Drupal.org Engineering Panel ![rw-book-cover](https://img.youtube.com/vi/HV0GCdhX-3c/maxresdefault.jpg) ## Metadata - Author: [[YouTube]] - Full Title: Drupal.org Engineering Panel - Category: #articles - Document Tags: [[drupal]] [[funding]] [[gitlab]] - Summary: Drupal.org Engineering Panel - URL: https://www.youtube.com/watch?v=HV0GCdhX-3c All right. Hello everyone. Welcome to the drupal.org update. Um, and thank you for joining us at an afternoon session. I know everybody starts to get a little bit tired in the afternoon, but it's nice to have so many people here and interested in the topic. Um, drupal.org is all about putting together the tools to serve the the community to empower all of you to then build Drupal. Um, so I'll start off with a little bit of introductions. The top row here is the staff of the Drupal Association on the engineering side of the house. You can see we're all here today. So, those are usernames and our names. Um, you'll see us as we go. In the interest of time, I won't do one-on-one introductions. I think we always run out of time. On the bottom row are folks who help with drupal.org infrastructure. They are with other organizations and either sort of volunteer or part of inind trade agreements. So, those are some other names and faces that you might see um in the Drupal.org channel or the Drupal infrastructure channel on Drupal Slack, that sort of a thing. I also want to give a special shout out to a couple of folks who've recently been involved in uh sort of volunteering to support drupal.org. So, uh Jamie Johnson and Bill Sprout especially have been doing a lot of work on the new blue cheese theme, new blue cheese edition, uh purely as volunteers and have been doing some awesome things. Um and Shane Collier was working on uh design work, um and revisions and all sorts of stuff based on the extensive community feedback. So, really, really good. But we're going to kick things off um starting uh by talking about strategic initiatives a little bit. And um some of this is information that you may have seen alluded to in the keynotes in other sessions so far, but I'm going to talk more about the Drupal Association's involvement in this and how the engineering team comes into play and we'll talk about some specific ways we've been involved. Um so first, one of the objectives of the Drupal Association has to do with contribution to Drupal. Um so one of our goals was to um help increase the number of organizational contributions and individual contributions to the project and that has to do with aligning with the Drupal core team, the CMS team, whoever else on identifying what strategic initiatives are currently important and finding ways to encourage our partners to support that by donating time, by sponsoring developers, any of those sorts of things. So we've tried a variety of programs over the last three years uh to do this including things like uh bounties on specific issues or waiting certain contrib projects higher because they represent something important to an upcoming initiative to uh the most recent one which has been these sort of almost contractual more or less contractual agreements with folks with organizations to take on whole tracks for major initiatives. um and that's how significant portions of the Drupal CMS um functionality have been put together, the Drupal AI initiative and things like that. And so despite I think a pullback in general uh in the open source and technology space uh in the past couple of years, we've seen a uh a really strong sort of refocusing of our community and a lot of people who are typically competitors with each other coming together to uh support some of these major initiatives. So that's really cool. Um here's the individual contribution numbers. The two differ because um an organization will only be credited once even if they have many contributors on that on that issue. Individuals are credited once per issue and might even be crediting more than one organization if they happen to be sponsored by more than one. So um there are three major categories of strategic initiatives that the Drupal Association is involved in supporting. So um Drupal CMS is certainly one of them. Um and obviously this is that combination of project browser automatic up automatic updates recipe discovery um the trial experience for Drupal CMS. Uh one of the things that we are doing following this 2.0 period is trying to solve the hosting problem of where people can actually use Drupal CMS uh once they decide they want to. Um the Drupal AI initiative that you've heard some about and the site template marketplace. Um, Lenny, if I could ask you to maybe talk about a little bit about Drupal CMS and that initial recruitment and then it's also about Drupal AI. >> Sure. >> Shall I just >> Yeah, why don't you just >> uh Yeah. say about the uh Drupal CMS starting there. I mean it's it's been very by now we can say long journey really because it started uh back in Portland right was first announcement about the Drupal CMS initiative and a lot of work has been done there in particular related to the uh establishment of the fruitful relationship like in terms of the uh contribution that is meaningful contribution that is targeted and um yeah myself and the rest of team we were quite