 Well, yeah, the problem is these are statistics, and one of the things about statistics, and... Well, that's why the study is called that one. Exactly. Okay, great. But, you know, there's, there's, there's translation of messages that the user sees every day, and messages that they never ever see. And, you know, I'd like to just have some kind of a... It's a hard problem because you have to look at the desktop and say, what actually matters, and, you know, you know, the duration of my lines. Yeah, okay, so it's good enough to maybe tell, you know. Well, it seems like that should be a pretty coarse-grained qualification as well. The difference between 50% and 90% matters. The difference between 90 and 95 really doesn't, especially since we have a great team of translators. Right, and I mean, we can already do this for DI. We pick, we go in and we say, well, these languages have, you know, a translation that's like 95% of the important messages or whatever. And we figure out, well, you know, and then these ones are down 50%. Well, I'm going to try to offer them to the user because they're not going to be usable or so. I think we can do that anyway. And it's definitely going to count it that way. And then you can count things that way. And then, so the last one would be... Just a bit of internationalization. Yeah. Usually, if there are a lot of different people who seem to care about what do new strings are internationalized, but doesn't they go into consideration to software itself in order to consider... Oh, no, no, no. I'm really only interested in the actual software itself, the packaging. There's really very little of it that I think the user would see if they're installing CQL desktop. I'm guessing. The important thing is the actual menu items and health packs and all that kind of thing. That's what I'm curious about. So, yeah, if somebody wants to clarify that on the page, that would be very simple. Are you nodding because you're doing other things? I'm nodding in front of our minds, I might get you. Okay. You found it? Yeah. It's a page that is at the top. Do you know how to start? I will fix. I will let you see if you're... I'm trying to remember the screen message. I see. Yeah, I'm just going to have to solve it. But if you get to the top of this page... Oh, is it... Sure, right? Is it new? Yeah, you can just see it. It's a game based on lines. Use the size that's there. Yeah, that'd be nice. That's not any bigger at all. No. Okay. So, yeah, the other one down here is Media Size, which is its own can of worms that I tried to word it just to say, is the CD team happy with Media Size? Is the live team happy with Media Size? I don't care what media you pick or what size they are. I just want to know. Yeah, we're first but not for me. What we need to be making a decision on isn't just does it fit on the CD, but which particular image size should we be pushing the people down with? And it might also be, well, we would like to provide a very small CD with a full, useful desktop, but it might not be the one that's the default if you just go and get the DVD. That would be fine. Yes? In this context, I think it's important also to consider not just in terms of physical media, but also people who may not have such good internet access who need to get security updates and maybe doing net installs. So, we're not just talking about physical. The physical restriction of a particular media size is almost a proxy for a bunch of other costs that come with larger... It's too large of a cost to consider. I don't know how we could make that an item. I would be happy if you just mentioned that as... because otherwise you'll get the objection that you shouldn't be considering media size if people don't want to download it or something like that. It's typical, wouldn't you, either? Yeah, I know where they are. I'm going to say Toadu and... Obviously, a bunch of other multiple people editing the wiki at once, but to be honest, as long as the days went up, we can deal with the complex after. Exactly. So, yeah, those are all the things I've been able to come up with. I came up with all of them. I mean, that's all the things. Following up on that one, so it's not more again. As Jerry says, it is entirely feasible for us to have multiple different reports depending on the architecture, depending on whether it's online or in-store or the size of a media that we want to push people, whatever. It's possible. I don't necessarily convince it's always a good idea because it does cause conflict between people down the road. So, it's possible, but let's think about it. Yeah. I guess actually there is one other thing that I kind of left off of. I have been thinking about which is something that the community mentioned in their message, which is basically, you want to have some kind of a continuity between releases if you kind of, we're not changing it. It's maybe a good idea too, if there's so, you know, if it's not a big change anywhere from 2 to 3, which I don't know it would have mattered if we had to stay with it or not from a community perspective. But, you know, there's no point in flip-flopping that with all the 32 things, you know, constantly. Yes. So, maybe that's something to put on there. Can't think of much of many other topics in the middle there, so if somebody has any thoughts. So, going on, I mean, sort of another one. Sorting out the choice of desktop and whatever it has to sort itself out. Ah, yes. So, I wanted to talk about that by the time, but I wanted to get this through this first year. So, I'm going to be really controversial now and say that your list should contain on it support for non-system data. Okay. Well, I think for other architectures, we do need that. So, also, non-system data on a handbook is 64. Okay. Well, for example. This is clearly something that a few people want. This is the default desktop and the default install already has system data. So, it's kind of a, if you