Skip navigation.

Sign up | Lost password? | Help

User Centered

Studying the design of everyday things

Posts tagged with "google"

Physical gesture based dialing

, , , ...

Way back in 2002, I found a little web browser called Opera that had this feature that just blew me out of the water...mouse gestures.

Since I've become to rely on them for browsing, I'm also looking at all the other plugins and extensions that extend this functionality to other pieces of software and interfaces.

Phones now have accelerometers that detect 3D movement, so it seems natural to use movement and gestures to perform functions on the phone.

I've just viewed the video for this phonepoint pen project by Duke students that uses the movement to translate and do OCR on captured movement to text. I think this is interesting, but not very practical. Instead, I'd like to capture the movement and translate them directly to commands. Keeping with the Opera browser mindset, I'm picturing a handful of easy to perform gestures that are one or two linear movements that can be turned into key phone functionality. Apple is tapping into this idea, they just announced "voice control" and they have a handful of physical based gestures- but they are just on the tip of the iceberg: they have "shake to undo" or "shake to refresh" functionality, but this can be extended...

  • Speed Dialing a phone number (one gesture for each favorite contact)
  • Launching a browser
  • Initiate a desktop-to-mobile sync/handshake
  • Move data from phone to desktop
  • Copy, paste, delete, add...
  • Refreshing the current view (existing)
  • Shake to undo (exsiting)
  • launch maps and use current location (ideal for auto based "where am I?" queries)


But mainly things you'd be likely to in an automobile or need to get at frequently and quickly...what I like about gestures, is that you can do them without paying attention to the interface- there's no mental/cognitive cycles spent doing mundane interface manipulation. You're using muscle memory. Like my browser mouse gestures, I can still be in the middle of finishing up scanning the text of the page and close the page without any movement to the "x" button (or any real thought). at the sake of going off on my mouse gesture tangent, I'll leave it at "if you get it, you get it, if not, you don't" and just press on: On the mobile device, this opportunity is greater. I think this is easier to use than even physical speed dial buttons since you'll have to orient your hand to the phone and find the right button. With a physical gesture, you can find a single "activate" button (which would be consistent for all gestures of course) and then just let your muscle memory do the flailing!

I can think of some contexts & settings where voice controlling would be more appropriate, and I can certainly think of times when it would be more discrete to swipe your phone quietly through the air.

Anyway, I'm almost certain that someone already has some form of this out that goes beyond the "shake to X" functionality, so I suppose I should just wait patiently for someone to point it out in the comments. I've seen the speed dialer app that lets you make a gesture on the screen of the iPhone to dial someone, but that has none of the advantages of abstracted UI that I mentioned.

2009- Location Based Services will come of age

, , , ...

I talked last year about my desire for a location aware task list,and having phone settings be context aware(although that link was not GPS based), so I was excited at the potential applications for Locale on the Android platform.


Locale allows you to create Situations, which specify Conditions under which your Settings should change. For example, your At Work situation might notice when your location condition is 1600 Amphitheatre Parkway, and triggers your ringer volume setting to vibrate. Locale is an innovative way to simplify your life.



This is really getting into what I think technology should be doing for us- context/location/situational awareness that adds value. I'm excited to see what location based services will come out now that mobile application development is reaching the masses, and mobile devices are growing to embrace.

I hope to keep an eye out on location based services here on this site, so if you run across some news, feel free to post it!

Usability Research on Federated Login

, , , ...

Scenario 1: New user from a trusted IDP
If an AOL user comes to the buy.com site with their current UI (as opposed to the suggested modified UI), and has never created an account at buy.com before, then they would enter their @aol.com E-mail address, and choose "I am a new customer." In that case, buy.com would show them an account creation form. However, let's assume buy.com is willing to act as an RP, and it has decided to trust AOL as an IDP. Assuming they switch to the UI model suggested above, then when the user visits the buy.com site, they would enter their @aol.com E-mail address, and choose "Help me sign in." Admittedly the phrase "Help me sign in" is not as explicit as "I am a new customer" however so far our usability tests have shown it works just as well (though we would like help getting more data to confirm that fact).

In this scenario, buy.com could detect that the domain name is for an IDP that it trusts. It could then redirect the user to AOL to verify their identity. Assuming the user approves sharing their identity, then the user will be redirected back to buy.com which can automatically create an account for them, and log them in.



Read the whole report.

