MariaDB’s Andrew Hutchings (@LinuxJedi) explained their Project: “Integrating MariaDB Catalogs with PHP Platforms” for the CloudFest Hackathon.

Transcript (auto generated)

Simon:
Andrew, what is your project called and what is it all about?

Andrew:
Okay, I’m going to try and get the full name of the title correct because we got…

Simon:
Everyone struggles with that.

Andrew:
I believe the title is Integrating MariaDB Catalogs with PHP Frameworks.
So really long, complicated title that probably doesn’t mean anything to most people. So MariaDB Catalogs is a way of containerizing collections of databases and users inside MariaDB Server.
And this means every catalog is isolated. And this means that you can do things such as have a single MariaDB Server for thousands of websites and have a single pool of memory rather than a thousand MariaDB Servers with one gig of RAM consuming lots of resources.
But also, you’ll be able to eventually do things like resource constraints per catalog.
So if you’ve got a low-tier customer, you can set a number of queries per second or something like that that they’re allowed to do versus a high-tier customer that can do a lot more.
So I think it’s great for cloud and web hosting where you’re sharing a lot of resources.
Now, to use it, you obviously need a way of being able to create the catalogs, administer them, looking at statistics of them and things like that.
App. And this is where this project comes in.
So it’s essentially a framework, a library, you can integrate with the admin systems of various CMSs.
So that when you create a new site off of that CMS.

Andrew:
Created it a new catalog, and you’ll be able to kind of drop the catalog with the site and everything like that.
So that’s essentially what this project is. Was that the whole question? I’ve lost track of it.

Simon:
That was the whole way you answered it perfectly. So was all of that in scope for your work this weekend at CloudFest Hackathon?

Andrew:
That was what was originally intended.
Oh, there’s a story here. There’s a story here, yeah.
And I’ll be telling the story at the final presentation. but so readdb catalogs has not been released yet it is a pre-alpha state we were using code ripped out of founder montewidinius’s basement um in finland to actually make it work so there was no binaries we had to actually kind of create a docker image just so we could actually build the thing and actually use it and then when the guys started using it they managed to break it in ways i wasn’t anticipating them so um we spent pretty much the first day just firefighting bugs and things like that so we could actually get the thing running so i spent the first day thinking we’re never going to get anywhere but even if we don’t then watching the team use it is still feedback we can use we can learn from it we can keep moving forward and use that you know it’s still really valuable as it happens in day two we completely knocked it out the park well i didn’t They did.
We got a foundation that worked. We managed to actually build the PHP framework.
We’ve actually got it working with WordPress single site, multi-site, WPCLI and Laravel.

Andrew:
Four GitHub repos so far and possibly a fifth coming.
And so we managed to get there, even though we really were not expecting it at the beginning.
And we’ve opened six tickets in the really big bug tracker for all the issues we found so far.
And I’ll be talking to the rest of the developers later this week to see what we can do to make the experience a little bit better in future for people who before we actually release this as a final project later this year.

Simon:
So it sounds like there’s like some follow up work to do still after the hackathon.

Andrew:
Oh, hell yes. Monty, Vicente, Eric and everyone else who develop it are actually going to be here this week.
So i’ll be meeting with them probably later today and going through with them how i observed it being used in reality compared to how we expect it to be used yeah and see if we can tweak things before we actually get a final release out of the of the really catalogs feature because it is actually a huge feature under it doesn’t it doesn’t look it on the outside but on the inside it’s a huge re-architecting of so many parts of marie db to make it work.

Simon:
This pretty much covers was it thank you for taking the time that was very insightful.

Simon:
Enjoy the rest of the hackathon …

Andrew:
Guess 45 minutes or so yes thank you thank you very much.

Lars Gersmann introduced us to his project “JSON Schema Field/Form Renderer” for the CloudFest Hackathon.

Transcript (auto generated)

Simon:
Okay, Lars, what is your project called and what is it all about?

