 Okay, so this will be a rough interview and introduction on what happened in the last year and a little bit more even in the past in the Italian community. Now it will be a brief introduction about myself just to give a brief idea of it. The state of the Italian fedora community one year ago and then my initial goals for the last year and what problems we have encountered and what is currently the state of the Italian community and obviously my plans for at least the next year. Now first of all I would like to point out that the fedora online.it that we will mention sometimes it's actually the Italian fedora community official fedora community so you know I sometimes use the fedora online name but it's just the same thing. So a brief introduction of myself I'm a fedora user on and off since fedora 11 and I have been a consistent fedora user since fedora 19 and I'm now ambassador and I also am part of other communities but not very active at least not as active as I would like it to be. Now I'm also administrator of fedora online since August 2015. Now what happened in August 2015? What happened was that actually we had a very old infrastructure the core part of it was pretty old and we were on a shit hosting so kind of performance issue we faced. And many user many very active user of fedora online actually have left recently and even more in August 2015 all administrators stepped down there was like this kind of you know change of generational change and that happened about one year ago. In the general fedora community there are a huge amount of Italians participating in it and this is great but the big part of it is that the majority of those people actually do not identify themselves in you know Italian speaking groups. So we actually have like four people in G11M that have contributed in the last two or three release cycles so pretty much no one I would say. And also since in the fedora 14 to fedora 18 kind of cycles there were a huge amount of translators we have a huge amount of pages and you know strings translated but the problem is that they have not been updated since that age. So we have a lot of stuff that is still you know fedora 11 fedora 12 guides that have not been updated after that. So we have this huge amount of documents that should be updated because has been translated in the first time in the first place but we don't have any more people to actually keep those content translated. So what were my goals one year ago in for in fedora online that is the Italian fedora official fedora community updated all the pieces because we had very very old stuff so it was not very safe to keep that way. To move to private slash private and virtual server because we had performance issues we've seen before and to keep at least the same number of user and posts because you know you don't want to lose even more people than you have already lost in the previous years. And to create a new administrative team because the previous one has just stepped down so we need that one. And also I wanted to to have the inactive stuff people to decade because we actually had some stuff people still around but actually have not been active for like four or five cycles. So the the admins the inactive admins I think it's like not having admins so I think we should have had a way to let admin to decade after you know couple of cycles that they don't perform any administrative actions. So in the general fedora community my personal goals were first of all create a list of Italian users you know just to know that they are there. So the main idea here was that if you have a list of Italian users and by Italians I mean people in Italy actually living in Italy. If they do some kind of event you can actually invite those people to come in. I would say personally in the sense that you send them an email or a message or whatever instead of the more general I write a page on the wiki and probably no one will write about it. And also I am advocating I was planning to advocate more heavily but at least a little bit on a simpler and more consistent g11 and tools that is translation and localization tools because the current state it's not the best one I would say. And also to reduce the number of pages that have been localized in the previous cycles but are not updated anymore because we don't have enough people. Now this is necessary because you know you don't want to have very old information out there and maybe people thinking that are updated while they are not and you know it would not be nice. So obviously as usual things are not straight so you have a plan but the reality it's a little bit different I would say. So in federal line the first problem was that all the software we had we had a website a wiki a planet mainly so four pieces of software that actually were very cooped up together. So we had a huge amount of custom code and both were hooking those pieces up because we don't have a federated user but we have the same user across all the platforms and how it was implemented was implementing in each one of those software a piece of code to actually be able to share the password together. But the problem is that this piece of code is not a model or a plugin but it was just a patch on the core of the product so makes update pretty much impossible to do. And even more the real problem is that the forum that is actually the database base of the whole thing is mainly written in PHP 4 and it has some kind of hooks to work on PHP 5. PHP 4 has been deprecated three or four years ago now so it's very very old kind of code infrastructure and even more the real problem is that the upstream of the forum is dead. So we don't have any more updates we have some patches like security patches some of them has been developed inside some of them outside the federal line but still they are not really upgrades. We have a huge amount of bugs and for instance also the planet is that upstream that too that did not help and so another problem we are now having with the forum is that all those PHP 4 and PHP 5 hooking stuff actually broke the email. So we don't have the emails sent to the users when the forum thread is updated. So we are kind of also losing traction and that kind of stuff because engagement because people do not know when they are not notified when there are new stuff. I apologize this is a specific reference to the Italian community's localized word. Yes because that's what I walked in late at