably have to be written to set those up Rolling a release and putting it on Drupal.org is a standalone project and then submitting a patch to actually get it included in core if it gets enough nods and and seals of approval Compared to what whatever what was being talked about several years ago in terms of like making Drupal Something that you could actually install and start using for a particular use case, you know the the work that was On the table two years ago to make that reality Was really daunting and seemed kind of insurmountable And now it's actually super manageable Which is cool Which is like sort of the I think the dual message of this session one. It's that out of box D8 does a ridiculously huge amount of stuff to We could actually start shipping actual use case tailored install profiles As a part of Drupal or even as a very very lightweight like adjunct download for you know starter purposes This is doable It also just the only real shift is we have to start thinking about those use cases Rather than just thinking about them as showing off specific features Because now we've got a list of core features that is too big simply just to Show off we could build a whole lot of stuff now It's time to think about what the big stack of Legos the bigger stack of Legos now Kind of gives us the responsibility To help people figure something out there with with with an example And this is that example Presenting to you You all are a quiet Yeah, we do actually Roy has actually been talking about Some interesting like philosophical questions about what Drupal has the potential to serve I mean do you want to Because not to put you on the spot right? No, but because we're still We're also that platform we've discussed So even this snowman use case is pretty much generalized right a group is a given project with some jobs that revolver on that project But snowman fits in the us or we use case We're a group. We're once again something done. There's also the I I have a portfolio. I am a writer. I am a blogger. I am a freelance Something and I need my own presence on the internet that could be well be another example of Okay, so you're a freelance photographer that would give us the A way to showcase how we could build a portfolio in in core and Secondary show off some of the new file and media handling Since we've actually got images and we can crop them and scale them That's been it's amazing. Yeah, that's visit since the dark ages of Drupal seven and the world use case I've always Kind of envisioned as that's Drupal itself as in world domination is a Running theme in in what Drupal wants to achieve I'm on and secondary I would see that the audience there as Engineers developers, so that would be a showcase for hooking up Drupal with all the other parts of the internet, so I'm not An expert on those parts of core, but I would imagine that Drupal 10 years ago was pretty hip with having in RSS feet Oh, yeah cutting edge And also comment. Yes. So what's what's in there? now that Hooks Drupal up to the rest of the internet. I mean like the fact that the services. Yeah, the the services layers that are there in Drupal 8 There's a lot of interesting stuff going on in terms of like web apps or lightweight websites that can pull in content from a Drupal 8 website now Just based on the fact that Drupal 8 can expose its content and data as Like raw jason and stuff That's cool Probably bardic It's got a bunch of regions We can color it We can enable color margins. It's responsive, you know, I mean it and That is admittedly one of the things that is Let's say a unique challenge about building something just with Drupal core, you know the the ability to actually make Any kind of visual design decisions about what it's going to look like is very limited because we can't put custom css in there You know template overrides and stuff like that aren't really, you know on the table So that's that's a challenge, but I Will say bardic actually has a ridiculous number of ways that you can Shuffle and squeeze content into its various regions and the fact that it's responsive out of the box now That that's that's cool. I'm you know Because people test drive Drupal core Maybe we should tell people Well, then we're then we're back to that whole small core thing I mean I made the case I made that case for years, but at the end of the day Moving the needle on People downloading Drupal and trying to figure out what it does is not going to happen Unless we just make it absolutely utterly unusable like it will punish you If you attempt to download bare Drupal perhaps hurt your hard drive people are going to do it Right. I mean it it's definitely more doable now than it was two years ago, and I'm I am deeply sympathetic to the you know Why not just use a distro that pulls in you know contrib tools and stuff like that So so you're talking about install profiles Let's say something like a piece of functionality like an image gallery Would be a very logical thing to have Drupal do um, is that Is that uh, um necessarily an install profile or is that a module that packages together A content type and a bunch of configuration like where's the line between that's an actual profiles and modules that bundle stuff? Well Yeah, so that's a good question like If we were attempting to build a large-scale project Breaking up any of that kind of functionality into discrete modules that were responsible for all of that Stuff would probably make a lot of sense um I think That's something that I think we're going to have to figure out again um with the rise of cmi in Drupal 8 because So who here has been nerding out and like talking to other people about the config management initiative Okay, roughly the percentage I expected excellent um, basically Historically in Drupal tons of configuration information about oh, I put a view here. I've got content types There these are the five roles that users can have stuff like that Um, those get stored in the database Mixed in to some extent with the content like nodes and comments and stuff like that teasing those things apart and figuring out how to capture all of the clicking around and configuring that you did and Reuse that has been tough modules like features Lived in that space for a long time and still do in the d7 world um In d8, there's a whole new subsystem whose entire job is to make sure that that stuff lives in Easier to distribute and easier to deploy configuration files rather than being mixed in with the database so Some of the things that drove like the proliferation of having a site