involved in the in this negotiations in this uh communication with the Drupal certified partner with the organizations with the individuals as well who are contributing massively to the prosperity of Drupal CMS right and uh yeah um long story short by now as Dre already mentioned we are looking towards uh releasing the 2.0 zero hopefully this January and maybe even as it has been said on the birthday of Drupal but we will see about that um >> sorry yeah um then uh when we let's switch for a second uh well not for a second perhaps to the Drupal AI initiative and as you have like so you can see over here all the companies that currently are involved already and signed the agreements as the uh as the partners of the initiative. I might I want to say that it's not limited just for to these 22 because there is massive amount of the individual um also contribution going towards their AI initiative and uh Marcus in particular and Valerie they are u doing a lot of work uh trying to help uh with any sort of questions that the folk that are interested but you know not quite at the level to be able to uh to put anything but their own time. Uh so they could understand it better and they could understand how they can contribute once again in a meaningful way. Uh they also are busy with the u all sort of things including the uh the code reviews etc. But let's talk about what we achieved so far with Drupal AI because this initiative is quite new but it is evolving in a very very rapid manner. So it has been announced, think about it, only on the 9th of June, right? So it's been what about four months, four and a half months still till since the announcement. And by now what we've got, we've got 22 sponsors that signed up and we have quite a few considering already and we hope that that number will be even more impressive by the time of Dupalcon Chicago. We have over 1 million investment. It's the most massive like fundraising that's that's Drupal known so far, right? And one when we're talking about this 1 million quarter of it like more or less quarter like I think it's around 260 or 270 100,000 it's in the in cash and the rest is in kind meaning that it's uh time that people actually are putting into the um into the initiative to performing tasks not only about the the coding the creation of the um of the module etc. er, but also it's enormous work around marketing. It's enormous work around uh project management or even for that matter fundraising because it's not like people running toward us screaming shut up and take my money. Sadly enough, we would love that. Um and uh also like what is happening from the technical point of view we have regular releases and we have I think that's weekly at the moment updates on what is happening what is changing in order to keep everyone in the community informed because we all understand that AI can be well can be percepted in a different ways and we want to assure people that we are not going towards you know anything absolutely mad. Um uh and also like the that's that part I absolutely love that uh AI module has been adopted by many um uh many companies many institutions. So we we had a call call to action uh around 1 month ago. >> Mhm. >> Before the dupon vienna because it was like one of the brilliant ideas that you have like couple of weeks before some big event. How else and within one month we got 15 use cases submitted to us that showcase uh how like what is being done using Drupal AI. So for instance they there is Southwalk council in London that are using it for the uh basically uh to convert uh PDFs into the into the pages right and there is like the the cancer research etc. So there is many great really use cases that show that you know we what we are doing it just not just for the hell of it that it actually have practical usage and people find it not only exciting but but helpful as well and in terms of the DA support so what we do over there our role is uh obviously down to the fiscal and financial uh fiscal and uh legal story uh support meaning that we are taking care of all the agreements. We are also um kind of being a bank for the for the initiative in uh of some sorts but it not it does not limit that. So uh myself and our marketing team our our communication teams uh we are deeply involved in the marketing and uh spreading the word to the community and even to the like outside of the community in particular uh by the like you know emails, newsletters, uh social media but not limited even to that because we also uh attend the um out like not the events that not limited to Drupal only such as AI summit for instance this year in London and there was like thousands of people I mean it was like I don't know around 10,000 or something like that not to lie to you and um it's it's amazing when people coming up to your booth and they are like oh Drupal you do AI guys so some did hear about Drupal some didn't and u you know talking to them telling them what we do providing those examples and getting the feedback it was quite interesting quite quite illuminating I dare say. Uh also in terms of marketing we help with the guidelines and the outreach as I me as I mentioned but obviously we also are um in charge of the technical guidelines because we want to make sure that like everything