Hello and welcome to episode 238 of Late Night Linux, recorded on the 17th of July, 2023.
I'm John and with me are Faelin.
Buenos Tardas.
Graeme.
Bonsoir.
And well.
Bonno is dinti.
Oh dear.
Anyway, before we get going today, a couple of things to mention off the top bit of an
admin section before the actual admin section.
First of all, friend of the show Monica Madden is looking for a job now.
She's looking for something in community management, DevRel or advocacy or events organization
and support and she's looking for full or part time freelance or contract and either
remote or in Atlanta, Georgia where she is and she can start straight away.
She's got a good network and networking skills online and in person.
She's super friendly and personable and she's great at seeing the big picture and the
steps needed to get there and she's part of the Ubuntu Marta team quite a key part
of it, I think.
So I put links in the show notes to her master Don and her LinkedIn.
So if there's any jobs going that you think she'd be a good fit for, then get in contact
with her.
Also, if you are a specialist in SEO, like if you actually do it for your job and are
willing to help us out, then we're looking for an SEO person to make the websites better
basically.
There's a small budget available, but not a huge one and ideally would be looking
to work with someone rather than just hand you the keys and let you have at it, hopefully
that'll help keep the budget down.
So do get in touch at latenightlinux.com is the best email address for that.
Right.
Let's do some news then.
Ubuntu make a canonical pulls in control of LexD, that's how you say that as an LXD
it's spelled.
Yeah.
Yeah.
That seemed a bit odd and then one of the LexD developers, Christian Broner, posted
on master Don, as of yesterday, the LexD project is no longer part of the Linux container's
project, but can now be instead found directly under canonicals control, blah, blah, blah.
I personally considered canonicals moved to be a hostile act and as far as I'm concerned,
the decision was unilateral.
It may be legally sound, but it leaves a very bad taste behind and then a thread about
that kind of stuff.
And that was like, what?
Okay.
Why would Red Hat do this?
Sorry.
But yeah, I mean, come on.
And then Stefan Graber posted time to move on.
After a bit over 12 years of working for canonical, Friday the 7th of July was my last
day.
And he goes on to say, as I've told colleagues and upper management, canonical isn't
the company I excitedly joined back in 2011 and it's not a company that I would want
to join today.
Therefore, it shouldn't be a company that I keep working for either.
And then he goes on about LexD, following the announcement of my resignation, canonical
decided to pull LexD out of the Linux containers project and relocate it to a full inhouse project.
So if I'm getting this right, he'd had enough of working for canonical because the culture
has changed over the last 12 years that he's been there.
And as soon as he said, I'm off, canonical said, right, well, it's our ball and we take
it home then.
And this just doesn't look good for canonical, basically, I'm afraid you might have to stay
a bit quiet on this one, Graber.
Yeah, I think I will.
Certainly, Joey, your interpretation of events is the same as mine from the outside.
You know, I haven't spoken to Stefan for many years, but that's exactly what it looks
like to me.
Who funded all the work and how big was the team?
Was it just him?
No, no, there were, I don't know, seven or eight people on it, maybe when I was there,
so I assume it's probably growing a little bit.
But he was like the real visionary and the leader of all of that work.
That still is.
Do I get to be devil's advocate then and do I get to say, well, they did pay for all
the work to get done?
Well, Stefan says, having the LexD community experiment to be able to failure within canonical
seems unfair to me and everyone who contributed over the years, because LexD itself is a very
much canonical thing.
But LexC, which it really depends on, is a proper community project.
I feel like there's probably some internal stuff that we don't know.
It sounds like something didn't go right and somebody got annoyed, and that's not me
trying to stick up for canonical by any means, but I don't know.
It doesn't feel right about it.
It does sound like somebody's had a bit of a, a pram throwage now that could be from
both sides, but yeah, yeah, I don't know.
Nessie, look, if it stays on source and it's the source is available to everybody, what's
the problem?
Yeah.
I suppose.