my own fault. No problem. This is about the Italian one but I have spoken a lot with other localized communities and they have the same problems. So you know because in the end the Fedora approach core product has a very strong infrastructure team and development team so that's easier there but all the local communities are usually very small as province said before. He intervened in the Matt question and answer part and he said that there were just 50 people translating Fedora 24 so it's like in all the languages total of 50 people so it's kind of a general problem. And also updating media Wiki it's a nightmare if you have a custom team. In fact I was speaking with Patrick because in the Fedora project we also have media Wiki and I asked him how do you upgrade with a custom team and he said we don't have a custom team because we don't want to deal with custom team upgrading. So yeah probably not very wise to have a custom team if you want to upgrade and also very big problem out there. Time is limited you know and I am not paid to work on this so I have to do it in my spare time so obviously it's very very difficult. Also in the general Fedora community obviously Italian side we were speaking before about you know removing all the translated stuff that is not any more updated but it's still removing content and when you remove content you usually end up with a huge flame in the main list. So this is what happened actually and we saw this like two days ago actually in the meeting on Friday and you know after six months probably that we are talking about this it's not so bad. Also G11N seems to be a second class citizen because you know packaging it's a big thing and also other groups like ambassadors are very very active. G11N is you know a little bit second class citizen or at least that's how it feels maybe it's not but it still feels that way and also again time is limited. So the current state it's first of all we are on a private machine with the city nooks enabled so all the security concerns are limited at least now. We don't still have all the software updated but at least now if someone hacks our planet it's only our planet that has security problems not the rest of the website that is not good but at least it's better. And Media Weekend has been updated we managed to do it also with the custom team that was a user that one of our user in the community actually effort a lot of time to update the team. And also we are ready to move from Bilbo planet that is the one that upstream to the Venus planet which is also that upstream but it's okay because it's a static site generator. So even if you know it's maybe not perfect we still do not have any security problem while the Bilbo planet it's a dynamic website. We have not yet been targeted by any serious attack this is good probably we are not good big enough you know the like the project has been a major target of spam because the project is big but we are not that big so that's good in this sense at least. Also the for me now is completely broken it over the over the months every time it was like a little bit worse now it's completely broken so I'm going to work on this as soon as the the planet part is ended. But it will be probably the only way it will be to move to a different platform because the current one is dead upstream so you know you don't get any update. And the the full migration tool it's actually under QA I'm testing it I have to some kind of script from the web and make it work for us. It's not yet perfect we still the main problem we have is that we have a few users that in the migration lose their name that is kind of a problem you know but out of that it seems to be very good. So we probably would be able to move before the end of August or September maximum. And also we have a couple of new moderators which is good because you know starting from zero or one me that I was new at that point was yeah. And also a lot of parts of our infrastructure use Ansible which is also the standard for the Fedora project infrastructure. We are trying to like copy the Fedora project infrastructure team because I mean if it was for them it was for us and also it makes easier for someone you know in one group to help in the other one and help each other. The two groups so in this is a graph of the post the post is probably our main matrix we can gather because it's easier to gather. This is created by release we started actually at Fedora 3 so it's we have a huge amount of history and as you can see we have had some releases like from 10 I would say to 18 mainly which we had a huge amount of activity and then a downward trend. Now I would say the very good part is that this is Fedora 23 is the first up peak from you know this general trend so I think we are doing something right here and this is good. We are still probably not good enough because if you look down peak it's like you know the down peak here was huge the up peak here it's not so big but still it's a beginning of a trend so I see it in a good way. And probably the fact of having new moderators that wrote new ideas new ways to engage people because the point is we can have the best technical solution ever but if we don't have any ideas to engage people we will not engage people. And also I am a little bit concerned about this data about the actualness of this data mainly because our form is mainly on support this is a fact probably 90% of those posts are about support. So one of the problems of Fedora that I think we have is that Fedora is too good you know in Fedora 10 I remember you had a huge problems with the VGAs and monitors and whatever else. Now we don't have those problems so people do not reach Fedora Italian community but also the other communities because they don't have problems. So if they don't have problems they don't engage well they don't reach us and if they don't reach us in the first place they don't engage. So I think that as a general community and not only Italian and not only localised communities as Fedora general community we should try to find different ways to engage with our users. Because we end up with like Windows user or Mac users they don't engage with the company behind it you know they just are users. But that is not the way we want to have users because we are a community so if they use Fedora but they are not part of the community you know we don't have any advantage I would say from you know the presence of these