that is used and how it is used as well it's within the um the values that we have at heart at at all times you know so for instance that all the code yes we are talking about for instance since the early access, right? So that was something that some people were very skeptical about. Uh but we we are making sure that all the codes within this initiative will be open after a certain amount of time. That's like basically it being polished up and then everyone can like the code will be open actually like even sooner. I believe it's more about the recipes that will be pretty much >> given to sponsors. >> Yeah. under lock and a lock and key for obvious reason that we want them to be nice and shiny when they see the world right and uh of course there is fair amount of the operational support um so when it comes to I don't know aligning the processes not only between the DA and the initiative but inside of the initiative we are very much involved there and uh provide our expertise case in that relation. Uh and yeah, so if we're talking about Drupal AI initiative, you can see the providers, the list of the providers that currently are available and um yeah, it's only growing really. >> Yeah. >> So hopefully we will also maybe double that at some point. >> May already be doubled. Yeah. Andh yeah that's it I think from the AI and due to unforeseen series of events I need to run to another session. So if you have any questions I will be here by the end of this session back in this room however ridiculous that sounds and I will be here all day tomorrow and rest of the day today. >> Awesome. >> Thank you team. >> Thanks. >> Cheers. So >> and then the last area of um significant strategic support that we wanted to talk about is with the site template marketplace. So Dre talked about this as being the the sort of next generation effort, the culmination of what's uh the Drupal CMS project is doing, what the AI initiative is doing, what all of the efforts to make the power of Drupal more accessible to regular users are sort of coming together to be is the creation of the site template marketplace and this concept of site templates. So the the DA specifically is tasked with the marketplace buildout. We have an approximate outline of how we want to achieve these goals. I'm not going to go through this like outlined road map, but we have a a sense of the different components we'll need, how we're going to handle tokens or license keys for these um uh for these templates, commerce solutions, whether we need a private composer endpoint for the installation of these things, whatever it might be that we might need, we're going to begin the evaluation of that. Um we're also working with Christina Schumalis um on contract to help support the wireframing of what the marketplace may look like. These are very very early. We weren't expecting any before Drupalcon, but uh Christina provided some initial ideas of what the marketplace might look like when we put that together. Um so I'm going to keep going. I might even accelerate a little bit because we never have enough time. But um I did want to pick up on funding our work. So, if you were at the um the opening message um and uh you heard Tim Doyle, our CEO speak, you may recognize this 10-year-old note. >> Understand that. >> Yeah. So, let me see if I can attempt to explain this. What we did is we looked at the budget of the Drupal Association, especially the things that are aligned to engineering services or infrastructure and the annually and then the number of sites. Um, and we looked only at modern Drupal, but the number of active sites out there and did some simple division and said we spend x number of dollars per site um on supporting everything to do with the technical side of Drupal to support the end user sites that are out there. So, keeping the update servers online, keeping the contribution tools online, all of that costs us approximately $10 per site. >> This is Drupal sites on the internet. >> Yes, on the internet. All of them supported >> modern Drupal sites. >> Yeah. Yeah. Yeah. again looking at looking at >> not include not including Drupal 7 >> not including Drupal 7 in this in this estimate >> that makes more sense >> yeah yeah yeah although even with it doesn't change the numbers too much >> a year this is about if you took our current budget and divided by the sites it's about how much we spend per site per year and then this was so yeah and so this breaks down where does that money come from like what what happens where do we where do we kind of pull that together and um in the slide that you saw from from Tim Doyle You saw that $3 of it come comes from DA revenue programs. This is like the supporting partner program, proceeds from ticket sales to Drupalcons, those sorts of things, that kind of fundraising. About $4.50 comes from our job as a staff negotiating free services or heavily discounted services for direct infrastructure tools. Um, and about $2.50 is things we would love to do or positions we'd love to have on the team that are still unfunded. And you can consider those deferred or delayed maintenance, right? When you ask why is X subsite still on Drupal 7 for Drupal.org, right? For example, that's where that is. We have enough people to focus on the the highest of the