I mean, like, if you do fund a project, like if you say there's like eight or ten people
there, then I mean, realistically, what was he expecting to leave and then, you know,
still be in charge of the team that is hired by canonical or something that just seems
a bit odd?
I don't know.
The problem potentially is, and I don't know this for a fact, but I can see that perhaps
this is something that may happen.
If it becomes a canonical project, then you are required to sign the canonical agreement,
whereas if it was a Linux containers project, then maybe there's an alternative more friendly
or less restrictive, I don't know, an alternative contract that you had to sign or a terms of service
or something that was like a more open source centric one that didn't sign your rights over
or your copy right over to somebody else.
But that is complete speculation on my part, but I can see that that is a situation which
may well exist.
Well, it does send his blog post, however, I don't intend to ever sign canonicals CLA,
contribute a license agreement.
So yeah, that must have something to do there.
In defense of CLAs, it does allow you to do stuff like real licensing, whereas if you
take a big messy project, like, for example, Pigeon, this is something that Gary from
Linux down time has talked about, it is impossible for Pigeon to re-license because they just can't
get in contact with everyone who has contributed code to that project.
I mean, some of those people are no longer with us, even.
And how does that even work with getting in touch with relatives?
It's just not going to happen.
Whereas if things are developed with a CLA, it means that there's one central body that
can make decisions like that.
Like if a new version of the GPL comes up, for example, that is better than two or three,
then you can just switch over to it, but you can't necessarily do that if you don't
have a CLA in place.
So that's always been the argument for it, but I can see as an individual contributor,
you're not really going to want to do that to sign your rights away effectively.
I'm generally in favor of the CLA.
I don't have a problem with it personally, and I think it makes a lot of sense from a
business perspective, but the canonical CLA suffers with the same problem that quite
a lot of canonical projects have, and that is being associated with canonical.
And that puts off a lot of developers from outside, just the way that it is.
Yeah, I guess that should point out also that contributors still retain copyright on
their code.
The canonical CLA says that you're basically granting a license to canonical.
I really hope Stefan writes more about why he doesn't want to sign the canonical CLA.
I'd certainly be really interested to hear his point of view on it.
He's done an amazing job on Nexte.
We should say that.
It's been incredible the work that he's done and he's responsible for its success and
so many people using it and it's so complicated.
I do wonder if he's going to fork it.
It would seem like an awful lot of work and burden to take on, but the way he's talking
in this policy kind of get that impression that he's at least thinking about doing that.
Choice is good.
Yeah, fragmentation is even better.
Oh, yeah.
Yeah, that's the sprinkles on your ice cream, that is.
Yeah.
Well, we'll have to see whether that happens or not.
There's no word on that yet.
And it looks like he's not looking for another job at this point.
So who knows what's going to happen.
Fedora workstation 40, considering to implement privacy preserving telemetry.
I really hope this is going to be a short conversation.
Good.
Go for it.
Intercaps.
Oh, my God.
I can't believe they're doing this.
Oh, come on.
Fuck anyone who says that they shouldn't be doing this.
If you actually read the proposal, the way it's going to be done is properly anonymized,
self-hosted by the Fedora project.
You can't properly anonymize data.
You can still tell who's who.
Ah, look, I trust Fedora to do stuff like this properly.
Now Raider's voice, that was his first mistake.
Please don't tell me you're going to argue with me on this for you name.
No, I'm all in on the Katie metrics, but I trust Katie.
I trust Fedora.
They've got to take away all the options and all it will be is just a questionnaire system
on your computer and then we'll close down.
Yeah, that is quite funny that that is one of the key reasons why they want to do this.
So they can find things to get rid of, setting and stuff, but you know, that's fine.
That's exactly what telemetry should be for the moving choice.
No, shaping the thing that you are making into the thing that you want it to be and your
users want it to be into the way that known once your computer to be.
Yes, correct.
That is the correct sentence.
And look, opt out, I think is fine, as long as it's easy to opt out, which it looks
like it will be, I have no problem with this at all.
And I think that the people who are complaining about it use a different distro then if you
don't like it.