want to install non-default, then you're probably picking the stop anyway, aren't you? I would echo a related comment there that the default isn't what everybody uses. The default is what people who don't necessarily have the experience or the knowledge to change the default should be. This could be clear. This is for people who don't know what the work is going to do, KDE, XFC, LXD, X, WIPLM, et cetera, etc. And that set of quotes probably doesn't want to be choosing a non-default. Don't care what they need. We're not going to, we're not moving. Okay. Okay. Thanks. You made your point. But I have a second point, which is post-installation flexibility and configurability, which is something that it seems people do care about. And apparently there are, you know, I don't really run any of these things, but apparently there are people who have views on this. Oh, you mean of the desktop? Right. So, yeah, there's nothing here about how old the desktop work is. You may use your UI approach. Yeah, I mean, there's nothing here about how configurable is the desktop, how old is the word, how popular is the desktop. These are things that we couldn't put on here, I suppose, that they're awfully soft things to try to get any kind of, you know, you can get a gut feeling about them. Okay. So, in that case, what I'm saying is, please don't make this decision based only on measurable criteria. Okay. Well... Because, you know, that way lies management. So, the problem here is if we, if we just do go with the gut feeling, then people are like, well, I'll test some of the data or Joe, he is just making some random choice in there, or rather, you know, bad magazine articles or something. And if I go with one of our material, obviously leaving out data, we're leaving out gut feeling data. So, one possible approach to this, two approaches I would suggest that you might use to deal with the problem is explicitly list non-quantifiable things on your Wiki pages, things that will be taken into account, but which we don't consider measurable. So, we're not asking for a report for, or which I'm considering. Another thing you could do would be to treat this page merely as an input that produces, you know, like a acceptable or not acceptable flag, and then choose somehow it's for the remaining set. If we get a set of more than one items that are acceptable, then I think there's going to be a decision that's made on some kind of other basis, because, you know, I don't think this page really says how this data is going to be used, just data that I think would be good to have, and I think it's good to get input from as many teams as you can who are related to it and somebody has to make the final point to answer whatever. I think the analogy to the architecture requalification process is actually relevant here. We're not necessarily trying to rank and pick the best. We're trying to say these pass, these don't. Exactly. Which ones actually pass? Yeah. Yeah. Exactly. We don't have a concept of which architectures are better than others. We just have what's on FTP. I mean, that doesn't work, and what isn't. Well, that's actually true, but we certainly have, you know, processes in place based on the ADC that's currently passed for an audit. It would be more popular than this, and things like that. Yeah. So, sorry. I'm not, like, familiar with the, you know, process, but for the time frame we're picking up, the time frame we're picking up. Oh. To make this decision. Yeah. Well, we have to do it for the freeze or we have to do it for half a year. We are going to be making a decision in the next four to six to eight weeks. Yeah. So we can actually get all the implementations before the phase. We have to have, you know, documentation, images, sort of, all that stuff. So it's sooner or better. I'd like to, if I can find the time, start building the teams during demo and get their feedback. And if you're thinking, if, you know, the project would like to get involved and produce statistics for graduation data or something like that. Yeah. I guess we couldn't try to do that. I guess I'd rather just let the, you know, the team figure it out. I'm sorry. I mean, is that the end? Also, through the page, I mean, it says, almost all items are, I'm going to refer them to every part of the page. System D is a part, because it's basically those civilian, therefore, mixed up batch students on that. That's kind of generic. And some other items, I'm going to ask you a question. Right. Right. That's a good question. Yeah. I have more factors. And, you know, and you can generalize system D and say it's something like, you know, does the desktop integrate well with that? I just wanted something specific, because clearly, system D is one of the main topics now. But if anybody has another idea, please raise your hand. A lot of those would fall under that first category of packaging of, does this have, you know, RC bugs or important bugs? Is this horribly broken? Right. Some of them are more integrations. First of all, maybe it's not something just packaging, but maintenance. In fact, do we have a reasonable set of packaging and maintenance to keep it going? Right. As opposed to the technical set, which is further down, does it work well with X, Y, Z? Yeah. The packaging kind of, you know, there has to be a maintainer team where they can't report all those. But yeah, I think, yeah, I think you have to kind of factor in the quality of the team, too. But also my feeling is, I mean, I know that the XFCT was a small team, but does that really mean that they couldn't be the default desktop, just because they're overworked? I mean, I think we have lots of small overworked teams and everything that's still managed to produce something. It's a default. Yeah. Without really advocating the content, I wonder if you'd like to be on the desktop's developers, by including, as anything, no criteria for evaluating them, irrespective of the user's privacy? Well, I