YouTube Opens Auto-Captioning to All

Image of the captions in use on President Obama's speech about the Chile earthquake.

If you've been reading my blog for a while now then you may have noticed my post back in November titled YouTube Will Automatically Caption Your Video. In that post I talked about YouTube leveraging Google Voice and its speech recognition features to automatically generate video captions. That feature was only available to a subset of all YouTube users, a "small, select group of partners."

On Thursday YouTube announced it will open that feature to all YouTube users (The Future Will Be Captioned: Improving Accessibility on YouTube). In reading the post on the YouTube blog, it looks like they will be working through all the videos on YouTube. Video owners who want to speed up the availability of auto-captions on their videos can click a "request processing" button, hopefully dropping it into a queue. However, since it's a free service, I wouldn't expect them to set a deadline with you.

The blog post lists a few things to keep in mind:

• While we plan to broaden the feature to include more languages in the months to come, currently, auto-captioning is only for videos where English is spoken.
• Just like any speech recognition application, auto-captions require a clearly spoken audio track. Videos with background noise or a muffled voice can't be auto-captioned. President Obama's speech on the recent Chilean Earthquake is a good example of the kind of audio that works for auto-captions.
• Auto-captions aren't perfect and just like any other transcription, the owner of the video needs to check to make sure they're accurate. In other cases, the audio file may not be good enough to generate auto-captions. But please be patient — our speech recognition technology gets better every day.
• Auto-captions should be available to everyone who's interested in using them. We're also working to provide auto-captions for all past user uploads that fit the above mentioned requirements. If you're having trouble enabling them for your video, please visit our Help Center: this article is for uploaders and this article is for viewers.

The obvious benefit here is the ability to satisfy accessibility requirements and open your content to a broader audience. This also can help in your ability to search for videos with appropriate content since the captions live in a separate text file that is pulled into the video, all powered by Google. In addition, with Google's free translation services, it will be far easier to translate videos into multiple languages, reaching an even larger audience.

W3C Releases 7 HTML-related Documents

The W3C has announced today that it has published seven documents related to HTML. I'm going to cheat and just bring in their description:

• HTML 5 and HTML5 differences from HTML4. In addition, some content that was part of the HTML 5 specification has been published in two new standalone drafts: HTML Canvas 2D Context and HTML Microdata.
• HTML: The Markup Language, a first draft. This document describes the HTML markup language and provides details necessary for producers of HTML content to create documents that conform to the language. By design, it does not define related APIs, nor attempt to specify how consumers of HTML content are meant to process documents, nor attempt to be a tutorial or "how to" authoring guide.
• HTML+RDFa, which defines rules and guidelines for adapting the RDF in XHTML: Syntax and Processing (RDFa) specification for use in the HTML5 and XHTML5 members of the HTML family.
• Additional Requirements for Bidi in HTML, a first draft. Authoring a web app that needs to support both right-to-left and left-to-right interfaces, or to take as input and display both left-to-right and right-to-left data, usually presents a number of challenges that make it an especially laborious and bug-prone task. Some of these are due to browser bugs, but some can be traced to a gap in the specification of the bidirectional aspects of a given HTML feature. And some of these challenges could be greatly simplified by adding a few strategically placed new HTML features. This document proposes fixes for some of the most repetitive pain points.

Don't let this trick you into thinking that HTML5 is official. It's not. HTML5 is still a draft. Let me repeat. HTML5 is still a draft. However, knowing the direction of HTML5 will make it easier for you to implement it, convert sites, and ultimately convert your existing skills. If you are already building sites using this draft of HTML5 for a browser market that doesn't yet support it, then good for you. Regardless, the two documents that you should pay attention to as you go down this path are HTML5 differences from HTML4 (W3C Working Draft 04 March 2010) and HTML: The Markup Language (W3C Working Draft 4 March 2010).

Related post:

• Too Soon to Advocate HTML5?, posted on this very blog.

RIP IE6 (Not Really, But Here's to Hoping)

CNN is reporting on a funeral today for Microsoft Internet Explorer 6. The funeral is in Denver, Colorado, so I will not be attending.

That the mainstream press is covering this is good news — somebody out there in the non-tech world understands that is newsworthy, even if only to a niche audience. Not that the funeral is newsworthy, but that causes pushing for the demise of IE6 is newsworthy. Granted, this is not tied to any official campaign to kill off IE6, but it is a fun way to draw attention to an annoyance many of us face.

