Damn is that just of SVGs or of built PNGs as well? Is it using git-lfs?
I think only the latest release logo is natively an SVG file. All of the older release logos are PNG and/or XCF files which account for most of the size of the repository.
$ du -hs releases/
488M releases/
The other big source is the wallpapers.
$ du -hs wallpapers/
250M wallpapers/
Looks like I hadnât pulled in a while. The repository contents are 765MB. 1.4GB if you include the git objects.
No, it is not using git-lfs.
Are we firm on having gaps? Without gaps it looks great and wonât have any issues when scaled down.
i also think no gaps looks fantastic.
Yeah and it avoid the short lambda neck! (I do like having them to when the logo is used in bigger size, but there is no real reason behind it, which in design is quite important)
Thanks to @zmberber and @djacu for their work on a new NixOS logo.
Iâm currently more favourable of the geometrically perfect design by zmberber.
On a small-ish side-note: When searching in the Matrix room of NixOS Marketing team, there is no message about djacuâs draft. Some about release logos (which are kinda gone by the new branding-repo as previously mentioned) and some more about users not wanting the logo be seen in pride colours. Meaning, I agree there was no real way for zmberber to get to know about djacus work, which now kinda appears to be set in stone.
Would all parties, mainly djacu and zmberber, be available to find a way to combine their crafts in developing the perfect rework of our beautiful logo? Thanks again to you both.
It has at least been mentioned here: 2025-04-04 - Branding Redesign Minutes.
I mean if I would start working on a logo I probably get in touch with the marketing team first.
I donât really mind whether we change anything about the current logo, but Iâm hoping that @djacu and @zmberber can bounce ideas off each other for future logo design efforts, e.g. future NixCons, local conferences, NixOS releases, etc. ![]()
Iâve taken a long time to look over this and think about it and the logo is going to remain as it is defined in the branding repository. Unfortunately, the triangular grid design described here is not compatible with the tooling I have built up and would require a significant amount of rework of the tooling and updates to the branding guide.
Still I think this idea is cool and want a place for community artwork to live. When I had the nixos-artwork repository archived, I created an Artwork page in the official wiki. I think it could look really nice with a little work and some contributions from the community. Also, there was some discussions recently in the marketing matrix channel to open a new artwork repository in the nix-community org. Both are great options to showcase community artwork. Neither is maintain by the marketing team and the wiki doesnât even require PRs to get artwork added. ![]()
Have fun
We can continue discussing refinements, but the final decision will be made based on what best serves the projectâs design goals, not on which version gets the most support in a forum thread.
It feels like the final decision was made by you, because you donât feel like investing any more time into your tooling.
I went and reread your communications and nowhere does it say that you are changing the proportions of the logo. We are always reassured that there will only be minor changes. I think that is the reason why there was no involvement from a lot of people.
Thank you for your effort, which is very important. However in my opinion it is not done yet, because the codifyed logo does not truthfully represent the real logo.
I implore you to reconsider your decision and let other people finish what you started.
Do you have any facts to back up this statement?
That is their problem that they didnât get involved, or at least watch what was happening. You canât blame the person who did the actual work that their work isnât good enough after everyone took no action. Peopleâs inaction is their own fault.
Then get involved with the marking team and âfinish the work.â
What is the ârealâ logo? What truth is there?
But reading the words of the original author of this thread, it seems they have no intention of contributing to the tooling, theyâve passively said that their work should be incorporated into the tooling; they did not say that they will do the work to incorporate their design into the tooling.
Weâve hit the very frustrating point that is all too common in open source: A lot of work has been done, but not to the satisfaction of those whoâve done nothing.
I got a little nerd-sniped with this ![]()
shows, both can be done with the triangle methodology. Likewise, I think both can also be done with @djacuâs methodology:
(Spoiler: If one makes the top 3/8 to 1/4, the 11/32 to 7/32, the 9/8 to 1, the 35/32 to 31/32, the 1 to 7/8, and the 31/32 to 27/32, one gets @zmberberâs version, at least modulo gap width, which is where the 32nds come from (but letâs not get hung up on gap width, which is a parameter of lesser importance.).)
I think the actual difference stems from whether one imagines the âtuckingâ or not. i.e.
That is to day, everyone agrees the top of the lambda and the legs should exist in a 3:2 ratio. (The sides of the acute angle : the sides of the acute angle = 3 : 2, to be a bit more precise.) The question is instead what do we measure to get those ratios.
-
With @djacuâs we just fill in the gap, and then we get that 3:2 ratio. The corresponds to a mental model where the lambdas are âabuttedâ, not âtuckedâ, and then the gaps are added.
-
With @zmberberâs we have to ârestoreâ not only the gap, but also the tucking, and then and only then do we get the original 3:2 ratio.
Speaking personally, I have always thought of the logo as having tucked lambdas, which âmorallyâ extended to meet the short leg of the next lambda as in the tucking diagram. I canât say way I thought of it that (maybe it has to do with everything being derived in my head from 1-D logo, i.e. the ones with the red line above?). So I am a bit sympathetic to this new version.
But procedurally, yes it is unfortunate that @zmberber didnât figure out that @djacu was also working on this sooner, and we are just hearing about it now. We donât want to have too much churn. It would be good to have @zmberber join the marketing team. If @djacu alone in his current role doesnât want to reopen this issue, maybe they both can instead work on per-release logos or something.
Finally, I want to say again that I think all this âBranding as Codeâ (reproducible logos from first principles) is really good, and really evokes just what the Nix community is about â the âalgorithmic journeyâ to the logo indeed I think really tells a grander story of Nix than the final âdestinationâ of the log itself, whatever it may be!
I told @djacu in person once that I thought the CAD drawings looked really good, and could be e.g. website backgrounds (if we can make them not too noise), because pictorially convey the journey really well â doing things reproducibly, with care! And I think that applies to @zmberberâs pictures above as well.
zmberber knew back as early as mid-May that I was working on this. It isnât sudden
indeed, i did know by then, i did reach out and we did talk briefly. i should have been more proactive back then with what i am saying here. i was already told that the tooling was there for me to try out and that several months of full time work had been put into it, stuff wonât be reworked, etc. i didnât want to pester around more, but i should have at least talked about the proportions, that is what i care about more than the tooling. only later did i fully realize that the proportions are being changed, and that is what i am writing in this post.
@paperdigits i am thinking about looking at the tooling and actually contributing. i tried to make it apparent in my initial post that i want to be careful and not complain too much or ask that other people do what i say after a lot of work has already been done.
i know that this is especially difficult due to the design of the branding guide, which is very elaborate and tied to the new proportions.
It seems, at least to me, that more would need to happen than just contributing to the tooling.
As you originally said:
I agree. So youâre either asking to change the logo, or this was an exercise for your own enjoyment.
yes and no.
i like discussing design, proportions and geometry in combination with aesthetics.
this post was meant to be for multiple things, also just to show the differences between the new logo and the old logo, which people are not aware of.
i named the post âa discussionâ, and most of my post concerns the existing design itself and how it (probably) came to be, and of course also other designs that come from these abstract constructions, as i tried to explain, which also include what djacu came up with. i did propose things, but my post is also aimed at just explaining the different designs themselves, just in terms of what they are.
i would like to repeat what i said in my original post that i am sorry for stirring things up.