 Okay, welcome everybody. Thank you for coming to the Drupal.org panel and Q&A. We're going to be talking about updates to Drupal.org's infrastructure, the way we manage the features for developers, the things we do to support the adoption journey in the community, and some thoughts for the future. And so we'll have updates for basically everything that's happened since Vienna and some interesting developments that you may or may not have seen in the what's new on Drupal.org blogs. As we get started, I do just want to familiarize you with some of the faces that you might see up here or recognizing other places in terms of the DIA engineering team. So I'm Tim Lennon, has been on Drupal.org, we have Neil Drum, Ryan Aslet, Brendan Blaine, I'm going in picture order. Daniel Girish has been sponsored to work with us and is remote in India and helping us out on quite a number of things. And then we have Narayan Newton from Tag1 Consulting who helps with infrastructure and from the infrastructure working group, Michael Hess. So these are the names and faces of the people who keep Drupal.org online, keep everything up and running, and we really appreciate your feedback and your support, which is why the first part of this conversation is always a thank you. So thank you to the members and community volunteers who support what we do. You directly fund our work, you fund our salaries, you make it possible for us to do any of the things that we do to serve the community. We also want to thank all of the organizations, some of whom you may work for, that sponsor this work, sponsor the conference, and make it possible for us to do what we do. Without that kind of support, it just, we simply couldn't do what we do. And I'll talk about some of the financial constraints that we have on our infrastructural side as part of this presentation. So in terms of what we're talking about today, this is the sneak preview that I gave during the public board meeting. We're going to talk about six broad categories of information, how we support adoption, how we support community, how we promote Drupal, some feature updates since Vienna, Drupal CI, the testing infrastructure for the Drupal project, and some infrastructure and security. And then following that, we'll talk about some plans for the future, both in features and in some proposed core initiatives and initiatives that you might have heard in the Dries note that are going to be a collaboration between the association and other volunteer parties, core committers, things like that. So firstly, to talk about supporting adoption, again, I mentioned this in the public board meeting, but we wanted to do an analysis of who the anonymous visitors to Drupal.org are to better understand our community, especially those people who aren't the engaged users yet, who haven't yet registered accounts, and who we still want to bring in and convert to the community. Anonymous traffic is 93% of our total sessions. So it's actually, arguably, the largest underserved part of our community. And so what we did is we used some audience insight tools wrapped in our own implementation of do not track to prevent anybody from any of these organizations from maybe playing fast and loose with the rules there, and targeted only at the anonymous traffic to learn things about the job functions of the visitors to Drupal. These job functions are about the role that an individual plays based on the data that's aggregated about them within an organization. It's not the industry they're in, but it's the role that their job might be associated with. So engineering information technology that could be in a services company, but that could be an immediate company, or that could be in a marketing company, or a nonprofit. It could be in any of those industry areas, but this is the role that the person engages in. And so we found out a lot about, you know, we know that our dominant users are in engineering and IT. That's not surprising. You can see that really half the amount of the technical users go to the front page is go to the rest of the site. They're usually involved in the issue queues working on projects doing contribution. But the front page traffic is mostly made up of these roles in business development, entrepreneurship, design, marketing, communications, and these other roles that we haven't historically served very well. So we've done this as sort of an initial analysis, and it's been part of the work that we did to redesign the front page of Drupal.org. I'll get into that a little bit further. In terms of supporting the community, and a lot of those efforts, there's a number of different things that we did. One is, we asked the question, how can we amplify the voice of the community? And in particular, the community working group, because with all of the conversations going on about governance and community issues, that voice needs to be heard. And whether or not you're familiar with this, in the past, most communications from the CWG came out in the form of like a Google Doc that got tweeted by their account. It didn't have an official home somewhere in Drupal.org. So this wasn't a technical challenge. It wasn't a big deal, but what we did is we revamped the community home page on Drupal.org, and in particular, created a new blog, gave access to the CWG so that there's now a canonical source for those communications, and so they can be included in Drupal Planet and all of those sorts of things, which will hopefully help foster the engagement and extend the reach of their communication. We've also been working on improving membership tools on Drupal.org. This is still a work in progress, but hopefully most of the people in this room are members. One thing you'll see, it's now actually hosted directly on D.0. The association website is sort of semi-deprecated, we're mostly using it just for elections at this point. And we have a new individual member directory with new filters so that you can learn about the community and the new members of our community. At the time of this snapshot, we've had about 2,400 individual members who are listed in the directory. And while the work isn't complete, we've got the donation system fully migrated over and the member system is sort of in beta, I would say. We're also adding some long-standing feature requests like your donation history and things like that that people often want to go back and take a look at. Anything to add on this? Okay. And then the other community support issue that was really important to us is how can we help people signing up for new accounts feel safe and included? As you, well, many of you here probably have had accounts for a long time, so you may not even remember. But the last time the gender field, for example, on user profiles was updated was something like eight years ago, nine years ago. And the state of the community conversation around that at the time was perhaps reasonable for the era, but not particularly good. And the state it was left in was, like frankly, offensive to some people in the community, it was a barrier to people wanting to participate with us. And so we've taken a first step, not a final step to try and make that better. Which is to say, on the left here, we started using demographic self-identification options, of course, a variety of axes. This is based on the big eight or big 10 principles that you might have heard of, if you've looked into this before. And these are what we actually use in our speaker submission forms. So anyone who spoke, you've seen these options to help identify the diversity of our speakers. And so we've put that onto the new account registration process to remove our updated form, our offensive previous form. But we want to take future steps. And so I've been having some conversations here with Nikki and Tara and other people involved in DD&I about the Open Demographics Initiative, which is a GitHub repository where people are collaborating on identity and ways to do self-identification and things like that. And it's mostly been standing up in hallways and talking about this. But I think the consensus we're coming to is turning that into an API that could be consumed in a module would help us be in a situation where we continue to stay up to date. We continue to stay up to date as these things change as missing identifications get added to the system. Instead of deploying a change and finding out five years from now, that we're behind again. So hopefully that will be an effort that proceeds well. In terms of promoting Drupal, you'll notice that the thrust of these slides gets increasing with technical. But in terms of promoting Drupal, we've done a few things since Vienna. We've launched a new industry page, a solution page about the nonprofit industry. There's some interesting case studies there that just in passing and some more that might be coming soon. Things around some pretty significant couple projects actually that I just talked to some people here in the conference that might be of interest. And we partnered with form one to produce the content for this page. And if you're curious about why we have the industry content that we do, we currently have health care, government, nonprofit, what are the others? We're working on e-commerce and travel education. Yeah, it's all based on it's all based on the partners that we found to our domain experts who can help us to come up with the information that helps an evaluator in that industry. So since we are a very small team, we don't have those domain experts or those content curation experts. I'm more or less an engineer, but I find myself writing landing page content. So we need to work with these people to get media and publishing is the other one that currently exists. So in addition to that, and everybody's aware of this because you're here and you've been using the website for quite a while. Around Vienna, we launched a new Drupalcon brand. And this is a unified brand for the event. So we now have a central brand identity for Drupalcon itself, with elements that become specific to the cities and locations that we go to. So we retain both local color, but also don't have to reimplement the wheel for every event, which with the level of resources we have is just not sustainable. And I think that's been very successful. And just just from like, you know, the production and print graphics and that amazing like window detail and all that, I think it's been phenomenal in this first iteration. And of course, everyone has seen by now the Drupal.org redesign. See if I can get this to do its thing. So I've been joking about hashtag lava lamp because of the cool fluid blobby shapes that appear in the background of the header and down in these persona channels that we are funneling our traffic through. We're trying to reflect the kind of like fluidity and adaptability of Drupal. It actually is intended to do the Drupal is the