CSS Working Group proposals

Jeffrey Zeldman thinks that I'm angry and writes that Anger can be brilliantly motivating; but anger is not a strategy.. If Jeffrey were to read from the phone book, I'd sit and listen, such is the respect that I have for him. But he's wrong, I'm not angry. Frustrated? Yes. Disappointed? Again, yes. I know that I'm not alone in having those feelings. I might not agree with some that web-standards are in crisis, but I do think, and know that others agree, that we must have a fundamental change in the way that these standards are developed.

Daniel Glazman does seem angry (at me) and writes that I'm incredibly naive or incredibly out of my mind. Daniel, you got me. Guilty on both counts. Daniel is a developer whom I respect a lot and is highly experienced in the standardization business. Coming from a browser developer, his comment on my last entry illustrates why change is so urgently needed far more eloquently than I could have done. (More on that later.)

I think that there are two questions that I need to answer that have arisen from my last entry. The first is why I think that Opera's action is linked to my call for for the CSS Working Group in its current formation to be disbanded, and the second is how I think that the CSS Working Group itself could be restructured to become more effective. So here goes on both of them.

Opera's anti-trust action and the CSS Working Group

Jeffrey and others have pondered my linking Opera's action against Microsoft with my call for for the CSS Working Group in its current formation to be disbanded. It's worth pondering.

But like others I don’t see a connection between Opera’s lawsuit and your call for the disbanding of the CSS working group. [...] (That’s because there IS NO CONNECTION. These companies sue each other all the time.)

Daniel calls me naive. Maybe. But if it's naive to think that a looming law suit can do nothing more than disrupt the collaboration between Opera and Microsoft on the CSS Working Group, then yes I'm naive. These companies may sue each other all the time, but when the action relates directly to the very standards that Opera and Microsoft are supposed to be collaborating on inside the CSS Working Group (rather than, let's say a patent dispute), then I'll naively wonder how Håkon Wium Lie can work alongside Microsoft in any meaningful way.

I can understand that actions in the past might not have affected developers on the ground working together, and maybe it will be true this time. But we all know that developers don't make strategy decisions. If they did, Chris Wilson would have got his way and Internet Explorer would not have languished in the doldrums for five years.

I could be wrong, after all I am naive. So I would like to invite Håkon Wium Lie, a senior representative of Microsoft and Bert Bos to publicly give web designers and developers their assurance that Opera's actions will not hinder the spirit of cooperation that is so desperately needed on the CSS Working Group.

Daniel didn't call me a fool, but maybe I'm foolish too. But if it's also foolish to think that Opera's actions will only cause Microsoft to be less open, less transparent and less cooperative, then I'm a naive fool. Daniel writes in his comment:

What do you think is a standardization body ? It's a BATTLEFIELD where vendors fight for competitive advantage. And that fight is not quiet or non-violent.

This is one of the reasons why I think that it is time for a fundamental change to, not only the CSS Working Group, but the whole process by which future CSS is designed, developed and released.

I know that I don't speak for everyone, but speaking for myself, I do not want tools that are the byproduct of battles for competitive advantage. I need tools that are well designed, well implemented and that I can rely on in my daily work. The demands that my work makes on me are constantly changing and at an ever quickening rate, so I need tools that help me meet those demands as soon as is possible. The web design and development industries cannot and will not wait five years or more for the W3C and its current process to come up with the goods. Our clients will not allow us too.

On HTML5 and the voice of the crowd

Before I move on, it's important for me to address one key point. I might be all about the tool of production [being] placed in the hands of the people as Andy Budd commented, but actually I agree more with Keith Robinson who said at Future Of Web Design, Consensus is for losers. Now I don't follow or participate in HTML5 so I won't comment on it or its progress, but what I will say is that I am not, and never will, advocate for an open, public decision making process for CSS3. Consensus is for losers. At least Daniel and I agree on something.

Read me well, because I am an old monkey, a very old monkey, in this standardization business. The more I see the crazy mess HTML 5 is becoming, the more I trust corporations in member-only discussions inside W3C. Bringing more individuals with no knowledge of the internals of a layout engine will NOT help improving CSS. We certainly need input from the designers' community, we need feedback too. But designers will always remain unable to say if a given feature is easily implementable or not.

I agree with Daniel in principle, but as a browser developer, Daniel is either missing the point of my argument or singing from the browser developer's hymn sheet. I do realize that the input of browser developers is essential so that new CSS features can be implemented. That is why I proposed the formation of a Technical Advisory Panel so that browser developers can provide this very valuable feedback.

However, browser developers are clearly the wrong group to be proposing or designing new CSS features. This is the job of those who use CSS on a daily basis. What is needed is a thorough rethink of CSS Working Group members (including myself), its structure and the process of designing and developing new CSS features. I believe that the time for that rethink is now. So how might the new CSS Working Group be organized and what are the different ways in which it might work?