Look, I think you're right.
I think also this is one of the huge, untapped potential resources we've got in the next
and open source to improve our software because I think we all get some share in the positive
outcome.
And it's something that is so useful, you know, just knowing how people use the desktop,
the applications they use where they're clicking, whether they're using menus or icons or
keyboard shortcuts or command lines.
And we don't trust anybody else.
We wouldn't trust Apple or we wouldn't trust Microsoft with this data.
It just goes out there.
But I do feel like I think we've almost talked about this before.
There should be a way that we have control over the data that we're sharing more so that
other companies could tap into it, you know, like some kind of open API and an open platform
that's hosted on something that we can trust so that we can all sign into it and all of
the apps that we want to share our data with.
I think it's really important and we could really improve things.
Well, yeah.
And Fedora talk about how they're going to have the whole thing open source, the back
end as well, so you can set up your own server and see exactly what your desktop is sending
to that server.
That's what I mean when I say I trust Fedora to do this right.
And you write even stuff like monitor resolution.
Is anybody using a 720p display or thereabouts 13, whatever it is with Fedora anymore?
Maybe that's 1% of people.
So maybe they can deprioritize that.
I know that was something that you cared about with Ubuntu will.
There's so much potentially wasted resources because we have no idea how many people are
using a certain workflow or setting or a bit of hardware or hardware combination.
And metrics collection properly anonymized is the way to find that out and direct limited
resources in the right places.
Does that still hold true if you are opting in though?
Does that not mean that the hard line opt-inners will opt in and the people who don't want
to opt-in will be potentially a large majority that won't, whereas if you have to force something
to opt out, is that not a better way of doing that where?
Definitely.
Yeah.
Then you get actual proper stats.
Yeah.
Okay.
So it's going to mean almost nothing.
Maybe potentially.
No, no, this is going to be opt-out.
Is the point.
Is it?
I see.
I misunderstood.
I thought it was going to be opt-in.
No.
You're talking about being opt-out.
Easy to opt-out.
Right.
But opt-out nonetheless.
Okay.
Well, that makes more sense, son.
Yeah.
I mean, that's the controversy.
Is that everyone wants it to be opt-in, but there's just no point, as you said, if it's
opt-in.
You are self-selecting people who want to help the project.
Yeah.
Whereas really, you want everybody using it to contribute.
And really, a few stats about how you use the distro that you're getting for free seems
like a fair trade to me, especially if it's all up-and-source, all anonymized, all
auditable, all testable by yourself and your company or whoever's using it.
Like I said, this should be a very short conversation.
It should be well done for Dora.
Go for it.
Okay.
This episode is sponsored by Entroware.
Go to Entroware.com.
Entroware sells computers with Ubuntu and Ubuntu Marta preinstalled.
They've got a range of desktops, laptops, and servers, and most parts are configurable,
so you can pick the CPU, RAM, and storage that's right for you.
If you can't find exactly what you want, then do contact them, and they'll work with
you on a bespoke solution that's perfect for your needs.
The ship to the UK, Republic of Ireland, France, Germany, Italy, and Spain.
And if you do buy one of their machines, there's a little drop-down at checkout, and you
can select late-night Linux, so they'll know that we sent you.
Also go to Entroware.com for all your Linux computing needs.
Onto a bit of admin, then.
First of all, thank you everyone who supports us with PayPal and Patreon.
We really do appreciate that.
If you want to join those people, you can go to latenightlinux.com slash support, and remember
for various amounts on Patreon, you can get an ad free feed via just this show or all
the shows in the late-night Linux family.
And if you want to get in contact with those, you can email show at latenightlinux.com.
All right.
More on the whole rail source code situation.
It's been a long two weeks since we last talked about it.
It's all going well, let's blown over.
Yeah, we'll never be talking about it again.
Hang on, no.
Let me just start with a little bit of feedback from Jean.
They wrote,
Sun said a while back that they were going to go to CentroStream, so the rail change
shouldn't matter.
I think somebody said that they were concerned about Sun.