people. So I think we should really try to find new ways. In the Fedora Italian community we had a couple of ideas over the years. One was a magazine that was actually mainly done by MAGA and it was a PDF kind of magazine so that you know you kind of have a different way to communicate to people. And also we had for a few years some kind of contest in the forum that was generating some kind of draft you can post and whatever. But you know it's still the same problem because you have a very small amount of people that have a lot of knowledge and then a huge amount of people that just come by, ask a question, get an answer and disappear. So I think we should try to find new ways to you know to share new ways to engage people. So in the general Fedora community we actually reached a consensus on what policy to keep to keep pages and not. Now we have partially copied the French community. The French community was pretty strict on this because we do not want to have to translate anymore with the pages. So we delete them all. We are a little bit, our policy is a little bit more you know loose. So we decided to keep between 20 and 30 pages because that is the amount of pages that we expect to be able to manage. Now we have at the moment about 300 pages. So it will be very long now to choose which 30 pages out of 300. But I think it will be you know few months and we'll be there. Could you give an example of some of the pages you're talking about? Yeah, well I have actually a post, a mailing list message that I was going to prepare during the flight, my flight here. Let's see if I can move it here. Okay, so for instance this is my list personally so it's really my point of view. Pages that I would keep, it's like the Anaconda. This is like a page where it describes how to do all the steps of Anaconda. So it's like a user page, so it makes sense I think. Or like stuff about documentation of the product, the distribution, the design. Well Fedora live CDs, how to make a CD or USB drives, the foundations for foundations, a little bit of FAQs, releases, roadmaps. So it's like user part. And then those are the pages that, well we have some pages that are actually in Italian, full Italian users and historical pages like released party Fedora 17 and 21 release party. So these are historical pages, we can keep them because they are not ongoing, they don't require ongoing maintenance. And then we have a huge amount of pages that are actually in Italian, not translated but just in Italian. And we probably will need to rework those a little bit because I would say the majority of those are just old content, double content. And probably we will need to update those and see a little bit about those. And then we have those two pages that are actually in Italian but are guides. So probably we should have those in English and then maybe translated but keep them in English also not only in Italian because they are how to create live CDs and live USB. Maybe there is already a page there but we have to check if we have more material, merge the two pages in English and then maybe translate those. But it doesn't have no sense to have a different Italian way. And then we have those are the pages that I would suggest you delete and are by categories. I'm sorry that it's in Italian, my comment because it would be sent in Italian mainly so it makes sense. But the idea is like all the ambassadors stuff it does not make any sense to translate because you want your ambassador to actually know English. If they don't know English how can they will be able to engage with your upstream community, with your federal community. I mean they will be maybe very good going to people and speaking in Italian with them but if they don't know anything about English how can they bring information from the federal project. So I think all those ambassador pages can just be dropped. But zappers, it's like but zappers are people that take care of bugs. If they don't know English how can they take care of bugs, come on. And then while we have few categories that probably we can drop. Also RPM packages, how to create RPM packages if you actually know English how can you create an RPM package. Because also a huge process to create packages and it is like a confrontation process with other people in the community. And it's done in English so it's like pointless to translate those kind of thing I think. And also translation, we have translated documentation for translating stuff. Now I would say that if you are translating stuff you should know English at least. So you probably don't need an Italian version of how to translate English content. And infra, same thing, it's like yeah it's okay to translate stuff but if you are kind of in infrastructure side, you should know English because you know. And then we have a bunch of stuff about helping in Fesco or contributing that kind of stuff still. So if you are going to contribute in the Fedora project, you know core Fedora project, you need to know English. So I think those pages are not very useful. Then test cases, same thing, if you are performing a test case the goal is find bugs and then open bugs on bugzilla. If you don't know any English you will have problems opening bugs. And you know dealing with the people in the other side saying okay what exactly is the bug, can we repeat those and that kind of stuff. And then very advanced topic like virtualization, ARM architecture, that kind of stuff. I would say the majority of people trying to do that kind of stuff knows English because you know virtualization is pretty advanced topic. Usually you are a CIS admin and if you are a CIS admin and you don't know English, you are not very well placed. Also because in the late 90s, beginning of 2000, we had a huge amount of books that were actually translated in Italian, you know, system administrator books. So in the last few years, not even the big corporation like Coralea or Erplie or whatever, they don't even translate anymore. They just import the English version and sell those because I mean Italy is only 50 million people and people speaking Italian it's like 50 million people. I mean like Spanish, you have a lot of Spanish speakers out in Spain so that makes sense for them to... Actually there are more Spanish speakers outside Spain than in South America. Yeah because you have South America. But in Italy it's not like that. It's like only Italians in Italy speak Italian so it's a very