I was involved in rolling out the new HTML5 logo for W3C; I wasn’t the person in charge, but I helped out with some aspects of it.
The interesting thing to me was the reaction to the logo, aesthetics aside. (I like the logo personally, but I also like the retro-futurism of Soviet Realist art, like the statuary in Memento Park in Budapest, and comic books.) I loved the parodies and jokes about the logo, but I was both unsurprised and disappointed at the some of the negative dialog (and in some cases, monolog) about what the logo is meant to represent.
Just like the Gap and Starbucks logos, there were going to be some people who would be unhappy because that’s what makes them happy. There would be no pleasing them, and it’s senseless to try.
But there was another set of people that we really value, and we wanted to make sure that they were comfortable with the logo: the perfectionists. As you can imagine, it’s not the easiest group to please, but they are the ones who make standards work, by reviewing the specs to make sure there are as few errors as possible, by creating exacting tests to ensure interoperability between browsers, and by teaching web developers the precision of terminology and best practices that lead to high-quality web sites and webapps… usually on a volunteer basis.
These are our people, and we wanted to please them. But at the same time, a logo has to serve lots of different communities… including managers and clients who don’t know or care that there is an “HTML5 specification†(whatever that might be), and people throwing around the word “HTML5†like it’s birdseed at a geek wedding. Even people who know better use “HTML5†as shorthand for “Finally, browsers are doing all this cool stuff I want to play with! WHEEEE!â€
So, how do you strike a balance between people who want us to say, “This logo is for the HTML5 specification only, and we will sue anyone who puts it on a site that uses HTML4.01+CSS3†and those who want us to give them the shiny and shut up? You make an “identity system†which includes sub-icons for specific technologies and general technology classes, and you say that the identity system, in toto, represents HTML5 and the peripherally-related technologies like CSS3, SVG, WOFF, Geolocation, etc.
And how do you screw it up? You oversimplify in the FAQ and use the word “logo†instead of “identity systemâ€, and let people infer that the main logo, the one for HTML5, is meant to represent not only HTML5, but CSS3 as well. (Interestingly, some of the more vocal people who complained on Twitter don’t mind the geolocation API or SVG being conflated with HTML5 nearly so much as CSS3… that seems too fine a distinction to me.)
But we knew we would make some mistakes. We went out of our way to explain to people that we were not trying to shove it down their throats… that we wanted to have a community conversation about it.
Once we had something we thought was ready to share, we released it to the public, rather in a rush. We took to email, to Twitter, and to blog posts and comments, to collect feedback and clarify our message. Turns out 140 characters is not enough to convey much nuance. After a couple of days of seeing if the message was clear enough, giving people a chance to have a second look and see if the idea of a “logo system†stuck, we decided it wasn’t good enough. We updated the FAQ to split out the HTML5 logo from the sub-icons. That seemed clearer to people, though it’s really saying the same thing in a different way… but the way you say something matters.
And that’s my point, here: The way you say something matters.
W3C started the conversation about the logo by clearly stating, and reinforcing consistently, that we were interested in feedback and we were listening to what people were saying. Sometimes people didn’t believe us, but we were. (And I sincerely appreciate Jeremy Keith taking the trouble to explain what was frustrating him about the “logo / identity systemâ€.)
But some people didn’t hear that we were listening, because they were too busy shouting. Shouting things like “Fire W3C Communications Person Who Led Messaging  I’m not the person that’s targeted at, but I found it a rude and mean-spirited thing to say nevertheless. (Pet peeve: I can’t stand when people make a public rant in a post and don’t allow others to comment on it… it demonstrates a certain lack of accountability.) Tantek Çelik has been in Web standards for a long while, and knows exactly who the (overworked) Comm person is whose job he was calling for, and I think that’s irresponsible.
Tantek compounded this with a tweet that somehow manages to pretend that W3C wasn’t listening, but all the shouting forced us to change our FAQ, then goes on to say in his next blog post:
Yet with W3C’s first draft HTML5 FAQ, we almost gave up. Almost. But apparently sometimes angry rants can effect change. Sometimes passion works.
Within 24 hours of the last of our posts, W3C updated the FAQ, removed the errant entry, added more clarification
You know what also works? Having a calm conversation, reading the FAQ carefully and seeing where a mistaken impression (or even an outright mistake) might have been made, and making suggestions to the people who have invited you to share your thoughts.
Folks, you didn’t need to get angry. Yes, there was some ambiguity in the FAQ, and that was a mistake… one we were sorry to make and happy to correct, only a couple days after the logo was released. But the result of all the shouting was that, several days later, I’m still talking to people who think that W3C is sending this mixed message with this logo, because that’s what everyone was shouting about… and all the clarifications and updates now aren’t going to put that spilled milk back in the glass, since you sent out the cry to arms. A few simple words and a modicum of patience and trust would have served your cause—the precision we all value—much better.
