hpr-knowledge-base/hpr_transcripts/hpr0218.txt

Episode: 218
Title: HPR0218: Source Cast Ep 5
Source: https://hub.hackerpublicradio.org/ccdn.php?filename=/eps/hpr0218/hpr0218.mp3
Transcribed: 2025-10-07 14:12:26

---

But I think I am working hard to work on that prize.
Thank you very much.
You guys returned?
There were lots of those on the display.
It's really a health advantage for the public.
It was looking like you years ago in privatized motion alone.
Also, the book says,
are there anyone living in the country,
can you buy it?
And isn't it?
Welcome to SourceKist Episode Pi, I'm JD.
And I'm Jeremy.
And this is Episode 3.14.
As we mentioned last week in Episode 3, we're going to do on the weeks that we don't
do disparate reviews, little mini episodes where we may be discussed what's going on
in the Linux community, or we take viewer feedback, or we review, for example, hate to spoiler
things, but I think I did already last week.
We're going to review KD4 here as well because we've been playing with it the past week.
And I would say instead of playing with it, we were avoiding the urge to take another
human's life, but go on.
Okay then, and yeah, this episode, these point episodes are mainly for everything that
really isn't a disparate review, and we'll be doing them every other week as I mentioned.
And we decided to call this episode Pi because it's the only time we'll get to have an
episode Pi.
So, you know what, Episode 3.14, you will like it.
Well, first we got an email from someone we will call them Skeleton with a zero instead
of an O, and they like the podcast, but they had a question for us.
They have an, geez, this is just terrible modeling.
I want to know who thought this was a good idea in marketing.
And HP, DV6436NR, Z, Alpha 219 or 8360, Tango, Foxtrot.
The actual model name of this thing is DV6436NR.
I've heard of it.
It's actually a decent HP laptop from what I've heard.
And he wants to know what distro would be the best because he's tried to Ubuntu, Susan
and Mandriva, and none seemed to work with his wireless.
He was going to try Gintu or maybe BSD to try and compile drivers from scratch, and
wanted to know what distro we would recommend.
I couldn't find a lot of information on the hardware in this laptop, but it looks like
it is running the latest generation of crap-tastic Broadcom Wi-Fi, BCM43XX Legacy drivers
totally hopeless on this platform for not dug up.
B43 may work, throw some salt over your shoulder and give it a shot.
But in terms of getting bleeding edge hardware support, if you don't want to wait a few
months for other distros to catch up, Gintu is not a bad shot.
It'll probably be simpler for you in the long run just because it's so ridiculously easy
to get the absolute very latest kernel or even an SVN of the kernel they're working on
if you're so inclined.
Also, I would also recommend Arch, which is both me and Jeremy are now using it because
first of all, it's just as easy to compile a module for your kernel as it isn't Gintu.
You don't have to deal with compile it, it seems like he only wants it because he needs
something to work with his Wi-Fi card and he probably doesn't want to have to recompile
all of KDE or Nome just to get a working Wi-Fi card.
But before you're not comfortable with those distros, the first place I would tell you
to go is go download the Alpha of the next version of Ubuntu.
I've had terrific luck with hardware support and Alpha's of Ubuntu for stuff I could even
get the work in Gintu or Arch.
And really nasty firmware driven software rates that really wanted to pretend really,
really hard that they were a real hardware rate and I could not get that to work in Gintu
or Arch when I played with it and it works just fine in the Ubuntu Alpha when I tried it.
So that's probably your easiest first place to go but of course it isn't an Alpha of Ubuntu
so don't expect everything to be working nice and pretty when you try it.
Yeah, the current Alpha of Ubuntu 8.10, it's using the latest kernel which is 2.6.26 but
I mean they haven't even made a live CD for it yet so that just shows you how kind of
not ready for prime time it is.
Yeah, you can find some unofficial sort of homegrown live CDs, give them a shot.
I can, of Alpha like two.
I think so, yeah.
They say like, next Alpha, Alpha 3 I think is next.
They will have a live CD but yeah.
Okay, so thanks for that email and keep them coming.
Yep, JDatSourceCast.log and Jeremy at SourceCast.log for you, Kim.
Yes and onto the next part of our little discussion here and that is retouching on our review
last episode of Suza11.0.
We actually had a sort of, I guess you could call it developer, he called himself a member
of the Open Suza project and we had him respond to our reviews.
I'll go ahead and take up his issue with my review and that was that I installed it from
a 103 disk which, well okay, Guilty is Charged there so I did sort of overlook any installer
features that were added in 11.0 so in the interest of fairness I went
and downloaded the 11.0 live CD and did a completely new fresh install from that and then
it was a mixed back.
