I like pizza's idea a lot, it's based on how most major (non-github) OSS development is done and gives people who have no patching experience a taste of that workflow, but in a less "stressful" environment then major projects like Linux or a BSD.

I would just want to make an objection here. Even if we create such a maillist, to send patches, expose new releases, yadayada... It will be useless.
We don't have enough participation yet to justify a maillist. That's a fact. Only 3 projects live:
* one personnal project from tmplt
* one "community project" with only 2 persons involved for now
* one dependant from all other projects

Please make this repo live! Contribute, push code, check out other people work, submit ideas! This repo is the opportunity to build something together, let's not just build a whole infra that nobody will use...
Get involved!

Yeah that's why I said we can just do this stuff on the forum for now. A mailing list would be nice, but it seems as if people did not consider enough the next sentence where I said a mailing list is overkill, and the last sentence where I said I wasn't specifically advocating for mailing lists.

Anyway, I'll continue working on the ports repo. Maybe I'll start my own project if I get an idea.

Something I wanted to do (but I need permission from the copyright holders, "the iotek dev team") is to relicense urnn as ISC, clean it up (make it put all it's scripts in /usr/local/libexec, run from /usr/local/bin, copy the data from /usr/local/share/urnn to ~/.urnn if it doesn't exist and use that, etc. It's a cool, useful program, and I want it to be packageable.

I could just do all of that and leave the license as-is, but then no-one could include it in their repos.

I'll get started on that. But let this be a warning to all users of git.nixers.net: pick a sensible license! There are plenty out there, just choose one instead of writing your own!! (and WTFPL isn't a license)

(02-08-2017, 09:05 PM)pizzaroll1 Wrote: (and WTFPL isn't a license)

It actually is, explicitly defining Public Domain in countries which don't have implicit Public Domain. I deliberately choose the WTFPL for all of my (software) things for two reasons:

1) I don't care about what happens with my software as I don't make money with it anyway, nor do I want to have to care about that.
2) I don't intend to enforce anything. A license that contains "you must" and/or "you must not" is restrictive by definition. Even the ISC License - a generally fine license, I admit - violates this to some extent by forcing developers to add copyright and license things into their code which nobody will ever read anyway.

Which license - except the WTFPL - does this and why should it be a better choice?

edit: Also, WTFNMFPL if you live in one of those "I'll sue you" countries; but seriously, that's already too much text. Note that licenses should actually be read before using a software. Nobody will read this. (This also applies to the CC0 legal text.)

The WTFPL has exactly one easy clause. Beat that.

What I perhaps meant to say was that the WTFPL isn't a sensible license.

The problem is, for anyone who wants their software to be used by anyone, the WTFPL is dangerous since it doesn't include a disclaimer clause. You say "one of those "I'll sue you" countries", but the reality is that basically every country is potentially an "I'll sue you" country, and if I ran some open source project, I wouldn't want to live in fear of being taken to court because I failed to choose a good license.

I am not telling you to stop using WTFPL because I don't like it or something (although I don't). I am advising you to stop using it because it puts you at unnecessary risk.

If you like the ISC license, just take it and delete the "provided that the above copyright notice and this permission notice appear in all copies." bit. The license is widely accepted, provides adequate protections to you, and places no restrictions on the users (if you delete that one clause).

I don't see why anyone would want to use the WTFPL. The OSI rejected it because it didn't add anything new to already existing licenses (it's just a more ambiguous and less safe version of some permissive licenses). It's plain to see that it was written as nothing more than anti-GPL satire and should be treated as such (i.e. not seriously).

Sure you can use it, no-one will stop you. But I would never want my name on a project without a real disclaimer and neither does any software author, really.

(02-08-2017, 09:18 PM)jkl Wrote: Which license - except the WTFPL - does this and why should it be a better choice?