W3C went out of our way to give you a voice. You’re long-time contributors to standards. You’re all insiders… use your inside voice.
But Doug, you can’t have it both ways. This logo (and accompanying message) was developed in complete secrecy. Not even the HTML Working Group was consulted. To then complain that, after a very public launch with much fanfare, people didn’t subsequently come to the table and discuss things before writing blog posts, strikes me as a bit of a double standard.
You say:
I’m still talking to people who think that W3C is sending this mixed message with this logo, because that’s what everyone was shouting about
Nobody would have had to shout if they had had they opportunity to calmly explain their concerns before the logo was announced. If it seems like people were shouting to be heard, that’ only because they needed to match the volume level of the logo announcement itself.
This whole episode has reinforced to me that for a standards body to develop anything in secret—whether it’s a spec or a logo—is utterly, utterly counter-productive.
To say that the person on the Communications Team should be fired might be an over-reaction, but it has to be said: this whole incident was the worst possible example of “communication” from the W3C.
I hope there won’t be any more such surprise announcements from the W3C Communications Team. I hope that in future, the relevant working groups will be consulted before there any PR proclamations issued about the technologies they’re working so hard on.
Heh. I said hard on.
Hi, Jeremy, thanks for your comment.
Okay, first, “the worst possible example”? Really? That strikes me as more hyperbole… “Could have been handled much better†seems more reasonable, especially if accompanied by specific examples of how it could be handled.
But taking the substance of your comment, let’s imagine how else this could have gone. Had the W3C Comm team approached the HTML WG to simpy ask if the group wanted a logo for the spec, it would have taken months to get agreement to move forward, if it ever did at all; I suspect most of the WHATWG loyalists would have be adamant against it. Then having gotten past that hurdle, there would have been a many-months discussion of what the general goal or aim of the logo should be, and what general style it should have, or at the very least, what process should have been used to select a designer or set of designers. It would have taken at least a year to settle on anything, and I doubt it would have been as eye-catching as the logo we have.
Then there’s the idea that we could have taken the logo we have now, and presented it to the HTML WG with minimal fuss and without the presentation site, a “soft launchâ€. I think that would have followed pretty much the same pattern as starting from scratch with the WG, and without any of the excitement that this generated. It would have been a complete toolshed permathread.
Then there’s the idea of a contest, which I agree sounds fun. Set aside for a moment the fact that the AIGA (for example) traditionally frowned on design contests (or “spec workâ€), since it amounts to a couple winners and a bunch of unpaid losers. How do we set up a vote with any semblance of fairness? Who gets a vote? Everyone? Just the HTML WG? How do we prevent gaming the system? Having run the contest for the SVG logo (a much more modest effort, years ago), I can tell you how time consuming that would have been. For the SVG logo contest, we got nearly a thousand entries… for HTML5, it would have been thousands or tens of thousands… basically, a non-starter. Pity, too, because I think that would have been really interesting.
I’m not trying to be dismissive or sarcastic. I just genuinely don’t see how, logistically, we could have made this work in a timeframe or with an outcome that anyone would have been any more satisfied with. If you have concrete suggestions, please speak up… this is almost certainly not the last logo W3C will design, nor probably even the last “identity systemâ€, and I’m sure the Comm Team is very interested in hearing ideas.
But as it happens, for this logo, we were presented with an opportunity (just a few weeks ago) to work with a talented designer for a limited timeframe, and we decided to move ahead on it, with the understanding that if the logo was rejected by the community, we would try again some other way (maybe by the means I mention above). But… that didn’t happen. People seem content enough with the logo, such as it is, and I expect it will be as much of a success as any HTML5 logo could be.
Finally, I think you may expect too much transparency at all stages of project development. If I had to report every hallway conversation I had, every offlist suggestion, every phone conversation, that led to the formation of a working group or the development of a feature of a spec, nothing would ever get done; it would get picked to death. There’s a point at which something is ready to put forward publicly, after it has germinated for a bit… and this is the same in every browser vendor, and within the WHATWG. Accountability is not always the same as transparency.
All that said, I agree that if we had had more time to plan and bring this to fruition, we could have avoided some mistakes. We’re sorry we pulled such a boner.
(Heh.)
Â
To quote:
« Just like the Gap and Starbucks logos, there were going to be some people who would be unhappy because that’s what makes them happy. There would be no pleasing them, and it’s senseless to try.
