Releasing this on April 1 may have been suboptimal, but I had seen the draft before and was already positively impressed back then. I'm especially happy about the scalability of the original vector graphic – why is the logo at the top left still a PNG, though?
Project talk:Proposal for changing logo of MediaWiki, 2020
I don't think MediaWiki's logo configuration supports SVGs.
That's not true! (Or at least it worked nicely enough when I used one a year or two ago)
The Timeless skin still shows the old logo for me.
Not the case for me, can you refresh/purge the cache?
Thanks so it is simply a cache issue.
does a new favicon come with it? I still see the sunflower as the favicon
This is probably due to caching effects, where your browsers assumes the old favicon still to be in place. If you clear your Browser cache (for a given page, Strg+F5 should do it in most browsers), you should see the new favicon.
If you are still seing the old favicon afterwards, in particular if it is on MediaWiki.org, please let us know.
Ah yep, a hard refresh did the trick. Thanks!
There will be a lot of caching effects related to this for a little while, not just on the mw end, but in proxy servers / acceleration services and browsers that are outside of WMF control.
That is all :)
Thanks. We are still waiting it to be cleared and registered by WMF legal so we can roll it out. Hopefully in a month.
Rather than jump right into voting on logos, consider starting the logo redesign process by writing a w:Creative brief. Consider asking for advice from the Wikimedia Design Team and well known like (w:Susan Kare, w:Jeffrey Zeldman, w:Simon Oxley). A lot of designers will see this as a significant opportunity since there are few opportunities to design a logo for a well known software project that will also appear on every single page of Wikipedia in the "powered by MediaWiki" badge. A design brief might include the following topics:
- Background — what is the background of the project? Why is it being done?
- Target audience — what do they already think about this subject? Is there anything that should be avoided?
- Profile of the typical user or consumer
- Key insight - what has been learned about the market's attitude to the company, brand or product
- Single message — what is the one thing to tell the audience? What is the single thing they should remember about the offering? How will they believe what we say?
- Desired customer behavior? (e.g., trial, purchase, recommend)
- Tone of voice (e.g. serious, humorous)
- Mandatories (mandatory elements, like the sunflower)
- Timeline
- Budget
- Approvals (who needs to approve)
I agree that soliciting logo ideas is a great idea. I wouldn't bother with a detailed creative brief, though, because getting people to agree on this kind of thing ("key insight", etc.) is probably even harder than getting people to agree on a logo.
I agree that it feels premature to start directly with a vote on specific outcomes. That said, if the vote was status quo vs. almost anything else, I'd vote in favour of changing away from the logo, so…
@Jdforrester, DVrandecic (WMF), ZMcCune (WMF), Quiddity (WMF):
Is it me only, or the new logo ("pink flower", no longer a sunflower) looks very much like the logo for BP (except it is green, not pink). In 2001, in response to negative press on British Petroleum's poor safety standards, the company adopted a green sunburst logo and rebranded itself as BP plc ("Beyond Petroleum").
Note that the BP green sunburst logo even predates the yellow sunflower logo used for MediaWiki (Mediawiki was created in 2002, one year after BP adopted its logo): we cannot argue that our use of the *generic* sunflower since years grants us an exclusive right, it was acceptable *only* with the distinctive "MediaWiki" wordmark added, making it really distinctive.
See BP in English Wikipedia (showing the logo w:en:File:BP Helios logo.svg hosted only in Wikipedia under "fair use", not suitable for Commons where it was deleted multiple times).
Note that the chosen File:Mediawiki logo proposal (gradient translucent, capitalised).svg logo uses translucent colors (not a problem), but also gradients, which can become a problem:
- Printing it will be difficult; the former logo for Mediawiki was already difficult to print.
- When registering it for protection, we'll have difficulty to protect it with a black&white only design (and distinctively enough from the design for BP logo if it was registered as well in black&white only); the former logo could not be protected and was not protected.
In monochrome version (or for colorblind people), the confusion between BP and Mediawiki will be evident without the added wordmark!
This means that the new logo only works with hi-color devices where it is sufficiently distinctive but will stioll not be a strong mark without associating it with the wordmark "MediaWiki" (this was already the case, but the horizontal metrics of the former logo allowed placing the wordmark easily below the logo to fit a square.
Finally the mere image of a flower (without the wordmark) does not qualify it for registration and a protection: there are **lot** of uses of flowers/suns/mandalas internationally, including on national or religious flags (e.g. in Taiwan or India). So we've not solved anything: what is protected is the wordmark "MediaWiki" (independantly of the presence of absence of the logo above it).
We could have designed something that was still a sunflower, but not needing high-colors, preserving the identity of sunflowers (notably the central area with the golden-ratio spirals, absent from the BP logo), but more symbolically represented rather than being a mere photo. We could have then kept the yellow/orange colors without requiring hi-colors, and with a black&white shape that was still distinctive abnd not requiring the presence of the wormark (still impossible with the new logo for its "favicon" at 16x16 logical pixels, rendered at 24x24 or 32x32 resolution on HiDPI devices).