I think we've discussed this already you and I when I made the research about licenses.
There's the CC0, Creative Common 0, here it is.
It's a bit more "professional" to waiver your right, let's say, than to have a "wtf" in your license.

Between, you even said the following, in the podcast discussion:

Quote:Note that, unlike other Public Domain licenses, the WTFPL does not have a liability clause so you could still be sued if your WTFPL'd application causes any damage. (Not proven yet.)

Which is your own arguments against the wtfpl.

If anyone is up for the tasks (simple ones), I'd like to test the ascii2svg script with different kinds of ascii/ansi files and update it.

It still needs to have configurations for some tuning like choosing the font and color scheme to use.
Maybe this could be made so that the colors are extracted from the Xresources or that there's a flag to use the default ANSI colors.

The font and other metadata could also be extracted from the SAUCE line.
Maybe the usual ascii-artsy fonts (Topaz, P0T-NOoDLE, MicroKnight, mO'sOul) could be linked somehow.

Once that's done I think it'll be good for packaging.

Aaaaah Licenses. What a waste of time :D

I used WTFPL for long because it's cool. Then gave up when I started to care about people seeing my projects and use them (having a real license makes your program look more professional).
I'm now down to use two licences: ISC (when I want to keep it) and UNLICENSE (when I don't). This way people are happy and stop bothering me about my license, so they focus on the code itself :)

(03-08-2017, 02:32 AM)venam Wrote: If anyone is up for the tasks (simple ones), I'd like to test the ascii2svg script with different kinds of ascii/ansi files and update it.

It still needs to have configurations for some tuning like choosing the font and color scheme to use.
Maybe this could be made so that the colors are extracted from the Xresources or that there's a flag to use the default ANSI colors.

You can use the motd and xero's lab for it!

(03-08-2017, 04:06 AM)z3bra Wrote: You can use the motd and xero's lab for it!

Most of those aren't ascii/ansi art, they are shell scripts.
Another TODO would be to add support for UTF-8 but that's not standard in ascii-art.

Bookwyrm is currently licensed under the GPLv3+, mostly because pybind11 is as well. Will this cause trouble during packaging?

(03-08-2017, 05:19 AM)venam Wrote:

(03-08-2017, 04:06 AM)z3bra Wrote: You can use the motd and xero's lab for it!

Most of those aren't ascii/ansi art, they are shell scripts.
Another TODO would be to add support for UTF-8 but that's not standard in ascii-art.

The MOTD repo only has ascii/ansi test files. For xero's, I'm fairly sure your tool is well designed and fits the unix philosophy, so you could do something like

Right?

(03-08-2017, 08:30 AM)z3bra Wrote: The MOTD repo only has ascii/ansi test files. For xero's, I'm fairly sure your tool is well designed and fits the unix philosophy, so you could do something like
$ ./xero-ansi-script.sh | ascii2svg > xero.svg

Right?

Sure, that's a great idea.
I only have to change 2-3 lines to accept input from STDIN.
However, I still have to support utf-8 for those to work.
At the moment I'll test on files from http://pc.textmod.es/ .

Tmplt Wrote:Bookwyrm is currently licensed under the GPLv3+, mostly because pybind11 is as well. Will this cause trouble during packaging?

No. Any widely approved-of free license is fine. Any license on this list is 100% ok: https://opensource.org/licenses/alphabetical. The FSF's list is also good: https://www.gnu.org/licenses/license-list.en.html.

But some of the licenses on that list come with a comment saying they don't recommend their use (all of the nonfree ones, for instance. And the WTFPL).

So you have nothing to worry about.

I've always used BSD 2-Clause aka the FreeBSD License, as it is (1) GPL compatible for those that care about such things (I don't) and (2) has a simple non-liability clause, which is really important as I once learned.