didn't get that. The idea is, could we make more of the criteria that the desktop respects the user's privacy? So, the team talked about the message that we got from the developers, talked about how they're working to do that, but I don't know any details. I don't really know how you can quantify that it could be a, it's obviously a goal of streams that are doing stuff. So, some examples, we're repeating, for example, a protection toward a sandbox that will be managed for isolating things. You know, maybe integration with things like this, or not sending the user's keys through a system server, and you use it. Right. Right. Anyway. Excuse me. Yeah. I think if we do that, we can have a privacy section and basically a checklist, and it'll be, like, disappointingly long, but that's fine. It'll just be like, within the privacy area, how much of these does the desktop actually do or not do? That seems, that seems like a pain, but also it seems argumentative too. Or we could just say, are there any, like, have there been privacy related bugs or something like that? The problem with that is, then you end up at the mercy of, you end up at the mercy of the really subjective evaluation of the question of, did anybody bother filing a bug about it? If it happened to be true that LXDE woke up one morning and added LXC jailing to every application and removed user agent strings from web browsers and so on and so on, no one would know probably. Certainly no one would follow a bug related to it. Right. Sure. Yeah. Well, I guess we could put privacy on the table on all four of us, hold on. Personally, I would advocate to figure out how to, yeah. Based on the number of different axes there from affirmative efforts to improve privacy to just don't have any crazy bugs, I'm wondering if, rather than making it a call of aid grow in the table, would we just put it under that set of, here's some soft criteria we should consider. Yeah. I don't think we have a part of quantifiable space to talk about which desktops are more private than others. Well, you consider them a checklist. Sure. But that checklist is going to look a lot more like, here are the features this desktop is working on. Exactly. The features that desktop is working on. It's hard to compare this item as much as a checklist. Exactly. Yeah. Anyway, at some point that checklist starts looking like, looks like, you know, more looks like KBE, but also it sounds like it's a whole lot of options. Do we have a, I mean, we don't really have a team in Debbie and there's the private C team. We have various, what, we could talk to the board. No, I'm not on the private C team. Huh? You know, I'm actually not on the private C team. I mean, the private C team can maybe use a little more heat to inspire them to maybe work a little bit faster than I think they can do. Perhaps we have a private C team and nobody knows what's going to happen. So, I have a question from IRC from Peter Van Halsten. Oh, yeah. He's asking how about making it easier for the user to choose among these desktops. So, I guess, yeah, there's not any other things to add here. I do want to get to that. Peter, I hope you can hear me and I'm sorry if you can, but like I said, yeah. Okay, yeah. So, yeah, I guess we can go ahead. Does this, before I do that, I just want to know if this basically makes sense to anyone. Is there anyone who this doesn't make sense to, who thinks there's some better way to do it or would you just do it more obviously and there's no possible ways you know. This is a really useful way of doing it. Good, good. Russ came up with the idea, it has to be a good idea. That's not good if I was totally, but yeah, this is a really working idea. And then to be honest, if there is anything that any of the various desktop teams think is really important to them and we haven't thought of them, there is scope for them to at least give a summary at the bottom of the page. Yes. And we can summarize and work through it later, but we need to have a central place for tracking the data and that's exactly what this is. Good. Go ahead. Yeah. At the risk of being mildly controversial, is it worth taking a stronghold of the room? About own choices. Yeah. Do you want to throw the camera off? I don't want to throw the camera off. I think if you want to use focus, I'll work with you. What's that? I just thought I'd ask the idea. To be honest, I want you to strongly know this room is self-selecting already. Fair enough. You know, I would personally vote no, all the desktops are crap, let's have none of them. What kind of specialist knows what kind of knowledge cameras are. Yeah, I think the point of having some kind of trajectory is specifically to avoid it from knocking into the structure. Yeah. Fair enough. We will end up using this as a basis for some kind of subjective decision, anyway. Or let's at least have the information available to make a sensible normalization. Yeah. So, you, I would say that one, the architect, the architecture, the reconfiguration process is not about choosing one. It's about choosing many of which are. Now, the question I have asked myself is, with those questions answered, are we going to be able to choose an example? Well, except with what did you say? At least we'll be able to choose from a set of desktops that go to criteria. Yeah. We will find out if any of the main desktops people have mentioned are clearly not suitable for some reason that we may not yet be able to deal well. So, we will be able to drop some, but yeah. Yeah. So, you know, the stuff that we've chosen. To be honest, if the people on the desktops we haven't yet listed, you know, I'll give us the information here. It'll tell us enough information to say, is it worthwhile, Joey, me, various other people putting the extra effort in to make sure there's an extra CD value for those people and all that kind of thing. Yeah. This is all useful information before the release, so let's have