Well, they did fix a few of the minor things I had.
The authentication method when you install it in the 11.0 live CD is indeed hidden by
default now.
Also it actually gives you a helpful message that tells you what to do before you reboot.
That's also good and it has a much much much sexier looking post install wizard.
The downside is pretty much everything that I really hit open Suza4 to begin with and
that is ridiculously slow and ridiculously painful software management.
He said, you know, the disk upgrading wasn't supported and I understood that and I went
forward anyway because at this point I don't see any reason why any distro, especially
one with no vells backing, shouldn't be able to pull off dist upgrade.
I can't name a major distro that can't dist upgrade, can you?
I can't name a minor distro that can't dist upgrade.
I mean, we're not talking about, you know, something all that new here, Debian's been
able to dist upgrade since Woody and of course you've had rolling release.
It's been able to dist upgrade since, I think, one of the least before Woody.
And then of course you've had people like Jin 2 and Arch where you just don't even bother
you keep rolling on so I don't see why they technically can't do that.
I think asking your users to go to the trouble, to download a CD every six months or however
often you release and do a complete risky might I add very risky install from a CD because
if something goes wrong, it's a much higher odds of being, you know, unbuttable, at least
when you dist upgrade you can leave fallback kernels and things like that.
Yeah, fallback kernels and you can still CH root, you know, everything still pretty
much works a little bit, you know.
But upgrading from a CD and also, I mean, I didn't read the documentation, but it's also
probably suggested that when you upgrade from a CD that you also reformat which, so you're
basically telling everybody back up your data every six months otherwise you shouldn't
upgrade.
It's just...
And the other thing he said was that, well, first off, my Wi-Fi still didn't work in
11.0 and it isn't even a Broadcom card but it's been around an awful long time folks.
It should work by now.
And he said, you know, hey, if you don't have an internet connection, how are we supposed
to, you know, go pull the repositories and my answer that is, well, all you're changing
in the path of the repositories between releases is a version number.
It shouldn't even have to go out and pull somewhere on the internet to begin with.
It should say, hey, these are official repositories, they're going to be there.
So whenever you get on the internet, just look here anyway.
Yeah.
I mean, seems like a pointless step to me.
Pretty much.
I would agree.
I mean, if you install Debian or Ubuntu or really those are the only two I've actually tested
this with from a CD without an internet connection, it will put the entries for the repos in the
sources.list.
I mean, on Ubuntu, it comments them out on Debian, it doesn't, but still, at least they're
there.
So do you have any particular other issues to address from where he brought up things
from your review?
Well, most of the, he only really brought up three things from my view.
One, and I will agree to this, App Armor is not proprietary.
I know it's open source because Ubuntu uses it now, not by default, but it's in the
repos.
And I just, I don't see why you couldn't have gone for SEO Linux.
That's all I'm saying.
I mean, what, you know, why reinvent the wheel, so I'm saying, you know, he mentioned that
the password encryption options are hidden behind the change button now.
I didn't notice that, but, you know, my brain automatically clicks those things, so it
just could have been that I forgot, I'll give them that.
As for Nano and Pico, yeah, he basically said Nano and Pico weren't on because a new users
don't really know about Nano and Pico, which is reasonably true, but if they've been taught
even a little bit about Linux, they know about them, and that they take up too much space
on the live CD.
Yeah, that's a, that's a such a load of shit.
You could, can't you fit Nano on a freaking floppy disk?
Let me, let me just be 100% bleedingly accurate here.
I'm looking at running 2.0.7 here.
Only 2.0.7 installed on I 386 is 1396 kilobytes, so just under the size of a floppy.
That's been, I mean, that's, yeah, I see zero reason to not include Nano or at least Pico.
I mean, those are tiny, but very versatile text editors that are far more friendly to a power
user than say Emax or Vi, please don't play me if you grew up on Emax or Vi, I'm just
telling you, my generation, most of them don't want to do with that all that mess.
They find Nano just fine.
Also he mentioned pseudo, because I mentioned that pseudo is a bit confusing, he mentioned
that pseudo isn't the default way to enter root.
I'm sorry, are you guys back in 1995 some more?
Because last I checked on every distro, pseudo is the default way to enter root.
No distro uses SU anymore by default.
I won't give them too much grief for that, for the simple fact that I, I freaking hate
pseudo.
It has its uses, but on a desktop they're minimal.
I'll give you that.
And B, it really doesn't solve the fundamental security problem of using root excessively.
