Episode: 1697
Title: HPR1697: FOSDEM 2015 Friday Night and Saturday Morning 1 of 5
Source: https://hub.hackerpublicradio.org/ccdn.php?filename=/eps/hpr1697/hpr1697.mp3
Transcribed: 2025-10-18 07:53:20

---

This is HPR episode 1697 entitled Fatsim 2015 Friday night and Saturday morning.
One of five and is part of the series interviews.
It is hosted by Ken Fallon and is about 51 minutes long.
The summary is Bradley M. Koon, Karen Sandler, Shryram Ram Krishna, Matthew Miller,
Rich Boen, Karen Bir Singh.
This episode of HPR is brought to you by Ananasthos.com.
Get 15% discount on all shared hosting with the offer code HPR-15.
That's HPR-15.
Better web hosting that's Aniston Fair at Ananasthos.com.
Hi everybody, my name is Ken Fallon and we're at Fatsim 2015 and we're at the Software
Conservancy event and we're talking to you.
This is Bradley Koon from the Software Freedom Conservancy.
People might also know you from the podcast that is free as in.
Free as in freedom with I co-host with my colleague who also works with
the Software Freedom Conservancy.
Karen Sandler, that's faif.us.
Okay, so what is the Software Freedom Conservancy?
So the Software Freedom Conservancy is a charitable organization that helps open source and
free software projects get done what they need to get done.
We take care of pretty much anything a project might need so that the developers can focus on
what they do best, writing the great open source and free software that they do.
So we handle everything you can imagine that might need to get done from trademark issues,
to licensing issues, any kind of legal stuff as well as the very simple stuff of just
handling donations and allowing the project to raise money to fund its developers,
fund trips to conferences.
So anything you can imagine a nonprofit might do for a charitable project,
we take care of for open source and free software projects.
What kind of to just do it themselves?
Well many of them do and they do that for a while and usually come to us to ask for more help
because it's very hard to get volunteers to be able to do that nearer to work that needs to be done
and especially for a software developer who doesn't necessarily know it doesn't have the
expertise and all these different things.
They need to come to a place where we can provide that expertise to them.
I have been to be a software developer myself but I haven't coded in a long time so I wouldn't
presume to jump into a project and so I know how to code for them.
When they join us, we realize hey we need help here, we need the experts and they come to us
to join us so that they can be part of us and join the other great free software projects in
our organization and get the benefit of having one place for all these projects to get the same
type of help they all need.
And say the project rules and they want to go out on their own, is that possible?
Of course, we've had a number of different great success stories.
For example, the Mifos project formed their own organization and they have an organization called
Cosm now, the Groupon Microfinance that has created its own organization after being a
Conservancy member. We don't want to force projects to stay in Conservancy by any means.
In fact, we expected when we started Conservancy that many projects would want to form their own
organ. We thought it would actually be a very common thing where we're helping projects form
their own organ. What we discovered was is we fortunately did our jobs pretty well and most of them
are very happy and want to stay. But when they get large enough and want more control of their
own destiny, they can of course go out on their own and all our agreements with our projects
allow for that and actually we help them do it.
How can people help the Conservancy?
So there's plenty of ways you can help the Conservancy. The most straightforward way for a lot of
your listeners is probably if they're developers to actually just go and contribute code to our
member projects. That's a pretty standard way to help. And the other ways they can help is making
donations to the Conservancy. We're a charity. So if you're in the United States, for example,
you can make a charitable donation. If you go to sfconservancy.org slash supporter,
you can become a supporter of the Software Freedom Conservancy and help us continue this work
helping free software projects.
Thank you very much, probably.
Hi everybody, this is Ken here at Fostem 2015. We're at the Software Conservancy event and I've
just mapped Karen, how are you doing Karen? Great.
So two things that I know about you. One is your lawyer and the other is you've got a heart
proving that lawyers have heart. You know, you're no idea how long I've been waiting to tell you that.
Normally when I feed in my talks, I say I'm a lawyer and I always hide behind a podium in case
someone has rotten fruit to throw at me. But no, everybody who knows how you're involved in
free software. So you're giving the keynote tomorrow at Fostem. What are you going to be talking about?
I am. I'm going to be talking about the identity crises in freedom. So it's like it's called
identity crisis. Are we who we say we are? And it's about the deep complex that people have in
freedom and serve software. I mean, freedom and serve software started as an ideological movement
where people were, and where people were making software for convenience for fun. And then also as
they went to change the world and then create something also that happened to have commercial benefit.