priorities, but not everything can kind of make that list. And there's another way to look at this. Um hopefully it's kind of useful. Um right this is the cost of activities approximately the approximate cost of of staff of infrastructure of just other IT services or the unfilled costs and this is where the money comes from. So among other things you you start to see that if we didn't do in kind trade just what we got from community memberships and donations wouldn't even cover the infrastructure right um all that kind of stuff. Um, so we want to try and give people a more accurate picture of what it takes to to run everything we do. And yes, our infrastructure cost if we were paying every dollar is about 1.5 million, not counting staff uh to keep everything running for Drupal. Um, so again, where does all of this stuff come from? We talked about this a little bit. Primary sources for DA revenue, certified partners, sponsorships, memberships, some services, etc. The inkind support comes from negotiations with companies. A couple of those are Drupal related organizations. Tag one helps us with infrastructure services uh on drupal.org. Palunteer helped with the recent redesign of of drupal.org. And then we have relationships with all of these commercial organizations for like I said either free or discounted services um to try and support what we're doing. Um and that's how we make up the majority of the engineering budget is in these things that we've we've negotiated. And then again those unfunded things have have to do with result in projects that outnumber the engineers being able to do the 80% but not the 20% and those hard choices about which things were prioritizing. So but speaking of those things we did get to prioritize we'll go through a couple of our our recent projects and um celebrate at least one that you've definitely all been waiting for. Um but we'll get there. So let's start with talking about showcasing Drupal. Um, so let's see. Neil B. Neil, maybe I'll hand it to you first to talk about the uh Drupal CMS trial a little bit. >> Um, >> I didn't work on that one. >> Oh, well, fair enough. You worked on the you worked on the form to get there. Um, >> oh, okay. I I did leave form on that. >> Yeah. No, fair enough. Um, no. But the two of the things that are new are the Drupal CMS trial and the Drupal AI trial. Both of which have just launched in new versions basically as of last week or even this week. So Drupal CMS trial was a hosted trial with a partner if you checked it out anytime since like June or something like that. It is now not that. It is now a new iteration of the desktop app. So you can download the desktop app, use it forever, kind of build out and play with Drupal CMS as much as you want. Um, but it now uh does uh collect some information when you're downloading it for us to use to help reach that audience to understand who's actually looking into Drupal um and begin to understand that folks and potentially connect them to partners and hopefully fund more of our work. Similarly, the AI trial, I'll mention this one too real quick before we keep going. Um the AI trial, uh we're partnering right now with Amazio, who is one of the um uh Drupal AI maker partners. um to provide uh AI demo experiences where you can select from multiple experiences with specific use cases because one of the early pieces of feedback that we've heard from many people not just in Drupal is that everybody has now heard that they need to care about AI and they have no idea why or what they're going to use it for. So these these the the Drupal AI trial experience says here's here's a trial that will show you how to use it for your like metatags and SEO or here's one that'll show you how to use it for content or how to use it for whatever gives you an idea of the actual practical foundation that you might care about instead of just the buzzword. Um so yeah so those are some of the things going on uh part of the revenue generation to try and try try and keep all of these things going. Um, but yeah, let's get into some some redesigns on the other showcase content. So, either Neil or Brendan, if you'd like. >> Yeah, I can talk about case studies. Uh, I know about those. Uh, so case studies are the first content type that we're porting uh from our old Drupal 7 site to the new site. Uh so uh yeah the uh chain uh you saw in one of the earlier slides worked on this design and then uh Bill uh did a lot of the implementation and then I realized uh we can't just do like a view only uh we do have that as a preview but we want we're editing case studies over so we're running all the uh uh you know we've been working on this project long enough for it to have technical debt. Like the fields were set up not quite right because uh we didn't realize u the the right fields to set up uh a couple years ago when a volunteer gave us a bunch of migrations uh for free. Um then uh yeah stuff like uh yeah I wish I connected this earlier but we should update the migrate the URL aliases and uh redirects. Uh so Fran got that done uh started uh this week morning. Uh so all all the a lot of the problems we'll like have to uh uh solve to make the next content types easier. Uh but yeah, this is available as a preview on new.rupal.org/case studies. Uh but