Lars:
It’s called the JSON Schema and Forms Project, and it’s all about the WordPress frontend and JSON Schema.
Right now, frontends in, for example, the settings pages and so on are handwritten using PHP formulas.
So it’s complicated to adapt, for example, another style or even Gutenberg controls and so on. And the second thing is, because it’s hard-coded, you can change the WordPress admin dashboard.
But if you used an abstraction level like JSON Schema to define which settings are required by a form, and the settings can be rendered to the form, then you have two benefits.
You have automatically generated beautiful-looking settings pages and stuff.
And you can even change the look and feel maybe in five years or ten years if the design changes.
And even integration of plugin setting pages or even the WordPress setting pages can be adapted in different systems.
For example, in account pages from hosters and so on.
But you can use it also in Gutenberg.

Simon:
So what is that sounds like? That’s not all. But it’s some of the key points.
It sounds like this is a project that will go on for some time even after the hackathon. what is it you plan on shipping by the end of CloudFest Hackathon?

Lars:
Actually the dream is to actually get it into WordPress Core, in some not actually the code from the Hackathon this is just a proof but it’s amazing what we did in the last two days.
So we get a form renderer, we got inside the WordPress and preview so you type on the JSON schema, which is just a few JSON snippets, and you get immediately rendered the form, including validation and all the stuff.
We did some Gutenberg blocks and so on. So the proof says it all works.
And the next steps would be to get in touch with the right people to create clean code, which can then be contributed to WordPress or to Gutenberg code.

Simon:
So your hackathon project is basically a proof of concept.

Lars:
Right.

Simon:
Okay.

Lars:
And it’s far more on completion than just a product.

Simon:
Thank you for taking the time.

Lars:
Thank you too.

Nils Langner introduced us to his project “CMS Health Checks” for the CloudFest Hackathon.

Transcript (auto generated)

Simon

What is your project called, and what is it all about?

Niels

It’s called CMS Health. So we are testing CMS system for their health, basically.

So there’s much more behind that, but…

Simon

Obviously.

Niels

Obviously, yeah. So when you normally test your website, it’s just the uptime check, right? So you do like a curl request and just check, is the server still up? But modern websites, normally, they hide their broken stuff. So it could be that things on the website do not work, but still an uptime check would say, yeah, everything’s fine.

So when we want to have more detailed checks for that, something like, okay, I can deliver data, but the database is down or the hard disk is almost full, something like this. And this was the basic idea, right? 

And then we started to think about, but this is not a WordPress problem. This is not a Joomla problem. This is not a Typo3 problem. This is a problem everybody shares. So the idea is, why not create a standard for that, that everybody can use, and just send the data to any monitoring system out there?

Yeah, so this is a problem every CMS system shares, right?

We now have somebody from the WordPress hackers in our team. We have Typo3 guys in our team. And we want to reach out to every CMS out there. So the idea is every CMS is able to create those JSON files and there’s a monitoring system, a standardised monitoring system that just can read those files and then can return the data or can then alert if something fails.

Simon

So that sounds like a bigger project that will go beyond the limited time of the hackathon this weekend.

Niels

Exactly. Exactly, exactly.

Simon

What is your near-time goal? What is the increment you would like to end up with after the hackathon ends?

Niels

So the basic idea is to have this standard. We are already having an implementation. We also have a mobile app that checks that.

So this is pretty cool. This is pretty cool and it works for TYPO3, for WordPress.

But the most important thing is the standard. I work for a monitoring system for WebPros and we would implement that standard, right? And then it would be great if we already can monitor everything out there, but therefore you need an open standard because otherwise nobody will implement it.

And we will also contribute.

Yeah. And so when the standard is out there, so everybody can take that standard and then hopefully all the monitoring systems out there, like Pingdom, like Koality, like 360 monitoring, like New Relic can take that standard and then build the monitoring on top of that.

Simon

Cool. So this is very much in the spirit of like cross open-source CMS collaboration and then also an open ecosystem basically for like different providers to hop onto this standard. Exactly. That’s super cool.