But there was another set of people that we really value… »
The last two sentences are quite telling. Pointless to try and accommodate a demographic of people you don’t value. But worth responding to criticism from? You’ve precluded yourself from this task: don’t say you won’t ‘sink to the level’ of these people you’ve already decided you’re going to ignore, and then respond to the alleged mud-flinging. As you’ve intimated, you can’t win.
Later on:
« the perfectionists … are the ones who make standards work »
The perfectionists are the ones who’re irritated at the vagueness, misleading representations and conflation of terms. These people ensure standards are of quality. The other people who make standards work, the ones who aren’t so vocally and pertinently distressed with the latest PR move, are the lowest common denominator. They are also essential, but if you appease them exclusively and leave the real perfectionists (those who won’t be happy with rigorous, honest and self-critical work) out of the bargain, you are left with confusion, bloat, uncertainty and the spreading of ignorance and bad practice.
Of course, all these things are the real danger when preaching standards — which is why we depend upon the  vigilance of people like Jeremy to make sure we don’t just swallow shit because it pleases 70% of people who won’t apply critical judgement and have no qualms with reducing the quality of the standards.
« some of the more vocal people who complained on Twitter don’t mind the geolocation API or SVG being conflated with HTML5 nearly so much as CSS3… that seems too fine a distinction to me. »
Geolocation API is a part of HTML5 spec. Methods of directly embedding SVG in HTML5 is a matter of HTML5 spec. CSS3 is not part of the HTML5 spec. The fine distinction is based upon the HTML5 spec, which by definition defines the specification of HTML5. It’s a pretty clear distinction and if you don’t understand it, I don’t think you should be telling people exactly what HTML5 is.
Â
Hi, Barney-
I said there was a set of people we really (read: “especiallyâ€) value, not that we didn’t value others. As for the perfectionists… I think we’re in violent agreement that perfectionists are an important part of getting standards right. As for the geolocation API, you are absolutely, totally, 100% wrong; it is not part of the HTML5 spec, and never was; I should know, I was the one who set up the Geolocation API WG, based on an email to the WebApps WG by Google. SVG is not defined in the HTML5 spec, even if HTML5 defines some ways in which it can be embedded; I’m in the SVG WG, and I approached Hixie about including SVG references in HTML5 when it first came to W3C. You are correct that CSS3 is also not part of the HTML5 spec. I guess 1 out of 3 ain’t bad… keep guessing, maybe you’ll win a prize!
I think the “identity system” is even worse than the main logo. Not the concept but the actual icons. http://hsivonen.iki.fi/sergeant-semantics/
Hi, Henri-
I can see your viewpoint. At the same time, the military chose those styles because of their simplicity and ease of quick recognition, which is what I think the designer was going for… the military shouldn’t have exclusive use of common icons, like the circle and star (which is also used in maps, for example), and most of the icons are not evocative of the military, so I don’t know how far you can take the comparison. There is the badge-like quality, but lots of non-military organizations award badges.
“Okay, first, “the worst possible exampleâ€? Really? That strikes me as more hyperbole…”
Â
Doug, to clarify: when I said “this whole incident was the worst possible example of “communication†from the W3C.” I meant “this whole incident was the worst possible example of “communication†from the W3C in the history of communications from the W3C, not in the history of communications.” I hope that makes things clearer.
“But taking the substance of your comment, let’s imagine how else this could have gone.”
You can imagine hypothetical situations all you want, but none of them would have been as bad as the reality of what the W3C Communications Team did. They’re going to have to work hard to win back they trust they squandered.
In any case, you are once again conflating criticism of the message with criticism of the logo. I like the logo. Tantek likes the logo. If you had come to the working group before launching the website for the logo, we would have responded with some changes for the *message*, namely changing “HTML5 means CSS, SVG and WOFF” to “HTML5 means HTML5.”
“there would have been a many-months discussion of what the general goal or aim of the logo should be”
It strikes me that asking what the aim or goal of the logo should be is a very, very, very good question.
Hi, Jeremy-
No, I understand that you and Tantek like the logo. I’m not conflating the logo and the wording; what I’m saying is that the conversation about corrections could have happened on the HTML mailing list, for example… and I applaud you for posting there first, before taking it to the Web. Ian (our Comm guy) had noted your concern and had it on his list of things to correct immediately, when the whole thing blew up. Perhaps he should have run the FAQ alone by the HTML WG first, sans logo… maybe it could have even built up more anticipation. But even that could have blown up, and as it was… well, 2 days later, the FAQ was changed just as you asked, and it was based on your feedback. I understand where you’re coming from… perhaps ironically, the logo was announced in too public a manner (larger than just the HTML WG), precipitating your public response. I still maintain that some of the comments (not necessarily yours) could have been rather more civil.