it, please. Yeah, definitely. So, there are many candidates for people who text who have some desktop manager who wants to pretend to become a desktop environment, meaning something altogether. Maybe on this page we could have a checklist of something which says to be able to go to this list. You should have what you should provide certain amount of features but you should have file manager you could have I would rather just say anything that is commonly accepted as some kind of a desktop environment and that is generally thought to be broadly enough usable you know, consider. It's a tradition that we've had by this. Yeah. If something does come up saying I'm advocating that you own desktop and I think we have all of the bits needed then to be honest let's talk about it. Yeah. We'll be back to when it happens. Okay. Okay. Oh, yeah. Do you plan on doing this every stable release? Yes. That's the idea. Oh, do you envisage this cutting process being stable? That's why it's about this. A comment on that is that if we think this is something which is mainly for the benefit of people in environments is and every stable release do we come you see where I'm going, right? Yeah. We come to a different decision which would be quite a disturbing thing for someone who buys a laptop every day. It's actually clear that we might want to have an item saying is it simple or it used to be a desktop environment? Yeah. It could be something which is worth explicitly considering. Yes, I would. It's the transition between the previous one and this one. Good. It's quite a disturbing one or not. So, okay. So I think we're about done with this page. Let me then talk about desktop choice and task cell or a rant about it maybe I'll know. So currently task cell when it comes up in the installer has 10 items in the menu which is a not coincidental number because it's the maximum number of items that I'm comfortable with having just because you can get a little overwhelming to be confronted with under 200 or a thousand choices when you're installing them in the first time. Yeah. We used to call people as deselected at the end of the month. Yeah. Right. Actually, I'm actually considering removing some of the items from the very often used and now that we have tasks cell data we can actually say well, there's many people choose a file which is actually surprisingly large number of people to look at it but we actually don't know which ones are doing more tasks meaning if you want to say a few words about it Sorry? Maybe you can say a few words about it. Yes, that's where I'm going. So what I do feel comfortable about this having is kind of an escape hatch to say I'll give you more. So the problem is that implementing that in the constraints of DI is kind of annoying because you're using DevConf and the current menu item is a multi-select and I wouldn't mind having a more at the bottom of it I'd be graded with more button but unfortunately DevConf can currently do custom buttons so it would have to be a menu item saying more and you pick that and then you're plopped in another screen that has probably all the main desktop environments listed here and also blends and whatever which in that list could be 10 or 20 I don't want it to be more than 100 or something here we go let's put it that way so there are some technical problems with that because the user may have selected the desktop environment first or they may not have and this is going to influence what's selected in the second list and also we're close to release it may have to get all this translated and as a designer test doesn't believe done with this in mind so I don't know how to do it so maybe not for Jesse but after Jesse a more flexible task so they would be really nice any time somebody wants to work well of course it's probably like I was saying it's too late for Jesse somebody wants to work on it that would be an awesome thing there have been technical reports I've been for 13 years saying we've had electrical parts and stuff and I'm like well we don't have a widget maybe we do now and you know yeah there's a completely separate discussion for a different off about whether the I should perhaps use DevConf as a backing store for and not a UI even too late yeah but yeah I mean this more thing it seems like it could be done you know within the time frame that we have if somebody wants to do a I don't know if I want to do it if I want to hack them it's probably something we should do as well is actually go through the existing set of tasks as you say we want to do it's probably worth slowing down yeah and then definitely also I mean these are things we should be reviewing regularly but we never get rounds until we're coming to release is work on what's in those tasks you know the file server tasks for a very long time has included Apple Touring stuff but frankly let's please kill it yeah yeah um yeah if you look at the number of changes in tasks all over the past well since January it's been two changes just pretty absurd really that seems worth asking explicitly if someone wanted to come help figure out what should be in task desktop or task something else where would be the please volunteer hero points okay so tasks all maintained by and so so the deputy because the deputy the people who had and and you know the we get a lot of book reports saying change this or change that and it's kind of worth figuring out what the production makes sense and then you know that way so if you're you know doing what's in the past that's always good but there's relations and why the change I guess that so yeah if somebody wants to hack on the more the more item you know which is full code can't be that hard except like I said it has a few wrinkles now a full flow of adding more options in tasks as well is right down to the bottom of this list again with media I've seen discussion from people saying yeah yeah we want to all get into so everybody gets here's the thing I