And then in the process of that developing, it turns out that the companies have become so
essential to the development of free and open-source software that we have all of these different
affiliations. And we often don't even think about them. Someone, I've heard many people have
conversations where they say we do this and we do that. And in one moment, they're talking about,
you know, they're sometimes, you know, on a board of directors of an organization. And the next
they're talking about their company. And they're talking about the purpose of the goals of the
free software project and the goals of their companies. And it's all intertwined. And why do we
have nonprofits? And are they important? And things like that. You know, as I became, I was especially
attuned to it because I was a lawyer, but it permeates all of free and open-source software.
It's figuring out how we're behaving, how we interact with each other, and how we make sure that
we do so in a way that's not misleading. And I mean, we know people that change jobs every two
years, but they're working on the same thing for different employers. Yeah, so these are
issues that come up over and over again. And I think that as long as we talk about them and we're
cognizant of them, then we're able to do this really excellent balancing act between them.
Right, we do that because we are in a situation all the time where we have to speak for different
parties. I mean, I am on the Canome Board of Directors as a volunteer. I was elected by the Canome
Community. I used to be their executive director. I am still their pro bono council. I used to be
a volunteer at Conservancy and pro bono council, and I'm actually a co-founder of the organization,
but and I was an officer. I was secretary, but now I'm employed as executive director. I'm a
co-organizer of the free and open-source software outreach program, which used to be called the
outreach program for women. And I have, I do pro bono, I'm pro bono general council of question
operator. Oh, I'm pro bono council, the FSF, like the number of affiliations just goes on and on
and on and on and on. And I'm actually in a situation right now, which I can't talk about in detail,
where I'm negotiating something, where I have an interest on potentially a conflict on three
sides of the transaction. So I am not involved in all. And like even though I'm very involved in
the groups that are negotiating with each other, I can't participate in any of it, because to do so
would be wrong. Like I wouldn't even know who to represent. Like if there are people are trying
to negotiate with each other for something, how do I choose which side I'm more loyal to? So I just
basically, I just bow out of the whole thing, I have to. And so right now there are negotiations
going on between people who I normally am like a team with, but I can't participate.
Not even on the side side. You know, I can do a very minor supporting role, but I have to be sure
that I'm not in a position where I'm making any decisions and I'm not trying to influence anyone
because, you know, I'm a trusted person in these groups, because I've been volunteering for a
long time with a lot of them. And so if I say something, even if it seems minor, at the end,
might have influence over what happens. That's wrong. Do we have enough lawyers in the
foster community? Too many. Why are all these lawyers here? No, I'm kidding. I don't know. It really
depends on how you look at it. The lawyers that are involved in our community do such a wide variety
of things. If I could clone Pam Chestic, who is standing right in front of me right now,
like I've downloaded, she doesn't know I'm looking at her. But if I could clone her and have 100 of her,
I would be really happy. Can't get enough of lawyers like her. And there are a lot of lawyers
that do pro bono work. You know, Aaron Williamson is here and Justin Collin, you know, is not here,
but tons of lawyers that do great work that I wish we had so many more of them. But you know,
there are lawyers who are actively trying to pursue corporate interests that are not any interests
of free and open-source software. And everybody deserves to have representation. And it's
perfectly, I don't have anything against lawyers that do that. But do we need more of them? I'm not sure.
How important is fostering for you?
fostering is incredibly important because it's enormous. It's a huge conference and it's community
rent. And the community rent conferences used to be such a huge part of free and open-source
software. We used to have so many of them and they were really vibrant. And then the commercial
conferences have become so polished and so well organized and such a part of the infrastructure
that now the community conferences have been slowly dying out. And now there are only a few
that exist. There are regional ones, you know, like Linux Press Northwest. And Ohio,
Linux Press does a great in their small geographic areas. But, you know, right now there's FOSTEM
and there's like Linux Conf Australia. Our two is the only huge community around conferences.
And those are fundamentally different than you for speaking. Like, you may have noticed some
of the press at LCA that has come out, has talked about, like, Linux four vaults talked about,
you know, people who say that he's a toxic personality and he should shout out a mailing list.