I agree that the aim or goal of the logo is extremely important, and that was the subject of many discussions. It would likely have been more satisfying to some members of the community for that to have been public… as it happened, we had a window of time (and energy) to accomplish this in. I would have liked to be more central in the discussions myself (though I’m not as active in the HTML WG list these days, being occupied with other specs and groups).
“Finally, I think you may expect too much transparency at all stages of project development. If I had to report every hallway conversation I had, every offlist suggestion, every phone conversation, that led to the formation of a working group or the development of a feature of a spec, nothing would ever get done; it would get picked to death.”
This reductio ad absurdum.
Being clearly targeted, even if not directly named, as one of the shouting wolves, I have to say I find your article is a superb example of political correcteness erected as a way of life, Doug.
“You know what also works? Having a calm conversation“. Precisely. And people who complain about the logo – including yours truly – are just sad this conversation did not happen before the logo went public. Even the temporary logo (I can’t even describe how weird it is from a comm point of view to make a temporary logo public and tweak it later…).
The W3C has a bad track record with feedback. In particular feedback from its own community. During ten years, the community has complained about XHTML2, the W3C did not listen and decided to go its way anyway. Led to WHATWG and the current HTML mess. I still remember the silence and the “dont’ show don’t tell” attitude when James Clark killed Schemas during a TPAC.
The logo story is of the same magnitude. The military look and feel of the logo with sublogos is, really, a lame choice. Speaking of my own domain CSS, the CSS sublogo is such a failure that I could not find a single person understanding its meaning in my daily work neighbourhood. I won’t comment the other sublogos not because I care only about CSS 3 (apparently your explanation), but because I let the corresponding WGs or spec authors do it.
You’re right, we are perfectionnists. But calling us perfectionnists here because we shout at the proposed logo is just a misrepresentation of the reality. We shout not because it could be even better, but because it is bad. That’s not perfectionnism, that just a request for efficiency.
Saying that the FAQ needed clarification and the recent changes clarify the situation is again false. The problem is not the FAQ. The problem is the release of a logo a large part of the community does not recognize itself in. The problem is not that the logo was designed in small committee, it’s the incredible lack of communication inside W3C after the choice and before the logo was released. AC Reps were not even notified of the logo. Wow.
“Hey, CSS WG, we plan to release a CSS sublogo for our OpenWebPlatform thingy and our choice is that one; opinions?“. Is that too hard? I don’t think so.
You said “monologs”. I agree. One cannot dialog if the other party is not willing to listen. That’s probably why there was anger.
Hi, Glazou-
I understand why there was anger. These specs are more than just technical documents to many of us… after so much work on them, our identity gets wrapped up in them, to some degree. I agree, personally, it would have been better to approach the groups in question for their feedback, and in a better world, we would have had time to do it, but there was a judgment call made… was it better to approach just the working groups, or the community as a whole? W3C opted for the wider conversation. (A small correction: the AC reps were notified in advance, 3 days before the release of the logo.)
I understand that you and many others don’t like the logo; I happen to differ (though I agree the CSS3 logo is a bit lame and unsymbolic. But we’ve asked for your help, and you’ve declined. That’s your prerogative, but it doesn’t help make a change.
At the same time… you complain both that it was designed in too small a committee, and that it was released to too wide a public review. I’m glad I wasn’t the one who had to decide precisely how wide the review was; that’s a tough job. (FWIW, I didn’t have you in mind at all when I wrote this blog post.)
I don’t quite understand how a launch like this one from was intended to be (or spark) a “conversation” with the community, rather than a declaration of how this is how things will be. I mean, the idea that things could still be open to change seems less plausible when there is already swag being sold.
Â
Hi, Bridget-
I personally wouldn’t be surprised at all to see the CSS3 icon change. The swag is not a sign of commitment… it would be even cooler in a few years to have that HTML5 logo that flopped on a T-shirt, in true ironic hipster style. We’ll see how much uptake there is of the logo in the wild, though I think the chances already look promising.
No. The ACs were notified that a logo was about to be released. The logo itself was never sent to the ACs. I guess most ACs learnt about it either through the press or through my own message in the AC Forum. Well done.
And I already explained precisely, although not in public, why I declined to help.
Total outsider here, but I think it’s clear from your comments that you already know what would have worked better than what transpired, and I don’t personally buy the “we just couldn’t decide on how to go about it” stuff.