drop. It's fluid, it's adaptable, it's customizable. And we're really happy with the new design in the new direction and also the ability to have a more complex top level navigation and IA. And this is the point where I embarrass 611 yet again by saying thank you to them for helping us to put that together. They have been partnering with us very closely to produce not only the Drupal con branding but also the Drupal that are redesigned and a number of other initiatives that have been really important to us. So we really appreciate their help. Lastly from a sort of design or promotion point of view, we recently did a overhaul of the way the hosting market place works. It's not quite easy to tell everything that you can do just from the screenshot. Neil worked on this quite a bit. And more or less, we have a better combination of filters and features so that people can find. I want something located here. I want something specializing in my industry, etc, etc, etc, like people actually find a hosting partner that's not just a generic choice, but one that will fit their particular needs. Okay, now we're going to get technical. And by technical, I mean kind of legal, which is the worst kind of technical. So there's been a lot of angst in certain parts of the community, the parts that care about what licenses mean and whether things are compatible. And can I use this library with that library and pull it into my module and that sort of thing in a number of important ways. And we've communicated this out already. It was done shortly after Vienna. But just to keep getting the word out, I'm going to go through what some of these important licensing questions were. So one of them was, can I commit GPL compatible code to a GPL repository? That's not already GPL. That seems like a simple one and it is more or less a simple one. And so we can now say yes, you can do that. It's just that once it's redistributed, it will be redistributed under GPL to or later as the standard. So as long as it's the original license was was GPL compatible, yes, it can be used. The next question was, can you use GPL incompatible noncode assets in Drupal projects? So this is this might be icon sponsor, things like that. And we clarify that you absolutely can package and distribute GPL code in aggregate. This is the language from the GPL. And it's in section three two, if you're curious to read more, as long as you have the rights to whatever that those assets are. And it does not affect the license of the code itself. The next question was, can a Drupal project have a dependency on non GPL code? So this is the new evolution of licensing concerns in a world of composer and dependency resolution, right? Because it's always been true that you can use GPL code with non GPL code in a personal use case. But the GPL stipulates that you can't then distribute it without either having compatible licenses, or one of them taking over something that or sometimes not at all. So what the clarification here, the important one is Drupal.org hosted projects can depend on or link to GPL incompatible code, maybe that's via composer, or the historical example is modules that like say, hey, you're going to have to go download a separate library before you can use this before we had automated dependency management. So that's absolutely allowed. It's just that we can't host those in as part of the module, we can't pull those dependencies in as part of module and distribute that together from Drupal.org. But you can run composer and get and get those things. And that's all fine. The other one that we got, which seems like it should be obvious, but wasn't written down anywhere, and hasn't been written down anywhere in most open source projects. And actually, to make a digression, when we started talking to lawyers about some of these questions, we discovered an interesting thing, which was that a lot of legal teams and I'm not a lawyer, so this is not legal advice. But we discovered that a lot of legal teams, who are at all familiar with open source, said when we would come to them with a question, they would say, well, you're one of the biggest open source projects. So what do you do? Like, when there's not prior case precedent, they're looking for industry standard practices across some of the largest examples, and we happen to be one of the largest projects. But anyway, one thing we don't explicitly affirm is if you're building a client site, and giving it to them, is that distribution? So can you start using those incompatible assets? And then the answer to that is, yes, of course you can, or else we couldn't have an industry around building Drupal sites. So we explicitly affirm that if you're a provider assembling a code base under contract, that doesn't fall into the restrictive terms of distribution, that's you acting directly as their agents to provide them the project, your proxy, not distributing a product. That's different than if you were selling en masse, like pre made images, that could be a different situation that had non GPL code. The last question, which is still unresolved that some people have, and this is especially relevant to anybody who works with distributions is should we allow some GPL three only projects on Drupal.org? And right now that's currently unresolved. The reason this is relevant to distributions is because distributions are packaged together, typically, and distributed from d.o. Or they're just a composer manifests, in which case it's a little bit less of a concern. But this is this is still an outstanding question. Drees has to weigh on it, more lawyers have to weigh in on it. And if we do do that, we need to make very clear that we're telling anyone who comes to evaluate a distribution or other project, which license is relevant for for some very large organizations, they won't touch GPL three, the GPL two is fine. And that has to do with something that they unfairly characterize, I think, as a patent poison clause in three verses two. So that's the kind of only unresolved question, but we're very happy that some of those other questions have been resolved, because if you've ever been waiting on something in the whitelist, that licensing whitelist, those things has just suddenly started getting cleared out now that we've had these questions answered. So talking about features on Drupal.org, that we had a few different updates since Vienna, more than these, but here are some of the highlights. One is related to Composer. You know, you've heard a lot of people say Composer's too hard to use, and it's too hard to figure out what the right way to do something is. So one thing we've done, and it's just the first step is to make it easier. It's on any individual release node, the composer instructions are included, along with a link to the general instructions about the proper ways to use Composer for managing dependencies. We want to start feeding that up to the project page level, so that's a little bit, you know, you don't have to dig to find how to get to the composer instructions, but at the very least that's a start and that's something that we can bubble up. The other thing is there's a new Composer initiative being thrown around here at DrupalCon, and that was part of the announcement in the Drees node, to standardize the way that Core uses Composer and to create a set of best practices. And I think once those are defined standardized, Core is updated, then we'll be better able to promote the sort of one true way or the best recommended way, instead of having to deal with a diversity of possible options. If you've looked at the current documentation, there's sometimes four or five different ways that someone might recommend doing it. Another really good quality of life feature, and I think Brendan was working on this one for us, was just fixing our friendly URLs for issues. It's not all node number anymore, for God's sake. So it's a small thing, but it's helpful and it's useful for search and for finding that particular issue, finding a particular issue you might be looking for. There's also some shortcuts that I just want to get the word out about, because I think they'd be useful for people at the sprints on Friday, trying to pass things around. If you have a node number, you can just put it in the search box and go straight there. You can also just do drupal.org slash i slash node, and we go straight there. So you can, if you're shouting across the table on Friday, it'll be a little bit easier than what these URLs were before. And the old ones still work, of course. Neil put together this one, the in-context issue tag explanations. Do you want to talk about it a little bit, maybe? Sure, yeah. These issue tags pretty extensively in drupal, especially in the course you do, and there's a lot of meaning in these tags that's not upfront, and there used to be a documentation page with all these explanations varied. Now it's right on the issue page. You can figure out who needs to review the patch next, or what needs to happen next, and have links off to more resources. And I think there's a lot of lessons that new initiatives or contrib projects that have a lot of issues to manage, or whatever, could learn from some of the ways that issue tags were used with previous initiatives. And being able, it's more or less tools for an additional workflow steps or additional approval steps. If you need usability review, accessibility review, whatever the case may be. You may have also noticed that along with redesign there's a new top-level navigation for drupal.org. We finally sort of broke down and said, you know, we have a lot of different audiences who have different needs, and we don't want to bury things that you have to get to a landing page before you get a whole set of other menus and such. But we do want to kind of reorganize into the categories of things that we think an individual user flow is probably all within this area, or this area, or this area. And that's something that'll probably continue to evolve, but it was just a nice change that came along with the new design. And now we're going to talk about Drupal CI a bit. So I'm probably going to refer to Ryan. But first, you want to talk about this? Yeah, Amazon Web Services switched over the first, second building back in October. For a while, we had to do a lot of things in Drupal CI. We kind of maximized the spare CP time where we would start in six, because we dynamically scale our CI instances. And it's something that the trip patch might take four minutes to run, and then we don't have 56 minutes of espositing around. And so we had a lot of strategies to try and not roll money away, spending them in the instances, basically. But Amazon went for a second building in October, which made it so that we could much easier not have to worry so much about. We could just follow scale exactly for our demands. So since that change is made, let's do something