Google News- Can we aggregate the aggregate?

, ,

Dan writes and asks for thoughts on Google News...

Originally posted by Dan:


Why is it that Google News can't recognize when a hundred different websites are simply printing the same AP wire story? When I search for a news item, I often like to sort by date (instead of by relevance), so I can be sure that I'm getting the most recent news item on a subject. Almost always, this results in me getting two or three pages of results of the exact same article, just reported by different TV and newspaper organizations. You can tell because the titles and snippets that Google provides are exactly the same! Shouldn't there be a way to collapse these results? When I sort by date, I expect to see this (I made up an example of an election result):

John Smith wins! - Today
Results were certified earlier today...
See 150 similar articles

John Smith projected to Win - Yesterday
Final results are still coming in...
See 150 similar articles

Polls Show Smith and Blow Neck and Neck - One week ago
Local residents are split on who will be Smalltown's next mayor...
See 35 similar articles

(a couple pages later in the results)

Joe Blow to Run for Mayor - Six months ago
Joe Blow announced his candidacy today for this fall's mayoral election...

------------------------------------------------
Instead, it looks more like this:

John Smith wins! - 10 minutes ago
Results were certified earlier today...
Smallville Picayune

John Smith wins! - 18 minutes ago
Results were certified earlier today...
Smallville Ledger

John Smith wins! - 2 hours ago
Results were certified earlier today...
WSMV - 5 ABC NEWS

(three pages of results later)

John Smith wins! - Today
Results were certified earlier today...
Delaware County Register

Am I the only one that thinks this is bad design? Why would I care if the Picayune or the Ledger (and a hundred other news publications) posted their article 10 minutes sooner, when they both just republished the exact same AP wire story? When I'm looking at this view, I'm hoping to see a chronological account of what was reported from when it was breaking news to old news. I'm not looking for 100 copies of the “old news” wrap-up. Is this unreasonable or off-base? I'm sure that Google is smart enough to recognize identical articles, and could group them together (they certainly do in "sort by relevance" mode). If someone wanted to find a particular news source, they could always expand out that list of similar articles.


(Update: I didn't "sort by date" when I took these screenshots/made these comments... see comments)
I feel for him.. But I'm not sure if I fully understand what's going on... it seems hit or miss. I've found both ways that Dan is talking about. First, is a search on "Hewlett Packard" gives me on the first page, the same article reproduced in differing links:

But there's also the "similar" or "related" links too:


...so I'm not sure if I'm fully understanding. But in the first case, I do find it hard to believe that google can't somehow figure out the articles have the same "content." I'm trying to get a handle on this from an "activity centered" point of view. Your goal in browsing is to quickly scan aggregated news articles from different sources. But... of course the AP articles, republished ones get posted around a million times that it creates extra noise you have to filter out defeating the point of the aggregation.

Thoughts?

Thanks for the email Dan.

Google on the go

, ,


So you've got Opera mini or Mobile. You're all fired up about widgets and can't wait to get 'em on your phone. Well, how about another tool that I don't hear too much about. This probably isn't news to most people, but well... I just don't hear anyone talking about it, so I figure there's got to be a handful of folks that have never heard of...

Google SMS. You can send your google queries via SMS to "GOOGL" and get an SMS message back with your query. Seems like a strange thing to do if you're thinking about how you use regular 'ol google, but this is optimized for mobiles with a ton of keywords and shortcuts. For example, you can send "movies 55901" just like the regular google and get a SMS back with the listings. Useful on the go (I use Opera Mini for this...but hey, I'm just spreading the word).

Try the examples in the sample paged linked. It might come in handy at some point. Make sure you check the FAQ and send it "tips" to get a list of shortcuts.

The more you know....

One step closer to ideal calendar usability

, , , ...

I was thinking Outlook, but clearly google is first to market on something like this:

From my early days*

Outlook Calendar and Mail- tighter integration Posted 13 Sep
Highlight a date in an email: “bob- let’s meet at 200 today to discuss the changes.” and be able to add an appointment to the calendar from that. Nobody sends appointments for things like that, and when someone who is not an exchange users sends you an email- you still might want to add it to the calendar. The appointment creation already takes a broad range of input values for dates/times- why can’t we tightly integrate that into email?


