Jenkins User Experience Special Interest Group, 22 Jul 2020

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Jenkins UX SIG Meeting - 22 July 2020

Description

This UX SIG meeting discussed the designs for the Welcome Screen which is showed to first time users. There was a short update on the Table and Tabs redesign which has just been merged as well as the table to div migration. There was quite some discussion about themes in general and their supportability. Tim Jacomb gave an update on the Black Theme which had just reached 0.0.5. The meeting finished with a short discussion about the future direction now that most of the generic components had been updated.

A

Okay, cool well welcome everybody to the next iteration of the ux sig meeting small group. Today um we have our typical items. Obviously nobody knew today. So we can skip past that and dig right into some updates here.

A

So the first one there we go hey jeremy, uh so the first one on the list here is uh called welcome, screen ui it's under this title of empty states. We'll see why in just a second here, so I'm going to share a brief design deck and then we'll talk about implementation.

B

Shall I take the minutes, joe just uh oh housekeeping notes, yeah, that.

A

Would be very sorry about that? I forgot.

B

Yeah no I'll uh take care of that then. Thank you.

A

You're welcome um all right, so uh the item we're looking at here is the welcome screen ui in jenkins. uh We had talked about this previously. I think real briefly uh how we wanted to improve this. So here is what the design currently looks like. There are a couple of different configurations uh possible in jenkins, depending on the user's permissions, um but here's the current state.

A

You know grayed out all the other elements here, so we can focus on what matters and here's the updated design important to note here that this is very much still a static mock at this point. This is not being implemented yet, and felix can speak more to that shortly, but this is a very open to feedback at this point still too so. Looking at this, some key takeaways here are that this.

A

The goal of this is to improve the usability by doing a couple things, making the style here more consistent with the other components that we've redesigned and also by providing additional context for the situation. Everyone on this call is familiar with this state: it's not an entirely um common state. Certainly, it's not even really, arguably that confusing, but we have an opportunity here to introduce a formal, empty state treatment, um and so that's what we're trying to do and we'll look at some more variations of that.

A

But any questions or thoughts on that before you dig further.

B

We have like an image or just something to.

A

B

A

This sorry, is this not showing up.

B

No, no, no, I mean what I'm seeing here is all text, I'm just wondering. If you know we have a lot of nc real estate. We just show a pretty picture of something gotcha.

A

B

A

So it's a pretty common common thing to do that with empty states right and it offers.

C

An opportunity.

A

To inject some personality in there, um I don't, I don't think you know, I don't see why not, I personally lean toward keeping things more minimal and not overwhelming, especially since the configuration here could be a little bit. You know we, depending on their permissions. We might be offering three different options here with with external links already, I could explore that if that's something we think would be useful or more beneficial, maybe.

B

We could get the chuck, norris plug-in, pre-installed and then have it show its first image.

A

That one I'm not too sure about, um but but you know, but something agreed, could there's an opportunity there, um whether it can be done gracefully in a way that doesn't uh push the the more important details here. Out of the way I can explore. uh So let me let me check into that and provide a mock-up for you in the next save or actually we'll talk about it sooner than that jeremy and maybe like kind.

B

Of a like a line-artsy kind of jenkins figure, you know like a light gray just you know, I don't know just thinking something.

A

B

Something subtle, yeah.

A

Yeah, there's there's an opportunity here, um yeah. I think I think, there's probably not much more for me to say on it right there or for now, except to say that that's worth exploring. So let me check into that.

D

Yeah, I quite like it like it's a lot nicer, I think, than the existing one. um I was wondering the one bit that I'm not so strong about is how it's got like, create an agent configure a cloud and create a job, it's kind of like the first. The first one is two different options, and the other bit I'm wondering about is whether syrian agent or cloud should be as part of creating the first job.

A

Yeah, it's it's a valid question for sure. um My goal here was really just to to to update the style of what we currently have and so to provide those same options.

A

um This could be an opportunity to to reconsider those a bit, but I think that is a slightly larger discussion, but what would you suggest there um so one thing just on the so.

D

The first one, it's quite clear um that you create an agent or configure a cloud, so this one here doesn't really convey that it's one of the first two options.

C

I could see that yeah- um I wouldn't get too hung up in this. I in copy and copy text, because this is just the same. It will need to go through copy editing process anyway. So.

A

Yeah, this is a good point. Thanks felix um it will, it will need to be verified. The copy that we see here will need to be reviewed by by the documentation writers to make sure that it's clear and consistent.