Like, he, Linux has had that conversation at countless other conferences, but it's always
been highly controlled. Right? And now you have a community conference where you start having
that conversation and you had people lining up to tell like the first two questions of the keynote
session FAQ were asking him about that specifically. And it doesn't happen that way at the other
conferences. I must say though, he did an episode on that. And it was the only episode that I
stopped and deleted. Really? Because I think this, I would have praised this.
I think they, I know, I just want to say that you censored me by deleting me.
I'm laughing. Yes, yes, you're, you're a lawyer, right? I got a shut up now.
I think quite a lot of the thing was cultural, but then since then, I'm thinking we have a lot
of issues and it goes no harm to bring into the forum. Look, I don't want to get into the actual
substantive discussion because I think you and I might, you know, there were some things that
in retrospect, I wish we had explained a little bit more and delved into a little bit more,
and maybe we'll do another cast on some of those things because I've gotten a little bit of feedback
and I think that there are good points to be made. But you're not, but I think what we're discussing
here is not really the substantive discussion that was contained in that particular episode.
So much as it's the fact that the corporate, at the corporate conferences, there's this
interest in controlling the conversation such that it diminishes controversy and makes companies
feel more comfortable with sponsoring and their, you know, and their profitable interest. Whereas
a community conference is about making sure that our, the interests are run by the community and
trying to improve our overall way that we interact with each other and our goals as a community
and as I often with an ideological bench and having a place to have those conversations where
it's not going to get shut down is incredibly important. And so there's a lot of quiet censorship
that happens at some of these corporate organized conferences. And I think these conferences are
fantastic events. I think that there's a place for them. I think that those companies that do that,
they're, you know, they're performing a valuable service, but we need both. And the community
conferences because they're at universities because there's less money sloshing around, you know,
at the beer event, everyone is paying their own way, these events become less popular. And it
falls down because it's such an institution because it's so huge. It's just an enormous event that
has a lot of momentum. And I'm really excited by that. And that makes Boston an incredibly
important conference. Okay. If you could change one thing about the
fast community, what would it be? Just one thing. I think that what I would do is I would,
if I could change one thing is I would, a lot of the problems that we have are historical ones,
like, people are so receptive when I talk about software phenom with an ideological vent.
And I think that we have this, we're coming from this historical place where our community was
very divided. Okay. So if I were to change one thing, I would, I would unify us a lot more.
Because right now, and, and, and where we come from is a place where we have so many options
for everything. And that's part one of the joys of free software is that everyone can do their
own thing. And we can have many, many answers to things. But we are such a divided community on
so many things. So many people, I talk to you think that free software is a fundamentally
different thing than open source software. And we have these arguments over and over again. And it
just serves to make us completely unable to, to really advocate to newcomers in an effective
unified way. And so one of those people or of us? I, you know, I, I think it's really unfair to
isolate one person. No, no, but I mean, a lot of, a lot of people look up to our
armature myself. Yeah. But I, well, I think our message is right that we should focus on freedom
and we should focus on ideology. But at the same time, I think we should, and I've said this before,
like I think that if you say open source as long as we're talking about freedom, as long as we're
talking about those users rights, then I don't care what language we use particularly.
He's right in that if we don't ever talk about ideology and we use loaded terminology,
then we're going to have a problem because people won't think that freedom is included.
But it's a matter of style, you know, and, and I, I think from where, from where I'm sitting,
I just wish that we were a little bit more unified rather than fighting so much amongst ourselves.
Okay, um, was there anything else I missed?
There are always a thousand things we could talk about and we'll have another conversation
another time. We will do. Okay, thank you very much Karen and thanks very much for organizing the
event. It's been very delightful to get free beer and food since the summer of the day.
I'm good luck tomorrow. Thank you.
Hi, this is Ken. We're at the Solfer Freedom Conservancy event and I'm talking to
Shriram Ram Krishna. How are you doing? And what are, what do you do? Why are you here at Fostam?
I am a director at the Gnome Foundation and I'm here to support Freesaw for and of course
the Gnome Project. And for those three people out there who don't know what Gnome is,
that's a new product that they're going to be releasing.
Gnome is the one of the more interesting desktop projects we have for Linux.
