Comments (7)
@eero-t thank you for the feedback! As you highlighted, the goal of the document is to explain why Clear Linux packages and updates software the way it does and point out the benefits. The goal and language isn't meant to accuse that everyone would have experienced these pain points on other distros.
I think the order in which the information in the "Versioning" section is presented can be revised to make sure the focus is on Clear and mention some missing benefits, like package dependency resolution. Of course it's important that we remain humble and not overestimate, so I appreciate you bringing this up and if you have any specific changes you think should be made in the document please let us know!
With the later points around package search and patches, it would be a better discussion if you can raise the issues here: https://github.com/clearlinux/distribution/issues.
from clear-linux-documentation.
Hi @eero-t. Thanks for your feedback on swupd-about. We're planning to revise some descriptions of the Clear Linux* OS features related to swupd
, thanks in part to your feedback.
Some questions you ask in the penultimate bullet list are already under consideration. Please be patient.
Swupd, and swupd update
specifically, offers users several benefits. One benefit of this is that a single number in a Clear Linux OS release confirms that all package dependencies are resolved at build time. This saves developers the work of resolving conflicts by themselves.
At the bottom of Downloads is a link to Clear Linux OS releases:
https://cdn.download.clearlinux.org/releases/. Feel free to explore.
If, for example, you wanted to see versions of packages for the most current release. You may go here: https://cdn.download.clearlinux.org/releases/current/clear/source/. Then download and view the packages-sources
file.
HTH
from clear-linux-documentation.
At the bottom of Downloads is a link to Clear Linux OS releases: https://cdn.download.clearlinux.org/releases/.
Ok, great, while this info isn't on the ClearLinux installation itself, it's at least now easily accessible. And to see what patches/changes have been applied, I can download and extract the source packages.
(I was surprised to see that you're still building Mesa with autotools when Debian/Ubuntu have switched their builds to faster meson/ninja already 3 months ago.)
One benefit of this is that a single number in a Clear Linux OS release confirms that all package dependencies are resolved at build time. This saves developers the work of resolving conflicts by themselves.
The cases where people need to do manual "package dependency resolving" are after they install packages from somewhere else than their distro's own repositories. If one wants to install packages from elsewhere to ClearLinux, they need to be building ClearLinux themselves. Yes, you don't then have dependency problems in your target device, but you need to do large amount of additional work before hand. To me that's a difference, rather than an advantage (avoid problem by hand-cuffing the user).
Or did I misunderstand the problem being solved?
from clear-linux-documentation.
(I was surprised to see that you're still building Mesa with autotools when Debian/Ubuntu have switched their builds to faster meson/ninja already 3 months ago.)
Clear Linux uses autospec to build mesa, and in this case, autospec selects autotools as the build pattern instead of meson. Selecting a build system when multiple options are available is autospec's choice, unless a build pattern is forced via the build_pattern
config file. As far as I recall, autospec didn't give preference to meson over autotools.
from clear-linux-documentation.
The cases where people need to do manual "package dependency resolving" are after they install packages from somewhere else than their distro's own repositories. If one wants to install packages from elsewhere to ClearLinux, they need to be building ClearLinux themselves. Yes, you don't then have dependency problems in your target device, but you need to do large amount of additional work before hand. To me that's a difference, rather than an advantage (avoid problem by hand-cuffing the user).
Or did I misunderstand the problem being solved?
In fact, much of our documentation is geared toward Linux developers who do build a Clear Linux image themselves for a variety of use cases. Developer Workstation shows a brief overview of some of user roles, and bundles relative to role, which may be of help to you. Using mixer, developers can build a custom OS for a specific use case and act as their own OSV, if they choose. We feel the flexibility mixer offers, along with our autospec tool used to generate your own RPMs, provides a lot of freedom. Yet, as with most things, βwith freedom comes great responsibility.β We're constantly seeking to make it easier to the customize a Clear Linux OS implementation.
from clear-linux-documentation.
As far as I recall, autospec didn't give preference to meson over autotools.
Ok, if a tool does that decision instead of maintainer, then it makes sense to be conservative.
FYI: Git versions of few projects (Weston, Mesa...) require now an extra option if one wants to continue building them with Autotools, and will soon remove Autotools support completely.
from clear-linux-documentation.
Checked the document changes.
Main error there is currently this:
"On a package-based OS, system administrators update each individual package or piece of software."
If it would instead say "... can update each individual...", it would match reality.
(Individual package update is done automatically by the package management SW for the administrator. System administrators do "apt update && apt upgrade" to get upgrades, they don't upgrade things package-by-package unless they really want to.)
I assume that "extremely efficient" comment(s) apply to a future version of swupd which isn't anymore significantly slower than the packaging tools in other distros? Document says swupd to be better in downloading the updated content, but the 60MB download (of metadata), which it seems to (almost) always do before it starts doing the actual update, makes it slower than the other distros.
from clear-linux-documentation.
Related Issues (20)
- RPM build errors when change the version of kernel to compile HOT 2
- is this file supposed to be missing?
- can i install clear linux on bear metal without Carry-less Multiplication extensions (pclmulqdq)
- broken links HOT 1
- wrong information HOT 3
- "this utility is not required"
- clr-fwupd-update
- mailing list link on clearlinux.org is pointing to the staging site (403 error) HOT 1
- nvidia drivers page lib64? HOT 1
- Screen Media Style for Terminal Contrast HOT 2
- Design Flaw Menu Folds
- Example password hashes in iPXE guide don't work
- Documentation Website Software Search not Returning Queries HOT 2
- Kata Containers - https://github.com/clearlinux/clear-linux-documentation/blob/master/source/tutorials/kata.rst HOT 2
- Deprecation warning on disable-ticketing HOT 1
- Installer fails if there is a previous LVM partition layout HOT 1
- Broken link HOT 5
- Broken Links on Clear Linux OS and Containers page HOT 5
- Broken documentation on the website HOT 7
- Follow Doc and deploy Clear linux in Google Cloud, login still need password ? HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
π Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. πππ
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google β€οΈ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from clear-linux-documentation.