Yeah, me.
Yeah, so you were wrong.
Well, I think it's actually a missed opportunity.
I think they should have gone to Debian and they wouldn't have been happier.
And I think that would have been better for everybody because that's a proper community
distro.
I think for all institutions like that, just don't go along with a corporate backed
product.
I think maybe get on board with the actual proper community thing and then drive the community
that way.
And pay someone to support you who is a community member.
Yeah, like the likes of Freaksian, for instance, they do the way I think that most support
should work where a block amount of hours is ordered up and they deviate out when a job
needs to.
And not everybody can afford to stick a contract on every single box.
It's just it's crazy money and people don't have crazy money.
It just seems like a wasted amount of cash.
So where is at least this way you can see the work getting done.
So I don't know.
What do I know?
Anyway, right.
So a lot has happened.
First of all, Oracle just, I mean, this is just comical.
Keep Linux open and free.
We can't afford not to.
Goodbye.
I don't know.
500 words of just fucking pure gaslighting from a couple of their quite well paid people
by the looks of things.
I mean, come the fuck on Oracle.
Who do you think you are trying to kid with this bullshit?
And then at the end, finally to IBM, he's a big idea for you.
You say you don't want to pay for all those well developers.
Here's how you can save money.
Just pull from us, become a downstream distributor of Oracle Linux.
We will happily take on the burden.
Yeah, let's put it in the hands of fucking Oracle.
I mean, Oracle is literally a punchline at this point.
The problem is, though, are the people in charge of the purse strings going up no all
the back history of this nut?
They're going to look at that and go, oh, they seem like a smart bunch of chaps.
Let's talk to those guys.
They must be really good because it's not always the people who know who are asked.
I mean, look, there are some reasonable points in this piece about the shit that read out
as done and how that's not ideal.
But then you remember, hang on, no, this is fucking Oracle.
Fuck you.
You can't just come out and say this as Oracle.
I'm just not having it.
And the idea of, if you are working on REL and you're not happy with it, come and work
for us.
Good luck with that.
Well, across one of those off the sponsors list, anyway.
That golden check has just been returned to sender.
Actually, who's left?
I've lost track.
Suza are left, so I have their blog post.
Suza preserves choice in enterprise Linux by forking REL with a $10 million investment.
Not brilliant idea.
It's genius, isn't it?
What could possibly go wrong with this?
Team it up with Rocky Linux by the looks of things.
When they hint at that, they don't tell us what it's going to be called.
Ruse.
Maybe.
They talk about wanting to try and maintain this one for one bug for bug compatibility.
And yeah, that seems like it's going to be fairly tricky.
And I mean, $10 million, it's not nothing, is it?
It's quite a lot of money.
You do have to wonder what that $10 million is going to be spent on and who's going
to be working on this.
And have Rocky just said to Suza, I don't worry, it'll be fine.
Give us $10 million.
It'll be fine.
We'll make a dance video on YouTube.
We'll throw in one of those dreadful, change the lyric songs that are just massively
cringe.
I can do that for you now anyway, don't worry about it.
That's true.
Yeah.
And finally, Alma have come out and said something quite different.
They've decided to play the game with Red Hat, stay on site with them, and not maintain
bug for bug compatibility.
What's the point then?
Well, that's what I thought, right, until I read it.
And then they said that they are at least aiming for ABI compatibility, which means that
any software that will run on REL will run on Alma.
That seems difficult to me to make sure that every bit of software will do that.
But it also seems to me that they are playing Red Hat's game, as I said, they are not
saying, fuck you, we're going to do what we're doing before.
They said, right, well, we can't just be a clone.
We have to be a separate distro, but we're going to work hard to make software run on
it.
And that's sort of what Red Hat wanted, I think.
They didn't want just straight up clones.
And they say that it means that they can potentially fix bugs and then contribute them
upstream and maybe even implement new features and stuff.
I mean, what you said for him is the kind of bursting of that bubble, though, like what
is the point of a distro that is not a clone of REL, because you've got a stream, which
is, I mean, it sounds like this is going to be somewhere between stream and REL.