We like to push the frontiers. Yeah, so I've heard you were a, okay, there was a lot of controversy
when you moved from 2 to version 3. Is that over now? I think mostly it's over. A lot of people
have accepted what we're trying to do. But in back in 2011 it was very hairy. I spent a lot of
time talking with a lot of people and getting their feedback and hopefully we incorporate a lot of
that feedback back into the project. I know that from what I see in social media that a lot of
people do like where Gnome is heading these days. So I like to think that a lot of the controversy
is over but there's a lot of still a lot of our luggage is still there. But do you think that the
codebase has benefited from it? I'm sorry, what was the main driver for was this old codebase
that needed to be updated or was it a new paradigm? The driver was that the codebase didn't really
give us the flexibility that we needed to explore the user experience. Gnome too was still a sort of
rigid. So if you look at where we are versus Gnome 2 versus Gnome 3, Gnome 2 merged the user experience
with the development platform. Whereas if you look at Gnome 3 we have Gnome Shell which is sort of
divorced from the development platform. Gnome Shell is its own thing. It's written in JavaScript
and we have a lot of ability to change how it looks, how it feels, it juices a lot of the
standard technologies like CSS web interfaces that people know. So if you see if you look we have
things like extensions that actually changes on the fly what the user interface looks like.
So originally it was there so that designers can modify the look and feel and be able to test
their interfaces. Whereas now it's also used for other people to say well we don't really like
the way Gnome did this. We want to do our own experience. And so you see a lot of things where
they override design elements and did their own thing and overwriting the design that we had in
Gnome 3. But if I stand back and look at it we've got you have the Gnome project and then
that's you had boom 2 and off and did the unity thing and then the old code has been taken,
the Gnome 2 code has been taken and been used in the Mate or Mate. Mate yeah. How do you feel about
that? It's not a waste to develop our resources. So diversity is our strength. So we have lots of
people who use, so we have Mate. So Mate looks at it in a different way. We have elementary OS
that also has their own set of designs. But what's important let me just go back to Mate. They're
actually reporting to GTK 3 and eventually they'll be their own project going in their own way.
But I think what's great about all these projects is they have their own idea what
user experience looks like. And that puts pressure back on GTK and some of these common
libraries. It means that they're much more universal. Instead of a lot of people used to
complain that well Gnome has taken over GTK. But when you have all these other projects that means
it could still be so much more universal toolkit as opposed to just being the Gnome toolkit.
So I embraced them. We had a hack fest last year where we invited the elementary guys and we
actually had a great meeting of minds. And we really want to continue at least like in Guadag to
invite people who use the platform. So if it's Mate we would love to have them go out there and
get presentations. If it's elementary OS, if it's cinnamon, all those people we would love to have
them out there and show us their point of view about what user experience is and what new things
are doing that we also can maybe steal or I guess steal is not the right word but incorporate
if it makes sense and maybe they can do the same thing. I will get attacked if I don't ask this question.
How do you think the accessibility of Gnome shell is at the moment? I really can't say I
know that initially it was not on par with Gnome 2 and I haven't talked to the accessibility team
of late so I don't really want to give a status but as far as I understand it's almost on par.
One thing that I've heard about Gnome shell is that you require a 3D graphics driver to run.
You do, it does require 3D graphics but it requires a hardware that's
well at the time when we said it 2011 it was five years old so you know you're talking about
something that's out in 2006 so if you have something that's out in 2006 it should work with
and an additional we actually have software software rendering as well so if you have a
powerful enough CPU we could still do the 3D stuff. It's not going to work on Raspberry Pi,
don't we? I think it might. It won't work on a Raspberry Pi then. I haven't tried. I don't know.
Maybe you might want to try that. Maybe I have to try that. I don't have a Raspberry Pi but I use
it for, what is it? The XPMC, right? Yeah, I had a new one. So what are you looking forward to
tomorrow at the pasta? So I have two talks. Tomorrow I'll be talking about our trademark battle
that we had recently. Most of people remember we had a company that had taken over the Gnome
trademark and we had a battle we had a remarkable, remarkable fundraiser that just went beyond
all expectations. So me and Pamela Chastick who is our pro bono lawyer will be talking about that
and then on Sunday I'm talking about the lot of things we do in Gnome actually puts
pressure on the Linux system. So we actually work up and down the stack. So unlike a lot of desktop,
they tend to stay within the middleware portion but we're known to actually work in the kernel space
and other places because in order to do something that says just works requires to be outside our
our stack. So you know we have people like Bash in this era that works deeply in Bluetooth and a lot
of things that make Bluetooth just works requires it to work in Bluetooth. So a lot of the stuff
is there are improvements in there and then you have things like D bus, we have KD bus that's
coming in there that came out of the Gnome project. You know you have Pulse Audio, you have all these
kind of things that we even even things like Compass, it was the Gnome project that came out with
the cube. It was by Sousa at the time with Miguel and everything. I remember we had we had brought
that out in Oskar. We showed the cube going all there and you know all those things that came
that started in Gnome. That didn't start pretty well anywhere else. And so we actually come up with
these things. Now we abandoned a lot of the Compass type stuff because we really was focused on
user experience. So we're very conservative on not the special effects but it's true that
we're the ones who had kind of explored that space saying this is sort of that compositing
type of thing. Compositing desktop started in Gnome. So you know we're always trying to push
the frontier because when you start with an overreaching goal saying we want to make great
user experience, you're saying how can we do that and how can we make changes up and down
the stack to make that happen? The free desktop.org initiative wasn't attempt to sort of harmonize
the desktop. How is that going? Are you still involved in that? I'm myself in Gnome. We're still
doing that. A lot of the things like DeepBus is still done in free desktop. And I suspect any future
things we want to do. We're basically working on sandbox applications. I suspect that will also
move to free desktop at some point. Not to put words in them but it makes sense that we want to
want something universal as opposed to something Gnome specific, something as important like that.