As you can guess from that...I *love* the new "Quick Add" feature for the newly beta'd Google Calendar (unfortunately, there are some quirks in Opera). I just typed in the string "test tomorrow at 3" and it made the appropriate entry. (also "raquetball tomorrow at 2pm next to the gym" got entered as "raquetball next to the gym" at the correct date/time)

I think the next natural step for google then, would be to allow you to somehow get text from gmail to send to quick add (the tighter integration I spoke of above). It could be as simple as adding a "Quick Calendar Add" box like the "Quick Contacts" box. You could copy/paste to quickly create appointments from emails sent. But ideally, I'd like to highlight text, right click and "Create Appointment" from the string.

*That "raw dump" post consists of some "rough thoughts" I had written down before I started here at the Opera site.

When will online maps think like us?

, , ,

Yahoo is primed to unleashed it's revamped volley into online mapping frey again with its beta yahoo map. So I've rekindled my thoughts on online mapping.

If you set an address as "home" in your favorites/bookmarks/whatever. Don't you think that a map should be smart enough to know that you know how to get out of your neighborhood?

Every online map I've used takes half of the space of my printed directions just getting to freeway two blocks away from my house. Look- I live here. I've told you that. I know how to get to the freeway. I'm going someplace I've never been before- give me the details on the destination end.

We should be able to set the scope of the search. If we are driving across country- we have a different scope than say- finding that new pizza place that opened up in town. You should get appropriate maps.

Even local maps- why is it that we can say "it's on 3rd Ave. Take 2nd to broadway and turn left, two block up on the left" One or two sentences is all it takes there, but online maps will probably spit out out least ten lines that have "0.1 Miles" and "Turn from 2nd Ave onto 2nd Ave W" or something when it's clearly the same street or the street changes name.

Worse- if you try to reduce the noise to focus on the signal by entering just a city name for the starting destination- you get some random point in the middle of the city that often produces even more garbage directions. I think you if don't specify, the directions the map should show the first road that takes you outside the city boundries.

I realize mapping the world is a difficult process. I realize that streets are poorly designed, not black and white, and you need to make special cases for "stay to the right as the road splits" kind of things. I assume it is no small task, and map makers are erring on the side of caution here. Generalization will not always work.

But you should be able to set a "Home" preference where you can set some level of abstraction like "I know where the major highways are" or "start all out of city directions from this freeway" so your directions start with "take I-95 South..." instead of starting from your driveway. Further abstraction should be possible with a "Details" or "Explain this" or "How do I get here" button that drills down.

Most often, I will look at the directions and then write down a simple three for four lines of directions from among the 25 or so. If what I'm proposing is too simlistic- then it would be nice if we could at least choose the directions that get printed out. That way- the user can decide what's important and what is not. I think that would be a good compromise.

Note- that's not really my address- it's a papa johns pizza place or something like that.

housingmaps.com- Ideal handling of browser checking

, , , ...

UPDATE: it appears this site works fine with Opera now, and lets you right in the front door!

A great service mixing google maps with craigslist: http://www.housingmaps.com/

But what I wanted to point out is the homepage. At least- the homepage that saw. There is a browser warming page that I got because I use Opera. The creator handles it pretty well I thought. In a nutshell (meaning- these are my own words), I got a

Hey- we've tested with all these browsers here: <lists them>.

Your's is different. Here's how to get one we've tested with...<link>

But if you still want to try it out- here's a link in. We can't promise anything though.



It's that last part that I really thought stood out... the &quot;you've been warned, but if you want to try anyway...&quot; type of link that is all too absent from websites. Mostly when I use Opera- I get &quot;this browser is insecure- use IE instead&quot; and I'm not permitted to go any further (without changing agent ID or spoofing of some sort). I understand that site creators don't want or need to test with every single version of every single browser on every single version of every single OS- but I'd be fine with telling me what you do test with and giving me a clear warning that you don't promise success with the tools I'm using- but still give me the choice. Often (no mostly.... no...almost always) I can get the job done just fine with Opera. Kudos to this site for letting me try.

Here's what I sent the creator:

...thank you for at least putting the option to let me try your service in Opera. I get REALLY mad when sites don't let me take the chance... Sometimes, there is a limitation with Opera (not full AJAX support)- but most often its simply because sites don't want to think about anything other than IE, and one or both Netscape/FireFox.

I understand that nobody wants to test in 100 browsers that less than 1% of people use. But at least let me have the option to try your site with my browser. You do a fine job of that- your main page is exactly what sites should do.