like a set of what we have. So there's a spread going on. People submit a lot of patches that they're all getting tested once. So it's pretty nice. Yeah, there were some pickups getting it together in the first place, and there's some things that we still need to figure out, right? Because we were managing our total budget. This is when I'm talking about kind of association budget, right? We averaged something like $4,000 a month just on the test bots to pay for testing for core. And we would manage that before by saying, okay, we'll put a cap on the number of test bots. We'll try and reuse test bot cycles until we can get them done and just kind of rate limit the total number of testing. We don't really, we haven't had to do that in quite the same way since we moved to per second. We don't have wasted cycles. But we're noticing, and Ryan may not know this yet because I just checked it this morning, we're tracking to have a $6,500 month for April because of the con and things like that. Yeah, so so there's still work to do to balance the easy availability and the speed of the testing infrastructure so that developer velocity is high with the budget that we have to provide that service. And by the way, if anyone has super strong connections in Google Cloud, AWS, Azure, and anyone wants to donate instances, that would be phenomenal. So, and we're talking to some people on that front. There's also a really cool new feature in Google CI that I'll let Ryan talk through. Yeah, Google CI in classical, you've been able to configure along Google Adobe for your projects to say, I want to be able to test with this type of environment, the virtual PHP, which version of the database, and it was, sorry, it was expensive, but it was a lot of options, but there wasn't a lot of control for maintainers and for maintainers to manage the build. And what we've done is this Google CI in the yellow file was always with driving Google CI. But now what we've done is expose it so you can put one in your projects and add additional steps. So, these keys here correspond to a code plug-in inside of, like, Michael, in Centroid and CI itself, so like CSS and ESLint, ESLint, PHP, CSS. You might not want ESLint to manage your project, you might not have any JavaScript at all, so why not the steps, you could just take that out if you want to. There's a couple of steps here that I haven't got on here, in need of better documentation, which is run arbitrary shell commands. So, you can run whatever you want, basically, it's like maybe you're testing something that requires you to generate some big keys to either testing API or maybe you want to experiment with something and just want to be de-dested. So, any commands that you run, that's why it's a little bit save the output of an artifact that will show up on database. So, it allows maintainers to do a lot more than they can do before. And, you know, the first value, this is that we're submitting on the core, we're submitting a batch of cores to break up the testing cycle. So, it learns from the test, and that's just the first that it runs the current test, and then it will fiddle fast. So, if you have a lot of code, you just go past and you want to test it for an hour to find out that, like, well, if you want to do it now. So, but, you know, there's more documentation to come on that. Yeah, and actually, if you're a power user of Drupal CI and want to help out with docs, you probably do so during sprints. Let's see. So, then I want to talk about a little bit about infrastructure and security, because that's been top of mind in the last couple of weeks, both how Drupal Auto Engage is up time and security, and how we support things like the security team's efforts. But, I do want to talk a little bit just about maintenance and things that are probably on everyone's radar. For a long time, obviously, PCI has been a thing. It's not new. The DA participates in commerce in three ways, memberships and donations, events and jobs. And so, we've been working through this year to try and reduce PCI scope, reduce the amount of, you know, the kind of questionnaire stuff that we would have to qualify for, and just reduce the overhead of managing that for the e-commerce side. GDPR is a really big deal. And something that's still, I think it's becoming better understood, but it still has some open questions when it comes to how that's going to affect, especially open source projects. So just to give you the general overview, it's the general data protection regulation. It's a European regulation. And it applies not just to European hosted data providers or data data controllers or processors, but to anyone who does business in Europe. And that concludes us. We have a global community. And it, the regulations take effect or are relevant whenever there is personally identifiable information about what they call a data subject, an individual. Or in some cases, whether there's pseudonymous identifiers being used. That pseudonymous identifier means like a user ID or username, something that stands in for another identifier. And the most basic principles is that you need to have explicit opt-in to communication and data collection. You need to have a right to be forgotten, sort of, in other words, account deletion process. You need to have some recourse for data portability. It actually doesn't specify that has to be automated at this point, but you do need some recourse for data portability. And the privacy policies in terms of service need to reflect what's collected, why, and how to contact an organization about that data. So we've been doing a lot of prep work, having some legal conversations. We've got blog posts prepped. Liz, on the Drupal Association staff, has actually been spearheading a lot of this. And you may have talked to her at the booth throughout this week. And so it's something, more or less, what I'm here to say is that we've got our eye on it. Fortunately, our community is so privacy and security conscious that a lot of the steps that they are now mandating as regulation are steps that we've already been doing in terms of affirmative consent, in terms of ability to lead accounts. Not so much data portability. That's a different question. So, and you know, I had a conversation actually with typo three maintainers a couple months ago about GDPR and how they were thinking about it as an open source project. Because the big question in their minds was, how does the right to be forgotten apply to get history? That's a scary question. And the truth is, probably what it's going to mean is just that people will have to implement a get agreement, which we already have, that says, you agree that this won't be forgotten. Can't be. It's immutable. And that's allowed under those regulations. It's just needs to be very explicit. And so that's, that's, that's the question for open source in general that I think is catching some people by surprise. So let's talk about security. Michael, do you want to talk about this one a little bit or show? Sure. So let me just first say, obviously, we had some, a little bit of instability issues during the security release. We had some 500 errors. Could have been much worse. It was actually not, not so bad. But this was my favorite tweet during that period. It's, it's, it's not bad. Certainly can't hack your board if it's down. So, yeah, why don't you? So for the, for the release of 2018-02, we did some things that were interesting on the infrastructure side. We're an open source product. And there is a delay between pushing the commits into Drupal and there are actually deep packages available for people to download and publishing all the bits and checking out the checkboxes. So we shut down and get publicly all of that out of there. So we, we shut down and get publicly all that happened, which went very well with the exception of people really like the F5 security page. We need to see if the update had been posted for two hours which made it hard to post an update. We've come up with some creative caching solutions in the future to prevent that from happening again. And so we, you know, we learned from our mistakes. We, we, it was interesting that I was in the middle of publishing that and the retrospective meeting got scheduled and then we'll publish again. Thank you, Jim. And so part of the retrospective was to figure out where, where things went wrong, what the queries were, where things were causing things to happen. Should we hopefully have to, but should we just get we won't run into that problem again. As far as the shutting down of the services, it seemed to work fairly well. We forgot to shut down one service, which I'll have noted next time. Using Perimeter X, we're going to talk much about what Perimeter X does. Yeah. Actually, Narayan, do you want to talk about some of the like malicious traffic and activity that was also correlated with the release window? Yeah. So we've been working on integrating something like Perimeter X as basically a border filter for traffic control and which is plot traffic for just misconfigured plot traffic, which is a lot of what we base paperwork as an API, but the API that you look at the line is sometimes incomplete. So people tend to scrape too low for information and they don't always do that and that's what they do and sometimes they actually cause infrastructure issues with the scraping because they try to do it like concurrency of three and not waiting all sorts of things you're not really supposed to do. Also, there are people that do it maliciously. So there are lots of things you can do to a group of websites, both for the pre-excessive load on a database server and during the security rollout there are people that were having fun doing that. So we were having issues with the security page and the fact that there was query on security page that Neo actually fits to very nicely also looking at some caching around there to bring that back up. But the other issue we're facing is without looking at the graphs afterwards about 50% of our traffic was malicious trying to cause issues on the site. So we had a kind of suspected that there might be issues during the security release so we had the filter being tested in some chosen net blocks that were known to be box specific so that we could tune it in so it wouldn't catch everyone who was actually at moving through and we had it ready to slide into place in case there are issues. So when the site basically came back is when that caching was put in place and we put in the filter and the filter for the duration of security release was basically walking up the features of our traffic. Yeah, it's pretty intense. Simultaneously with that a few folks in the community including myself and Neil and some maintainers of the Drupal Slack admin channel and things I think by 1230 I released add like 15,000 emails in my inbox that came from a sort of spam attack to try and