On proposals for a new CSS Working Group

I'm only a naive fool, but as I have been asked for practical, concrete suggestions of what I think is necessary, here is my own rough outline. I would love to hear your thoughts and suggestions.

In today's fast moving web, the W3C CSS Working Group should not be in the (as Daniel puts it) the standardization business. Standardization is about looking for the common ground between different approaches and writing a specification around it. This is not what web designers and developers need.

What we need are new, creative solutions to our problems. Coming up with these solutions requires a creative as well as a technical approach. It also requires a new way of designing, developing, testing and launching of these solutions, all within a defined, transparent framework and with timescales attached. We can no longer afford to treat the design and development of new CSS features as as academic exercise.

So onto my concrete proposals.

Auditing

The W3C should begin by employing the talents of a highly qualified, independent individual to look objectively at the current status of CSS2.1 and CSS3 modules. This person would be experienced in managing the development and launch of large software or other commercial projects, not necessarily related to the web although that would be an advantage. His or her first task would be to investigate the current status of all of the work of the CSS Working Group and to (with the relevant consultation) determine the priority CSS3 modules and an overall timetable for the design and publication of these modules.

Core staff

The CSS Working Group needs to employ or second people from the commercial world who are highly experienced in certain key areas, to enable it to operate more like a commercial entity. These people should work full-time on the CSS project and not alongside other, employer related activities for a minimum period of twelve months.

The Working Group needs a skilled project manager to coordinate key activities and ensure that every person involved is contributing what and when they should.

The project also requires education and out-reach personnel who can both clearly explain the new tools and manage feedback from web designers, developers and others in a controlled and structured way.

Invited Experts

The CSS Working Group needs to design and develop new CSS features that truly meet the needs of the people for whom they are intended. On that basis, the project requires dedicated invited experts who are employed or seconded for their specific areas of expertise. Experts will include people who are highly skilled in page layout design, typography and other specific design areas.

Experts will also include people who have real-world experience in working with CSS in large teams and on large scale web sites. They will also be skilled in the practicalities of accessibility and internationalization among others. These experts will build on the work that has already taken place and help shape the modules to provide the solutions that web designers and developers require.

Technical Advisors

The knowledge and experience of browser makers is vital to ensuring that new CSS features can be implemented with the limitations of today's technologies. Browser makers will form a Technical Advisory Panel to look over the Project's proposals, feedback on areas where problems might occur and suggest solutions to those problems.

Clearly defined timetables

Unlike within the software industries, no clearly defined timescales or firm deadlines for the release of new CSS features seem to exist. After consultation and auditing, the CSS Working Group should publish its priorities with publication dates attached. For example, the modules closest to completion such as selectors, backgrounds and borders or multi-columns could be published within six months of full-time effort from all those involved. More abstract concepts such as Advanced Layout or Grid Positioning may be published within eighteen months.

Terminology

Many web designers and developers (including myself) misunderstood the fact that CSS specifications currently follow implementation rather than new CSS features being designed and then implemented by browser makers. This, and the terminology used needs to change to make it more understandable to design and development professionals. In the commercial world we are now well used to terms such as Alpha and Beta, and this terminology can easily be applied to the products that the CSS Working Group design and develop.

When millions of people use it, its a product

I believe that the design and development of CSS should not be a part-time or academic activity. It is a serious undertaking to develop CSS as a product that thousands of web professionals will rely on and must be treated as such. I believe that designing and developing new CSS features is conceptually little different from developing a piece of commercial software. Customer needs to be well researched, new features designed, developed and tested and made available within a reasonable time frame.

Again, perhaps I'll be accused by some as naively not understanding the way that the system operates, or has operated until now. What I do know is that as a W3C customer who uses their CSS product, I and I know many like me are becoming frustrated and disappointed with the service that we are getting from the W3C.

The W3C is lucky to have customers like us who care that it does the right thing.

On the need for leadership

I’m disheartened by the general lack of leadership. I wish The Web Standards Project would either disband or get meaningfully busy.

[...] The Web Standards Project has been silent about HTML 5 and CSS 3 and lots of other things. In the absence of strong leadership or any leadership from The Web Standards Project, frustrated designers and developers are leaping to support anyone who (a) is competent at design and standards (b) voices the frustration we’re all feeling and (c) proposes some sort of alternative, however vague. I think highly of Andy and if he’s ready to fill in the leadership gap, that’s great.

I could not agree more that strong leadership is essential in what I hope will be the start of a new era for the development of standards and in particular the CSS that I am so passionate about. I hope that we will see that leadership within the CSS Working Group and at the Web Standards Project too, and I wish the people who take up those challenges the very best of luck. As Johnny Cash sang, It ain't me babe.