A few years back, the org I worked for at the time was dealing with a major issue where legal was doing an audit of our software licensing. At first, the legal team was trying to get in contact with the respective authors of our open-source licensed software, and ask them to issue us a special license removing the liability clauses, making the author legally liable if their software caused any major issues in our environment. As I'm sure you can imagine, they had a 0% success rate with that. The "reasoning" behind this was that Microsoft and Adobe offer liability and support, so everything else should.

After a few months of no progress, the lawyers gave up with the special license crap. For a while, we were afraid that we'd lose any access to open-source licensed software, but in the end legal gave up on OSS and focused on rich companies that would actually sign liability modifications. Regardless, it was a stressful ~6 months for IT/Dev.

The forums could also be a convenient way to talk about the nixers projects. Though, mails are universal and well known by git.

We do probably not want to see patches in [ code ] blocks.

A public mailing list with a web front end could be more suited.

(05-08-2017, 08:51 AM)josuah Wrote: A public mailing list with a web front end could be more suited.

Before having a mailist/web frontend, we need collaborators. So yeah, let's wait to see a few patches come up before adding yet another mean of communication :)

Sharing patches for review over iopaste has proven good enough for me in the past btw

(05-08-2017, 03:15 PM)z3bra Wrote: [quote="josuah" pid="18798" dateline="1501933899"]
Sharing patches for review over iopaste has proven good enough for me in the past btw

So let's go! :)

There are a lot of symbols not being found when pushing:

I just recompiled stagit / stagit-gopher so it should be fine now. Try to run "ssh git@nixers update-stagit bookwyrm" to confirm.

Confirmed; no errors now.

(03-08-2017, 04:06 AM)z3bra Wrote: You can use the motd and xero's lab for it!

those scripts are all unicode block characters. the fonts venam wants to test are amiga / CP437

@venam, you should test it on your new bugs study from the last impure pack

(17-01-2018, 01:48 PM)xero Wrote: amiga / CP437

@venam, you should test it on your new bugs study from the last impure pack

The script it working fine so far, I wanted more edge cases.
I haven't worked on it in a while.

Apparently, the web front end is down (at least for me). Gives

Yop dudes.

I'm bringing back this topic to life, 'cause I plan on rearranging my web services over all my servers.

Is anyone still interested in this community git server, or should I bring this down?
I see the last contribution was from May, and most other contribs are from 2017.

This was created to ease contributions within the community, but I think it was a huge failure in the end (at least, from my POV, and the project I've shared through it).

There could be multiple reasons for its failure, be it a lack of interest in code made by the community, a preference set over github, or simply being shy?

I don't know. What's your idea? Should this server live? Die? Do you miss something that would make it more appealing to use?

Please share :)

Hi z3bra, using community developed software myself, most of if on github and rarely on own git repos. Still thinking it's worth to save it for future even if you planning to commit there alone. And if it's not big maintenance costs in hours and cash.

Or move to github organisation and forget about maintenance. The simpler - the better.

(29-11-2018, 10:03 AM)z3bra Wrote: This was created to ease contributions within the community, but I think it was a huge failure in the end (at least, from my POV, and the project I've shared through it).

I just think it comes down to the fact that when a platform like github has become the norm it is very hard to break that habit.

It's like newtons-whatever-law where he says: "Objects at rest will stay at rest unless acted on by an external force" unless something comes along that motivates people off platforms, they will likely continue to use what they always have.

I'm a newer member so this could be dead wrong but it seems to me as if a community git is a neat idea, but we are too few in number and too disorganized for it to work well. Think of Conways game of life, the project has too few adjacent squares to continue to live for the next generation.

Well, I just though it was cooler to host our shitb ourselves :)
I trust every member here better than microsoft for hosting my projects and intellectual property.
We are most likely not safer, but more honest, and that is what I care about.
Now that is my personnal POV, and I understand why people would prefer github as it is more of a social network than a code hosting platform.

Regardless, unless someone stands against it, I will consider tge service to be dead, and bring it up again if it goes down (be it cause of a reinstall, or hardware issue.

I can provide archives of projects if people ask me :)