All you're really doing is forcing users to go through a little extra hassle to keep
doing things as root.
If you want to address that problem, educate them on why they shouldn't run in as root
and just let them use normal SU.
Right, but I mean still, it's, it's not like it's changing anything.
I mean GKSU or KDESU has pretty much the same functions as pseudo, it works pretty
much the same.
So I find what I do a lot of times on machines I have is I just leave the terminal running
in root.
Yeah, so do I.
And just so I can do things quickly and if I'm on Ubuntu, the first thing I'll do is
I'll just do pseudo issue and get right in so I won't have to be bothered with it anymore.
The other problem that I mentioned that he brought up was the computer menu, also known
as the slave menu.
It was done by usability testing, bra, bra, bra, that like, you know, it's meant for six,
most people use six to eight applications daily and the rest are hidden and, you know,
so it automatically puts your favorite applications, the ones you use the most from the past week
up on the main window, but a favorite applications did not work for me and b, it's just, I mean,
I really dislike that you click more applications and opens a fricking window.
Why can't you click more applications and it transforms that little mini pan, that mini
section of the panel where your applications are to a menu like the way mint menu works.
Or kick off in KDE 401 or kick off because I mean, that's what I like.
I'll use mint menu, but kick off works almost identically, you know, by default, it's,
you know, the basic favorite applications or, you know, recommended applications or whatever.
And then you click a button and it goes to write in that same slave menu.
It goes into, you know, accessories and multimedia internet, all that stuff.
I just, I just don't see why it couldn't have been implemented that way.
Okay.
So thanks for the, for the feedback, Kevin, I don't want to say his last name for fear
of butchering it due to point, maybe DUP UI, but we, we encourage this sort of feedback
from people, you know, in and around distros.
And I know when we get to Fedora, which won't be too very long, that we've had a Fedora
ambassador already volunteer to file bug reports on anything we find.
So we, hey, we encourage you to just sort of interact with us.
Hey, it's, it's ultimately good for everybody.
Pretty much, pretty much.
Okay.
So on to KDE 4.1 and I will take the lead on this one simply because I wouldn't say that
I, okay, KDE 4.1 as, taken as a whole is better and an upgrade than, and a legitimate
upgrade to KDE 3.x, however, I don't just dislike the panel.
And KDE 4.1, oh no, I have a raging, psychotic foaming at the mouth hatred of the panel
and KDE 4.1.
I want to know who the hell thought of this so I can take out a titanium face golf club
and whack them in the nuts repeatedly until I feel better.
Yeah.
I, I, I would agree with you.
I'll go ahead and touch on what's good before I get to the panel simply because the panel
is full of so much hatred.
Yeah.
First off, the new compositing window manager, KWIN, it's, I like it, it's nice.
It's, yeah, it's, it's good.
It's like 80, 90% of the functionality of Compass fusion with a tiny fraction of the overhead.
It's great.
Exactly.
I am kind of anti-compass simply because it's a little more over, it's a little bit too
much overhead for just like pretty graphics, but yeah, I, I really like the KWIN compositor
and as I mentioned previously on the show, I do use gnome, but I might start using gnome
with KWIN 4, which will just be an incredible mind fuck.
And, and yes, when you see it, you will shit for X.
I, I do like the widgets, the, the widget concept and the fact that it works with Google
gadgets.
One way, it may not work with Google gadgets yet.
I'm not sure, but I do know it works with OSX widgets or gadgets or whatever you call them.
Which is funny.
OSX has like 20,000 widgets available.
The other change that we talked, which I already touched on was the kickoff menu.
At first, I hated it, but I've sort of grown into it, it, it does sort of work for me.
After, I put, I used to put the little icons of apps that I used all the time into the
actual taskbar.
We'll get to why that sucks so much in a minute in the panel, but in kickoff, I can just
set them in my favorites and I can, it's 1, 2, 3, 4, 5, 6, 7, 8 spots for favorites
before you have to scroll.
And that gets 90% of what I want to do in there.
And the applications menu's not that bad, it's, it's kind of like,
Also, I like how easy it is.
You just right click on the K menu and you click switch to classic style and it's switched
back to classic.
It's just very easy.
You don't have to like remove, remove K menu, add kickoff, remove kickoff, add
kickoff, and, now easy, yes.
And the, the applications browser is sort of like a bizarre,
in place version of the, like, panelled file browsing in OSX, where the column style
file browsing in OSX, where instead of going into a new column, it's like, it stays in
the current one and like moves over.
That's the best way I know to explain it.
It's a very unique way to browse through your K menu or start menu or whatever you want
to call it.
