Re: [Infotropique] notice

Delete this message

Reply to this message
Author: Nils Gillmann
To: infotropique
New-Topics: [Infotropique] temporary log location (was: Re: notice)
Subject: Re: [Infotropique] notice
Nils Gillmann via Infotropique transcribed 4.8K bytes:
> Quiliro Ordonez via Infotropique transcribed 535 bytes:
> > El 2018-09-17 05:00, Nils Gillmann via Infotropique escribió:
> > > Hi all,
> > >
> > > this is just to let you know that I am still busy with the project. I
> > > have decided I will make the next message (after this one) the one
> > > where I announce a first release you can download + useful websites
> > > and mdoc/man pages.
> > >
> > > I made good progress, but it could still take a long time. Sometimes
> > > I write not so short notes about the work at
> > >
> >
> > Nice. Is there a way that those notes are sent to this mailing list?
> Hi Quiliro,
> Not really.

This is really bad for the start, but I'm in-between websites (at least
for the easy uploads with autogenerated rss). Would a folder with text
files per day work out for you?
In the long run I'll improve it, probably use one of my old static
compilers, or hastysite or something else.
Mainly I'm looking for an easy way to write:

2018-09-23: ports: Added slrn, icecat.
            plant: Did ....

And write 'make' or whatever and it upload to the folder on my server.
This would just be an extension to what I write in the git logs
and the many TODO files in the repositories.

Later on I would just merge this into html files.

> That's pleroma with ActivityPub federation there. I briefly
> recall someone pondering about hooking Email in there or fixing old style
> rss.
> The code is currently still moving around alot, and the communication
> about it is mostly between Devan and myself offlist in a way which
> would not really fit into something you can adjust to Email or Blog
> posts. I tried Blog posts for a while, but this takes time.
> Mainly what I'm doing right now, and what can be followed on
> for most repos is this (without further
> explanation, which will follow):
> - fix up plant to compile itself
>   - once that is done, merge in Guix work after the 0.15.0 release
> - move and work on packages in ports and ports-wip.
> - recently I made the discovery that you can even place build-systems
>   whereever in the load-path, so now I'm closer to an initial ports
>   layout.
> - Dissecting guix' (gnu packages) modules in ports and plant,
>   leaving only the bare minimum packages in plant (infotropique packages)
>   and moving the rest into ports.
> - Move my local backlog of patches+packages into ports-wip
> - Getting familiar with ways to build at least FreeBSD and OpenBSD kernels
>   from source in non-native environments. This is part of a mutilevel
>   approach with these goals:
>   - run plant (and eventually also Guix) on *BSD systems as a native
>     application ("porting plant")
>   - build *BSD-kernel based systems with plant system command (and
>     later on templates)
>   - build *BSD userland software in non-*BSD environments and discover
>     how to work around the issues which will appear.
>   - build GNU userland software in *BSD-environments (...)
>   => *BSD (OpenBSD, FreeBSD etc) here are just trampoline projects to
>      learn more about porting to these platforms. personally I want
>      the support and modularity as described above in the very long
>      run.  We've been talking about Genode support after one of these
>      steps and once we understand Genode better.
> - writing build-systems to comply with some of the changes wrt
>   hardware support etc.
> - dissecting Xenocara to see how/if it can be build in a different way
>   than OpenBSD builds it
> - porting a couple of applications
> - translating a bunch of Texinfo manuals to mdoc so that in the long
>   run we ideally can drop Texinfo as an requirement for the base
>   system.
> - writing the websites
> - experimenting with co-existence of multiple package managers as
>   an ... well, experimental... feature.
> - gnunet-service for shepherd once the blocking bugs in gnunet are
>   fixed
> - reading into lexing and parsing for the templating system.

> just off the top of my head, there's many more items I work on.
> Updates on some of these: ports is becoming increasingly
> selfhosted.
> I have this best-effort upstreaming approach. I apply changes in
> a different way (like for example trying to default to libressl
> when I know it is safe to do so). Changes and inovations will be send as
> proposals to Guix, applied to Guix structure. For example we have been
> working on systemwide integration of Nim. Once this is done and shaped
> to the point where I move it from ports-wip to ports I will eventually
> send it to Guix (eventually: once I have time etc).
> I hope to move my personal system from the hybrid state I have right
> now to infotropique OS/core with my personal template additions in
> the next 3 to 12 months, but I don't like to set deadlines. Obviously
> I really want this, but I can't work 24/7 on this.
> gnunet fs itself is the real blocker for the really exciting features
> I want. We have to spend some time thinking about my original ideas
> which based on the past explorations and discussions of guix + gnunet.
> Devan has done some quality work with discovering ways you can use
> gnunet cadet instead. We had an distributed (and because of my ISP
> connection: slow) shared video stream :)
> Ideally I get to speak at FOSDEM 2020 or 2021 about this project,
> maybe earlier. who knows.
> I mean, hey I have send nothing since April, so obviously some
> people are interested even though I'm working right now only
> for myself.. which is cool.
> Essentially, what I learn and work on is all over the place, so
> sending Emails (or setting up an rss to email) is not the best
> way to communicate I guess.