A

um I do see your point. There tim I'm not sure that we want to solve it right now. um The the goal really the scope here is just to it's just to change the style.

D

Yeah, I think I think it's a big improvement um that welcome to jenkins is always bothersome, but it doesn't doesn't feel right.

A

Yeah, I actually I ironically, I actually like the the bit of personality with welcome to jenkins personally, but um but I think, it's more important to provide some context here and say what would actually be here and why are you seeing this message? Yeah great.

A

All right cool, so uh just to reiterate for my own notification, so jeremy, I'm going to take a look at what we could possibly do. There was something more visual uh and tim. That's a good point. uh Let's keep in mind that we're gonna have to review with documentation here about the copy, but I think we're not gonna probably solve uh how these relate to one another right now. This is just more purely stylistic for the moment.

D

Yeah I mean even with the when not to come too much detail, whether you can separate the top and bottom section a little bit with the agents and jobs as separate things.

D

A

Okay, I will take this into consideration.

A

And uh this is something you know. I mention here that this kind of introduces a formal empty state that could be useful throughout the rest of jenkins.

A

So this is in a way just the start of a new type of treatment right that could be implemented in many other places and we'll actually talk about that pretty in depth in the next sig meeting.

A

But that's good feedback for for taking into that consideration. How we could have options relate to one another um more more intelligently than just a stack of items, perhaps in the future.

A

So just like, with the current ui, depending on the user's permissions, these options might need to change. So just a different iteration of this is just the option to create a job or an agent, or even depending on the sign in state the options to log in or sign up again not trying to introduce anything different in functionality here, just an adjusted style from what currently exists.

A

um And I think I may have- I may be repeating myself, but really important. Takeaway here is that just like with the rest of this initiative? Up until now, we are trying to improve the user experience through improved style, improved design. So it's really key that the interactive states are as consistent as possible, with the other components that we've implemented as well. There's some slight deviation in the treatments here, but this component also exists in a different context within the ui and it gets a different background than say a particular nav item in the sidebar.

A

So it does require some different treatment, but using the same color palette, the same textiles. That sort of thing was somebody gonna, say something a minute ago. Sorry about that.

B

It might have been made, but it's okay.

A

I'm sorry about that, um and the last item here was just seeing it in context. You can see that it relates back to the other improvements uh pretty nicely and so we're overall, just like with other components, making a more consistent, ui piece by piece.

A

Any anybody else for now.

C

If we can move ahead to the implementation, detailed discussion for sure um yeah, so I was doing some tech analysis of this improvement, and I did found find out that this is something that, if we, if we approve and go ahead with this design, it will be tough to support the existing translations for the empty states. First of all, the copies change.

C

We cannot do anything about it, but also the way they are the current uh translations are set. They are the each line is online and the the hyperlinks are in line within the translation.

C

So what that means is that it's going to be tough to actually be able to implement it without breaking translations without breaking a internationalization for existing localists. I don't know if that's acceptable or if or if it's not what an alternative could be. So I appreciate any ideas on this one, especially udm tv.

E

Screen, what do you think it's only for the welcome screen broken or for everybody who is using that concept.

C

I know it's for the work on screen. I would guess it's for everything: okay, the user decided that states uh so.

B

Don't the translations just default back to english? If the text is missing.

C

Yes, but I mean it may be weird for everybody to just see james river to a huge english block. I don't know.

B

If people would be bothered by it,.

C

I wouldn't right.

B

I mean, I think, that's kind of you know just the reality of making changes, new elements that you know new translations will be required and we have a fallback to english and the localization teams, if they're sufficiently motivated it will.

D

This is one of them. This is one of the least same screens by users. It's our it's only seen by the admin when they first install us.

C

Okay, okay uh yeah! I just wanted to mention it then so so if everybody agrees, I can go ahead with the poc.

D

Yeah, I don't really understand the problem.

C

I mean I I just don't want to break all like I'm, I'm looking at them right now, almost er 20 to 30 translations. Look at it! So if it's okay with everybody yeah it's okay by me, I I will go ahead with it, then, and also something that only you may have noticed. I think you, because you maybe it'd be the most interesting that I think this is an opportunity to sort of introduce some sort of card component here.

C

uh We will try to make reusable components, not ad hoc stuff.

C

We will try to do reusable stuff that everybody can use later to fight, so that jenkins can finally have some sort of carp, and um I dreamed a bit so the other day, and maybe I had the idea that, for people with a more skill of extensions point at night, this could be a visit in this sort of screen could be a good place for, for example, imagine that maybe the list of call to actions could be an extension point so that plugins could register another on call to action.