Yeah, let me get more tears.
It's great.
It's like, I don't know, I don't get.
I think I'd love to know inside the Red Hat boardroom, if this is, you know, how they
proceeded things to be planning and are all very happy about how it all went, or if
they're like, shouting themselves now, because they've unleashed like the other two main
competitors against them, I would really, dearly love to know, but I just don't get what
the point is anymore by the looks of it, you still have to have either a Red Hat license
or you're going on one of the other two, because Alma is, I can't see that as contender.
Like if you log a bug, say, say you're, you've got a DB, I don't know, whatever, and it's
not working.
And they say, well, what do you run on Alma?
And they're like, no, sorry, that's not, that's not supported.
That's not a proper release.
I mean, ABI compatibility, yeah, that's lovely, and I'll, but I bet you could probably
make a Debian that's close enough and would run the application if they had the libraries
at the right level.
Hmm, I don't know about that.
I think it'd be a lot more work.
No, my point is the fact that if you've got the right library levels, whatever that happens
to be, then that is as close to ABI compatibility as would really run something.
To get something that's going to pass the software provider's checklist, if it's not
the actual thing, then it's not the checklist, it's, you know, the binary, it's one or the
other.
Well, you said that you'd love to see inside the Red Hat and IBM boardrooms.
And for me, all I can think of is the gift that comes from Breaking Bad, the face off
episode, whereas on the phone, Skyler, and it just says, I won, Red Hat of one, because
they've signed chaos.
And out of that chaos has come this weird, uncertain future for Rocky and whatever Susa's
going to get involved with there.
And even Oracle admit that they probably won't be able to be a straight up clone.
And Alma is like this weird in between.
And so they have achieved their goal of making real, but it's own thing with no straight
up cloning of it.
They have just won as far as I can see.
I don't know about that.
I mean, if I'm a vendor of providing software, I'm not going to just want to be working
with one user at the fire end.
I mean, I'm going to want, you know, whatever happens to be the choice between everybody
else.
And maybe that is no longer going to be the Red Hat ecosystem.
This is why you're not a very successful business man.
No, but you have to look at, I don't know, I, I, I, CRL is great now, but I don't know
whether that's actually where the future software lies.
If you were to strictly look at the likes of 80 Western containers, Red Hat's not in
it because you're not going to have license fees for that.
I think that's why this got pushed through.
No, this is a totally different market.
This isn't cloud and containers and Kubernetes and everything.
This is banks and on-prem stuff, old-school IT.
That's the market that we're talking about here.
That's cross-seater.
That's not going anywhere.
That's the problem.
Exactly.
It's not going anywhere.
Yeah.
It's not going anywhere.
It's not up.
It's slowly dwindling away.
And the problem is all of those new people are not going to be learning Red Hat because
that's just become a shit show.
They're going to be run something else.
Well, that was already happening anyway.
And so now they are securing the medium to long-term tail of old-school on-prem IT, which
will never go away because there's compliance issues and stuff.
And yeah, the cloud vendors will try and say that you can have a private cloud with them,
but there's certain banks and government stuff that has to be on-prem.
And you have to have certain compliance checklists and everything.
So I don't think that that old-school enterprise Linux business model will go away in the
next like 30 years plus.
I think you're wildly optimistic on behalf of Red Hat there.
Look, how many coreball programmers listen to our show and do that shit every day and
get paid shitloads for it because there's so many old systems that still need those
people to work on them.
I'd say it's two.
Well, right in, show it like nightlinks.com and prove me right and prove failing wrong again.
Need a post full address there, Joe.
OK, well, all the Java programmers right in as well until us.
Yeah, and all the people who are maintaining these on-prem systems for banks and government
departments and everything.
A lot of that stuff is dying off.
Even Bank of Ireland, my crusty bank over here finally got rid of its old mainframe.
So that is all gone now from them.
I don't know.
I think there's something, I debrue this in Red Hat where they went, oh my God, we're
in the shits.