fill it up and just disrupt our operations I suppose. So this is the kind of challenge that we tend to face. There are various kinds of scraper or DDoS either not without malintention or deliberate that happened frankly maybe once every two months we have some some kind of event. We've had one during the comm that Neil has been dealing with the last couple of days and the the positive side is for the most part that's been completely transparent to our community for the past several years apart from the the glitch during the the actual security release which was probably the most intensive instance. The fact that we've had this every two months it's not really been anything that anyone's noticed and we're really proud of that fact and it's due to the work of this team so. You heard of these? Meltdown Inspector Narayan can you talk a little bit about how that impacts our infrastructure in particular? Yeah, so we have a a melting pot the hardware they all bring something different so after this in the beginning is kind of this was somewhat simple of doing it for people to work out your interests on the other points as well and so we're kind of just going through and figuring out how the the initial taxes might impact performance and figuring out where the performance is going to be what's it going to be on the kernel path for the microchip path so the initial caching went out from that melting inspector pretty quickly and then Red Hat immediately rolled it back and then the initial caching went out from VMware pretty quickly VMware immediately rolled it back so it's been the interesting path but at this point caching is going well obviously all the public all the public nodes were immediately padding no matter what the had to be and then the path at this point is the kernel of the path is distributed out and applied but the microchip has to be are working on getting completely out because we have a lot of back-end servers that have interesting processors and some that would be the center and can you talk a little bit about what impacts that might have had on our resource overhead and performance I don't know how much well that's very honestly because the initial patch said an initial microcode patch we actually took a pretty large performance hit and then they rolled back the microcode patch and that would be a bit somewhat we're still in testing for example under high load we're still seeing probably about 10, 15, etc. we are working basically on re-arranges and resources to bring some more resources to bear on a web node because that's where we're going to see the most impact so basically climbing around that and look at what I met and everyone is doing yep cool so now I'm going to start moving into talking about the future and some other initiatives that we have moving forward and this is not an exhaustive list and it's not a list that is set in stone it's always subject to change we work as an agile engineering team with an agile roadmap process and things come up from time to time like a big security release and our and our kind of standard plans can change but here's some of what we've got in the pipeline so one is off support in particular we want this support for camps there's the variety of efforts at different kinds of chat organizations that have had concerned about identity management and understanding that hey these are actually the person in the Drupal community that you think they are Drupal Europe is coming up as an event right and that's not run on our infrastructure so they would like to be able to let you you know log into their site with Drupal.org account so it's something we're looking into we're currently basically in the middle of technical evaluation code review and security checks for some of the modules that exist on Drupal.org the next I already mentioned this and kind of we need to detail so I won't go too much but the open demographics initiative again for more affirmative identification of who you are within the community control over what you want to make public and keep private all these things it needs those two components it needs the open demographic folks and any contributors who want to help them to turn that from a basically repo of plain text into an API with JSON that we can consume and then we need to build the module to consume it if you missed the session on Tuesday about pull requests on Drupal.org you'll be really excited right now so we are planning on moving towards a new merge request system on Drupal.org you may have seen a series of blog posts that I made in December about the options we evaluated our requirements everything we had there I'm not going to read this slide but it's all summarized in these blog posts and it talks about the open source open source slash enterprise proprietary enterprise build it yourself options all of these different things and then also writes a sort of user story about the way the Drupal community collaborates and says how can we adapt the way we collaborate to the models that can be provided by these various tool sets and at the conclusion of that of that first set of posts that was not what you expected to see which is to say we looked at GitHub GitLab at Bitbucket and said okay GitHub has hard blockers and we talked to their engineering team and they weren't going to fix it GitLab has hard blockers and we talked to their engineering team but we couldn't really get in good communication and didn't seem like they were going to fix them either and Bitbucket which does not have the curb appeal of those any of those other solutions really people people never get excited about elastic projects unless they're Michael but checked all the boxes we could architect the workflow that allowed the many to one collaboration of the Drupal community using that tool set as a back end so when you read that post in December that's kind of the conclusion we came to then GitLab CEO showed up in the comment section and said call me so we've been talking with them for the past several months about clearing those blockers and they've been heavily committed to that they have shared slack channels with our team that we gave them a list of issues they fixed them they proposed support to hopefully help fund some of the migration work all this other kind of stuff because they from their point of view they were really committed to think Drupal would be a better project to show that they can be part of open source so it's not absolutely final but it's looking like this is the direction that we're going to move into and yes I'd be happy to take questions sure um I'm happy to hear that you know how much of the cultural I guess I would imagine that we're kind of going to do whatever it's going to be that we think for just a good amount of time probably going to break up with where we kind of go with but sure what I know about GitLab culture gives me a more robust I guess than you know using the bucket I mean the fact the fact that they have a CE a community edition that is open source at all in many ways makes that potentially a better fit so just just even that fact alone considering that GitHub is not open source the bucket is not open source like so I mean I think that helps I think also the way that we're going to do the implementation is going to be interesting so let me I'll try to visualize it I'm we I don't want to repeat the whole session that happened on Tuesday so Tuesday at five o'clock we had a whole session with much more detail and a lot of Q and A so I recommend that you watch that session but the rough idea is Drupal.org retains project pages retains issues retains kind of the home of any individual project but then allows creates on the back end a repository per issue in GitLab and allows you to create one to end branches that people can collaborate on for each proposed solution they might have to resolving that issue and then we get code reviewed tools inline editing all of those things exposed to the UI and we do this in a progressive way we initially just transparently change our custom get back end to the GitLab back end and nobody would really see that change just up you know hosting the repositories and then after that we'd start exposing functionality like replacing C Git with the code review tools and then introducing merge requests and these new branch models so it's something that I think is pretty exciting and there's APIs to do things to let us say okay you go into the code review side interface and start doing these things we can call back and say okay that branch was updated and maybe there's this diff and we can decide we're still negotiating how much do we pull back from the code review into the inline issue stream or how much does the issue become about the problem statement and the sort of architecture of the solution and then the merge requests about code review and specific detail so those are all details to be worked out as we keep moving forward on that but it's really exciting and this is that session I was referring to that happens Tuesday at five o'clock is recorded so if you have a lot if you have any more questions on that I'd highly recommend checking out the recording finally we are hopefully going to be collaborating with on several sort of core initiatives or initiatives that are not traditional core initiatives but overlap with some of the areas of responsibility for the DA one is the proposal from the Dries note about simplifying Drupal's documentation the notion of perhaps having an official documentation and providing the tooling for that if it's needed the question around this one right now all of these are early stages I'm not going to go into much detail but there's a content and editorial question that I think is more the harder part to solve than tools when it comes to documentation how do you make it official how do you how do you promote it in the right places is much more difficult than just how can we write documentation that looks good that that part we can solve so anyway we'll be hopefully supporting those efforts and involved in some of the architecture of that another one is the notion of better composer support and core we talked about this briefly earlier this is a precursor to things like automatic updates initiatives and site builder tool and project browser which is to say right now there's five different ways to do something in composer with Drupal and maybe half of them are the wrong way or maybe none of them are the right way and the point is core could be updated so that there is a standard best practice method and once we have a single process for using composer that works well for all the use cases then it's something we can automate then it's something we can abstract away and build tools on for the non command line using site builders another initiative and this is going to be kind of a a more complex thing it's not really a core conversation but it's sort of