RHEL 8 Beta - Half Baked Release

RHEL 8 Beta version was announced a few days back and I was utterly disappointed with the direction RedHat or should I say IBM has taken.

Maybe it is the fault of the upper crust in RedHat that have purposely delayed RHEL 8 to drum up their "new" technologies that were just not ready for prime time. The technologies that RedHat were peddling and showcasing to be included (again IMHO) in RHEL 8 were DNF 3, GNOME 3.30 / 4, GIMP 2.12 / 3.0, Pipewire, Silverblue, Removal of Python 2 entirely, and the Latest and greatest Wayland.
These all were described in my previous article here -> https://aajkyakaroon.blogspot.com/2018/06/rhel-8-release-date-update-2-chasing.html

But, these "new" technologies were never going to be stable enough even in Fedora 30. Wayland is 'Way' behind in terms of features present in the current display server. There is not even a stable remote desktop connectivity option enabled by default in Gnome in Wayland.

RedHat wasted at least a year to drum up these new technologies and delayed RHEL 8 for absolutely no reason at all. What IBM did is a sane decision of releasing RHEL 8 and then updating the release to include  these "new" technologies may be in RHEL 8.2/8.3.
If they decide against updating the release to include these "new" technologies in RHEL 8 then I think that RHEL 9 will not be much far behind.
Or, my hope that these "new" technologies will be stable enough in 2 years is misplaced. Hmm, may be this is the case and that's why Canonical has decided to support Ubuntu 18.04 for 10 years. May be these technologies are just in their nascent stage and will require a decade to mature.

In this current form of RHEL 8, I do not think there would have been a major difference in basing the release on Fedora 26/27. This would have saved approximately a year of time.

IMHO, a RHEL refresh was required in 2017 and this RHEL 8 of 2019 is just not justifiable. May be we can add a few repositories in RHEL 7 to get a new kernel and newer versions of software and just skip RHEL 8 till at least RHEL 8.2/8.3.

Best Open Source and Free games and tools for Linux

I was quite inspired with a series of posts on opensource.com that listed five open source games on GNU/Linux systems, in six different genres.

The links to these articles are:

• 5 open source strategy and simulation games for Linux -> https://opensource.com/article/18/8/strategy-simulation-games-linux
• 5 open source role-playing games for Linux -> https://opensource.com/article/18/8/role-playing-games-linux
• 5 open source racing and flying games for Linux -> https://opensource.com/article/18/7/racing-flying-games-linux
• 5 open source puzzle games for Linux -> https://opensource.com/article/18/6/puzzle-games-linux
• 5 open source card and board games for Linux -> https://opensource.com/article/18/3/card-board-games-linux
• 5 arcade-style games for Linux -> https://opensource.com/article/18/1/arcade-games-linux

While these articles are quite good, but they are also very limiting. These articles mention only those game that are fully open source, i.e., the games that have their source code and all assets freely distributed. Also, these articles cover only those games that the author knows or are very popular in the community.

While I am a GNU/Linux, i.e., Peppermint OS, user myself, I am not so strict in my preference of the tools that I use. I am a GOG and Steam user and my games library in these two together has swelled to more than 100 games (This is mostly because of the excellent sales on these two sites and the bundles through the Humble Bundle site)

In the coming articles in this blog I would jot down the many games and tools that can be freely used on GNU/Linux systems. Also, I won't be strict in following genres and would list a few games/tools in each post.

RHEL 8 Release Date Update # 2 -> Chasing Perfection

RHEL 8 Beta has been released and my take on this is presented in this blog post --> https://aajkyakaroon.blogspot.com/2018/11/rhel-8-beta-half-baked-release.html

Please remember -> This blog post is just my speculation.

Something is amiss with RHEL 8. RedHat is chasing perfection or may be the current technology stack that RedHat wants is just not ready yet.

Some of the technologies that RedHat wants to include in RHEL 8 are (from my understanding):

• DNF 3
• GNOME 3.30 / 4
• GIMP 2.12 / 3.0
• Pipewire
• Silverblue
• Removal of Python 2 entirely
• Latest and greatest Wayland

DNF 3 development has just started in March 2018 according to the article on https://rpm-software-management.github.io/
What will DNF 3 give that DNF 2 does not give? Frankly, the answer is what we all want to hear. Better performance, consistency, save the world, and this is just the tip of the ice berg. And, we definitely want to whole ice berg, don't we?

Why (always) the next GNOME. Because, even in GNOME 3.26 the remote desktop capability is not enabled by default (on Wayland). This is the biggest bottleneck in releasing an enterprise OS.
When will GNOME have this capability. The answer is always the next one. And when the next one arrives, the answer will still be the next one. I hope they get it right before GNOME 4, otherwise, as many have stated, GNOME 3 series is just a pile of hot *sauce*.
Who gave these people a million dollars? Hmm, beats me, cough RedHat cough.

But, this capability is tied to the Pipewire back-end. Pipewire is an under development multimedia wizard that will handle all video and audio use cases. The first priority is for video since we already have PulseAudio/Jack/ALSA for audio. In the future, Pipewire will replace these audio middleware and be an all encompassing technology for all things multimedia. The current state of this project is written in this article -> https://blogs.gnome.org/uraeus/2018/01/26/an-update-on-pipewire-the-multimedia-revolution-an-update/ (homepage for the article is https://blogs.gnome.org/uraeus/)

Oh, and the remote desktop capability in RHEL is also tied to Wayland. Because, RedHat has its heart set for using Wayland and no other.
Now, Wayland does not specifically state the way that can be used to enable remote desktop capability. Why? I do not know. May be Wayland can go different Ways. Wayland does not care which Way you take to get from point A to point B. Just pick a Way man.
So the good fellows at RedHat have chosen to dig holes, I mean Pipes, I mean, they chose to carve their own way and made Pipewire. So, at the moment, Pipewire is the only way that remote desktop capability can be achieved using Wayland in all Desktop Environments. Even the folks at KDE are using Pipewire to implement remote desktop.
The maturity of Wayland and Pipewire is the main bottleneck in implementing foolproof remote desktop capability.