C

For example, the pipeline create your pi and create a new pipeline, or a cloud agent create register your your agent or I don't know if jake could have something. So that's just an idea I wanted to float. I don't know, if necessarily that's a good one, but maybe if this sort of list format could enable us to do it.

E

Yeah one thing: I'm not understanding is that the bottom line is this the title or what is the the idea of the learn more about distributed builds so typically, you have a title for a chord and not a bottom line.

E

C

It's I I understand it as a card footer.

C

And it's more further more bottom.

A

Okay, exactly yeah, um it would be the the equivalent of this uh item here. So if there's a relevant documentation or external link, it exists in in a different place in the anatomy. But it's a card footer essentially.

E

Yeah, I'm not sure if, if this kind of component will be used by so many plugins, because um this is a the idea of jenkins- is to use everything by code and not by clicking around in the near future. Hopefully. So. This is something which is more a kind of wizard thing.

A

E

I'm not sure if we have so much uh requirements for providing such uh wizards.

C

E

I think it's a lot of people are using jenkins configuration as code to set up the whole instance. They don't want to use any ui to configure the jenkins instance.

B

I think this is more for the first time years ago I mean this is not. This is probably not so much if you're setting up your 28th master, but I mean I think the problem is you know if you you heard about jenkins, you download it, you install it and you have the kind of the now what problem you know.

E

B

And this, I think, is a good start. You know.

C

B

That- and I agree with you that I mean if you're experienced you, you won't be using this you'll, hopefully be using some kind of automation or you'll, know what you're doing, but.

F

Yeah to that I I was also wondering then um just for this particular screen. Instead of the title no jobs created, I think that's already very specific. um I mean like okay. We know that this should be where the jobs are listed and it's nothing there. So we want to create a job, but maybe just you know, let's create a job or, let's start building or something like that to be more towards what users can do or want to start doing. Instead of trying to figure out, you know jobs with our jobs um yeah.

F

You know these kind of questions if you're, really a first-time user, like jeremy just said this screen will be mostly used for.

A

Yeah, I think that's, that's, really good feedback. Thank you. um It depends a lot on how we're how we're thinking of this- and this is the first time that we're introducing like this, this type of card right, but if it's, uh if it's specifically an empty state treatment, there are different ways of coming at that, but either way this one does establish how we, how we label these these types of states- um and I think it could be some a title- that's more geared toward something.

A

A little more actionable, maybe a little more positive as long as it's specific um in say in explaining what why this state exists. um So I agree and keep in mind too. This is placeholder text, but um but let me update the map with something a little bit more um appropriate and we'll see what the talks people have to say on that.

A

A

I think that- and somebody stopped me if, if we're not ready to move on and that's totally fine, but I think we may be um for the next item- normalization of different colors.

D

C

Okay, I just wanted to talk about I've been working the past few days I was looking at the use of different colors different grays, especially especially the grayscale scale, on jenkins I found, but also about all of the hard-coded colors not currently stored in variables on jenkins codebase, and it's a bit of some sort of an overwhelming task. uh Basically, uh uh my initial goal was to try to okay, I'm going to get all the grace fitting into these five colors that we have available in our palette.

C

That's yet not possible, because many cases, the designed existing design of circuit and components, depend on that. For example, the job configuration or the new item list option is the new colors that don't look right um so that sort of needs a bit of a redesign and that's okay. So that said, I'm going to probably be landing a pr with some sort of normalization of colors uh or that I think it's.

C

It can be a good chance to see how it will affect and sorry how it could inspire other plugins uh to do it um yeah. It was just an update or what's what's going on and what you can expect and it will mostly be focused on the grace greater scale, great scale and not only all the english success rings and craps yeah.

B

Can you give a few examples of this.

C

Yeah, let me look at it. Let me look and see if I can have.

D

A something example of it is, if you search the css and the jenkins code, there's like so many different grays, and some of them are using the variables, and some of them are just using very slight different.

B

Yeah, but I mean, is this kind of I mean I, I guess what I'm fishing at is. Is this a is this a problem from a code neatness point of view, or is this the user experience problem for the end user, but there's so many grades? I mean I understand that it's obviously not nice to have css with part variables. Part number.

C

No it's more more of the more performer, but in the end it leads to inconsistency.