been called hello Drupal but it's a concept of reimagining the tri Drupal experience because we know that that's something that is friction for a lot of existing Drupal users because it feels just too much like an ad and not like a good demo experience and so thinking about trying to create a product forward here is a beautiful demo of out of the box mommy blah blah blah but while preserving unfortunately the existing try Drupal program well fortunately provides a quarter million dollars to the association it helps fund a lot of things so we want to figure how to marry those two in a way that gives a good user facing demonstration of what Drupal can do and also supports some of the revenue that lets us do these programs another one this is again not really a technical one but an editorial one is providing evaluator product comparison information like there's no one Drupal.org that says why we're better than psych or or adobe or whatever like we just so the promote Drupal initiative that you've been hearing about throughout the week is important to that is important to finding the domain experts that can help us write the content and featured in the right place on Drupal.org to get those 93 percent of anonymous users you know 20 percent of those being CXO director business development types to actually see the information they're looking for which is not the same as what our developer projects wants to see and then as I said building on something like the composer the initial better composer support initiative we can start understanding the automatic updates initiative and again there was a session about the Drupal core auto update architecture and some proposals about how to make that a secure process focusing on the balance between usability and security and the things that we need and I'd recommend taking a look at that as well if you're interested in that topic I mentioned before this work could be built on for a site builder tool or project browser and lastly well two more things we want to do a telemetry initiative we want to gather data about who's using Drupal we get we get a little bit very little bit of information coming back from the update stats because everybody phones home or just about everybody who doesn't turn it off but we could be learning more we could have this opt-in to send anonymized analytics to Drupal.org and let us know what are the real use cases what modules people care about like and provide information back to the contribution community to know where their efforts would be best focused and finally our reach on Drupal.org is massive but it is a fraction of Drupal sites right so being able to have some limited carefully curated appropriate communication about what's going on in the project within Drupal itself could be extremely valuable to reaching those people who use Drupal but don't realize that they are part of a community they're already part of a community and they don't know it and being able to say hey there's a Drupal con and hey there's this event and hey there's this important release happening all those sorts of things from somewhere in the administrative interface could be quite valuable all right so that's my long spiel I think I went longer than I intended but are there any additional questions or other topics people would like to raise maybe we should okay so go ahead um I want to thank everyone for pointing out the original but or it's a tribe to be involved in helping to lead people around and about it might work in Drupal and fixing issues in the quarter with mostly defined intimidating so those people are here in order I mean it's all ahead that really takes great deal patience and persistence I really appreciate that I know it's the most people who are here I think are attacking the people but I don't think that most of the problems we're dealing with is about or the archetype they're archetypal collisions so we really need to have people out of work be here and have people look at design thinking visibility we can't just be this the lubricates we're here right now going back to the initial slide and comments on the trend that people are involved in Drupal or something that's that's really clear that's missing you're not asking them not all of us users to enjoy it yeah and I don't know why and how that isn't happening this is something that is has been going back for I don't know this goes back to the purring issue yeah is it right was it going on looking at ways to help get people involved if you don't ask people who are on our site to join how long did you know they know that they shouldn't be doing things yeah and I think I think it's a really good point and I think so for example the hang on a second so some are your point thank you for the record uh we need to be focused on not just technical details on implementation your board or all of your stuff UX, UI, documentation, community, creation, and specifically why are we not asking about the users to join maybe not all of them but it needs to be focused on yeah I'm not saying we need to need to be focused on things now yeah so somebody needs to to be thinking about these things and I think that's absolutely true in terms of where we are and whether any steps have been taken on that front as small as they may be I think the first one was the new front page redesign and I'll say it for this reason what we discovered in our research is I don't know if you remember the old website but if you looked just above the fold at the previous home page I think you could count something like 17 or 22 calls to action just above the fold in one spot and that makes it really difficult regardless of your persona to figure out what the hell am I supposed to do next so it was very very tricky so one of the one of the key elements here was okay we're identifying three core personas that represent our core audience of existing users and the most present audience of some of these anonymous users so what we've done is I think we've created the top of the funnel but we haven't solved and this and this is largely content and usability questions we haven't solved the problem of where is that funnel going to lead those people and where we've started it but I think we're going to definitely need help to figure out how to make that a complete well there's that too yes so instead of the game that's you often end up having the contents that's just presuming that are like 22 like Google in a specific marketplace that's not necessarily or what's being done to reach out to people who are using Google and say dedication and gets those people to actually revise the content supported so the question is how are we seeking out domain experts particularly for content writing in areas where we don't have that knowledge hopefully that's not the servers and the answer is all of those so for the industry pages for example the solution pages all of those were made in concert with partners so far so we were like we would reach out to okay maybe we know that there's an agency that specializes in that particular industry and we'd say hey can you help us make this content that's how the nonprofit page came about that's how a lot of these came about and it's typically been kind of one-on-one outreach although like I would call out anybody who knows those areas please do that I've talked to several people in the higher ed space just at the con random stand-ups here who said hey we've got ideas about this or we've got ideas about additional levels of this conversation so one example for higher ed was right now it's focused on maybe I run a university or other organization that hasn't looked at Drupal or has heard of it but isn't sure whether to adopt it and that's kind of the persona that that's targeted to but there's equally this persona of people who use Drupal within their university but aren't connected to the others who do it and aren't taking advantage of a community of Drupal users in higher ed that exists and aren't learning how to they're not sharing techniques on how we got to allocate our time to actually contribute as part of our jobs right and all of that would be really useful but it's a matter of finding the people you know individual contributors consensus building recruitment like most like most efforts to find a maintainer for something you know I believe there was a question yeah you mentioned that in the personas that haven't been called out now in the new site one of them you identified as really the existing Drupal right sure I mean right back it encouraged me that I think a big talk of the existing Drupal yeah and I don't see that present is there is that something to search or sort of that's a good question I think it goes along with some of these other initiatives about the like the automatic updates and the site builder tool and the standardizing the way core uses composers that we can build an abstraction above it for the site builders the problem one of the problems is I don't think we can solve right now we can't just solve content for the site builder on a content level we actually need to solve some technical problems to make it possible for that to not be frankly a crappy experience and so I think we want to unify that so that the underlying practices that us that a the hardcore developers doing are actually the same things that the site builders are doing they just have a tool that abstracts it for them and then at that point I think it will become easier but there's definitely work to do yeah that makes a lot of sense to me and is exactly what I would have said this time in two days you work at I was not no that was very interesting that a number of site builders stood up and said we don't we're not developers we are and actually we love it is easier for us interesting and it's only going farther in that direction but we really feel like you just know that we've already would not know that outside of that and that's interesting because I think since frankly before the release of it while it was still in development the community has told the narrative of this is going to be super hard for the site builders and that's that's an interesting point I should check out that recording yeah okay okay yes please yeah I'll bring some technical solutions like I was working in the middle in the middle I think I think I was also at this small portion I heard you think about the post you're going to emphasize that we've got technical solutions we also have to provide the editorial contents which we've communicated to companies I think we need the composer as a community to be able to communicate the importance happily because there's a certain part of the community I agree if there's going to be a 20 percent that needs your site needs a UI for it sure and 40 percent on that to get it then we've got the gap there we've lost a bunch of people who need it and you need to say okay it's a little hard when you print and