broken up into like 35 different feature modules that were You know, this one is where we exported all these views and content types And this one is where we exported all these settings and stuff like that um If features module isn't as necessary just for maintaining sanity I'm not sure what I'm not sure where those lines will end up being drawn Hopefully The snowman install profile concept would stay small enough that it wouldn't be like hitting that. Oh my god There's 300 views. How do I keep track of them point where that usually makes sense? Was that dodgy enough? Oh, yeah So i'm going to summarize this a little bit differently Two years ago or two and a half years ago to plakon chicago You got up and said this is what we need to do for core to be able to pull this off And core development team said challenge accepted Yeah, I they didn't actually literally say that at the time. They just went and started coding and I didn't notice It was shocking when it happened so I'm going to put it as as a core developer. I'm going to put it back on you and say Yes, please your turn Indeed will you say challenge accepted back because this needs You know the way to deal with you know, what's the exact scope? This needs a clear product owner to to drive it I nominate jeff eaton This has been the story of my week hasn't it? To do So challenge accepted jeff challenge accepted excellent Hey, that was an ominous round of applause Okay, so uh, this is a kind of idea our kind of snowman replacement of snowman or you could say like It might be a snowman sit in there So basically up to Drupal 7 we were selling some presets now Drupal 8 is something like You know like a thing that has a lot of things you can do a lot of preset So snowman can be one of the preset So imagine a world you just go there Drupal dot r click download Drupal you go to a nice page Where it is placed Download with preset and select a drop down snowman and it automatically checkbox all the You know like a core module that need for snowman plus additional snowman stuff that can be either your views or any Config schema stuff or it can be a custom module or it can be custom theme that comes with snowman. I mean it's a snowman specific And core stuff it would be so cheesy and filled with snowflakes. It's I love it So in that case It works a lot of ways. So either you change you just select something and then Uh, I want snowman plus on top of it. I want this this this I can select it It's not just restricted to snowman closely And then I can click download and that works exactly if you do that with some mechanism That works for our github download as well. So basically instead of downloading drooper dot git We may download drooper high fun one zero zero zero one one one one one So basically each one represent each digit represent your module in an order So one means you want that module zero means you don't want that module And then dot git Is he arguing for flexibility? I I'm tempted to say challenge Because I just walked into it. You may have just no No, I mean I I think the funny thing is is like around the time that snowman the initial ideas for this were being cost around like one of the other really Really engaging ideas that everyone was excited about was the idea of like Turning the droopel dot org profiles listing page Into an actual primary spot where you would go and say, oh, yes I want a droopel that does x I am downloading that and you know, essentially De-prioritizing the idea of the core droopel download is what you do and you do stuff Is what you download and do stuff with That ran into intractable arguments around How much do you start to help, you know, what should live in core or what shouldn't live in core I was one of the people involved in many of those intractable arguments But to be honest, it doesn't matter right like it doesn't matter because if you give a option to say that like It's again, it may lead into a component based system Even you have a lot of things in core if you can choose what you want It can still with stay in core that the problem is small core I mean, I don't understand the whole concept But I feel like the problem with small core is it's hard to maintain and keep Tracking and up to date with all the other components and make sure that you do a release It then it might be end up with a nightmare You cannot do a stable release because all others Plugins are in a different status or has some problem So it still can be a big core But if you if like if we can make it like a component based like okay, I'm not going to do anything with web services So I don't want that It's a part of core. I don't want that. I don't want aggregator Now what that mean you wouldn't actually get it in the download or just wouldn't be turned up I don't know if this is a target that can make that Distinction or decision and and I think that that's got a scene anything That's got a lot of infrastructure Questions that would go along with it Mostly I'm saying in a developer perspective, but if you are not not technical person, that's where the preset come So they see okay There is a separate project page for snowman where they can see all the fancy stuff It can do and then when they click download We end up with the score page with the things pre-filled just for snowman. You don't need to do anything if you don't know Be careful. Don't touch anything. Just click download I mean I I Like that It also requires I mean it's Another extremely long bullet list of a lot of infrastructure pieces that need to be in place for that to be An actual implementable reality so I think it's a vision that I think would be cool for us to be building towards But like in the same way that you know two years ago Making a thing that listed things with core was Yeah, yeah me at the end of the day we all same family, right? Community we use php. We've got to deal with all those arrays. That's a very big family. I don't want to go that extent as a dribble Thank you. Thank you So this might be a stupid question Could you use basically drush make files? And therefore not actually put stuff in core just have a make file list all the stuff you want to use They don't want it in core No, nobody actually wants to see this thing get into core. That's the message i'm saying All you're putting in core is a file that says go get this Make file from somewhere and then that make file has list of all the modules and themes and you could use any theme you like Well, I mean what is this drush? I mean this is the audience You don't actually need drush. You just need