mean they all get in that doesn't mean they're all loose exactly so it might be yeah you can pick open stack you can pick W and science but it's going to go down to the water stuff I think this is a big deal for any of those use cases maybe in other use cases you'll want to have a separate CD and say one of the things that we've done for a long time is and it's typically sure that the things that tasks sell once things are pulled in by the core tasks all fit on the first CD the first DVD however you want to whatever you want to target if we do start adding a lot more potentially way out there packages which the blends people are very keen on but maybe not everybody else in the whole district wants you know the concept of a first CD is dead the concept of a first DVD I think we need to get and know some of the more short story that seems to make sense we do have a concept of what is it essential tasks and non-essential tasks so for example when we come to checking sizes and things or we will make sure that we put all the essentials of a particular task on first and then we fall back to essentially recommend later things to do depending on the people who put together the metadata in the first places they have all the information to say well actually this might be recommended but without all of this data you may as well not bother with the first set whereas for other people very much the recommend or suggest so frankly just sugar that really makes things look prettier but we don't care about in terms of making it install CD right right the definition of at least the recommended should all be installed it is it is kind of not as clear as it could be making it all fit but personally do the packaging metadata that is down here okay yeah pretty much other time here I'm glad that this makes sense to people yeah maybe just comment do you plan to do some architecture may not support this is primarily difficult big stuff oh yeah for for choosing a simple best choice oh you're saying for other architectures yeah well I think what was key previously already used XFC as I recall is that right Steve I think yes it was a lot of discussion on this yeah I've slept well I mean yeah I think that was already the case and I think to some extent it's really up to you know reported at that point although we obviously have this information they can say well we support two different things and we know which one they have before we go and then we'll talk it over and see what we know it's entirely feasible for the default sets of install media to have a different desktop documentation that people look at then changes depending on what you're wanting to do or which architecture you're using that's hard we just need to make sure that we make it clear to people what's happening when they make things because when you say that gaming thing maybe we should try to have consistent second pages best choice yeah so I have another question comment from Peter he asked if there should be an option to install hardware specific packages depending on what hardware is connected to the machine like if you have a TV card install a TV application or whatever see this seems to be where we're getting into don't want to put a new task to solve I guess you could say the task is undoing whatever the TV cards but someone posted a project to maybe and then develop back specifically about that go detect all the hardware in your system and find the appropriate package that's the one yeah yeah I don't know if it needs to be attached to such one that's from previous that's definitely just random yeah so I think that's probably probably so I guess the very last thing before we back up who's driving this before we do guessing because somebody doesn't need to make sure that this actually happens the main reason that I'm unsure if I'm going to be able to work on the more tasks I'm going to try to send out me or some teams I'm not sure so does anybody volunteer to do the more tasks thing the task selection thing because this is the way making sure this happens and without just having hot air in the room here I don't see any obvious happens I am sure it's held not volunteering I've got too much to do already nobody shows up I'm maybe interested to do it I think and then yeah so it will be lovely to have this some of if anybody's listening to the on video or is on our see or whatever please dive in you know talk to us on Debian Boot and we'll and we'll go with this we can't promise that we'll be able to help very much but we can definitely help review things I have a stream is currently stuck so what is the stock scope I can relate to the IRC people looking at the stream but it's just down so what's what's the scope of the task oh that's a hard question the the idea though is that it's a task it's something that you're going to do with your and so you could make it into a mail server or a file server I think the scope of the task is it will be great if someone could make it possible when you're looking at a selection of tasks for example which desktop to install to be able to say no I want something else bring on more right and then see okay so it's adding another tough question and coming up with some kind of metadata to say this is on the board screen or it's on the main screen that's really all we have to join we we we we we we we we we we we when you say might actually help to design I didn't like what it has a short term thing for Jesse for that matter review the sort of existing tests and tests so the 10 that we currently if it turns out there's two or three that there's don't make sense anymore default desktop parent going back to wanting to have the default things in the CAS cell actually make sense to somebody. Yes, I'm not sure about much about those talks, so I'd rather not mention the way of it, and this one way to do it. Okay, I think we're complete that in time, so I'm just going to wrap that up. Thank you all for coming. Thank you.