I use which was recent it's really incredibly valuable yeah you have to communicate that I put out there when we're talking that you need a video that says okay this is composer don't be afraid it's command by it's okay yeah and I mean composer I'll put out something that really is a trusty example for the book composer should do this but someone does their first composer out there on the site which is a congratulations on your first composer company you are entering the world at THP it's not yeah but a bunch of a bunch of developers get upset they don't understand how talkable people I keep with that attempt to be to be as we do the small initiative to communicate and emphasize communication I think that's absolutely true and I think unfortunately one of the tricky resourcing constraints just in the association right as you pointed out engineers is this is this whole room and you know the promote Drupal initiative is to help us get some more resources in that's focused on promotion not on communication about usability and onboarding which is another area in which we have a lack but I think we're going to at least in part be relying on some community resources and frankly like if someone came to me with a wireframe that said or a video like we don't have video production facilities but some people to actually do within marketing and media organizations yes yeah yeah yeah another session do it sooner when you point right you didn't bring this up I want to have this I want to have this okay I got to back up a second time and you can put it for module so do videos in the way for my experience to be incredibly positive to the point where I was just saying I don't think I go watch I'm at like 40% maybe watching videos fine people approaching they're incredibly smart they get to it for months they're like thank you for the videos I know they never watch them but it made them feel better when they started using the module and know that they're doing things yeah there's there's hold on videos and communicating I did some research with the videos and I want to have a list of links so I started linking materials and I said I'm gonna link to anything that's relevant I went out to find a big content for you yeah because it's the most relevant and maybe the day to explore my videos keeping all the people producing these outside resources and getting our people that want to get making money from my offering to video you know maybe it's it's placement where you're having two multiple resources being on this yeah they are able to get their kids you know they're quick for you know video I guess I'm sorry yeah when we start generating good content I don't think you want an issue to see when you know it's a paid produced content that links maybe to these three quality resources because this is the best way yep it's complex spending $50 a month for you to develop this amount so yeah no I I think I think it's really interesting and certainly something we've talked about before I think even talked about it with you know the two plays me folks to a certain extent one of the things I will say that's interesting is you may remember when we announced the industry page initiative I think there was Dublin and of course we have we launched with we launched with three at the time I forget so and of course we had to produce those three before Dublin so it was probably a year and a half two years or something of to get us to five pages where we are now working with content partners who are domain experts in their field but who aren't necessarily 100 percent committed to this initiative because I've got their own businesses to run and things like that so one of the things we need to crack and there's value at right there's there's placement of their case studies and things there right but yeah that's one of the things we need to crack is is this coordination of people who are those experts who have to market Drupal and realize hey if you three do this together you don't have to all do it over and over separately like we need to get people to realize that there's an efficiency to create the best materials and that's that's proving tricky but it's something that I hope we can figure out how to crack sorry I'm going to go over here I guess just a thought that came up from the start because the answer is making it a little bit and it's ruling as well like I I went down around the pole on something I got mislead by some doctors you know they got computation on a lot of work complete and I went in and edited it it's very good to give you yep workable is going to go well we haven't discussed page yeah I got you know I guess but I just went in and changed totally yeah which is mostly what we want as opposed to like 70s documentation and boss on the whole maybe model category can go eventually the difference of course being that I was going to make that change from 70 I was going to pull request and sort of somebody else would have to curate look through the name of whatever you know so that no the answer might be a little bit tooling that would might be better got if it was you know with the pull request please it's interesting because we don't we don't go so far as the pull request thing although we've experimented that with the duplicate user guide the duplicate user guide is actually aggregated in from a repository as opposed to other types of documentation and that tooling around that is a little bit of a contraption so it's not like ready to scale but the documentation in general has a concept of maintainership has maintainers are alerted for example when these edits happen and what they are and they can actually can they revert them we may have yeah they can revert them and these and these different changes without but we also have a lack of priority around documentation like again I'm of a mixed mind about whether it's tooling because there's some documentations it's phenomenally well maintained but Joe Schindler is the only person on the documentations working group is active really right now and it's like it's very easy to say well if only someone would just maintain and curate that's great but who's going to maintain and curate because we have 12,000 documentation pages so so but your point is absolutely valid it's just that even if we provided that tool we have to find the people who will take on the maintainership yeah so it's just a neuron that firing my brains totally totally we're really going to talk about the the real question I have is about the virus I see that everybody's pretty everybody's exactly in this whole whatever this I think issue with that yeah um is there an issue with that because I'm I'm starting to know when the people are not happy yes so there's there's that it's also not working right right right exactly it's so it's so here's the thing about communication media and I'm going to borrow uh a metaphor metaphor no it's just a a way of communicating this that Ryan came up with which is that when it comes to how we communicate together as a community it's not mandated it's an election it's wherever people choose to spend their time right they're going to go where they go and we can say I'm sorry everyone we're going back to bulletin boards and that's the official Drupal way and that's not going to make it happen people still use Slack or still use rocket chat or still use any of these other options and there's an interesting divide because you have the I think you have a lot of people particularly in North America who use Slack and say a corporate environment and it's already there and they just want another Slack channel because it's so much easier it's just a matter of convenience and then you have a lot of people who are like the bleeding hearts of open source who aren't going to touch anything if it's I mean they're running Linux on their laptops if you're if you're saying I won't use proprietary software and you're running a Mac let's have a conversation but like but other than that like and so when I don't think we're ever going to solve the problem of unity in terms of our chat solution as long as new chat solutions exist we'd all still be on like AIM or Skype or something like it's going to evolve it will change and it's not necessarily a bad thing and I think it might even be worse for us to restrict it and say sorry we're doing this we're doing this we're doing this and then maybe something suddenly gets better and we're choosing not to use it but it's tricky because Slack is also disabled their IRC bridges or is about to which is making it harder for those people who still wanted to communicate that way so so it's not a cut and dry problem and it's one of the reasons why for example the OAuth initiative to be like well at least we can have identity in these multiple areas that's unified will maybe help to resolve some of those issues but at the same time if a vast majority was on one solution and it happened to be a proprietary solution if we could fund it should we you know should we make sure there's history in Slack even if we're not trying to bless it as the one true way should we be trying to pay because there's 4,000 of the core community there truthfully that's pretty that's pretty much the position at the moment again we want to make sure that like we're supporting that we're not a blocker for people to try and rally around their own solutions and again the OAuth thing right some of the other solutions especially ones that we're trying to bridge across communication mediums we had to kind of shut down because especially given what was going on in the community at the time there was impersonation stuff going on and things like that and so unless we could solve the identity piece we were really concerned that that could be exploited in a very negative way but if we could solve those sorts of things then we can let people duke it out and may the best chat win you know so yes Damien not too not too big more than two separate sure discussion that I mean that's a five years ago that the answer yeah yeah we I really try not to promise anything around dates I will say you'll see some details in this in the other session so we actually well really Michael ran with a ball on this one quite a bit and prototyped a bit bucket integration with a Drupal dot org development site a GitLab integration with a Drupal dot org development site a little bit with the GitHub stuff before it became clear that that just wouldn't work and so like some of the problems are solved user syncing repository syncing like a lot of those things are actually already checked off the list so and we'll have to be usual long tail problems like get lab has different rules for what a username is then Drupal dot org so we have to figure out what's due the 30 people who don't have a valid username anymore right and putting one for America so yeah there's some there's some tricks to solve um I would say God I would hope I