uh it's not launched yet. >> And then another area you probably saw mentioned in any of the marketing related talks is the relaunch of a bunch of industry related uh case studies. So there's some for um there's some for AI specifically. I think 17 dedicated pages for industry use cases around AI have launched and we've relaunched some of the general feature pages about industry. So B, do you want to speak a little bit to some of this? You reviewed some of the that code. >> Yeah, I um it's been a pretty big overhaul. Clearly it's a a big visual improvement. It's a lot more engaging. uh behind Drupal is people and it's easy to forget that uh on the Drupal 7 site and the Drupal 7 theme. This really brings that forward. Um as Neil said, Bill did a lot of the uh a lot of the >> uh Jamie Jamie. >> Oh, this was Jamie. >> They both do a crapload. So, thank you, Bill and Jamie. Um and we've been reviewing so much code, it's easy to forget who did what. Um I don't actually know what to say other than it's gorgeous. Yeah, it looks a lot better. It's a big improvement. Um, you've also been working on contributor support. So, one of the things we talked a little bit about creating new mechanisms to try and incentivize strategic projects and contributions. So, be often handles the waiting of projects where we need to put our thumb on the scale and say they're more important. So, I mean, you can talk to that a little bit, Brendan. >> Yeah. So, this right here is uh the full current list of projects that are credited the same weight as core uh if you contribute to them. Um that list has been growing for a few years now and it should because we have a lot of strategic stuff we're doing as a community. Uh thanks Neil. Uh >> yeah. >> Yeah. So 10 credits uh essentially versus one because uh we really want people to focus uh on this to move move the needle forward uh in our various initiatives. >> Yep. And then in addition to this, right, we've uh as we've started with Drupal AI or the Drupal CMS tracks, we've also said we've also begun doing these things where we're like, okay, in addition to the issue credit, we're going to say we want some thousand plus credits for knowing there's a dedicated FTE by this resource who's going to work with the CMS team to accomplish this goal, which is something we want to expand. And there's another program. We want to have a standard program based on what we've learned for the CMS initiative and the AI initiative for general core positions um or key contributor positions. So release managers, subsystem maintainers, all of these folks who need sponsorship to do their work. Um uh we talked again this is about showcasing Drupal and and marketing stuff. We have a we're sort of reactivating the promote Drupal folks. Um so there's a new content editor model on the new version of drupal.org. if you could speak to briefly. >> Yeah, so uh on Drupal 7, we were using organic groups to uh kind of control who had access to edit sections. Um and that was actually pretty pretty nested. Uh a little bit complicated. Uh we're now using workbank workbench access, uh migrating from organic groups to workbench access, which is a much more simple, uh way to give people the ability to edit things on drupal.org. Um, any of the references are pretty cool. You all know that I'm preaching to the choir. Um, but yeah, this is the first like uh was the first step to making it so we can start giving content that people can actually edit that isn't us uh on the new site. >> Uh, and then very briefly, there was an overhaul to the just main information architecture on pl.org that you may or may not have noticed depending on um how closely you're paying attention, but can you talk to the motivation for that? >> Sure. Yeah. Well, a lot of people in this room probably just know the path and type it, but um our our previous our previous main menu um the the user journeys were a little bit scattered throughout the menu. Um and this is much more focused on user journeys. uh when you hit one of the top level items, the things that come down are are much more related to the thing you clicked on, which is good. Um there's now also room for uh the heart button, which you can use to send us money. Please do. Um yeah. Oh, and this this uh architecture has also been updated on the Drupal 7 site. Uh but you know, looks really shiny on the >> Yep. Okay. Okay, now let's get into some stuff that I think might be uh the real candy for some of the folks here. Let's talk a little bit about some developer tool stuff. Um so between between uh Fran and Neil, maybe you can run through these. I'll uh I'll drive the slides, but let's start talking about contribution records too. >> Yeah. So we we try our best. And so obviously many of you may have seen forms like this already. A few weeks ago, we changed radically how contribution records are stored. >> Very nice. >> So yeah, we we really changed. It was a a kind of all or nothing sort of change. We moved from the old system into the new system and contribution records. It stores the same information as before. who did what for which client, which company, whether they were volunteering or not. But it does it completely on the new site. This is not data that we migrated from the from the old to the new. We we are