A

Yeah, it is sorry it is indirectly a ux problem too right, because it's really important that so each each one of those subtle, different grays that we have in the palette we've established here with the sig means something and carries, carries connotations so having that inconsistency does harm the experience. Generally speaking over time. All right, sorry.

C

A

C

Introduce fair enough yeah, so one example uh to continue something I touched was uh light backgrounds, for example, like we have like stripped rows like the side panel, hover effects and that's a light gray color, which is f8f8f8.

C

There were really. There were a lot of similar, several similar light, gray colors that were not the same used across the engines. So what I'm doing is trying to consolidate all of those into using the palette one. The palette version of it um also something that's something I I was actually. I was able to do something that I skipped and I ignored was, for example, the colors of input elements, form elements the borders, because I don't know how to do it without making them.

C

Too, similar to the secondary buttons to the default buttons, and that's something: it's basically something that needs to have proper design attention. So I'm skipping it it's it's not it's less straightforward than I thought.

C

A

Makes sense, I think, unless you have more to share on that for now felix next item, I believe.

C

Yeah uh they will change some of these iterations okay, so we merged the tables thousand tablespoons feedback. Congratulations!

C

Thank you. We had some feedback feedback is spearheaded by yuli.

C

uh I sort of think I implement a few a few of those items. For example, I created a compact variable variant of the tables for the for them.

C

How do I say the project view the main view and I'm wondering if we should how we should proceed with that feedback or because to be I I don't know if we should gather more feedback, more more opinion on that, because uh there are many. There are many variants here, so somebody prefers the white header with with attached uh white header, with the touch like a great light, great header, with a attached. There are many combinations that I don't think not not.

C

I don't think everybody any everybody's agreeing on one so and I'm happy to just go ahead and do any change. So one thing I can do is try to make to add variants in jenkins with css, like I did with a compact option for the big table. I can add white header options for the big table or a light or a lightweight header option, so that would allow table users to sorry plug-in authors, slash core developer users to add whatever flavor of table they want.

C

Maybe that's not the best for generic ui consistency, but they they could have use. I can see how that can be useful, so in general, would this help or any idea how to act on all of these different feedback.

E

Wouldn't it make more sense that we start to provide more than one theme, so we have one theme now we have a dark theme thanks to tim and then maybe we have a theme that has some kind of light borders theme and we have a theme that is some kind of you know black header for table. I'm not sure.

C

Hey to be honest, I'm I I'd like to try to touch chord changes from themes as much as possible. I'm doing well, I'm doing everything because the more we support things, the less we can move on and break stuff. Basically and right now, everybody knows that dark theme is experimental, but if we just get everything into themes, uh so themes should be overrides, but we just still need to provide same defaults and if everybody prefers, uh if the consensus is like a lightweight header, with the touch tabs yeah, I'm fine with that.

C

uh I I really don't have a strong opinion on that one, but I don't know creating a theme just for that who should create it who should host it. So there are many questions there that I don't know.

G

Yeah there could be one path again, it's not something I propose for your to-do list felix, but we have now two team, which is light theme and it's widely used.

G

So if there was a contributor who will be willing to start migrating near to theme uh to the new engine provided in recent weeks, then it could be a convenient way to have another light theme, because it's over the existing theme, which has some concepts which could be adapted but.

C

uh Yeah, but what I don't understand is, I think it could can be confusing. I mean jenkins default is the light thing. Why is there? A need for a jenkins default is a light version. A dark theme feels a void which is of the um darf version of the engines.

C

For example, I don't see how it could be useful to spend personal power just creating a theme and supporting it, because teams need support uh just for a live variation or something that can be done, injecting scores. I.

D

Think one of the main parts of those other things is: they fill gaps in the in the ui or ux experience.

D

It's things like it's, the font, it's the sizing, it's the icons and the icons are like a major part, because the icons and jenkins are very dated, um and some of it was just kind of giving out like a material. Ui, look as well.

A

But within the context of of just uh within the scope of the the header treatment of the tables right, I don't. I don't know that that my two cents anyway, that that justifies pulling pulling that change or that variation into an entire separate theme, that's not to argue against themes in general or anything like that. Just uh just that, I don't think this particular um feedback justifies a whole different theme. I think one solution would be to do to to do what you're suggesting felix and make it customizable in that way.

A

uh For me, I I'm a little more opinionated about the style than like felix and that you're saying you could go with with either one and we just need to kind of I decide I think, but for me what it ultimately comes back to is which approach or which style, which one of those recommendations is most accessible visually for the most people, um and I think in this case it was the darker one, I'm happy to revisit that and make sure, um but that that's really what should be our deciding factor, because, with these style discussions it can be highly subjective right.