Niels

Yeah. Yeah. And it tests this commercial part, but you don’t have to do that. So we already implemented an open source system that can do that as well. I mean, all those professional systems out there, so where you have to pay money, I guess they will be better.

But for 80% of all those agencies out there, the open source solution will be enough. So, and that’s important.

Simon

Thank you for the insight into your project. Thanks for taking the time.

Pawel Suchanecki explained the ins and outs of his project “Securing more infrastructure by easing OS upgrades” for the CloudFest Hackathon.

Transcript (auto generated)

Simon:
What is the title of your project, and what is it all about?

Pawel:
All right. So the full version of the title is Securing More Infrastructure by Easing OS Upgrades.
But I would say we use this working title, Easing OS Upgrades, and it is working better.
It’s shorter. There was another part? What is it all about?

Simon:
Yeah.

Pawel:
That’s what I thought. Well, as you imagine, it is about removing the pain points from migrations in general.
But since we are enterprise Linux-centric, so Alma Linux project is now ABI-compatible with RHEL, so obviously we’re attacking this segment of the market.
And while we would like to make the world a better place for everyone, for other systems distributions as well, we’re focusing on our playground.
So we have a tool called Elevate, which has two first letters capitalised and it is enterprise linux of course what it does it really elevates you in a version of your operating system so it’s a migration in place that doesn’t require that many resources and i know that in the cloud computing times we have many ways to upgrade seamlessly because you start another instance do the configuration there and just switch to a working thing but there are cases where you don’t have this privilege you just need to do this right now under the pressure time pressure and of course you can be less fortunate countries with emerging markets and the the new shiny equipment that is pushed as um.

The road to towards the bleeding edge may not be your choice this is the other aspect of the project because our project actually had three legs the first leg was um and it was alldepending, on number of participants that we can secure that we could get during this event so the biggest thing that we had on the list was CentOS 6 to CentOS 7 upgrade.
But this is a huge thing. So the two alternatives were more towards this community, hosting community.
The first thing was to test how, let’s say, a real-world application, WordPress application behaves with our Elevate migration system.
So it boils down to, will your application work after the upgrade?
We thought it will be easy, but we found actually three bugs in it.

Simon:
That’s how life goes.

Pawel:
Yes, yes. But we are very, very happy about that just because we’re able to identify them to actually expedite a fix for one of them. So one will be fixed very soon.
The other is just a known issue. So we know that with the default config, it can happen, but we’ll add it to known issues and this will be resolved that way.
The last one is actually a missing feature in elevate which is about supporting third-party repos because that was the test with remy repos and i enabled so getting back over the the trackthe second bug was with maria db being upgraded to 10.4 and we missed server package there so So let me interject here.

Simon:
What is the target audience you intend to use the thing you built here?

Pawel:
Well, it is the Enterprise Linux users community.
It is hosting community because these are tightly coupled.
And you may be just in the position that you have an older system, forgotten instance somewhere on the internet in the cloud and you just need to upgrade it right now you don’t have extra sources and then you go with elevate possibly if it works for you because it supports migrations from the major versions so well if you are in between you have to help yourselves but this will just make your downtime shorter minimal downtime that’s the that’s the the goal for Elevate in general.

Simon:
I guess that is, that is pretty much it. Thank you for taking the time.

Pawel:
That was very nice to be here.

Anne-Mieke Bovelett introduced us to her CloudFest Hackathon project “Can everyone use?“.

See caneveryoneuse.com for more details.

Transcript (auto generated)

Simon:
So what’s your project called and what is it all about?

Anne-Mieke:
Well, my project is called Can Everyone Use? Which is derived from caniuse.com, which most developers know asthe source to go to if you want to check whatever you’re trying in your browser is going to work.
Yeah. So as an accessibility advocate, I’ve observed that many, many developers actually are criticised for not making whatever they’re building accessible.
Well, no one likes to be spanked over something they’re not aware of, and we have to put aside our frustration about the unawareness in general.