Many people (web sites, developers, forums, etc.) have been calling for the demise of IE6 in some way for quite a while now. Google joined the fray (Modern browsers for modern applications, from the Google blog) when they announced that they would phase out support for IE6 in Google Docs and Google Sites as of March 1 (just a few days ago, the date on the grave stone). You can see other sites (far smaller, for the most part) who are trying to push IE6 out to pasture, just visit ie6nomore.com. Whether or not this will speed the end of IE6's reign is to be seen. Catch up on some anti-IE6 articles at Mashable using their IE6 Must Die tag. My post showing January 2010 browser stats broke down the IE versions thusly (what a fun and odd word):

Internet Explorer is the troubling one in the mix. IE8 is now up to 22.31% of the market, but IE6 still beats out IE7 (20.07% and 14.58%, respectively). That equates to 1 in 5 users is still surfing on IE6, known for its security holes and buggy rendering.

That I haven't seen this event fly through my regular flurry of tweets and RSS updates from web developers and developer sites is a bit startling, but this is a small, very local affair after all.

If you are interested in attending (and are in Denver), or just want to enjoy the humor, go visit the IE6 Funeral site. There is a Twitter feed (@ie6funeral) in case you are interested what I can only hope are live tweets throughout.

Don't Let Social Media Get You Robbed (or Stalked)

If you are one of the millions of people using social media to report where you are, you may have been tuned in to all the buzz lately about the site Please Rob Me. The concept is very simple, when you use applications like Gowalla, Foursquare, Brighkite, Loopt or anything else that broadcasts your location, you run the risk of telling the world that you aren't at home and it's ripe for the picking. You can compound it by advertising to potential stalkers (not much of an issue for me, but certainly if you're popular).

The geolocation features built into these applications not only tell people where you are, but if they have any familiarity with you, the area, or just spend a few minutes reading your history, they can pretty quickly posit for how long you will be away. This isn't limited to just geolocation-broadcasting services — you can just as easily use Twitter or Facebook (among others) to inadvertently tell people you are away from home or at a particular place.

That it took all the press around the Please Rob Me site to cause people to consider this is almost laughable. Granted, the site is pretty overt in how it conveniently aggregates all the data for visitors, but it's not doing anything that an interested party couldn't do on his or her own in a few minutes. All the site does is show all the Tweets from Foursquare users as they check in — not very complex. To quote from its Why page:

The danger is publicly telling people where you are. This is because it leaves one place you're definitely not... home. So here we are; on one end we're leaving lights on when we're going on a holiday, and on the other we're telling everybody on the internet we're not home.

So how can you enjoy social media with geolocation features and still minimize your risk?

Techniques for Safe Check-ins

These are just three options. I employ each of these and find them fairly easy to follow. None of these will stop the old fashioned thief (who doesn't use social media), but they can make it a little harder for those casing you online.

Time Shift

Remember, you don't have to check in to a location as soon as you get there.

Let that sink in for a moment. I see people check in to locations as soon as they walk in the door, but they don't need to. Sure, you may not get the immediate benefit of knowing which of your friends might be nearby, but maybe you should sort that out in advance.

For example, I am known among my friends for posting photos of my meals through Brightkite. What many of them don't realize is that I often check in to a location after I have already left or (if it's near my house) as I am leaving. This has caused a couple awkward moments when a nearby friend is notified of my proximity and comes looking for me.

Hide Your Check-ins

Hide your check-ins from all but your trusted friends. And I don't mean those Facebook friends who you met once while in a mosh pit or that you haven't seen since second grade.

You don't need to Tweet to the entire world where you are. Sure, if it's a major event and you want to brag then go for it, but understand the risk. If you Tweet or check in for every morning visit to the local coffee hut, not only will people know where you are right now, you will be telling them where you are tomorrow or next week. Establishing a pattern of behavior makes it easy for someone to predict your moves.

Consider hiding your Foursquare check-ins and not Tweeting every one; you can still participate in the overall game. Lock down your Facebook profile from the general public. Consider disabling the geotagging from third-party applications that feed to Twitter.

Get a House Sitter

Some of you may be aware that I was just in Houston speaking at a conference. I was more than happy to post my progress via Foursquare check-ins, Brightkite photos, and an assault of Tweets. Anybody could have wandered over to my house for some pillaging, but would have been surprised to find that someone was already there. In fact, I was able to split the duties across two people, so it was better populated than when I am in town.

If you have the luxury of a roommate, house sitter, guard dog, militia, etc., then you may be in good shape already. But consider whether that person at your house is someone you want to put at risk (granted a rather remote risk), or if that person is comfortable with being on watch.

Related articles:

Continue to read up, you may come up with some ideas that work better for you than my suggestions.