small sector. And then we have a bunch of pages that are actually to be archived because it's like Fedora 12, 13 kind of stuff. So that's mainly the criteria I'm using. It's like if it's something that you need, some pages that you need to contribute to the Fedora project and your contribution will be in English, it's going to translate the process. Is that important? Yeah. Yes? Can I ask about a nuance of that which would be could this community help determine if an English language page should be simplified in some way? Yes. Because a lot of the writers of the English language material are native speakers in English and they may reach for idioms or specific adjectives and adverbs that don't work well on the central language. We have had this kind of debate in our huge debate about keeping on deleting stuff. And the general idea is that a lot of pages are already written in 30 plain English, I would say. Some of them are not. And the idea is to help, you know, making it more plain. Because I would not say simplified English, but at least plain English, not some idioms or kind of stuff. Because it's kind of a problem for non-native English speakers because we learn English in schools. And in schools they do not teach you idioms. So it's kind of a problem. And a lot of people, at least in Italy, have scholastic English because they have never used English outside school. So at least we should keep, I think, I would say one B2 level kind of thing. It's Cambridge levels, but still it's simply not very weird phrases, not very weird constructions and not very weird idioms. But keeping it not... Yeah, the other time is that I live in a non-English community. And a lot of English speakers use specific definitions of words. And in some cases it's the last or arcane definition, which is in American English. So point those things out because the majority of contributors write it in American English. But the problem I would say in the Wiki pages is that Wiki is like a powerst. Everyone can go there, create a Wiki page or edit a Wiki page or whatever. And it's as official as all the other Wiki pages there. And so it's probably something that is to be dealt with from a technical perspective. It's like probably Media Week is not the best tool for that kind of job, I think. But we will see in the next few years if something changed from that side at least. And also we have a huge amount in the list. If you look there is already a huge amount of pages that are already archived. And we have something like a couple hundred pages that have already been archived because old stuff if you want to archive those. So something has already been done. And so my plans for next year from Fedora online, so Italian Fedora community. First of all replace Media Wiki because updating Media Wiki is a nightmare because you got an update every month. And since Media Wiki is the same software that is used by the Wiki Media Foundation for all the Wikipedia's, there are a huge amount of people trying to look for security vulnerabilities in those softwares and you don't want to have an old Media Wiki round. Migrate the forum to SMF which is a simple machine forum, a different platform, way more stable, way more secure, way more updated. Good choice I would say. And then some more people to the staff who are actually three people active in the staff. I would like to see five to six people because you know. And then sort of the bus problem, you know, what happens if I'm hit by a bus? I'm the only administrator. I am the only one with the keys of the server. And if I am hit by a bus, well, the Fedora Italian community will die in one month because the service paid monthly. But I think we should solve this problem with the second administration. But with big communities you don't have this problem. In the Fedora general community, if someone in the infra team has some kind of issue and has to go outside the Fedora project, it's okay. We have plenty of those people. So it's okay. But in small communities, kind of a big problem. And I know that there is at least a couple of localized communities that disappear from the web, you know, overnight just because the administrator did not pay the following month of service. So it's kind of a big problem I think. And in the general Fedora community, push for more, you know, we have this thing where every single community, localized community, has its own rules, its own best practices, its own everything. Maybe we should share those a little bit more. It's like, this is also the intent of my talk. It's like, let's try to speak with each other and say, we are doing this, what you are doing. You know, let's try to find a way to communicate more so that we have more, you know, consensus there. Also, simpler G11 end tools, we actually have some kind of meeting on Friday. And I told them, guys, I will be on the floor next week. So if someone has some kind of, you know, complaints or would see some kind of feature on G11 tools, please tell me so that I can report to people that are actually developing those tools. And I have had a huge amount of, you know, points there. So people are not very happy probably with the actual tools and also the fact that, for instance, we do not actually use Xanata. We don't use Xanata for everything. We have seen some stuff on Transpex. It's like a little bit of a mess there. So maybe if we would be able to simplify things, people would be happy. Also because I tried to, I registered in Xanata at DevConf this year. And I actually had to do it in two days and with Patrick held in between because I was not able to, you know, register to Xanata properly on the first time. And I am actually, I think, a pretty savvy user. It's like, if I have some problems there, what will be, you know, the average standard user or even a low skill, low IT skills user. It's like probably we need simpler tools and with simpler GDIs, you know, more straightforward stuff. And then, as I said before, most standardized best practices are more different teams because it's true that we have different problems. You know, German people are huge teams. We are not so big. There are teams that are even smaller than us. So obviously we have kind of different problems, but we still have the same goal. So if we have the same goal, I think we should be able to find, you know, best practices that fits all of them or all of us or at least partially. So this is my review of the last year if