So we are working very hard to bring this awareness to the world, especially because next year in June in Europe, allhell is going to break loose with the European Accessibility Act.
But we do not want to come from no and from fear.
We want to come from yes. So this is why we’re creating this project where everyone can try and see if the componentsthey’re using in the library of their framework of choice, if that component is accessibility ready by itself, and if it’s not,what they can do to make it generate accessible output.
Output because accessibility is always about the output to the front end for assistive technology to read and the site will at the same time educate about the parts that the developer actually cannot influence but that he or she or they can tell the designers and the content creators they work with about.

Simon:
Okay, so your project is, the end result is for developers?
Yes. And it sounds like it’ll be pretty much platform and software agnostic.

Anne-Mieke:
It’s platform and software agnostic, and it’s being run on GitHub.
It has a main WordPress website, because we would like to inform the public in general.
And of course, it needs to be easy to find on the web.
With a regular website, I think we can do more about search engine optimization.
And we’re also going to need sponsors for this project because it’s really big.
There are a lot of projects going on. There’s a lot of frameworks, a lot of libraries, and everybody benefits from this.
So we’re taking it out of the voluntary atmosphere, for sure.

Simon:
So now you’re doing this in the framework of the CloudFest Hackathon, and time is kind of limited this weekend.
So what is your short-term goal? What would you like to end up with after this weekend?

Anne-Mieke:
After this weekend, we have the foundation standing in English.
There are already being high numbers of components being added to GitHub the way we want to see it.
And the website will be also done in the basic form in English, but we’re adding internationalization, localization to it, bothon GitHub and on the web, because there are so many developers in other countries that are not very well versed inEnglish, and this is a problem that we also would like to solve.
So that is the goal we’re trying to reach, and as things look now, we’re going to make it.

Simon:
I’m looking forward to that. I guess a note to end on would be you mentioned that you were looking for funding.
So where should people head to learn more about that?

Anne-Mieke:
They should head to the domain caneveryoneuse.com.

Simon:
Okay, that’s nice and catchy. Good. Anything else we should mention?

Anne-Mieke:
I think we covered about everything. But I would like to mention that I want to thank the CloudFest Hackathon onmy knees for giving us this opportunity.
And I would like to thank my team that has been working relentlessly, giving goosebumps at the speed and the love withwhich they are doing this.
And I would like to thank the other teams that have been interviewed by us by talking to us about how they deal withaccessibility, and of course again the organization thank you so so much it’s great and I’m looking forward to another onenext year thank.

Simon:
ou so much thank you for taking the time.

Anne-Mieke:
You’re welcome.

Syde’s (@inpsyde) Christian Leucht and Thorsten Frommen introduced us to their project “Managing Multilingual Content with WordPress Multisite” for the CloudFest Hackathon.

Transcript (auto generated)

Simon:
What’s your project called and what is it all about?

Christian:
Our project is managing multilingual content with WordPress Multisite.
And basically it’s to reduce the technical complexity of Multisite and make it more available for everyone.
So the normal user, not a technical engineer or developer, should be able to set up Multisite and then manage its content,which is in multiple languages, basically.

Simon:
Multisite mode in WordPress has been there for like, since 3.0 and WordPress, WordPress Coastal, basically like forever.
And it’s always been quite hard to deal with.
So I think I have a slide deck about multi-site and the first slide is, don’t do it, get out of here. So this is what you’re tryingto solve, right?

Christian:
Kind of, yeah. Yeah, I mean, it’s really hard because you need to edit multiple files like the WP config, HD access,server configuration, maybe.
So that’s quite hard to do for a normal user, especially when you don’t have access to the hosting environment, to the filesystem.
Them so um our goal is to like yeah make it easier as easy as possible like just have a button in the back end convert singlesite to multi-site and here we go and then based on that that’s the starting point and if we solve that then the rest is like noteasy but we can start from there right we can set up new languages new sites the back end easier to manage for the usersand it also includes like updating the documentation so the awareness of multi-site um it’s not really there So yeah, we needto teach the people, basically.

Simon:
Multilanguage is in the title. Why does that matter for languages?