Drupal Europe I'm not just showing slides let's let's put it that way so I have a question about the well a concern actually was raised to be that like nature is brought forward I like that there is an average age that not cross the age of communication right there are agency biased so yep there are like where are we highlighting members who have never seen one that are creating their own staff yep created agencies are involved and we really need to look at this isn't something that will be agency focused yeah I think so and I had the I've had some of the same conversations possibly with some of the same people while I've been here about in particular the higher education page and again I think that same model that we have where the front page now the primary CTA is identify your persona and then we'll take you the right information from who you are I think we can drill we can iterate on that and drill down through that recurse on that with some of the industry stuff and be like okay look if you're here looking for someone to help you do this this agency kind of content is what you need if you're looking to understand the success or how to increase your success with what you're already doing identify yourself there and find this content about how to participate in the the edgy slack that exists and things like that and again I don't know those things or those communities so I'm working with max I don't remember his last name in the high red community and he's going to try and come back with a proposal for a landing page for that set of personas for higher ed and we're going to see how that goes so yes I was thinking about the site over it it would be this is an observation it's completely in test and you know sure the borderline opinion the but it seems like we're moving toward the like getting the composer requires stuff on the project pages which I think is good but the site builders where I work they don't use any of the like prescribed ways doing things from the that that are on that are promoted on those pages right they like as developers we use composer to do things which is not also on the page and the site builders they go through they find the short name and then they go into into the desktop and they go the update page and put the project in and that you know using the FPP system right to bring it into the projects through the the Google site itself and it's like either one of those workflows is advertised as the way to like move forward with these things yeah so it seems like we're really like somehow supporting a a workflow of that you know ten years ago when I started was the way to do it download downloading tar balls yeah and you know you know point your V host at it and you're good to go yeah that but that doesn't seem to be really like wait anybody does anything you know I think that's true and I think I think the reason that we don't just suddenly that we don't take the next two weeks and go through and change to make a new box that says do this is because that's actually still not to find it like it's certainly not what's there but it's still not to find it and I think a lot of the conversations with some of these initiatives are about let's get the exact one and then put it in place and it's so it's this weird balance between well it's not relevant anymore when do we update it but that's why I'm saying it's not tested it's more totally you know but it's one of those things that we really need to find that and just do those you know do that you know we have like five different ways of doing all these different things let's pick one or two that yep those two personas and there's some there's some great uh conversation currently going on in the official documentation initiative issue about how to identify what is those those one true ways of doing things so yes I know question good not really it's it's like the ultimate how do you scale you know so as I said there were conversations and when you talk about these great ideas and there's a huge amount of resources in the realm you need to own up to that issue I want to contact if I have a way that the Drupal Association is just trying to write to you that this year which is totally respect yeah but we have these big ideas they're not on paging about they should be a content editor role curates this and they're like a clued co-op to run into this issue yeah they run into certain departments that this can't be stopped you get a content person and their job is to curate but also bringing revenue yeah and I guess it brings up a staff initiative scale like is the Drupal Association looking at how they should scale to the department what direction because also if you have that research and you approach the community it's that we need this role to to accomplish this penetrary initiative people might not understand I mean given that exactly we're making amounts to $100,000 to promote people and to ship us for a coordinate position some money they do that's a temporary I mean it's a one time so but the notion the notion being right so the idea of finding a sustainable model isn't going to come from hey guys we need it to be sustainable so well you put $100,000 dollars every year it's got to be it's got to be bringing in someone who can't as part of the role both identify the problems and identify how to create enough revenue to make the it's sustainable I know it's tricky yeah well I also I was thinking I just I was thinking about that I think about for you you know I'm sitting there and I think I'm trying to use it and you have to look at there's a limited resources and you know you know I understand you don't have a immediate hold on how to fix that but absolutely and I agree I mean as as the person managing the team I would love more resources here too so but um and I think you know it's being thought about it's a matter of priority and it's a matter of not overstretching ourselves because in the last we attempted to do deficits bend to for a team scaling and that collapsed back down so we're trying to do it in a more deliberate more careful way so that we're sure you know it's rough when you go through letting half your team go so we want to make sure that if we do it again we can we can keep everyone around I'm not taking those but I guess no yeah yeah we have we have us we have it till two in theory although we don't have to go that long I guess along the time I think that at least we could use you to increase your right weight loss like I mean you know I'll put a lot of express build at this point but there's nothing else exciting to me about $6,000 that he has to run tests or anybody who wants to run tests now I'm not saying that we shouldn't do that but the WordPress does not provide a test box that there are modules and all the rest and I'm and I don't know I guess I'm just kind of free-associating about like it sounds like there are parts of the spend side of this they're very uniquely brutal it might not be appropriately scalable and you have some projects like that are doing Travis you know like getting on where they kind of jump off the island and that's their thing and that they're also running tests on through the block so well um I mean the speaker most most of the contrived tests are like that they're small they're short and they're not that many it's really poor I think we run about $60,000 for tests a year so it's most of what it is is poor testing and poor testing is a giant can't explain any article speaks to just two in tests I mean we've just got a lot it takes 36 core machine you get about and it takes an hour all fully parallel lives 56 yeah oh 56 well that's good it's a lot it's good so it's so it's huge it's a lot of a lot of testing and a lot of compression testing and a lot and we we have testing policies to see what we have down on that but you know everything out there was like Travis or Kit Lentz yeah everybody that's like we support the source projects might tell them what about it's gonna go oh oh oh oh so and it's it's a significant amount of cost that isn't free anywhere you know I'm talking about the user and truthfully I think well that's why you're doing quite you talked about earlier the run was a unit test yes still have a scale passed so that'll reduce the cost there we've done some naturally at the costs we think you're there like there's a membership past tests test results had sponsors for that and past did I hear you correctly though you're willing to give Mr. Crack Crack because I will take that out certain rules like I think I've got the issue turning the press a lot off if they want to sort of order picture things I wonder yeah if we're piggybacking on our tool and we can kind of you know make it more so there's a few things like for example we're trying to change the way that testing configuration works so that people are not doing the full suite of regression tests against every patch they're not it's not until it's RTDC right do we need to do them the rest of the environments that kind of stuff so there's definitely efficiencies to be found in terms of well you don't need to test that yet you're gonna test it later like let's not do it three or four times before we actually get to the end of an issue yeah we could probably communicate a little bit better on something really about I've I've thought that we'd shoot before just want to tell us the path for some that's kind of new to the community that's a good patch and then I'll go through and just check everything with the environment box and it's like you don't even know the path you don't know if it applies we're not going to get everything out there but we're not really telling the path yeah there's there's something yeah so the there's there's a combination of policy communication and just the ways we can enforce that a little bit better yes we're so good at preaching to the choir if you want to go on crazy money we have to start asking the community of users not yes community of associations and then and yes shops for the local shops right yep absolutely yeah we ask for local associations for individuals and for for organizations but well why not just have a building a kitchen why not a kitchen well we do have one but yeah we do have a public place where people can go out and you don't have a cabinet that would would you know this going to be because customers don't come because they generally have an error because they have an error and they want to type that error to a new location they see it I'm sure a lot of those are people who are just the officers that land there and if we can say hey can people go out and help make people better we need to have more help but this this is where your money goes blah blah blah blah blah blah blah I think we could communicate that better the other thing though is I think we need to do it on at a larger scale right like so the we do have some examples of this right we actually have some