Okay. Thank you very much for the conversations. There's anything else that you want to
say? No. No. I'm good. Thank you. Are you heading over to the pyramid? I am.
Right. Okay. I won't keep you any further.
Good morning, everybody. My name is Ken Phalan.
Foster M has started. The hangovers are still in place. And we're at the Fedora booth we're talking to.
Matthew Miller. How are you, Matthew? How what are you doing here at the who are you and why are you
here? I am the Fedora project leader. I am here to talk about Fedora and meet people and talk about
their projects and distributions and connect up with everything that's going on. Do you work for Red
House? I do. Yeah. And fortunate enough that they pay me to work on something full-time that I
would be working on for fun otherwise. Can you give us an idea for the two people listening who
don't know what Fedora is? Fedora is a Linux distribution that's been around for about a decade.
It is the upstream basis of Red Enterprise Linux and CentOS and some other things. It is sponsored
by Red Hat but it is a community run and community led project. And you recently had a release
Fedora 21. Fedora 21 out a couple months now are best released yet as each release is of course
but this time I really mean it. It's a very solid release with a lot of good stuff. You broke the
distro up into three different things. Can you tell us a little bit about what they lost
behind that? Yeah so historically Fedora back in the olden days when Fedora started it came from
Red Hat Linux which you could buy in box sets on cells and that was kind of a general purpose
distribution for all kinds of things and people use it for servers and desktop and everything else.
So Fedora inherited sort of a community of users from these different user bases but over the years
Fedora had kind of sort of drifted towards being a desktop as sort of the most obvious form of
the distribution but we still had a lot of system in users and people using it for even in
production server uses and those things kind of pull at each other so we had a lot of disagreements
about what the defaults should be, how things should be made should we even have defaults
when people are using it for different ways and so we decided that having actually different
different focuses but have a workstation release that's meant for you know being a you know
power user desktop and then having a server release is actually meant to be a server as a top
level target would sort of help diffuse some of those tensions because you can say okay this default
makes sense for a server this default makes sense for a desktop we don't have to try and find
something that makes sense for both of them we can each have their own path and but you know
run the risk of the two things divergent and needing more people to maintain this whole thing.