A

I have my biases, someone else might have their biases because they're using different tools. So that's all good and well, but it has to be the one. That's most legible for the most people, ultimately uh is is my two cents on it. Yeah. So yeah.

C

Basically, what I, what I wanted to go is: if people, if I'm fine with I, encourage having a discussion with and to see if to iterate the point of iterating of merging, fast and iterating is to continue discussion.

C

uh I just don't know what the best way to discuss those designs are or how to. I don't know how to get rather feedback or how to how to react to that feedback. So I'm I'm a bit at all there.

C

I I I welcome many recommendations on how to upload that feedback, because if the stream object right now is different, separate opinion, I I don't know if I'm explaining myself my english works, the way to be honest but yeah, I I don't know how to gather it or if there's any experience on this before on some sort of similar things of the english project.

A

I mean we've we so I have not been in in the the jenkins picture as long as everyone else on this call. So you know this is a very limited perspective, but we've had similar situations to this, just within the ux, the sig right.

A

When we look back at the header bard the fact that it was certainly an issue of um there were some subject subjective preferences, and it was an issue of compromise a little bit from my side, certainly and ended up being a much stronger, designed element in the end thanks to the feedback of everyone here. um So this this may just be you know it's.

A

No one design design choice will make uh everyone happy about it, especially when it's something stylistic like this. So I think this is one of those where um we just kind of have to that. That's why I try to come from the objective perspective of who, which one is the best for the most people, even more so than which one do I like better?

A

I guess I'm not entirely sure how to proceed on it either, but but willie and tim, I guess I would ask. um Does that sound like a an acceptable approach, given that we have kind of stylistic differences? What do you think on that idea of? Let's just go, for which one is most visible to more people.

D

Yeah, definitely um just we, we don't need two options. We just need one, that's good for most people.

E

But how do we obtain the feedback just from the mailing list or.

A

So sorry, which feedback in particular I'm getting turned around a little bit.

E

Yeah, for instance, to go for white colors or black colors on the borders, so I think in the pull request it was most of the people or half people said it must be black and the other half said must be white.

E

So I think this is something which which is screaming, for we need our theme for that, and we don't need an option that we have one table in this design and another table in another design. That makes no sense for me. I think we should toggle all all tables in jenkins with one switch, and this should be part of the theme I think so.

A

And yeah it can, it can be done right. um A theme can happen. I I know like this is just me speaking personally bandwidth wise. I can't design additional modifications. I don't have have the capacity to do that to support and constantly update a different theme, so that's totally possible, but I think I think, for now the goal is just to decide um which one will be the choice by default, not whether or not that that uh that one style justifies a different theme which it may.

B

Yeah, I think mainly what joe says is right I mean it is a bandwidth. This year I mean yeah and obviously two themes means that you got on pretty much twice as much work, maintain it and test it.

E

Yeah, typically, if we would have used bootstrap, for instance, as framework we, we have 100 of themes available already, so I think yeah. We have really a problem. Writing it on our own. um So I think that it's okay, we don't need to support more than one theme. That is okay for me, so we should use one major theme for jenkins. That is okay, that which one then, which one yeah.

C

Motor wouldn't work because the way it would set up ccc supply wouldn't work here.

C

I can provide you with some variants: send you some screen captures you're getting some docker containers going on. If you can start a poll or something on the on the main list, yeah I can go. I can you provide you with all of those options.

C

uh Personally, for example, I find the white header to be have poor accessibility.

C

The action is, I find difficult on some monitors to see, what's header and what's not if it doesn't have a contrast in in the background color for example, and I think defaults should take into account accessibility. That said, maybe the dark gray is too much. Maybe a light gray option is it could work, possibly yeah, that's more of a maybe that comes in a stylistic choice, also something that can warrant a change on heather color is if we prefer to go to further iterate on the tabs and move ahead and create a detached tabs version.

C

I want to remind everybody that we did not think there was a possibility or we did not. When we worked on the designs, we did not consider the possibility of detaching the taps, because I thought it was a community choice or something such a strong and it was to be. If we want to iterate on the tabs as well yeah fine, uh I will try to provide you only with those variants: containers, branches- whatever we can or maybe just send screen captures and you can.

C

We can get a maid in this thread or something going on.

E

I don't know I don't want to change it on my own, so please uh go ahead. If everybody else is fine with it, it's okay!