I like it.
Like I said, at first I didn't like it so much, but it really grew on me after a little
while.
And some of the widgets, especially the ones that could on default, I really liked devices
recently plugged in.
I love this.
You pop in a CD, you plug in your MP3 player, your external USB drive, and it pops into
this little device manager in the panel.
And it's real good.
And if you hover over to give you a little eject button, it's great, it's simple, I love
it.
And get this, I plugged in my Sansa and it showed an icon of a Sansa on it.
Little touch.
I like it.
Yeah.
It's nice.
It's nice.
I did notice a performance increase, not as palpable on my old ThinkPan laptop, but it
really showed on my desktop, on my desktop, it really sort of spread its wings and flew.
It is noticeably faster, especially logging in for the first time, you know, it's, I would
say the load time to actually go from the splash screen to actually, it's done loading
is probably cut in half, noticeably faster.
All that said, there's the panel.
Well, I want to mention something.
We both used, we both used KD mod 4, and I used KD mod 3 for about a week before they
moved over the 4 by default, and you used KD mod for about, for maybe two, three weeks
before they moved over the 4, like the super sexy default theme is not a sexy in KD mod
4.
Yeah.
I mean, it's good, but it's not, but it's not, um, wet chair inducing, shall we say, you
don't need to towel yourself off now anymore.
Yeah.
Um, to the panel, um, first off, it took me several hours to figure out how to scale the
panel.
I'm not talking about how wide it, it takes up at the bottom.
I'm not talking about, I'm talking about like, how, if you, in the, in the past in KD,
you could right click on it, you could size, and you could say to custom or presets, and,
and cause I like mine to be the, the 32 pixels, I believe the preset they call it as small.
It's, it's the one one up from tiny, it's 32 px icons, all up and down the test board.
It's nice and small, it gets out of my way, I like it like that.
It took me probably four or five hours to figure out how to get it like that.
And of all the places you would have thought I would have found that, Google, nothing,
the Archlinics Forum, nothing, the Archlinics Wiki, nothing.
I found it on freaking YouTube because somebody had to make a video to explain how to do it.
That is so incredibly awful.
You, there's the little cashew, only far right of the panel, and you click it, and you go into quote unquote edit mode.
And it looks, at first, really weird, because what's calling your attention in edit mode are the widget buttons
to add and remove widgets to the panel.
I mean, a lot of the functionality there is very not obvious.
First off, to rearrange things, like if you wanted to put your clock on the far left instead of the far right,
you have to, if you hover over something in the panel with your mouse, then you get the little arrow icons that pop up
to let you move things around.
I probably spent 10 minutes, you know, because I used to just right click on it, and there'd be an option there that said move,
and I could slide it up around with my mouse and just click and let it go, and wherever I dropped it, it would be.
Apparently, such things are too obvious now.
Apparently, drag and drop is simply not sexy enough.
I don't know.
Go figure.
Yeah.
In terms of scaling how wide your panel is by default, there's a system of sliders, and the blue ones are how wide you want it maximum,
and the green ones are how big you want it minimum.
And if you hover over and it takes a while, you got to hover over it for like four or five seconds, it'll explain that.
So that's not that bad.
There's a middle slider where you control the position.
Again, it's not a bad way to do it, but certainly not obvious.
It would have been great if the first time you hit the cashew, there was like a tool tip that comes up mandatory that explains this BS.
Yeah.
Because this is very unintuitive, like, you know, you either have to be a complete genius or a complete idiot to figure it out.
And here is where you, and when you're in edit mode after you've clicked the cashew, that's how you control the panel size.
At the top is a very, very thin ball. It's got three EDB little dots in it.
And if you put your mouse on this razor thin, it's got to be only like three pixels wide if that.
Your cursor will change to up and down errors.
And if you click and drag, you can scale how up and down, how big you want the panel to be.
But that, first off, that is like completely hidden.
You would have to accidentally stumble onto that.
Like, oh, I watched the YouTube video. It shows you how it is.
But making things worse, if you just slide that around, not even, you know, just to see what it does, the panel, at least for me, didn't scale in real time when you moved your mouse.
So if you move your mouse and hold it, exactly. After you move your mouse, it will take a little bit for everything to redraw.
So if you just play around and move it, expecting something to happen, you can, oh, well, nothing happened. I better move it back.
Yeah, you'll be confused for like five minutes there.
And here's a particularly galling chain. In KDE3, the interface mantra, as I understood it was, if you wanted to configure something, right click it.
Don't care where it is. Don't care what it is. Right click it. And by God, you can configure it.