Christian:
I guess it matters if you have a lot of content and a lot of languages.
And so you have a lot of posts translated into multiple languages.
And you have maybe even editorial teams, so users who should only access a specific language.
And multi-site is a way to manage that. So you have the separation.
You have user accounts. you can define what role does a user have in a specific site or language context do they haveaccess at all and then you have like clear separation for the content so if for some reason you want to move out onelanguage you can do that because you only have content in that language and content is like posts and pages but also termsanything that you can see on the front end or for administration in the back end if you think about users there’s metainformation for For example, an author might have a biography. You want to be able to translate that.
So it is about managing, administering multilingual content, but also, of course, performance.
So those are, I guess, two primary reasons.

Simon:
What’s your goal for this hackathon? Because it seems to be like this is going to be a project that will continue afterthis weekend.
So where do you hope to end up after the hackathon or by the end of the hackathon?

Thorsten:
Yeah, that’s true. So Multilingual for WordPress Core is phase four.
And we are at the beginning of phase three. So it’s really just conceptual at the moment.
And our main goal was really to have a group of diverse people ask all the questions like how have you used Multisite,what was easy and what is not, and why do not more people use it?
Like what problems do we need to solve or hurdles do we need to like lower or really move out of the way?
So we have done a lot of talking, a lot of writing documentation.
Implementation um and we are like we would like to have like a proof of concepts click dummy maybe even code projectsum but yeah the the plan was really like to kick start this idea um and we also are thinking about writing a proposal for likethe make block for the wider wordpress ecosystem to just get awareness buy-in and of course people um telling us whatcould all go wrong and why why that’s a bad idea.
So yeah, just start that and try and keep momentum.

Simon:
As an avid user of WordPress multisite, I mean, literally this recording will be published on a multi-site.
I’m really looking forward to that and I hope this goes a long way to improve the experience. Thank you so much.

Christian:
Thank you as well.

Matthias Pfefferle (@pfefferle), Automattic’s Open Web Lead, explained his Project “Enable Mastodon Apps for WordPress and its Plugins” for the CloudFest Hackathon.

Transcript (auto generated)

Simon:
What’s your project called and what is it all about?

Matthias:
The project is called Enable Mastodon Apps. And the idea is to bring the Mastodon API to WordPress.
And the bigger idea behind all of that is WordPress is not known for its possibilities to have social interaction and to haveeasy ways to publish posts in a modern way.
Kind of like microblogging, short content, some images, focus on images, no titles, hashtags.
So we thought it might be a good idea to have a possibility to reuse some of the more modern publishing apps.
And the most open and most used app in decentralized communication movement is Mastodon.
So we decided to start with that to profit from the big app community so far.

Simon:
I know that you’re working on the ActivityPop integration for WordPress.
Is this something you can only use in combination with that?
Or could I also use Mastodon apps to publish posts on my WordPress site without federating the blog?

Matthias:
The idea was to decouple both plugins.
They work work nicely together but you could use either or so if you simply install the plugin you see all your posts in themainstream and you can publish new content you can also search and view by hashtags to see older posts or find olderposts of you you could also comment that if that makes sense yeah if you have a bigger blog with some of some of yourfriends and or your family it’s kind kind of small social social network, you could have an easy access with the with theplugin.
And one of the biggest goals of the hackathon project was to make it as extensible as possible so that also other pluginscould hook into the EnableMathodonApps plugin and provide their information or hook into some actions from the app.

Simon:
What’s the final result you want to leave the hackathon with?

Matthias:
The final result would be to make the current implementation solid and working and fixing some of the latest bugs.
And in the best case, we would try to have some example implementation of other plugins, like, for example, a big RSS reader to have the RSS feed as kind of a timeline, social network replacement thing.
So that you could see your subscriptions from WordPress in the Messelein app, for example.
Like similar to if you follow someone on the fediverse um yeah.

Simon:
I’m very much looking forward to that i guess i will be user number one after this hackathon for the very site this this interview is going to be published at i think that about covers it right so yeah i would say so thank you for taking the time thanks.