RHEL 8 is also targeting a full transition to Python 3 and a full removal of Python 2. Well this cannot be done if they choose to include the current GIMP 2.10 version. The 2.10 version is still dependent on Python 2 and the GIMP devs have stated that it will require at least till the next version to be fully free of Python 2. What will be the next fully Python 3 GIMP version, I do not know, may be 2.12 or 3.0.

Silver bullet, oh I mean Silverblue will be the silver bullet that RedHat hopes will kill all the werewolves of package management. So, why all this work on DNF 3? Hmm, covering all bases I see. Flatpak all the way is the RedHat answer to Windows 95 in 2019/2020. So, we really are around 25 years behind the proprietary world. Who knew.
I always thought that GNU/Linux package management was the superior way of sharing open source software. Oh, now I get it, rpm and apt is not helping the proprietary software gain ground in GNU/Linux. So we just bend over and present our "better" side to them.
So, get ready for one off software products from the leading proprietary software makers that will not get updated or patched ever again. You know, the same as Windows 95 to Windows 10. We will become so privileged.

So, when will RHEL 8 be released. When the stars will align and the Sun and Moon are dancing. Well, obviously when all the technology is matured to a point where they can be included in a Fedora release. But, when will that happen. Well, Fedora 28 is gone and Fedora 29 is due sometime in the last months of this year (30 October 2018 - from https://fedoraproject.org/wiki/Releases/29/Schedule) I am more inclined to say Fedora 30 will be the one that may have all the technology stack included. What is the release date for Fedora 30? From this page -> https://fedoraproject.org/wiki/Releases/30/Schedule the date is in May 2019

In conclusion, RHEL 8 alpha/beta will arrive after Fedora 30, may be somewhere in 2019 and we may get to see the Sun, I mean RHEL 8, may be in 2019, if we are lucky, or most likely in 2020.

My Peppermint OS Upgrade to Version 8

I had been using Peppermint OS version 7 for the past little more than a year and I just decided this last week to upgrade my setup to version 8.

Peppermint 7 cannot be "upgraded" to version 8 but we require a new install. This made me curious to try a few other alternatives as well.

I downloaded 4 distros:

• Manjaro XFCE 17.1.9 64bit
• MX Linux 17.1 17-Apr-2018 64bit
• Linux Mint 18.3 XFCE 64bit
• Peppermint OS 8 Respin-2 64bit.

I "burned" these distros in 4 different DVDs. As you can see in my choices above, I had selected the distros that have nearly identical user interface. My HP Notebook has low end hardware that has a low end Pentium processor with no other graphics card. But, the notebook uses proprietary Broadcom Wireless drivers.

Before I jot down my experiences with these distros, let me first state what I like about Peppermint 7: Firstly I like the low requirements and snappy performance. The screenshot scripts based on the Scrot application. These scripts can be replicated in the other distros manually. I do not like the XFCE 4 Screenshot utility - it just gets in the way and does not help in getting a quick screenshot. I also like the familiarity of the debian based package installation. The tiny things, like single click to select, double click to open; the dark theme, and more.

I thought of seeing and comparing these 4 live distros.

First I tried Manjaro. The interface is very good. The system is quick. The software installation was not too hard for someone like me who has been accustomed to the apt commands. But, the problem was that there was no way that I could figure out how to install the proprietary Broadcom Wireless drivers. Maybe this requires a more in-depth search and digging into the forums. So a no go on Manjaro at this time.

Then, I tried Linux Mint. I was a Linux Mint user for quite a few years previously. So, this new 18.3 version was very familiar. But what did not work was proper dark theme integration. I thought that the dark theme was one of the appeals of the Linux Mint 18 series. But, the XFCE version does not get proper dark theme integration. May be after XFCE is fully ported to the new GTK version, we will see proper support for dark themes.

Next, I tried MX Linux. This one surprised me. A very easy application management interface helps in installing popular apps. Just select the app and click install. The conky script adds character to the desktop. The dark theme is perfect. I changed the screenshot thingy to the Peppermint script - just for the Print screen button and it worked as it should have. Wireless worked out of the box. So all was good. MX Linux also has updated LibreOffice. So I decided to install this distro.
But I just could not figure out the Partitioning tool given in the installer - GParted. This just seems archaic and is just not intuitive. I was not able to do anything in GParted. May be the MX Linux team should use an established installer to appeal more to common folks like me. So a no go.

Back around to Peppermint OS for me. The version 8 is more snappier, more 'darker' - in a good way. Every thing is where it was - the screenshot scripts - the dark theme with a nice upgrade - apt package management - proprietary drivers - everything. So, this is now what runs my HP Notebook.

Just a little help for those like me: You definitely need a separate home partition. This will help in keeping your files safe in cases of OS upgrades/re-installation. My hard drive now has these partitions:

• 250 MB EFI (The recommended size is 100-250 MB. I think 100 MB is more than enough if you have just one or at max two OS on your system)
• 8192 MB Swap Space (I have this at double of RAM size - Same as RAM size will also suffice)
• 26624 MB Root partition (You could make this a little less - I think 15-20 GB is enough)
• The rest of space is the Home partition.

Peppermint OS is a good option for low resource systems. Good proprietary driver support. Amazing dark theme. Gives a clean interface and it just works.