I challenge you to figure out how to change the clock from 24 hour mode to 12 hour mode.
Can I right click the clock? Doesn't work. Can I right click the clock?
Well, you can. Well, can I click the clock?
Yeah. It does take you to some settings, but none of them are from, to change it from 24 hour to 12 hour.
Correct. And you say, hmm, okay, so I'm taking you through the thought process here. Maybe I should right click on the panel.
Nope. Okay, maybe I should click the cashew and go into edit mode for the panel.
New. Maybe I should right click on it while in edit mode.
No. Okay. Hmm. And of course, Google and the typical, you know, places look for this were also very devoid on solutions for this.
So out of desperation, I went into Katie control panel. And on the general screen, the first screen, you see, aha, computer administration, date and time.
Oh, we're getting somewhere. And you click it. And it brings up the locale like change your time zone, change your date and time and NTP sync.
Nowhere to change the formatting. So you kind of scratch your hands.
I mean, at this point, I was guessing I was just clicking stuff at random.
As it turns out, to change the clock in the panel from 24 hour mode to 12 hour mode, you have to go to the completely WTF location
of go to the KDE control center, go under the personal tab, personal, that makes sense.
And then go into regional and language, that makes even less sense.
I would go there if I thought, hey, I want to change like from English to Spanish.
Or dollars to great GPP. Yes, written pounds.
And under country, region and language, there's a whole bunch of little tabs over on the right.
And one of them is time and dates in there. Finally, for the love of God and everything holy, you can change the time formatting.
And what makes that even worse is you actually have to use the drop down because it's something very unintelligent.
Like you can't just, you know, it's HMM and SS if you said seconds.
It's not as simple as removing like one of the H's or adding AM to the end or something like that.
It's like you got to add something fucking weird to the drop down.
Yeah, the way they sort of zero pad is to use a lower case versus a capital case.
It's not really your typical kind of deal, but it would, but it did allow me to delete the AMPM specification on the time format to get rid of the freaking one of my huge pet peeves of the AMPM thing at the end of the clock.
Because hey, I wake up in the morning, I can see the sun. I know whether it's AM or PM.
So, I mean, that's like to change the clock from 24 hour format to 12 hour format, you like, I struck out probably, I guess had 12, 15, 20 guesses before I finally found it.
Just a whole bunch of almost everything dealing with the panel is just poorly thought of hard to configure, not intuitive to use.
It's just bad.
For me, I mean, I hoped I really, really hard core hoped that copied in KD4 would be awesome because I'm a big copied fanboy.
I used it under known for long while until I moved over this to this Debian install test bar flashing broke.
Completely blaming Debian, I'm not blaming copied, I'm not blaming anyone for that, just saying it happened.
And I was giving big hopes to copied with KD4.
Now, I add my Java account, it crashes.
I try going back and removing Java, adding my Amazon account crashes.
Like I mentioned last episode, I expected KD4 to be really nice, but now.
Yeah, I'm using KDE mod 4.1 on all of my desktop machines.
And overall, I'm happy with it. I'm much, much happier with all my desktop.
For some reason, the speed difference doesn't seem to be that palpable on my laptop.
But on my desktop, it's a much friendlier, much better experience.
It is better despite the fact that the panel is fail in almost every way imaginable.
Okay, so on to Xorg.
And this is an interesting topic because it is obviously you gotta have it if you want a desktop Linux.
And not a lot of people report on it for Onyx is one of the good places to find out the details on what's going on with X.
But not a lot of people report on what's going on with X.
Yeah.
And if you've been following X, they've been, they've had a lot of big things planned.
You know, things like curl mode setting, they want to leave the Xorg config file in the dust,
so you don't have to deal with it anymore.
It's so called bulletproof X to where it'll just configure itself properly to begin with.
Yes.
But there have been a lot of development challenges with X.
I know that X server 1.4.9, I wanna say, maybe 1.4.1.
It's something in the 1.4 series.
Yes, was over a half a year late and didn't ship with the feature set.
It was promised to have pretty much anything major within the X project is slowly moving behind date and behind features.
Yeah, 1.4.1 was supposed to basically fix all the crashers, all the major bugs in 1.4.
Because they basically rewrote everything for 1.4.
So that's understandable that, you know, there were bugs, you know.
But it took them, it was over six months late and there were still, and there are still two very major bugs,
as either major or critical in the X bug tracker that were slated to be fixed in 1.4.1.
Before we get too much in-depth with X, it's probably, I find this terribly interesting.
It's good to know who's actually working on it to begin with.
Yes.