storing all the information now on the on the new site. And most importantly, we did this to unblock the possibilities of contributing from different sources. Uh we have jupa.org issues which we all know and I think we are all kind of waiting for vit issues which we will talk in a few slides. So with this new system we can plug an issue from drupa.org and it will read all the information from it and allow crediting that people it will allow plug in a issue from gitla and allow crediting but most importantly it also opens the possibility for new contribution sources that could be localized to drupa.org or people translating strings or why not a github project or a gitlab.com project. So obviously those are not implemented yet. We are going baby steps. I mean they they're kind of giant baby steps but uh but yeah we're focusing first on our internal issues. We are going to migrate them to to GitLa. But as I said, this system is um it's allowing you to pull information from issues even from merch requests related to those issues and it's kind of fault proofing um contribution records um storing for the future. Uh yeah. Okay. Listening. So this was as well relatively new. So when we did the transition, we actually were using JSON API endpoints from the old side to read from the new side. then process it and then show it. And you may have seen that for a few weeks that was a bit slow. If you were checking credits for your organization or even your profile, it would take a few seconds to to load. But this is something that now we move completely to the new site. So we are displaying now all this information on Twittering from the new site loading it via Ajax into into the old site. So all that information that you see on company profiles on your user profiles is actually reading from from the new site and yeah the listing pages are kind of brand new like couple of weeks old but so far it has improved the performance of page loading and that's the idea like move more and more and more things to to the new to the new system. Um next slide. Cool. So, I was typing something on my phone before. I wasn't being rude to Lenny while she was uh talking. Um, we're going to talk a little bit about the is migration. This is a topic that we started not exaggerating three years ago, I think, on Prague. I did a demo of a is migration that tric showed in the keynote. But then so writing that demo it was was a lot of fun and we could start to see what the future would look like. But then as we were pulling from that code we realized that we needed to do a lot of things in between. It's not just it's just migrating issues. We needed to actually keep all the credit information. Gitlap is good for individuals but it doesn't have the concept of volunteering or organization. So we needed to actually work on the contribution records and because we needed to work on the contribution records then we needed to start checking about migrations 7 to 10 migrations we need to have all the users we need to have all the organizations we need to have all this data so we were pulling more and more and more and I'm happy to say that we are closer than ever we were actually aiming to have some projects uh using GitHus for this week But then kind of many things got in the way in the last two weeks. >> Then we had to get on a plane. >> Yeah. These kind of things. So but but yeah, I mean as I said we are like all our staging servers are completely set up. They are talking to one another. We are just running tests and yeah I'm happy to say that as of one second now cool the optin period for um projects to move into GitHus uh it's open h issues will look like something like this I mean this is a this is a screenshot that we took ages ago and it still looks the same we can see how most of the meta data on ja 7 is will be translated related into labels but then we be getting the title we'll be getting all the contents or the comments uh we will relate uh yeah it will look something like that I mean we are kind of uh maybe changing the format to make it a bit uh friendly but yeah you will see all the information that is coming from the other side and it will look something like this go ahead once we have issues that means that we will be able to have can boards ham boards will be defined per project so Each maintainer can define what's important for them and their project. Obviously, the needs of Drupal core are not the needs of a small Drupal project or canvas or any other project. So, they will um they will be able to define canon boards and organize the issues and and obviously I mean if we've been helping on other projects uh on whether it's on GitHub or GitLab itself, this UI is more familiar to to many of us. So it is a big change but it's not that we're changing to something completely new that we redesign. We actually wanted to use these tools because they are designed for that. So yeah we're going to we're going to get issue boards and most of the data will be migrated as as labels and most importantly the contribution record system is exactly the same. It's not going to change. So that's why we did that change before and once we migrate to it to to GitLab is that's going to remain the same. Um how to opt in. So I just put the message on the GitLab Slack channel. So