Â
“We had this opportunity…blah blah designer… here is what we got from that. Â Barring majority rejection from HTML-WG based on <criteria>, which we will discuss over the next 7 days, this will become the new identity system. Â The soft-launch site, based on the identity from designer X unless decided as above, for review/comment is: YYY.XXX”
Â
PS: Your text is flowing out of your sidebar. Â http://dl.dropbox.com/u/15519230/flow.jpg
Hi, Jeff-
No, I have my own opinions on what may have worked out better, but any of those choices might have led to even bigger problems I’m not seeing (with the benefit of hindsight). Even within the small team that was working closely on it, there were many different, strongly-held opinions, and magnify that by 400, or 4000, and I think it’s hard to make a realistic judgment on what the outcome would have been. I wasn’t the final decision-maker, so these are all my own thoughts (as I hope is obvious). W3C, like any organization, has a lot of individuals with different views.
Thanks for the screencap… it’s a known problem I’ll be fixing in my next site redesign.
Thanks for the inspiring, and rational, article. While it may not be as exciting, calmer heads always prevail.
Knowing how the things are happening inside the W3C Team (communications) included. I see the logo effort as remarkable. I have seen and participated to many communication actions at W3C. The thing which strikes me in these events and discussions is always the same. The lack of respect in *people* in the discussions over mistakes.
Â
Yes, staff at W3C do mistakes, but so we all do. Yes there are things which could be done better. Each time a mistake has been made, the people in the staff listened, and improved. Collecting feedback, articulating discussions. The staff is here for putting people around a table. Breaking News: There will be more mistakes.
Â
Each time, someone says W3C, do not forget it includes W3C members. A big group of people with diverse opinions and interests. The W3C organization precisely exists because there are conflicts between the different players of the industry. The feedback, the changes in W3C are the results of these interests played by different groups. The move from RAND to RF for the W3C patent policy, the slow abandon of Web services, the move from XHTML2 to HTML5 (integrating many features of XHTML2 in HTML5). Sometimes there are groups with different interests and will inside W3C. I read again recently something in Weaving The Web: “When organizing the conference and wondering about setting up a consortium for the Web (aka future W3C) The conference was the way to tell everyone that no one should control it, and that a consortium could help parties agree on how to work together while also actually withstanding any effort by any institution or company to “control†things.” – Tim Berners-Lee
Â
Having disagreement is fine. Making mistakes is to avoid, but will happen again. Attacking people, being violent, etc. has never solved anything. (And I know exactly what I’m talking about by having been rude in the past.)
Â
Â
Â
This is an exceptionally beautiful logo.
“I can see your viewpoint. At the same time, the military chose those styles because of their simplicity and ease of quick recognition, which is what I think the designer was going for… the military shouldn’t have exclusive use of common icons, like the circle and star (which is also used in maps, for example),”
I think there’s no way to spin the sergeant insignia as not military-themed. The star in the circle is then seen in that context.
“and most of the icons are not evocative of the military, so I don’t know how far you can take the comparison.”
Exactly as far as I did. (You’ll notice that I only discussed a minority of the icons in my blog post.)
On a different topic: I think the “People are making stuff!” and the “Your logo” titles on the logo page have an astroturf feel to them.
Hi, Henri-
I’ve spoken to Michael Nieling, who designed the logo and icons, and he denies any intentional military reference. I understand that you see it that way, but I don’t see why your spin is any more valid than the designer’s. My own first impression was of an orderly line of angle-brackets turned on end, for example, and that’s still what I think of.
To be precise, though, I think US sergeant stripes are curved lines, not straight, and point down, not up, at least according to Wikipedia. It’s also the insignia used for a police sergeant, which may be a more apropos association in terms of enforcing rules, especially along with the “badge”. But while I think the main HTML5 logo is probably set at this point, the icons could still change if that’s what people want. Personally, I’m not as concerned about the color of the toolshed, nor do I fear ill effects from the style even granting a military interpretation.
As far as the astroturf, I agree, but maybe I’m inured to stuff like that in promotional materials.
How a decade changes the tech landscape! When I posted this, there was a lot of discussion in the comments, much of it through a sort of pingback service by a company called Topsy Labs. They were bought by Apple, and shut down 2 years later. Sigh. And now, personal blogs are out of fashion, it seems; I know I’ve had my mind on other things, and don’t update this one much anymore. It seems like the consolidated social media services, like FB and Instagram, are where people are posting. I got contacted by a company called BrandMentions, a in some ways replacement for Topsy, that asked me to link to one of their posts, Best Time to Post on Instagram in 2021 – A 2.2 Million Posts Research in exchange for access to their service. Though I’m not savvy about marketing, they seem to offer an interesting service, so I’ll check it out, and the article seems well-researched, so I don’t mind posting about it here.