And this is really surprising, a lot of this is very, very surprising.
First off, the graphics makers, they, the major ones, don't do squat for X.
AMD slash ATI was less than half a percent of changes in X.
But see, AMD slash ATI, whatever you want to call them, they may not merge many changes in,
but at the very least they say, here are specs, you know, you all can work on them, have fun.
Yeah, but that's, that's strictly for 2D though, they're still holding their cards on 3D.
I thought they gave out 3D.
As far as I know, they, 3D is no.
2D is pretty much completely open, but 3D is still pretty much a proprietary landscape.
Hey, I could be wrong, but that's my understanding at this time.
If somebody out there, you know, thinks I'm a bleeding idiot, email me and tell me,
and give me a link back up your evidence, show me how I'm wrong.
And I will do a medical bonus episode.
Yes.
NVIDIA was expectedly a little bigger, but still pretty small, about 1%.
And speaking of any of one of their, one of their proposed changes actually got rolled back.
What they wanted to do is, and I actually agree with this, they wanted to change X so that when it goes to configure the X server,
you know, your resolution, all that good stuff, and if it sees that you have a vendor ID that corresponds to an NVIDIA card,
it will automatically try to use NVIDIA's binary driver if it's installed before defaulting to the open source NVIDIA driver.
I really don't see a problem with that because the NVIDIA driver isn't really, well, first off,
there's nothing out there to replace the NVIDIA driver for 3D, anything pretty much.
Well, yeah.
It's a nice theory, but it sucks.
Yes. So far, it's still woefully behind.
I mean, hey, I am all four open source and so on.
But before, I mean, what is more important than open source is your hardware actually working,
because if your hardware doesn't work, it doesn't matter how open it is.
And if you want to do some 3D gaming or some serious 3D CAD work,
and you've got an NVIDIA card, you've got one choice, and that's that binary driver, period.
So I didn't see the real big deal with this, especially since on most displays,
you have to sort of go out of your way to install the NVIDIA driver.
Presumably, if the NVIDIA driver is resident on your system, and you have an NVIDIA card,
you want to use the driver to begin with.
And basically, the X developers rolled back the change that the NVIDIA guy made
and said, hey, we don't want X to depend on proprietary blobs that we can't develop.
I mean, he made a good argument in principle, but in functionality, I think the argument is pretty lacking.
Simply because the user would have to go out of their way in almost all cases
to install the NVIDIA binary driver, and if it's there, why not use it?
I mean, it's not like they were trying to force the NVIDIA driver on everybody.
And if you're using the NVIDIA binary driver, and this is the way it is now,
bulletproof X, it helps you a little bit, if you get a kernel upgrade,
you don't install the NVIDIA binary driver, but yeah, you can't use X without a config,
because it'll use the 2D driver, and that's just not fun.
So, I mean, I see both sides of the argument, but taking into account that, you know,
your N user is the goal here, I don't see anything wrong with changing the behavior of X to do that.
It's a minor change, and it only helps the user in the end.
Yeah, but back to the contributors, the actual distributions of Linux were much smaller than I thought they would have been.
Well, let me rephrase that, the community-based Linux distributions were much smaller than I figured they would have been.
Debian was about 1.5% all the BSDs were about 1.5% Gen2 was a little under half a percent.
And what's really shocking is the heavy hitters, the people really doing work behind X,
and we'll go ahead and get out of the way 31% of the commits were categorized under other,
that's, you know, a whole bunch, you know, professor working in spree time.
Up to like a quarter of a percent making like, you know, a change to 1.25.
Exactly, exactly.
Lots of people doing one change, small changes, things like that, and that was about 30%.
The real heavy hitters.
There's nothing against that, just to mention, there's nothing against that, it's more or less that, you know, yeah.
Any work is good.
Yes, any work is good, but, you know, the people who drop in and make one change aren't the people who really sustain the project or give it long-term vision.
The people really, the number one contributor to X was Red Hat at 16%.
I think Red Hat probably does far more work for Linux than anybody really appreciates.
They don't shoot their horn.
They don't shoot their horn on what they do at all.
I mean, they took the huge money hit for the patent and basically gave the patent away to everybody in the open source community.
And they're pretty much the number one contributor to X in terms of an entity.
Probably a quarter of the code in Nome is probably Red Hat.
Maybe 5-10% of the code in KDE 3 at least is Red Hat.
Shit, what else?
You know, a lot of kernel code is Red Hat, I know that for a fact.
It's just that, like, Red Hat has, yes, they're kind of the original, you know, Linux company, but they've got so much stuff and nobody gives them really any credit for that.