Yeah there's a risk and so we have a Fedora based design working group that's sort of kind of
curating the common core between the things to try and make sure that you know where things can be
shared they stay shared and we don't go too far off but I think there's a balance between that
divergence and commonality that we want to try and get right. Yeah you're only one release in
anyway so what could possibly go wrong? What could possibly go wrong? Yeah it's exactly is this
your first time to foster? I know I was here last year as well so I don't quite feel like a veteran
yet but yeah you know you're aware around tell me how did you end up here what is your story what
did you do? So I worked yeah I dropped out of college and helped start an ISP back back in
the you know dot com days and then my friend was working at Boston University and said you should
come out here I can get you a job so I worked at BU in Boston for about a decade on the Boston
University Linux project because back then at that time if you would go and get you know Red Hat
Linux from you know the store shelves or whatever someone did a study and they showed that if you
put this on university campus Red Hat Linux was the most secure because it was 15 minutes before
it was broken into and owned and everything else you know it was was worse and so our security
teams I worked for the central IT department and our security team was running around telling everybody
you can't run Linux it's not you can't have it on our network it's not we can't we can't we can
handle it and so but we know we love Linux but we so we decided that we would make a version that
was tailored for the campus that had better security defaults and also would tie into like our AFS
file system infrastructure and curb roast and things like that so put together sort of a derivative
distribution that was meant for the university so I worked on that and got involved in Fedora as
the upstream for that for about a decade and you know by the end of that decade those initial
problems you know now if you go by rail you can pretty much trust that security is at least a
concern even if nothing's perfectly secure it's a you know people distributions care about
security now so that need wasn't there as much and so I moved on to another job and things but I
stayed involved in Fedora and when a job opening at Red Hat came up I took it excellent I know you're
here at Fullstown okay well thank you very much for the interview is there anything else that you
want to announce or what are your plans for the coming year no big announcements Fedora 22 coming
out in May we're back on our six-month schedule after a long longer or 21 release so I'll talk
about that tomorrow at my talk a little bit in the distributions dev room I think it said
early afternoon sometime all these talks I think will be online on the Fullstown site so
you can catch that later well thank you very much for the interview and enjoy the booth
it's possibly the coldest booth here right beside the door the best booth perhaps what you're
going to have the flu next week I think right by the door but it's worth a little cold we can handle it
okay thank you very much
Hi everybody this is Ken Fallon here at Fullstown again right beside the Fedora stand is the
CentOS stand and I'm talking to I'm Rich Bowen Irich and you're at the CentOS stand what's
your involvement with the CentOS project well I'm kind of borrowing space for my colleagues here
I work on the RDO project which is a distribution of open stack for CentOS ah so tell us about that
first of all tell us about CentOS so CentOS I'm sure that my colleagues can tell you more but
CentOS is a rebuild of Redhead Air and Prysalanx and a large community of people developing or
producing packages of a wide variety of things for distribution on CentOS so CentOS is a
rel without the Redhead funded branding more or less that's right and you know there's there's
a close relationship with the Red Hat people but it is operated as an independent project and it's
got lots of community contribution from outside of Red Hat okay so what's the RDO when the white
is a decision so open stack is a cloud computing platform and there are hundreds of companies
that are involved in this but as with most open source projects it releases source code and RDO
produces packages for use on CentOS and Fedora and Redhead Enterprise Linux and so we're a
community of people that are that are working on the packaging aspect of it making sure that it
actually works making sure that the packages have all the necessary requirements um don't to run
on CentOS so the idea would be you install CentOS you add repository or these available that's
correct so you add a repository and then you run some puppet scripts which stand up your open stack
cloud because open stack is it's complicated and it can run on one machine or thousands and so
it's not just a question of installing one RPM you actually have to orchestrate all of the different
machines and get them talking to each other and that's done with with puppet scripts and that's
that's the piece that we provided RDO okay so it is very very much tied into purpose um that's
or the philosophy that's correct and and it's called pack stack but it's a it's a set of puppet
scripts that orchestrate standing up the cloud and uh how difficult is it to run uh well if you do
a simple install you run pack stack and you answer some questions and in 20 30 minutes you have
your cloud um and of course open stack being open stack you can make it as complicated as as you
need to and that can be that can be difficult but uh but hopefully the you know hopefully the
installer takes a lot of the work out of that but there's still there's still some decision making
and is this uh is this uh been used quite a lot now or is it still there it is at this point
well we are um we're about two and a half years on in the RDO project and uh you know kind of our
poster child is surn the nuclear research facility and they have got 70,000 nodes running
RDO but and then there are many other smaller installations so yeah it is actually being used in
production a lot of a lot of uh research organizations as well as as well as companies are using RDO
in their in their production clouds and you are you also employed by Red Hat or are you I am I do
work at Red Hat in the open source open source and standards department okay okay since since we
a lot of people working for Red Hat are over here yeah we've got the the cento s guys are under