Unfortunately, the contest for the 2020 Mediawiki logo occured in August only, and it was a too short time to get many people contribute to it (and many were in holiday) so the short vote (2 weeks in September, still many people were taking their holiday only at that time, and it was not anounced to anough people on enough Wikimedia sites or other channels) was biased by the limitation of choice and even if it was "adopted", the WMF legal department has still not been able to formally announce its adoption.
Designing a logo should be a long process with a long period of propositions and discussions: voting immediately after one month of proposition (and with a procedure that was not formalized before the final vote) is not a good idea (this has already caused a problem with Wikivoyage that was required to change it logo in 2013 (after an legal action initiated by the WTO).
For this reason, the period of submission should be at least 3 months (since its announcement to as many places as possible). We should have the time to prepare at least 6 very different designs, discuss all technical and legal problems, fixing them, discussing of alternatives for specific uses (e.g. monochrome, low resolution, printing, search for similarities).
Just a tip for the designers: what about just testing an high-definition sunflower?
The problem is that the sunflower looks best only with hi-color devices. For printing, or on monochrome displays, it looks very bad (as the needed details have very low contrasts): you then jsut see a flower, all yellows are flattened. This makes a very poor logo at small sizes (e.g. favicons).
And as it is just a photo, it has no distinctive visual identity (only added by the surrounding square brackets, and the fact the logo is actually always used with the wordmark).
A better distinctive logo was needed since long, taking into account the experience. The old logo was designed fast, and in fact could not even be protected, so it was effectively in public domain (except when combined with the registered wordmark) and the logo could then be used abusively, but still legally (without the protected wordmark).
If proposal 6 was successful, proposal 5 also should have been successful. Seems like another attempt by the initiator/closer (yes, they are the same person) to use their own point of view
Proposal 5 have more than 2/3 support, this is classic wiki border in votings.
Why proposal 1 is not closed yet with a % of support?
Yeah I think someone needs to reevaluate this closure. The fact that it was closed by not only someone heavily involved but '''the person who started the proposal''' does not help either
@Iniquity - who can we ask to settle this?
It's not just the percentage, it's the number too, 60 support vs. 12 support is much smaller in borderline cases. I would be happy if someone double check this. How the actual number is a mistake?
13 pro / 9 contra / 1 neutral = 23 total
13/23 = 0,5652173913043478
In Wikimedia (at least fawiki and enwiki where I edit) we don't count neutral as total otherwise it would be the same as oppose.
or we do not count neutral votes to calculate %?
i'll fix this
I don't think we should accept anything lower than 70% TBH. At least for RfAs, RfCs and other things, nothing below 70% passes unless a very good reason (as a 'crat)
It don't change the picture, there is only one proposal with 74% (63 votes pro), the other two options with the most votes have 56% (29 votes pro) and 61% (25 votes pro). In ruwiki we have 2/3 = 66,(6) as usual border (for admins votes) and 3/4=75% for buerocrats votes.
Why was it changed from 71% though? At least proposal 5 is a variant on the current logo. All of the options in proposal 6 are terrible. People should be able to vote between specifically those two.
I don't know where 71% was from
13 pro / 9 contra = 22 total
13/22 ~ 59%
I would have read this as proposal 14 having enough support to warrant consideration in round 2.
Yes, but 1- 60% is still pretty small, I don't recall any successful RfC in any wiki that would pass with that percentage 2- The opposes in this proposal seem to be pretty stronger than the other proposals.
Proposal six has obviosly won. There is ONE top proposal. Others are below 2/3 and didn't make it.
My point is that we don't need round two at all now.
Round two is supposed to be between different variants of this logo – solid vs. translucent, which color etc.
Yes indeed. See https://lists.wikimedia.org/pipermail/wikitech-l/2020-September/093854.html You can test the voting gadget too.
File:MediaWiki logo 2018.svg This one is so amazing. This one should be reopened and should be selected.
this option has too low rating, please do not vote in closed sections
Could this be an alternate logo? RIT RAJARSHI (talk) 13:02, 23 September 2020 (UTC)
I also have some subjective preferences of my own, let's not put personal views before public ones.
Okay thanks youRIT RAJARSHI (talk) 13:08, 23 September 2020 (UTC)
I'm really surprised that nobody picked up on the fact that sunflower seeds make neat spirals in the center of the flower. No takers?
))_-_panoramio.jpg)
.svg){kind=link}
If you find good and free vector spirals - I can made variant of "Proposal five" with it.
maybe something like this can work as a basis?
I taked old monochrome version, but (1) haven't manage to pick right colours, and (2) I think that smaller pieces inside with bigger outside fits better, (3) I was moving this inside a center of the flower, but it always seemed that center is not in the right place.
Some of the logo proposals have the MediaWiki wordmark as explicitly lowercase "mediawiki", some are using the current capitalization of "MediaWiki" and others are using a font that looks as all-caps so "MEDIAWIKI".
It would be good if we could clarify if we're sticking with the current MediaWiki capitalization or whether that too is up for discussion/change.
The plan is to have different wordmarks as different variants in the second round mostly because MediaWiki might good in one logo and horrible in another. So having the discussion about the wordmark independent of the logo doesn't make much sense IMO.