Yes, they have just a shocking amount of output and changes and influence in the major projects.
And hardly anybody knows about it.
Like I said, they don't shoot their horn on what they're doing.
And I think they should do that a little more because I think they're horribly underappreciated.
I mean, they don't shoot their horn, but like, for example, and I think I kind of talked about this on another, on an earlier episode.
I'm going to, that was the wrong one.
I'm going to my Nome panel and I'm, shit, this, I fail.
Well, there's always KDE mod.
True, but if you get past the panel, it's really good.
Well, now I'm, because I did this another time.
Maybe it was all removed, which I really would not like that much.
Yeah, wow. It's like it was almost all removed.
You know, you right click on the trash in Nome, I think probably 2.20 and earlier, and you go to about, it says, you know, copyright Red Hat 2004.
You know, you go to the clock copyright Red Hat 2004.
So after Red Hat, the second biggest contributor was Intel at 14%.
I sort of expected that because they were, you know, they have the most friendly open source, good driver, GPUs on the market there.
They're not the fastest.
They're certainly not going to go, you're not going to go play crisis or far cry at a good frame rate on them.
But they're pretty good GPUs all around.
All things considered.
And of course, they're top notch with their open source support, you know, to the point of making Microsoft angry, in fact.
Yeah.
And after that is one that I bet also nobody would guess.
The third biggest, and I ask you about this before the show and you couldn't figure it out either.
It makes a little more sense when you consider their subsidiaries that they've acquired, but still unexpected.
Nokia coming in at 10%.
And for those who don't know, that's part of that is because Nokia bought TrollTech, the people who did QTE, which is the technology behind KDE.
And then after that, it drops way off.
The next, the next three group together were four really.
Apple at 5.6%.
Hey, here I was thinking all Apple did was buy cups in terms of their open source contributions.
Yeah.
Sun at 5.4% Novel, 4% IBM, 3% and then from there, a bunch of little thinking, nothings, you know, HP at 0.01% things like that.
So overwhelmingly, it seems to be a sort of sparse community and predominantly an Intel and a Red Hat affair.
Yeah, I mean, between them two, what was it like 50% between Red Hat Intel and Community, you're at over 50%.
Yeah.
That's interesting, but I mean, the thing is, is that Red Hat Drive in this, they've got money in the bank.
You know, it's not like they're doing this in their free time.
You know, they're actually getting paid for this.
And yet, there are so many DelGrey after DelGrey after DelGrey.
Yes.
You know, and things that people have been like wanting, for example, MPX, multiple point of X.
It's not the coolest thing ever, but it's kind of nice.
That was supposed to be an Xorg 7.4, which is in release candidate now.
It isn't.
It won't be until 7.5, which could be and probably will be another year before that comes out.
I mean, you know, one of the targets for Ubuntu 7.10 was bulletproof X.
And they had to actually cherry pick it.
Cherry picks up the bulletproof X fixes from trunk, because Xorg 7.3 wasn't ready in time.
It just feels to me like, you know, one of the most integral parts of Linux.
And yes, it doesn't get the love it deserves, but it's so integral and it's fun so far behind.
Yeah, I would agree.
It's not like they've lacked vision.
You know, they've had good features, announced, you know, they, you know, real noticeable improvements.
But it seems like after they, you know, have their little, I believe they have like a little developer summit.
It seems like after they meet and they announced the upcoming release and the upcoming features,
you don't hear from them until, you know, months after release.
And then it's saying, oh, we're going to cut this.
We're going to cut that.
And they vanish again and hear from a couple of months later.
And it's finally released and a little more is cut out.
And stuff's pushed back even further.
I mean, it's basically the vista of Linux software development.
Oh, no, he went there.
I mean, that's exactly what happened with Vista.
First, you know, it was supposed to come out like 2003.
And as we all know, it didn't come out until early 2007.
And it was supposed to have like WinFF and like, you know, a really new, completely like redesigned UI.
And great security, all this amazing stuff.
And it's pretty much got none of it.
Yeah, if there's any X developers out there listening or if you know somebody who's an X developer,
put them in touch with us.
I would like to know what their take is on what's going on.
Is it just that there's so much work and their understaffed.
They don't have enough manpower or it's just, you know, really, you know,
they're have to doing a lot of rewriting of code.
It's just really hard.
I mean, I know one of the biggest challenges with X is that becoming a developer is just a 4X is just a huge pain.
There's so much you have to learn just to become just to have basic competency in what's going on and keeping up to date.
It's kind of like the Linux kernel.
Yeah, you can come in and make a bug fix.