E

I don't want to stop your development right now. No.

C

No, I mean what I mean is I have locally some style variants prepared some css changes and some style variants prepared, and I can go ahead. I can just share it.

C

What I want is to see I I should we ch a questionnaire. Should we change something we if we change which option of tabs attachment and header color? Is it uh right? Now we don't have the bandwidth to go through a poll, collect all the feedback over everything, but and that's the hard work, the easy job, the the easy job is just to make the changes, and I have those almost ready. So I I think I also shared some some docker containers with that.

E

Yeah, maybe it's just sufficient to have some screenshots so.

C

Yeah, okay, I will, I will get those for you and you all. If you ask bootstrap theme, plugin thing is we already go? We already have some styles sort of similar to bootstrap.

C

Bootstrap is also changing continuously and.

C

Bootstrap is more of a lower common denominator and that many tools can use having bootstrap just removes many some sort of personality for everything, because everything everybody can smell. Bootstrap.

G

I don't propose how to implement it for the record. It just referenced it as possible option. If somebody wants a challenge because yeah you can take themes from bootstrap, you can basically replace existing templates. Similarly, how we doing dark themes and potentially get a theme plugin which allows to use any bootstrap theme most likely. It will look like strange, but technically, if someone is up for challenge, it's possible.

C

Yeah, but I want to remind you, um yeah, the thing is bootstrap: it's not a stylistic set of style of opinionated style, it's more of a boost, efficient framework. It's a tool uh and a tool does not fit well with the waiting cases done. So it's really awkward to integrate it into jenkins core, really awkward, and it will have no backwards compatibility options at all for anything. So um there is the bootstrap styles, uh there's no such thing as bootstrap styles as a design guideline or something they are just stylus.

C

The default styles of a framework and everybody is supposed to if they want to change. So uh that said, if anybody wants to create as much themes as they want the customizability is there lots of css variables. Everybody can go ahead and change everything.

C

Okay, um so do we want to talk more about this topic, or can we move on because.

G

One thing from the jenkins core maintenance perspective, so we integrated the current implementation of tables and the weekly release. This weekly release, unless something goes wrong, is likely to become the next lcs baseline.

G

So it means that current dark headers are going there by default.

G

If there is a decision by the seek to change that, due to whatever reason, then we need to know about that. Let's say by the end of august, because this change can be reported, but in principle, all stunning changes which were integrated in the today's weekly most likely they're going into lcs.

C

Yeah I'm the way you see it is iteration will happen every version, so we don't either. We can go ahead with. In my opinion, we can go ahead with this and then maybe we will for the next lts or for the something uh we can. Maybe we can try new headers. We can also try new icons. uh Hopefully we will try. The new icons, maybe are revisited header and breadcrumbs, so iteration can happen across versions. uh There are certain stuff that I don't foresee.

C

We will change that that we will remain static, for example um buttons hyperlinks the sidebar, so there are enough elements that are going to be and that I conf I have a high degree of confidence that aren't going to be changed. Yeah, uh okay, if we so yeah my point, I'm trying to think in circles. Sometimes I think we can safely go with this, even if nobody's against it and we don't iterate on the november release no problem.

G

Yeah yeah: that's why lcs? Because, let's say if there is strong negative feedback coming in the community, we still have an option to change it in lcs. Maybe even keep it as this and weekly and keep experimenting there, but yeah. If you want to do that, it's really nice to know about it by the end of august.

C

If yeah, the great thing about this is these changes can also can be safely applied on incremental on the releases and be vapor. If the. If, if we are only talking about variable changes, we can change it for a 0.2 or something maybe yeah.

C

As long as it's only variables.

A

Cool uh thanks for pointing that out, like you know, I think we have different uh within this group. We have different views on on the stylistic preference and that's okay, but since we, since we are at a this tiny, not a very big one but a tiny impasse there, um I think we having having it in its current state, where it's it's more accessible of the solutions to more people is, is fine and you know we can change it in the future. If that is correct, it's the right path forward.

C

um Because we have 15 minutes, that's yeah.

B

Yeah we haven't put on topic future work.

C

Yeah uh future work. I just wanted to share something that yeah. um So at the beginning, we well within claudius. We we all together. We created this roadmap and with roadmap I mean, or this set of goals we wanted to achieve a headers iconography economically, when push down the line. Headers typography footers buttons tables, hyperlinks side panels.

C

I said we wanted to focus on core elements that would affect the whole ui.