the bit that passes the make file Well, I mean like that's how install profiles on d.o. Actually get rolled You know you you don't actually check in copies of all the contrib modules you had So isn't that a moving part in that proposal? Well, it could achieve that because then you just have you know base install Profile one profile two or do you want to call them demo? Maybe small small Skinny core um no, I mean I I think I I like all of those ideas. Um I think one of the biggest challenges to overcome is first getting past the vision of a generic Drupal Being the best, you know foot to put forward with Drupal and I mean it took a lot of Back and forth you would have arguments over, you know if you're using contrib and there's choices Which one makes the cut and all that kind of stuff that would happen But it just means that you're not completely limited to what's in core you can add in a theme Or and that that theme might be a demo theme that you know is specific to your use case Um, the reason I say is we work for a media company. We have 30 other sites. We actually do this basically in Drupal 7 our Distro is one file, but aren't those scenarios the one that's that come into playing after two weeks or after After two weeks of using a snowman you realize. Oh, I much rather have Another look. I'd like to add this Yeah, but that need uh I'm I'm thinking about Dumb normal people who probably run into this not before But all it's doing is downloading a bunch of stuff for you So you can yeah Yeah, yeah I think yeah, I think these to me sound like framework platform Yeah, and considerations and a lot of which is the idea of this is also to be without using that custom code Yeah, I mean and and ultimately that kind of an approach to functionality that people would experience as a one-step process rather than a multi-step process would require Um a lot of buy-in and a lot of infrastructure work Which isn't me saying it won't happen or it shouldn't happen But also like as part of an iterative learning process This is something we actually can build now and continue learning In building towards something like that, you know, this is At least one of the baby steps towards that kind of a vision that we know We still have to learn these lessons even if we build the support infrastructure for that So where do I sign up? There is an additional slide Um, yeah There isn't I'm continuing the questions. Okay. Yeah, I will read the slide, but if you are interested in this kind of thing Friday Friday Friday Friday So say for instance, we're choosing to install profiles What's our process for beginning to decide what goes into heated arguments I mean and and again, this is the biggest challenge like when the idea of so what should Drupal's Show, you know showcase install profile do It I mean every time that conversation has come up in an open discussion It's ground to a halt after about the stuff it won't yeah that we also could yeah It turns into months of those kinds of questions and I honestly don't know of a way to resolve that Other than saying okay. No this Yeah, so that's that's the core principle Thanks, Larry Now I'm to blame What do you think I was doing well played well played vest The best loves you too But yeah, that that is why we need to have a product owner that trees Will support say okay What goes in whether whether we put this field on the content type that's the product owner's job. You just decide Um, I'm going to disagree with a couple of the other people. I yes, I want this in core I want this in core on its own straight up For a very simple technical reason it serves as as a actual test of our install profile system Precisely, there's probably a lot of stuff right now. We think we can do with install profiles now But we actually can't oh a lot of people try to build a real website with it and we'll see what gets broken Exactly, you know that that can find where do or where does the install profile process still suck And we've still got a little bit of time left to fix those things You know Does the that part of cmi that we think works not actually is it actually still require a custom php to work If so, let's fix that now. I'm pretty sure that Drupal truth just tweeted at me that cmi was fully functional I'm sure it is but could it still be less buggy it could be functional or yes Um So also, uh, feel free to fill out the uh The ratings on this session If you disagree with me feel free to rate me a zero. I won't be hurt. It just means that I'll cry I really agree with larry's last point there. Um, I'm just wondering if uh, if there are all of these, uh, websites that are live um, it just it seems a bit strange if they all have this sort of Other profile hidden away in their code base that isn't being used. I mean, it's it's not something that that really matters But it just uh, I don't know. I don't feel easy about it for some reason Yeah, well, that's I mean, would it make sense to delete it from your install if you're not going to use it But then what happens when you when you update? Yeah, I mean it I think it would probably we were looking at like What's the raw weight of that stuff that would make this up? But I mean, yeah, I don't think it wouldn't have any performance problem because You know, it just wouldn't be used But ideally it also would be a very small quantity of actual code It would mostly just be cmi export, you know Config exported cmi config data That would be the bulk of what's sitting there in a profile directory and technically we've got multiple profiles today just one of them Doesn't really do a lot and one of them purposefully does absolutely nothing so I just had one other point, um, I I think we might be moving to dribble seven for dribble.org Hopefully quite soon. Yeah, totally. Um, so is there a greater opportunity to improve Profile stuff on dribble.org once that has happened um, I can't speak for the software infrastructure teams, um, but like Getting the d7 upgrade out the door is one of their biggest priorities because It's a blocker for working on lots of other kinds of infrastructure and deployment upgrades and stuff like that. So like that that's They have not said yes, we would like to do that thing that was just described, but you know It it's impossible until the d7 upgrade goes out just because that's a huge blocker This evening nine o'clock at the second floor of the korean theater will be music Drinks and testing dribble seven issue cure tools Doesn't that sound like a fantastic evening? Thank you Thank you for joining us and uh, have a good one