But do you really feel competent enough to code on it some?
Yeah, no.
And I mean, that's the other thing that I was going to mention.
Outside of the kernel, what's the most important piece of Linux?
Outside of the kernel on the C compiler?
Sorry.
Yeah, outside of the kernel in GCC.
I could make a pretty good argument for Hal and Debos and those other project port and sort of initiatives for, you know, open.
Well, here's the thing.
We lived without Hal.
We lived without you, Deb.
We lived without Debos.
Painfully, but we lived.
Yeah.
It wasn't actually all that painful.
It was just like, why weren't these here in the first place?
But that's, I mean, that were not.
But outside of like, you know, pretty much the big four are kernel, C compiler, bash or a shell, you know, if you're not a bash fan.
And X.
Mm-hmm.
Because I mean, you don't really need KDE because X shifts with a window manager that's enough to get things done.
Not well, but it'll get things done with the GUI.
Yeah, I mean, you could even just do, you know, straight start X and have a whole bunch of nice, big, you know, terminals that aren't 60 columns or whatever.
And you could get a lot more done that way.
And that's like the most primitive GUI there is whatsoever.
Yeah.
And then like, you know, the X kind of stuff like X clock and shit like that.
Yeah, I mean, it looks like something straight out of like pre-development of Windows 3.x almost.
Yeah, really does.
And if anyone kind of uses that as their main desktop, please use like, rather something and screenshot it and put it up with our forums.
They'll then get to us an IRC or something.
Yes.
So that we can see the awesomeness.
That would be terribly amusing.
And I bet somebody is doing that too.
I would exactly be surprised.
But yeah, if you know anybody out there who's next developer, please, please send them to us.
We would love to interview them and see what's going on and get a peek into the mindset of what's really going on in the X community, what they think is a problem, what they think isn't a problem.
I mean, all we can do is speculate that here.
You know, we're just two guys looking at the evidence and you know, talking out of our ass and so you can enjoy us.
And so I think that will about wrap it up for our pie episode.
Let us know what you think.
Haven't heard anything good or bad about the audio quality.
So I'm just going to assume that it was so amazing it put you out unconscious.
Yes, yes.
It's so amazing.
It might have made you crap yourself, but then we probably would have heard from it.
Like your podcast made me crap myself, your bastards.
As usual, you can find us on our forum sourcecast.org slash forum.
You can also reach us on IRC IRC.sourcecast.org that wasn't entirely working the first few days after our last podcast release.
We apologize that is working now.
And we're both eye-linging that channel all the time.
JD at sourcecast.org and Jeremy J.R.M. White sourcecast.org.
So contact us any one of those ways.
We always love hearing from you.
What's good.
What's bad.
What you disagree with.
And you know, hey, we love debating with you.
Yep.
And our forum sourcecast.org slash forums.
And as one little tidbit of something I plugged last episode.
We now have a call for papers and a preliminary sign up for the Southeast Linux Fest.
So if you're in the Southeast or interested in traveling to the Southeast in the summer of 09,
go to southeastlinksfest.com slash wiki.
And even better still, Chris DeBona.
I'm sure if you're listening to this podcast, you probably know who he is.
He has expressed extreme interest in coming and boring other conflicts will probably be there.
Hopefully as a speaker, but we're still talking with him.
It's also it is possible.
And I'm not really a part of the Southeast Linux Fest because I'm not in the Southeast.
Yeah.
Yankee.
Yeah.
Yeah.
Yeah.
Yeah.
But some employees from Red Hat and iPad might be there because they're just up in Rollin' with Carolina.
So if any of you guys, if any iPad or Red Hat developers, people are listening, you know, take a look.
And maybe, maybe just maybe.
We may sneak you down here and we may do something like a live show there.
It'll all just see how it shakes out.
But if we do that, of course, it will be like, log radio live only minus all the BS.
Yes.
Mine's all the BS.
Mine's John O'Bacon.
Oh, that's right.
I should have said minus the John O'Bacon.
I should have gone with our running joke.
You should have.
You should have.
Although it would be amazing if he was actually there.
Okay.
Hold on.
Here we go.
It'll be like log radio live minus the Ubuntu Love Fest.
There we go.
That works.
So contact us any of those ways, forums, comments, email, IRC.
We love interacting and talking with you.
And we'll help you out any way we can.
Until next time, you've been listening to SourceGast. I'm Jeremy.
I'm KD.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
Flame the hardware vendor out of existence.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.
So until next time, remember, if you're having trouble with your hardware and Linux, just post a skating review on Dig and have 10,000 people there.