the red hat umbrella now and and Fedora is affiliated with us and uh we've got an overt table over
there as well which is a virtualization platform that's that's under the Red Hat sponsorship
to a lot for the community I guess yes so we we believe that uh and this is what really attracted
me to Red Hat to begin with that we believe that a strong upstream benefits our customers and so
we have lots of people that work exclusively in the upstream and you ask them how much our
product costs and they say I don't know I don't work on that part of it so so yeah it's really
cool to be working in the upstream so you're going to be able to booth here the whole day or you're
giving talks I'm not giving talks but I am also working on the open stack booths um yeah and then
I'll be working on the open office booth as well because I'm associated with the Apache software
foundation and so I'll be I'll be jumping around quite a bit today okay well enjoy the show
and thank you very much for the interview thank you
I'm just moved on from the cento s project and I'm talking to KB saying and what's your involvement
with the cento s project oh I got involved back in 2004 uh when the project was starting off and
since 2009 I've been the project lead for for centos okay so um and just to remind folks what cento
s's um cento started off by being a community rebuild of the Red Hat Enterprise Linux sources
and over the course of the last 11 years it's become pretty much the de facto standard on web
hosting and high performance computing and basically if you're looking for a community platform
for stable computing centos is literally the default choice at the moment okay and you've been
um you've been brought on to the Red Hat on brothers obviously they're paying for you guys
well I mean the easy way to explain this is that there is a joining of forces as in Red Hat
has come on board as the biggest sponsor um but the project still runs independently we have our own
board um Red Hat has some level of influence on the board but it's not exclusively on Red Hat's
as it were so what Red Hat allows us to do is to bring a few people on board at the moment three
of four guys um who can then focus on the day job as being you know helping the communities grow
on a technical side of things but the community still run independently okay now when we were
discussing this last year that move had just happened and I think a lot of people were questioning
what they what the reason behind it was and people were saying well it's the it's the selling
of the contracts you you get it's like the first sample is free and then we'll sell your Red Hat
Enterprise contract later but I was told that that wasn't the case no that's still not the case um
what what has happened is there's now a clearer definition of what is a centrist project versus
what is centrist Linux as a distribution um and centrist Linux as a distribution is still
carrying on being what it has always been there's no change in that at all now from the centrist
project side of things there's been a lot of changes in that previously it was very hard to get
involved as a contributor it was very hard to set up a trust relationship all of those barriers
have not come down we've got a bigger contributor footprint we've got special interest groups
coming on board that allows anybody with any project you know anything to do with infrastructure
or otherwise come on board with centers consume our resources consume our build services consume
our testing infrastructure consume our release infrastructure things which are not possible in
the past we can now do um one of the key reasons why we're able to do this now is that in the past
we've been a completely donation sponsored driven community and we've never accepted financial
donations escaping the money out of the equation made it possible for us to focus on the
technical side of things it also meant that the people who got involved were people who actually
wanted to get involved um so in the past and as we do now as well we stay away from commercial
endorsements we don't endorse vendors we don't encourage people to go out and you know try and build
services on our centers which aren't themselves open as well um now would that one of the
after effects of that was that we never had any money to spend so we ran our our core infrastructure
ran off 120 machines donated by I think about 116 vendors in 83 different data centers
and across 17 countries so this is a number I remember from Christmas of year before last
because that's when we didn't audit um and I remember sitting down and speaking to some of the red
hat guys at the time and they said you know so what are your assets then we said none
the like what do you sustain you know x number of millions of users and we like thanks to the donors
and the sponsors and every time the donor or a sponsor comes on board we make sure that they
understand there is no transfer of ownership they retain ownership of their assets they retain
ownership of their hardware of the software that they're giving us and everything and they allow
us to use it for community purposes now fast forward year down the road red hat has budgets
so that what that allows us to do is now justify requests for money where we can spend things on
things like setting up a large community testing for structure we could have perhaps done this
in the past as well but it would have been hard to sustain it because if a donor goes away
a vendor goes away they take the hardware with them and it's hard to you know like for example if
you run a project I don't want to go and tell you to come and use my infrastructure with the caveat
that I can't tell you if it's going to be here tomorrow or not having red hat on the sponsor list
now means that we're able to do that and that helps the overall community around centers as well
I mean we don't push for us success isn't about centers having you know doubling its user base
for us success as we can double the number of part projects that consume centers to deliver
service to their users I mean like if you're in the database uh writing software
software writing business we want you to use centers you want to be able to help you
get a better platform out for people who are consuming your applications
what's on the for red hat? um so red hat has an interesting situation in 2003 they
exited the community distribution built curated by red hat fedora became the community
representation of their distribution which is built curated sort of released by community fedora