if you just go to that channel you will see it. But yeah, that's the that's the issue number and the idea is like please send us your module if you're interested to to be an early adopter of GitLab is we will not migrate all the projects that I mean say that Drupal canvas come tomorrow dup core and say I want to >> core is not migrating tomorrow >> not going to happen so obviously we have some prerequisites because they are going to be the first ones we're pretty sure that there will be some bug packs or or maybe some things that we want to improve. So by migrating uh your project to big issues, you are helping us improve the scripts but there are some risks uh that come with it especially at the very beginning. So just go to that issue, read the description, read what you need to do and yeah just please help us be um >> yeah be brave of your project in and help us move forward and pace in. Yeah. Are the risks documented on the issue? Like >> uh yeah, there is a big warning message that we put in there. >> So the question is >> sorry yeah the question is are the risks of opting in early documented and yes they are documented within the issue description and they're things like we can't revert your migration. So you might expect you might find that some metadata didn't go where you expected or maybe there's a partial migration problem if something were to fail. We've never seen that in testing, but who knows? The first time first contact with the enemy, we will see see some issues. >> So, we'll see. >> A very small followup question. >> Mike, >> sorry. >> Uh, do does the original uh do the original issues still remain available? As in if things go wrong, can you still access a readonly version? Because that would be kind of like a worst case fallback. >> We've tried to bulletproof the scripts as much as possible. So that even if something catastrophic were to happen at the beginning, we can still do some sort of revert and then try again. Initially we are just setting the drupa is drupa.org issues as unpublished. So they we're not going to remove remove them. Everything is going to be there. So the idea is that we're going to process it sending to GitLab issue then and publish and then create redirects so that if you go to the old issue you are redirected to the to the new one even if the process were to be cut in the middle uh no information will be lost so all the information is going to be there and we have some tools to kind of reverse I mean we we don't want to promise we don't want to over promise but we've developed the tools to to be able to reverse If something goes wrong, >> if you have slightly elevated permissions, you would be able to access to still view unpublished uh issues essentially. >> Yeah. Yeah. 100%. >> It's not slightly elevated. Is this >> doesn't even sound like >> I'm not sure this is >> uh but yeah, the goal is to completely remove the issues from the old site. Uh so we're not maintaining any uh well yeah we're maintaining like our regular backups for catastrophic failures but eventually the issues from will be gone from the Drupal 7 database. Um if you haven't used if you used to be able to see unpublished issues on Drupal.org years ago you probably can't anymore. It's gotten much more granular now. So if you're not like on the security team or the infrastructure team you probably can't see unpublished issues. So I wouldn't rely on that as a fallback. Um I I so I've always assumed that before we migrate core we're going to like will we have I I assume I have access to do that like on some sort of staging server first >> for core itself. >> Yeah extensively like a long testing process. I I I'm assuming we won't >> Yeah, I mean the testing process is the smaller projects go first and we see how it goes and how long it takes >> because core has you know all the massive metadata with all the special tags and all that. >> Yeah. Yeah. Core will probably it may not be like the last of the projects but core will be after we have had a pretty strong success rate of small to medium scale projects for sure. Um and we do have a full staging environment where this kind of thing can be tested. Um, it's just, you know, when we're talking about opt-ins, this is your production opt into this. And to be clear, when Fran says we have a we have a revert tool, it is not a revert on request tool. >> It is our emergency recovery tool. So, um, yeah, you can't you can't change your mind is the point. If you're an early optin, >> thanks for and yeah, we've we've organically rolled right into the question period, so we're going to continue there. We we've got about well that was the next slide so we're perfect so we've got about four minutes left please >> um do you have some documentation what's the difference between the issues on Drupal.org work and the new on on GitLab because kind of the Drupal community has learned this process to setting issues to its work in RTPC. >> Is that kind of supported or or what how do we do it? >> So the good news is that those there are there there are some some differences but those workflow states will be re-represented as labels. Um Neil Yeah. >> Uh I mean the big difference is that GitLab's permissions are uh different. So the updating the issue summary and updating the labels