We need to make money quick.
No, I think it's much more long term than that.
But we'll see.
Right.
You've snuck a KDA corner into this.
It's very quick.
By stealth.
So you've got one minute KDA EV report 2022 is out.
Yeah, this is colossal and it's worth a read.
But the interesting thing I thought around the end was the finances where patrons and corporations
account for 23.5% of the money that KDA has.
And the great thing is that the members are 42.6%, which I think is really cool.
And their expenses are 54.5% on people and 24% on events.
I mean, I think that's possibly the best way I think could go get invested by the people
who use your product the most and then give it back out to the people who are doing the
work on the product.
So that's a really healthy financial statement, I think.
So nice to see that.
And they had the first academy in person this year, steam deck.
They hired some new engineers and all the gold stuff that's going on and the prep for
plasma six and the whale and the work.
I think it's been a really good year and yeah, academy's going on right now.
Yeah.
And there's raw videos available at this point.
There is, yeah, provide a link to the cockatoo, I think it's called the video pages.
There's a feed there with the various days.
I think there's four days worth in the two rooms available and yeah, they'll be all broken
down eventually.
But if you're really jumping at the bit to get in there, go for it.
Before we get out of here, quick update on the dynamic ad situation.
We're going to put that off for a bit.
Things are not looking great when it comes to ads.
But for various reasons, we've decided that we'll wait a bit on that.
Primarily because the company that I was negotiating with, they were going to sell a combination
of host red ads.
So me doing the old, okay, this episode is sponsored by blah, blah, blah.
Along with the programmatic ads, as they call it, which is the just random person reading
it.
It came up in a meeting that those ads would require a tracking pixel.
Now I don't know with 100% clarity what tracking pixels are, but it sounded shit to me.
And so that has somewhat put the kibosh on the whole thing for now.
And it's made me realize that you look around and you see the shit that's going on in
the IT industry and wider.
And you're not allowed to say in shitification anymore.
Did you know that?
Let's say that that's like hackneyed and overused at this point.
But it's fucking what's happened.
And it made me think, is that what we were thinking of doing to this show in shitification
with the dynamic ads?
And it's maybe going to happen eventually, it probably going to happen eventually.
But I think let's just ride out this period and see what happens.
And I'm a bit concerned that maybe some people signed up for the Patreon specifically
to avoid these dynamic ads, which are now not going to happen at least for a while.
And so I wanted to be as transparent as possible about that.
That is something that has been delayed for now.
Yeah, I think that's the right way to go.
Like probably as you say, you probably will have to do it, but we don't want to pick the
wrong place.
It's all very well.
Let's talk about some sort of software and then partnering with the worst in FinTech
bastards.
Well, it's ad tech in it rather than FinTech.
Ad tech, sorry.
Look, they're all the same bastards.
Right.
Yeah.
I mean, it's probably inevitable, probably, but maybe not.
Maybe things will recover.
Maybe there'll be a few more ads sold and, you know, I mean, I looked the other day.
It's been four years since we got ads on this show and we've not had a single episode
that hasn't had at least one ad on it since then.
And looking at my doc of episodes and what ads are going where, there's some gaps.
And that's got me thinking, hmm, maybe we could have a couple of episodes.
Off this summer, maybe instead of a proper summer holiday, I could actually just skip one
episode of each show, maybe if there's no ads sold on it.
I mean, I don't know if the patrons would feel short-changed about that.
Let us know, I suppose.
But, you know, we'll see what happens come the end of summer when things start to hopefully
pick up and get back to normal, whatever normal is.
I just don't want to insure to fire these shows that we do here.
It feels like just the wrong thing to do right now.
And that's definitely influenced by looking around at what other companies have done,
let's just say.
But as ever, let us know your thoughts on it, show at latenightlames.com.
But we better get it out of here then.
We'll be back next week when, who knows what we're talking about, probably some discoveries
and stuff.
But until then, I've been Joe.
I've been Phantom.
I've been Graham.
And I've been Will.
See you later.