anyone has some questions that has not been asked before. Yeah? No, go for it. We have time. So if infrastructure, and I'm not making any promises on infrastructure as part of it, I'm not saying this is ever going to happen. This is my particular delusional view of the world. If infrastructure were capable of providing, say, A4, AWIC, A hyper-pedia install, that you could then either run the localization that is provided by that upstream or help work with that upstream to get a localization to the Italian. So that you, in essence, didn't have to do as much of the infra work that you do, the admin work. Is that something that you feel like the Italian community would be responsive to or is the act of doing the infra an integral part of being in that community? Well, I would say the Italian community would be happy. But I'm speaking, I have an ongoing conversation with Patrick about this, because I mean, in one hand, the infra team does not want to have an overload of different platforms, different stuff, and I obviously see their point. And on the other hand, we need something different from what Fedora has, because Fedora does not have a forum anymore. We move to ask. No, and what I'm saying is that it needs to be a white software platform. But in essence, we could cook it first. Yes, it would be perfect. You guys, the French, the Greeks, the Dumavres, everybody got the same pieces. That would be perfect. It's also my perfect perspective, but I think we are very far from there. Because first of all, I was speaking with Patrick about ask, specifically ask. And there are some translated version of ask. But if you look at those, they are not really translated. They are like half big kind of solution, because the way we are doing it now is we have one installation of ask. And then when you go to different localized version, you have a different template, and that is why you have different strings there. And you have a different query in database that will only query, in theory, your language. Now, the query is probably not perfect, or ask is not perfect, or somewhere in between. So you don't actually get consistent results, I would say. So, yeah, that is one thing that we are working on. Also, as a Fedora online, so Italian Fedora community, we have a problem. That is, we studied in Fedora 3. So we kind of have a very strong name, at least in the Italian community. So it, for us, is okay to be more closer to the infra and general Fedora project. But we still want to have a little bit of naming there. Consistent names, or consistent domain names. That kind of stuff, because we are known locally with that name. I would say I don't want a radical implementation, but what I'm hearing is it would need to be something that the community, the Italian community, could still brand for themselves, and could still say, our region really prefers wikis instead of asks, or emails instead of forums. And so that it's not, you must take these three pieces and use these brands. Exactly. Because, I mean, obviously from an infra's point of view, this means more work. But I think that a good deal would be like, I am actually the only one-half people working on an Italian infra. And we are okay joining the infra team. So it would be, yeah, you have a little bit more work, but also one-and-a-half person more. So it could be some kind of deal this way. We are trying to find a deal with them, too, so that it works nicely. And also here, I think that from the other localized community, should be some kind of reach there. Because if it's just Italian one, okay, we are one-and-a-half people. But if it's like maybe the French one, the German one, and other communities as well, it may be enough people to have this whole infrastructure up and running and working. When you say that there are very little group of power users and a lot of just people who make a question and disappear, maybe attracting users that... I don't know, maybe they use the computer for designing or doing child's kind of things. Maybe they won't just... Oh, my driver is not working. Ah, thank you. I mean, something in the middle, not in the middle between just user and power users. Yeah. Well, I would say that if we are able to integrate more with the Fedora infrastructure, this will arrive because, as I said in the beginning, we have a huge amount of Italian speakers actually working on the Fedora project, but not in the Italian side, I would say, but just in general one. So if we are able to get closer to them being more integrated with FAS, for instance, the Fedora account system and other stuff like that, probably it would be easier also to get some of those people to also help out a little bit here and there. And some of people in marketing and websites and packaging and so forth are actually Italian. But we never see those in the Italian side of the project, so in the IOC or forum or whatever. Could it happen that they don't know the forum? I think they know. This is my perspective, I think they know, but they prefer to stay in the international side of it. Yes? Not into question. Okay. At first, you may have this in your first slide, and I apologize for being late. Is there a rough estimate of the size of the community? The Italian one? We have an estimate in post. I saw your post slide. Yeah. I wanted to ask about that. I didn't know for you how many people that was, even if it's transients, how large is that? Well, we have on our forum, there is a little bit of statistics in the... Okay. In the end, it's 8,300 users. But this is the total amount, you know, because they created 235,000... Some of the people who, after Fedora 10, you made the number 10. Yeah. Exactly. So my second question was, does this community roll up through ambassadors or conversations successfully with Fedora and the Council or the core, or does this community not feel like it's got a great path to get into conversations centrally? Well, yeah. Now this is kind of, you know, weird situation because in Italy, we have 19, I think, or 18 or 19 ambassadors actually written in the Wiki. Now the active one are like three people. I am one. He's one. And there is another guy that is actually active in this project. So it's like... And all the others are disappeared. So all the people that could be there to bridge out are there. But, you know, we are still very small... So in practice it could work. Yeah. Today there are challenges. Exactly. Okay. Thank you. Thank you.