those are only doable by maintainers as far as I know. So, but I mean the rest is you know there a comment thread with uh where you've uh yeah all the kind of social interaction other than issue summaries is basically the same >> but that means as I as a maintainer need to work on all the states of all the issues >> not the states the yeah the label changes yeah so >> once you start working on the canon board for your project. I mean as I as I mentioned before the needs of of one project can be different from others. So they will define the different columns of the canon boards and you can automate the label creation. So if just by drag and dropping an issue from one column to another it can create the labels for you. I mean this this is documented in the graft documentation. It's because we are effectively using that we are migrating all the metadata from D7 to GitLab. But then once you are on the cloud, it will be down to its maintainer how they want to the workflow of the project to to flow in that sense. >> Yeah. Because as a maintainer, I like to go to my project and just ignore everything and check the RTBC issues. But I know somebody tested it and approved it and then I just look at those. But now I have to do this search myself. >> Yeah, I mean GitLab has uh been adding custom permissions. Uh so that might that access might change. The other thing uh I know for gitlab itself uh they built a contribution site kind of like we have the dual drupal.org versus get.rupcode.org and they have a tool to an outofband tool to update the labels for community members. So we could end up building the same thing uh since there's a need for it. >> Yeah. or a custom slash command for slash RTPC maybe um that might be a way to do it. >> So there'll be there'll be a few things like that but we again if you're opting in you're going to help us find which of those are critical and which of those are just a difference in how we're thinking about it once we've used it. >> Uh but yeah Gitlab's a product so the permissions and set uh setup it has is what we're using. >> Yeah, we're not hacking core on GitLab. Yeah, if you have any project on gitlab.com or github and you maintain it, then the flow is going to be pretty similar. I mean, >> uh that said though, we do still have the fork management that contribution record UI also has the fork management stuff. So the the one difference is we have set everything up so you will not have 10 different forks of contributors all contributing to one GitLab issue. they'll be able to collaborate on the same forks um and open merge requests that way, which is one of the really big things we did want to keep from the Drupal workflow. Um but yeah, I think the idea of slash commands for key things like RTBC might be something we we end up doing. If you want to see a demo for your own project, just film me in the contribution room and I can just run the script on my local copy for your project so that you can see how it translates or even just on a random project so that we can see kind of the implications of it. I'm happy to to demo the whole script process that we're on. >> Honestly, that would be a great video on D. Uh if you find the keynote the uh segment that I did for for Prague for the Disc I mean we've got free address command where we just put the machine name of the project or the node ID. It just goes through the issues and then creates the redirects for you and then that's it. I mean anytime that you go to the issues pages on a Drupal Drupal.org project. It will just take you to GitLab and then you see all the issues. >> Uh I mean I'm happy to video but not a lot will change. >> Maybe we'll pull it out though. You did recently do a series of blog posts about the contribution record changes. Maybe we'll just do another series about the issue on and include a clip. >> That's our plan. Yeah. To to write some blog posts about it and we definitely include a video. So yeah, that's a good idea. >> So we have reached our official time. Um as usual I like to include the joint contribution highlights. Um and in particular in this case because if you would like to see as Fran has kindly offered and may come to regret um an example of your project being done on a local >> um >> yeah um or any of the other contribution activities or if you want to speak to any of us now that you've seen our faces and know who we are about drupal.org Um, feel free to find us uh in contribution or at the Drupal Association booth. And please help please help support us with that. Um, what did I do with a€ 10 bill? Uh, help us grow the little bill. Um, that would be really really helpful in in general. But um, in the end, we're doing all of this for you in particular, for you as the community who's building Drupal and trying to do our best to to foster and accelerate what you do. um you've probably you know what we were celebrating in previous years was testing getting incredibly dramatically faster and improved and we're hoping to see similar levels of positive impact from these changes as well. So thanks to those of you who are brave enough to opt in and thanks to those of you who've given us feedback so far and um yeah we'll we'll see you online. Cheers you., Drupal.org Engineering Panel, YouTube, 01k9yvhnkx92v3sfqrjw6nkj5b.