• The dark side of geo: PleaseRobMe.com at CNET.
• Are We All Asking to Be Robbed? at Mashable.
• Google Buzz: Privacy nightmare at CNET.
• Facebook Privacy Complaint Ignites War of Words at PCWorld.
• Twitter Your Way to Getting Robbed at Mashable.
• HOW TO: Make Your Small Business Geolocation-Ready at Mashable.
• How Robbers Did Their Dirty Deeds Before Foursquare at Mashable.
• Friday Poll: Do Location Check-In Services Freak You Out? at Mashable.

History of Eye-Tracking as Research Tool

If you've ever wondered what eye-tracking is and where it came from, there is a historical breakdown in the article A Brief History of Eye-Tracking over at UX Booth. The article covers eye-tracking as a research tool, not as a user interface input method, something I think is worth mentioning given the devices I've seen and used that track eye movements to allow the physically disabled to use a computer.

In the article the author goes back to 1879 when a French ophthalmologist observed that readers' eyes often pause on words instead of reading at a steady pace. He covers some other developments without dates until 1931 and then jumps ahead to the 1980s when eye-tracking was used for market research for magazine ads.

The article isn't intended to be a history of eye-tracking as a usability tool, but the author does get to that point toward the end and references Jakob Nielsen's book, Eye Tracking Web Usability.

The Science of Trust in Social Media

I am one of those people who always needs to see proof of some assertion, evidence to back up a claim. While I can accept anecdotal evidence in the right context, justifying decisions beyond how good the local taco hut is requires more than just musings from friends or co-workers.

In an article posted to Mashable today titled, "The Science of Building Trust With Social Media," the author attempts to explain trust in social media through the lens of psychological research. He references a Wikipedia entry on fundamental attribution error and then uses that to frame some of the dust-up between Kevin Smith and Southwest Airlines (at CNN, at Huffington Post, at LA Times).

The author does make a good point at the end about how the lack of body language (facial expressions, overall body cues) can make the medium very difficult to navigate. He even cites a video featuring the Domino's Pizza president overlaid with a graphic of the audience response based on body language. See the video below.

The article itself is anecdotal, but at least it's interesting.

MOM 2.0 Summit Notes

Keynote and opening remarks by Gretchen Rubin and Heloise. I have no photos of me in the panel, except the one I found online where I look like a startled rabbit. Which may not be far from the truth.

If you've paid any attention to my blog (as in, read the last post) or even seen me in real life (or noticed my absence for a few days), you may know that I was in Houston, Texas for the second annual MOM 2.0 Summit. I was a speaker/panelist in the web site usability and user experience session. I was paired with Angela Schmeidel Randall from Normal Modes and Sania Khan (blog at DesignBySania.com), both of whom were very bright women with a lot to offer. We managed to pow wow a couple times via Skype and again before our session, and the two of them had put together a PowerPoint presentation in advance (to be fair, they had more warning than I).

Other than some awkward microphone handling, we had fun getting our ideas across and fielding questions from the audience. After the session I fielded a few more questions from audience members, and even continued answering questions that night and through the next day. Based on the turnout, and the amount of questions, I'd say this was a popular topic — and we were up against the social media session, which was quite the draw I hear.

The MOM 2.0 Summit blog mentioned our session (Website Usability – Do The Expected.), and other than misspelling my name (easy to do when you are mostly talking about women), summarized it pretty well. What I thought interesting were the main ideas people seemed to take away.

For example, the blog mentions this point that I made:

Find people that don't like you or don't like what you do to give you honest feedback on the usability of your site

Out of context this advice can be a little tough to understand. Angela and Sania had both outlined different (very valid) methods for testing your site — free automated testing, focus groups, asking your grandmother, etc. — but I piped up with another idea to build on their points. I suggested you find dissatisfied customers or users, people who may not like you (and are less likely to gush), people who've had a bad experience with you before. They may be more likely to tell you what is wrong, what didn't work, why they were peeved and to do it without worrying about sparing your feelings. I'll distill this to two general outcomes:

1. They talk crazy and in the end you discard their feedback; or
2. They raise some very good points, you learn something and you may very well win back a formerly dissatisfied customer/user.

I'm hoping these general goals weren't lost on everyone. I don't think too many people were live Tweeting (there were only a handful of Tweets from our session, all the live Tweeters were in the social media session), but we did run through a lot of information in a very short time.

If you were in attendance, you got a shiny hand-out that Angela and Sania put together that contained a general web site usability checklist. If you weren't there (for shame) then you can view the web site usability checklist that Angela keeps on her site.

If you were one of the kind folks who sought me out for questions afterward (and didn't confuse me for another bald guy in the hotel), feel free to keep them coming. Hopefully I'll see you next year and we can see how you're doing on your project.