supporting partners who are end user organizations Pfizer J&J there's a couple others and those are people who understand that investing in Drupal is an investment and not an altruistic spend for their business and that's a case that we think I think we need to figure out how to make better but it's very tricky because they're with when we're talking those super large organizations it's these super long tail process of trying to high-touch getting them to change their corporate policy around the notion of contributing as part of what they do I mean I think that's necessary we can't keep squeezing blood from a stone from the agencies is yeah to Drupal and you know like on the DA again well and I I'm behind you we need in the marketing materials change the tone when you do the first ask for 200 fucking sorry for 200 dollars to support the VA where when someone's in Drupal it's communicated that we are a community it helps to just be an initial member when you start using Drupal whatever project you should be a member of the DA of the $200 dollars it opens the door to start this process of the task when Jonathan and Jonathan it might happen for type years ago what happened for the title project and at least they would be being it would open the wall initially and we understood the concept from day one and then it gradually grows and when we ask that we're going up and saying here here's a big thing you're using and it's never I'm going to have to just say this which I can't get part of this what people can get extended to say we just did they're not getting your message but the next thing I've been telling you is if you're not getting this you contribute it builds and grows yeah I like reading an article where Sysco was like we saved 450 million dollars which you know Drupal and I'm like do you think I have some yep yeah yeah and this it sounds kind of blatant a lot like the you know the giant yellow arm with the V's and you need to donate a dollar that day yep but it's so on the that's growth if we find out what the average cost of per test is and put that into the check between 26 and 34 cents by the by the way like pressing that button to submit all of those tests cost us you know it just costs me it's okay you know five dollars yeah you know I honestly I think the people are more aware of that yeah those things cost money so we pay for that what I'm hearing and it's pie in the sky stuff that we'd have to find some implementation time for hey Brendan how do you feel about some more projects but but would be things like yeah let's start let's do that like ching ching ching every time you hit the test spot button and increase the the scroll I think that would be cool I actually think it would be really interesting if you're not an authenticated user or if you're an authenticated user who's not a member to put ask on download pages or something like that you know like hey we see you're not a member could you contribute to the projects yeah we want to moderate our masks to mask it the right time not let it how Jimmy Welsall's face everywhere yeah we have experimented with that a little bit absolutely sure we're waiting for perfect so we don't have to we have to experiment and hear it like thank you Carl for that that's yeah that's how we do you know yeah but yes absolutely sure we know so do check put it back put it back put it back put it back like you know we have what we did the new Jenner field people were mad at us for doing the good instead of waiting for the perfect like you know by that a little bit so we have to be careful because you need to be you're gonna suck something right I mean that's true but yeah we but we but we're saying like it's easy to say perfect is the enemy to do it except you can easily start a forest fire by trying to do something good and not being perfect in some way that is perfect is like how many of these things so and I hear you it's also really easy to put it into the stack when you're going and you're trying to do something that's anything like okay we're you know we're gonna do this I know we're gonna offend somebody so let's just go get into it you offend a lot of things and things can lead you off it could absolutely yeah I mean even the small steps watch our even the small steps need to be deliberative action but they don't have to be the big monolithic solution which I think is totally fair point um for the good point not only is this an all-in-all but it's as long as it's a unique moment which also does some of the work to in terms of where we find the first week how we do this and we're great yeah over here well done I'm happy with yeah Donya Donya can't couldn't be here at the con but I'm but yes yeah no no totally and we had before before our team size collapsed through diversity on the team was definitely better and it's really I think about that fairly frequently actually and it's it's straight I believe it or not I was actually the last higher on the engineering team so in my role leading the team since then I've never had the opportunity to hire so it's been it's been a little bit interesting but sure for people who are interested that's a good question I kind of want to throw it to the to the list here bags and bags money so Drupal.org gets it's the Drupal site huge with the sites actually so we have websites that needs spun up to work on issues we have a pile of custom code that can get used to be more custom for some things or we've got things in other areas and so got Drupal's involvement is a thing that you can do people can talk with for larger initiatives we want to make sure that the planning is done with us all the time we have to usually have people build this whole grand thing and hand it to us like here you go dude I hope you want this because first now and then we have to you know take the cut tech if we do decide to take it we have to technical that forever and if we don't decide to take it they're very upset you know I've Ryan mentioned earlier documenting this will see I know you have a while sure we want to ask for to is it isn't a coding task but it's a technical task requires technical understanding testing in QA especially around the get that when that comes up yeah but you know hopefully especially for someone who has it I believe you've self-described family no it's we're getting the enterprise version of it and there's some custom code that is not public yet but will be at some point which you in theory could download and run on your own but I don't think that I don't think the the testing is there the testing is you know how things work now and how today's you know when we spin up this staging environment my thought is we'll be open for a few weeks and give everybody access it's just pretty helpful literally yeah just functional testing help us everybody four commit you know access commit to four on the staging environment just because that'll give people a play with it just help help us find edge cases and keep an eye on the marketing initiative that's yeah yeah isn't really a infrastructure technical thing but it is something that requires extra eyes on it I mean I think 50% of the feedback from this room excuse me which is the engineering geek types for the most part like 50% of the feedback was still about content and presentation and editorial like and I've had to step into participating in that myself and I think we can do that to a certain degree or we can find who we know our experts and point them to us because that would help as well yeah certainly doesn't hurt oh yeah right here that was an example of how there's some okay it is oh it is yeah I mean that's because I I work pretty closely about one or two to make these a for example I don't want to be able to see how I stuff but I think I think the larger issue with what we want to help on a lot of work is that like 95% of the work here is in coding you know it's not the development it's like applying use cases coming up with the the plan and you know architecting and orchestrating and doing all of that but then oftentimes when we come on and there's whites of code and the ones we can do that you know it's it's more that just figuring out what it is that we need to do yeah getting our arms around just the problems is often more difficult yeah question with the people see I am but will that allow us to actually turn off certain aspects of the test bottom to say like I don't care all about any of these exactly just from this exactly and that can be done now yeah yeah today here if you know if you know how to use it yep and there is a docs page up yeah there is a doc page up the one thing that's not quite document is the custom commands and really the one thing missing with the custom commands is the bash environment variable you can include in that custom command and the first task something you could do is spread that word tomorrow that you can turn things off you know it's gonna be a lot of test from tomorrow right we're looking at the building this is a relatively new thing in the past yeah well there's there's actually a patch in for like it's going to be yesterday that that's it's going to be about so I showed it to the Jess and she was like going back and she said this is very important in the market major so they sort of want to look at it yeah that would be that yeah we'll be around experience tomorrow yeah I can take it to any specific yeah we'll be on triple.org yeah and you're interested yeah and then the other way is like there's a long mission it's where we are bridging over before order you know we have a lot of issues like you know when you do stuff when you do updates we're going to be changing the API for how about bit board there's going to be a lot of patches for we're going to need reviewing we're going to need people lies on it that's also going to help or you know if you're in one of those groups that's particularly focused like if you participate regularly in DD and I and that kind of a thing helping that group to API if I the open demographics initiative would be really helpful writing a module that consumes it would be really helpful like those are things that we're going to try and work on together but and that's nicely scoped compared to these things that touch a lot of other stuff I think that was a great final question unless anyone has anything else burning so I think I'll wrap it up here thank you very much for your time everybody I encourage you to check out the recordings of any of these types of sessions if you haven't seen them already they're relevant to a lot of the things we talked about here so if you see something there that you didn't have a chance to look at please look up the recording and of course join us for sprints tomorrow and leave your feedback on all of the sessions you attend all right thank you yes thanks everyone thank you thank you thank you thank you thanks John thank you very much thanks thank you thank you