C

And we believe we are getting there and we are comfortable ending this first phase of css only changes this to end. This is only changes uh or the css base changes after we maybe do the empty states buttons.

C

Maybe we can revisit the header breadth terms and the table headers, because the header and records needs for visiting, and then I would consider this phase finished. I I don't know. I cannot.

B

See the left is empty states. You said sorry, I just wanna get this.

C

Iconography and revisiting uh heather and restaurants and table headers, maybe- uh and I think we are at a great place, especially if we look at our before and after of where the yankees ui was at january, and now those bigger overarching changes. I cannot think of anything else, that's better, and then we can. We could start looking into deeper surgery, as jeremy likes, to say to more important changes. What do we want to work on? Do we want to focus on on forms because we have momentum working on form?

C

Do we want to work on the job history page? Do we want to so that sort of uh sharing? That's not. I don't want to start a huge discussion on what to do next. I just wanted to mention that we are needing the end of what we want to call phase one and that's okay. uh I lie for one like.

D

C

B

Like to have some closure.

C

And to be able to say we are done, we have achieved.

B

Yeah yeah, just by the way, I'm in the process of writing a blog post for the cdf's blog, which we'll be talking a lot about the work. That's done so far and you know trying to get people to join the rest of the efforts. So I think it's important that we have a good roadmap of where we want to go to next.

B

F

B

Seems like a logical one, given the table to div kind of moves um should we include table to divs on the list of? What's still there.

C

No table to dips is a different initiative. I don't. I.

G

C

Want to I we have this goal, I I would like the the seasons when it changes. We have this.

G

C

This goal, I don't want to have a scope creeper. We just want to be able to have to compartmentalize, maybe and say these different sub projects. We are done with them. uh I think it's better for everybody to avoid having a scope creep. There.

A

Makes that makes.

C

A

Sense and I'm sorry, I'm just watching the clock here, uh tim you added a couple items. Yeah did you want to share a screen for these? um Maybe.

D

For the dark thing briefly,.

C

Okay, I'll stop sharing here. Yeah, I'm done talking the team. If you want to just go ahead.

B

But how are we gonna? I mean, I think it's important, that we do try to define what's next, because I mean I think it's important for well. First of all, our stakeholders at cloudbees that pay your salary, that we have a clear you know pipeline. This is what we want to do next, so they don't reallocate you to some other project, which is always a danger, and second of all I mean I think you know we try want to get people excited about this and to join.

B

If we have a clear road map of you know, we've done all this work. It's great we've proven we can do. This is what's coming next. Come join us. You know, yeah 100,.

A

B

Don't think we have time on this.

A

Call but 100 agreed.

B

Well, maybe we should put it as the main agenda item for in two weeks where we try to think of you know collectively brainstorm.

C

uh Correctly brainstorm, we can also look at the roadmap and we can see if we have ux research capability to see what the actual user user uses actual user interviews to see. What the pain points are. There's lots of ways we can decide, but I agree with this topic for the next second.

D

D

Okay, let's quickly do a brief demo of the new dark thing release, so we released zero zero five. This morning um the main change in it was aligning with the tables change um some now support, so it works with the latest weekly um and without this change, didn't look very good. So that's done and also added the text area handler, which is the drag down and um adjusted. Some text adjusted the dark link, colors and the I'll just quickly show that um so this is the.

D

This is the new alert it used to be quite a light yellow. It still stands out, but kind of experimented a bit and it was quite difficult to. um I think it looks.

B

Quite good yeah, not everybody's gonna, like orange, but I mean it contrasts well with black. I think.

D

um So we've got success.

D

Info uh warning and danger. Thank you.

G

G

Okay, I definitely don't like uh warning scenario carlos, but I guess it's uh the best thing for warnings, because you may want to get rid of them.

D

um Yeah so 100 open if anyone wants to do any experimentation, so the theming capability for it's now in jenkins core. So anyone is welcome to fiddle with the colors there um more than more readable now than they used to be. um But it's something.

G

D

Yeah I started with this.

A

I I think it might be worth this is nitpicking tim, so feel free to tell me now to not. um It might be worth bumping down the the brightness on the orange and the blue just a bit so that that white text is a little bit more legible, legible and it won't look quite as as appealing, but but this is definitely an improvement from where it was so good stuff.

D

Yep yep, you can certainly give that a try.

B

Cool, so that's that's the alert how the hell they're making a screenshot. Thank you got it.

D

uh It's alerts so the next bit was the.

