Web Standards at Google
Post from October 2, 2008 (↻ October 28, 2016), reflecting Jens the Developer.
This and many other posts are also available as a pretty, well-behaved e-book: On Web Development.
As an exception, I’m writing as a Googler here: At Google, we care about web standards. Officially, that’s no news, but given repeated criticism for the code of our pages (see posts by Joe, Roger, Matt Webb, Jeff Starr, Monday By Noon, Standardzilla, or others), maybe it is.
As someone who’s responsible for the quality and performance of Google websites I’d like to point out that on the one hand, there are particular reasons why some of our products do not formally validate or, say, have potential when it comes to semantics, while on the other hand, there are well good examples for Google sites that adhere to web standards.
Only focusing on live sites and without giving too much details for now, here are some random examples that hint at these efforts:
- 23 Days campaign
- AdWords seminars (German)
- Anita Borg EMENA
- Brazil Women in Technology Award
- Doodle 4 Google Netherlands (Dutch)
- Doodle 4 Google Germany (German)
- Google & Space (German)
- Googler for a Day Competition 2008
- Privacy Center U.S.
- Wimbledon 2008
I suggest this small sample to make clear how we very practically care about standards. There are a few Google products and pages out there that deserve some extra attention and special care, yes, but maybe this post sheds some new light on where Google really stands.
Update (April 28, 2013)
The majority of sample sites above is not available anymore. However I’m very proud to add that in the years after this post, we on Google’s Web Studio (formerly Webmaster Team) have stepped up our efforts so much that of thousands and thousands of Google web pages, the majority is now not just valid, but of generally very reasonable quality. Take Google’s corporate pages, Press site, or even Ads pages as examples. Tony, I, and others will continue to share more about our team’s work on quality Google sites on both our team’s Twitter account as well as the Webmaster Central Blog.
About the Author
Jens Oliver Meiert is a German philosopher and developer (Google, W3C, O’Reilly). He experiments with arts and adventure. Here on meiert.com he shares and generalizes and exaggerates some of his thoughts and experiences.
On October 2, 2008, 22:23 CEST, Dave said:
Why do some people make such a big deal of insisting that they know better than Google? You get so much farther in life if look for reasons why people do what they do than if you focus on tearing down.
Wimbledon 2008 — 404 error.
And — yes — we in Yandex (search engine and services, Google’s rival in Russia) taking care about web standards too Not so much as we would like to, but…
Having web standard compliant pages is the safe bet for assuring the pages work well across a wide range of browsers. Also take a look a this blog post titled: Web Standards and Search Engine Optimization (SEO)—Does Google care about the quality of your markup?
From my impression, people express the majority of their “Google vs. W3 standards” critisism while they judge the search page’s quality, and not one of the smaller Google properties you mentioned.
Most of the critics do understand/suspect that Google has to be scarce on just about every byte of bandwidth and relate markup errors to an attempt to reduce page weight, but a closer look reveals that Google Search’s invalid markup won’t save a lot and even causes the opposite effect on occassions.
A technical explanation of the reasons behind these design decisions would make for a very interesting contribution to this debate, even if some trade secrets would limit the amount of detail.
Google is certainly producing every single element of their markup on purpose, isn’t it?
On October 3, 2008, 20:33 CEST, Dave said:
“Google is certainly producing every single element of their markup on purpose, isn’t it? ”
As a total outsider, I would assume that the HTML for Google’s most important property is deliberate and perfect, or a 20% project would have already fixed it.
I assume “particular reasons” is not synonymous with “airtight or even rationally defensible reasons.”
If code concision is the goal, you can sure leave out a lot of tags in HTML and still produce a good document.
The people at Google may care about standards, but Google-the-company shows no serious commitment.
Flagship products don’t validate; services like Gmail have a long history of bad cross-browser support (heavy irony for the Chrome team); we’ve been told that valid sites won’t index or rank better than crap sites.
So I’m afraid Google’s track record speaks louder than ten valid microsites.
Google’s sin is one of omission. Google has the world’s attention and could make standards compliance into a “must have” just by telling people to do it. But, Google does not do that. So, by omission, Google-the-company does more harm than good for standards.
All I can really say is keep fighting the good fight
Вадим, maybe there was a temporary problem, the Wimbledon page seems to work (again).
Robert, indeed, focus often lies on Google’s homepage as well as the results pages, but Google’s code base is a little bit larger
Ben, well, I’m not sure if actively contributing to standards organizations, standards advancement, and standards implementation can be considered “no serious commitment.” And taking into account this as well as the examples, another “perspective” is legit I guess. Anyway, thanks, and yes, I’m not lazy either
On October 22, 2008, 2:22 CEST, Kim said:
For better or worse, Google in particular has become a role model for Internet web design behavior. Therefore the underlying message (the one people pay attention to) is that web standards are a nice idea but not practical.
I truly think they should make this a priority. The other big role model (and offender) in regard to web standards is Amazon, who seems to totally disregard the issue. “I’m too big to care about your silly rules”.
This is the future of the Internet guys, it’s vital.
Any chance Google might be able to take on the hosting of the validators? http://www.molly.com/2008/12/11/w3c-validators-in-jeopardy/ …seems like something Google’s infrastructure could handle without even noticing
Ben, that is an interesting idea. I followed up on this internally, thanks!
On April 21, 2009, 10:19 CEST, Francesco said:
Since when does the Google homepage have a HTML 5 doctype?