has never really been released by communities been released by red hat but is developed and
curated by a community red hat enterprise Linux is targeted at the business users
people who need some level of assurances two o'clock in the morning your machine breaks
you want a phone number to call um and my pro my my answer to that would be you know
have your credit card ready um fire up thunderbirds and an email to centers develop hope for the best
yeah yeah or you know so that doesn't that kind of works for a lot of people it doesn't work for
everybody so over a period of time I think we've tried really hard to build a firewall between
centers and red hat anybody who joined fedora anybody who joined red hat as an employee was
automatically disconnected from any privileged levels in centers it is perhaps in the in retrospect
it seems a bit overly pedantic however what it allowed us to do was to build that firewall
between centers and red hat um and when the conversation with red hats started off it was pretty
clear that they benefited from having centers in the ecosystem because it allowed developers
it allowed third parties it allowed contributors to build stabilize to test against an archival
compatible platform without actually having to spend the five hundred bucks six hundred bucks
to get on to an archival platform so while they would never certify a piece of software which
was built and test on centers it meant that people could build and test on centers
and then go to rel whenever they need it to or whatever point they need it to so having
centers be a part of the family is like the third leg of the stool is not kind of complete
in that there is now uh an area or a or a group that you can go to for upstream innovation
where if you're working on g lip c if you're working on the kernel if you're working on gc c if you
working on you know the core competencies of what is Linux you have Fedora which is a constantly
moving constantly evolving constantly stabilizing environment I would I would probably say it's
slightly off the bleeding edge I think bleeding edge is probably you know LKML or you know just
it's not far off yeah it's not far off you still cut yourself yes yes um but I think what what
centers now does is it allows um people to curate content that they care about most within
understanding that the platform isn't going to change on them every six months so for
example if you're in the business or if you know if you have a great idea on your way back home
from work thinking hey I want to write a piece of software that does one plus one is equal to three
now you can go away you can do that on centers and you know that your gcc isn't going to change
for a few years your kernel isn't going to change for a few years your gnome isn't going to change
your kd isn't going to change your zed live you and all all of the libraries that you need
the core platform isn't going to evolve on you so as to speak because most people who are doing
Apple-level stuff or user-facing stuff and this is open source projects are not in the business
of curating zed live they're not in the business of curating gcc and they want that static target
that they can build against stabilize against deliver a user story against and I think that's what
does the second leg of this tool as it works you've got the fedora for platform innovation you've
got sent us for innovation on the platform and then you've got rel for people who need a nestle
let that come along with it um and in many ways I think we all complement each other but also in
many ways we compete with each other and I think and that's good that's healthy for everybody
because it gives the users a choice if somebody wants to build on centers he's got his app up
everything is fine but he wants to keep an eye on what's coming next he may every Saturday decide
to run it on fedora as well okay yeah and then if he has somebody who comes down the road and says
hey you know um I'm happy this is fantastic software this is making me money I won't support
on it he could then reach out to rel and say look if my guy was to use rel could I work with you
on supporting him on the rel platform so it kind of gives you the whole give you the whole picture
the whole story so are you going to be stuck here on the booth so you're going to give him some talk
so um we've been coming to fosdom for about eight years in an official capacity I've been coming
to fosdom I missed one fosdom unfortunately the first one um and so fosdom for for us as a
group and for me personally isn't really about the table as such it's about re interfacing with all
of the relationships we built up over the years so um over the next two days you'll find me pretty
much at every other table except for the center table talking to the guys out there you know and
also it gives us a great opportunity to talk to the guys you know one-on-one about what we're getting
wrong you know stuff we get right it's fine because that's already right the things that I care
about are stuff that we're getting wrong and then collect enough of uh you know enough for
feedback to go away and then basically build out a six month plan and hopefully come back next year
and you know you you hope uh the list changes I'm sure we get enough stuff wrong next year as well
but you don't want the same thing to keep coming up again and again and that's that's basically
what my personal individual plan for fosdom is to I've got my list I've got my list on last year
I know who I spoke to last year I'll be seeing all of those guys again and hopefully the list
have changed excellent thank you very much for the interview and good luck with you your uh
fosdom thank you very much thanks
you've been listening to Hacker Public Radio as Hacker Public Radio dot org
we are a community podcast network that releases shows every weekday from Monday through Friday
today's show like all our shows was contributed by an HBR listener like yourself
if you ever thought of recording a podcast then click on our contributing to find out how easy it
really is Hacker Public Radio was founded by the digital dog pound and the infonomicum computer club
and is part of the binary revolution and being rev.com if you have comments on today's show
please email the host directly leave a comment on the website or record a follow-up episode
yourself unless otherwise status today's show is released on the creative comments
attribution share like free dot org license