D

I shouldn't pick this page, but it's almost.

G

G

Yeah one question about the left's um buttons: do they have colors suggested toilet colors or are they independent? uh The buttons haven't been changed at all. They just use the standard okay, um so basically it will be darker blue buttons on any warning screen.

D

C

Which is the key, I guess it says same as in jenkins. Yeah. There is nothing there are no, but other button variants at all.

D

That's why I was asking.

G

Because, after that, I may have missed this change.

D

Yeah, so the next bit is this little handle that you use to drag down text areas is now themed and matches the rest of the input.

D

Colors, um the dark link colors changed from a the blue primary color to the um to the regular text. Color. um It's mostly was done because the blue primary was very light on the timestamp. Here it wasn't very eligible, but now it's quite a lot easier to read.

D

I've rebased all the changes against the pipeline stage, viewport request. um I think I must have made a mistake in the red here, but apart from that, um it looks it's looking far better yeah. Definitely.

B

It'd be great if we could get wow in as much as the pipeline stage. View looks good at all, but that's a different problem.

C

Looks better looks better than in stock. True.

D

Yeah I mean it's used a lot um yeah.

B

I know it's used a lot, so I mean it's something that you know, even though we're going away from it would confuse a little bit of love. I think.

D

Yeah- and I think is that a text thing contrast yeah, so that's it so not very much uh in there, but I'm just mostly paper cuts just trying to improve it and the junit plug is still open with the for the junit results, um so need some review over there as well, um and the I think the e-charts and the um bootstrap api plug-in are all merged and released. um So each hearts work out of the box and I think so warnings ng um looks fine.

D

Now um so yeah warnings in g was fixed, but that wasn't- and there was no change to the dark theme plug-in for that, um and the cloud-based support um plug-ins also been fixed as well. um This has been a few issues closed recently.

D

um The questions.

B

Jen, if you don't mind um what changes were made to the core to for better support for them ability or word only because the last thing I didn't quite follow.

D

In this release, or in general and.

B

John well, in general, for the black theme.

D

In general, it was um changing from um css changing from sas variables to css variables so that we could change them at runtime. I've got an upcoming post for the cdf newsletter, which I can share with you in case you're wanting to include any of that in your update sure.

B

Share it, and I will read it at least yeah.

D

Yeah, okay, um but yeah, so basically it was changing from sas variables to less and overwrite and adding variables where there wasn't so some places weren't using it. That was some of the gray that we were discussing earlier.

B

It was less not css variables but less variables.

D

You're moving from list to css um and in some.

C

D

Needed changing um bootstrap needed removing from some places.

B

D

B

Kind of just like cleaning up core a bit to make it easier to make.

D

Yeah core in plugins, um the number of plugins have been changed for it as well.

G

Yeah, for example, in cloud, we support the plugin. The changes are allowed to get rid of custom, javascript and other e-bits, which complicate the maintenance, and it's not the only plugin, where we were already able to optimize how ui works.

C

Do you have anything really quick to share about the tables to dips.

D

Yeah really quick to share is that the css pr has merged for it. So both the css and javascript changes have been mainlined. um The css changes haven't been merged back into it. Yet so it's conflicted, um but once that's done, the diff should be just jelly files and very small css and js changes.

C

And are you still going to go ahead with um to try to go ahead with the forward option, sorry to the upping option.

D

uh Face to toe.

C

Yeah or she will just try to power through it, um for the oh, in.

D

August, I'm not sure if I can be bothered doing the feature, toggling approach at the moment. I don't have time, but possibly to just taking a look at some of the earth issues and trying to fix them yeah, better use of the time.

C

Yeah, I agree with that. If we can, I say if we can unlock that layout breaking by issue. We can just match it at the beginning of. Obviously, in my opinion,.

D

Oh yeah, yeah yeah. I think the only way that I can see to fix it is to adjust to change the layout and then change the layout everywhere and then patch the javascript, because I was able to get it to not break by removing one layer of nesting. But then a lot of javascript needs to change, and god knows what the impact is well, if it's well, it's needed.

D

Oh, we just changed a whole bunch of plugins. It's not.

C

D

Break very many of them.

C

Okay, so I think we are almost over the time. uh I think we don't call it a meeting. uh Thank you. If is there anything else? Anybody wants to say.

C

Okay, then. Thank you. Everybody for attending and see you all in august, in two weeks, enjoy the.

B

C

B

Yeah, exactly bye, bye.

A

Thanks everyone.

B