Thought Pattern

2006-06-29T09:25:00.000-04:00

Idea [#5] - Microformats and Mobile Registration

I've been thinking about security and logins for mobiles and peeling away the conventions that we use on the desktop. I wanted to see if there is something else that can be done that will work better for mobiles. From recent articles the future mobile usage will be the primary means in which emerging markets and eventually everyone will interact with the web.

In this vain i've been trying to make sure that on Wampad you don't need to use a desktop to configure the site. My first "concern" is registration, should I just keep the fields down to the bare minimum and give the user the ability to add more if they want.

As more serivces for mobiles come online the requirement for registration to use these services will increase. I can see this being a pain for mobile users, it's a pain for me on the desktop as it is :).

As an example, you see a poster with a url or QRcode that takes you to a website, but before you can use the information they promised you need to register to access it. Now you have to decide if you are going to fill out the 5 field form or just forget it. Even worse you have to wait until you get home so you can register on your desktop machine becuase they need the 10+ fields but didn't want to burden you with typing that in on the phone.

Now here's where microformats can come into play. Kind of in the vain of Openid you could provide a url where your hCard is and the service your registering for will pull in the information it needs and all you need to do is specify a user id and password. Pretty much all the personal data they could ask can be stored in hCard format. Heck, it's even got a area for a photo that all the web 2.0 site want you to upload. I would suggest that if this is going to happen the service would show the user what data they want to use so that the user can ok it.

I was also thinking that for users who don't have a url with an hCard, someone could build a microformats server that could host hCard as well as other formats. If it's standard the users could enter in an short code that could translate to a url like http://microformats.mobi/u=wampad and all the user would have to enter is wampad.

Microformats could be used for a number of things in the mobile world. With Nokia's idea of a web server on a phone the service could get your hCard directly from your phone, so all you would need is to provide the phone number. Maybe the caller id information of the future is pulled via the callers hCard on their phone's web server.

2006-06-27T19:40:00.000-04:00

Thoughts About: Security and Logging into Mobile Sites

One of the things that I find hard to do is type out my password on my mobile phone. The way Netfront handles password text boxes is by replaceing the last character entered with a *. I'm not complaining about this as I think it's a good feature. Being security conscience though I tend to use an alphanumeric passwords and not having a full keyboard to type my password has caused me some minor head/thumb aches, but not enough to forgo logins on my mobile.

My concern is for the masses that will be using their phone to access sensitive data like checking accounts as seen in the recent Chase commercial. There have been tons of whitepapers and blog posts about users having insecure passwords. What type of passwords are people going to choose if they have to triple-tap the password. The guys accessing Paris Hiltons sidekick account and getting her contact list and pictures was portrayed as comical, but I'm sure if it was her bank account it wouldn't have been portrayed the same.

Biometrics could solve some of the problems, but they were supposed to solve them on the desktop but I haven't seen them really used anywhere.

Well this post was more to toss out the topic for discussion. I've got my brain turned to this now so maybe I'll have more on this topic soon or some links to sources of information on the web.

2006-06-23T12:25:00.000-04:00

Bloggers and Enterprise

I just read a great article, Multiple Architectures For Mobility by Daniel Taylor. The part that really hit home with me was this,

"How can there be positive press about a set of technologies that customers don't exactly want, aren't buying in droves and have few plans for? It's a very simple disconnect between markets. FMC and VoIP have a great deal more appeal for small businesses than they do for enterprises. But most analysts, editors and bloggers don't work in enterprises; they're in the SOHO and SMB environments."

This is so true, and i'd extend it to new hires also. I couldn't tell you how many times i've sat in a go no where meeting with someone who came in from a much smaller company. They have all these ideas that just aren't going to work when your talking about the enterprise. Most of them require resources that are just not available, and then when you bring up roadmaps and CMMi the meetings turn into a barage of "what ifs". I heard them say that they would build it and then you have to talk about support. Are they going to come in at 1am on a friday night if it's down? Of course they want IT to handle it, but then IT only support apps they build or third-party apps with a 24/7 support structure which then tends to lead back to roadmaps, rinse, repeat.

I've been watching blogs about Enterprise 2.0 and I see most of it as being a bit to optimistic. Usually it never takes into account HR issues, where some of these apps can be distruptive to corporate culture if not handled right. My views might be skewed because I've worked for the same company for ten years and usually the cutting edge stuff is more like a butter knife when we get around to implementing it. We'll get to Enterprise 2.0 some time when Enterprise 3.0 is said to be here.

2006-06-13T22:44:00.000-04:00

Geo-Location for Mobile Browsers

I'd like to see carriers and mobile browser manufactures produce a way to get location information to mobile websites. Java apps on phones are able to tap into your exact location using GPS, but sometimes all you really need is a zip code. I do think that mobile websites will be more popular the Java downloads in the future. I am a little biased because I'm building a mobile website. Mobile web browsers will eventually provide the types of interactions that the Web 2.0 sites are displaying now and kill the interface advantage Java currently has. Opera is already pushing ahead with the mobile AJAX ideas.

Oops, sorry back to the topic at hand.

What i'd like to see is a bit of granularity in what you provide to a site in the way of your location. This might lessen some of the resistance in adoption from users who are concerned with providing to much personal data. I can see that Zip Code would be the most generic level that you would want to go. Going out to county or state would probably be to large of an area to be of any use. The next level might be the GPS coordinates of the cell tower your connected to. It gives you a closer proximity then Zip Code without giving away your position. The final level would be exact location using GPS in the handset or through radiolocation.

To get this information to the server of the mobile site your viewing, it could be handled by a special request header like "mobile_location" or "user_location". If you wanted to send the site your exact location then your phone could send the header "mobile_location:38.897625,-77.036661". Getting less specific would need the cell tower to modify the request header as it came across turning "mobile_location:zip" into "mobile_location:20500". It could also be possible that the browser will let you hardcode a zip code to send, so that you can use sites as if your always in a specific location.

If a mobile user isn't trying to get sports scores or querying imdb to settle a bar bet, most of the interactions on the mobile web are going to be for geo-relevant content like weather, directions or something like dodgeball.com.

Personally I don't want to login into mobile websites for them to provide me with some relevant content. If I go to espn and i'm in Chicago, I want White Sox or Cubs information. I don't want to have to fumble through a menu system to get that information, or heaven forbid triple tapping Hurricanes into a little search box. Dare I suggest "Mobile Where 2.0", Um na forget I mentioned it.

This might even be useful on the desktop side as well. Either your browser or the router that you connect through, could append the request header to all requests. From an intranet server perspective it could provide an easier method to identifying users location then maintaining a list of ip segments and where they are. On the internet in general it could be used to give your location rather then site owners doing lookups based on your ip address. That way if a backhoe takes out a comcast line and they route you through their Chicago lines, sites will still give you content from Milwaukee.

2006-06-06T23:28:00.000-04:00

Idea [#4] - Video Camera + Slingbox + Mobile TV = Mobile Nanny Cams

I was thinking that there has to be something cooler/useful then watching tv on your phone. To me if it's a must watch show i'm going to want to see it on something a little bigger then a 2 inch screen.

I was reserved to the fact that mobile tv technology could really only be useful to watch sports live. Then one night when my wife and I were out, we made our obligitory call to check on the kids and it hit me. Nanny cams streamed out through something like a slingbox could be a nice way to check up on things at home, not that I don't trust my mother-in-law.

Well that thought got tossed around in my head for awhile. I figured nanny cams is a bit of a niche market, so who else would want to watch live streaming video. Well the most obvious is a mobile version of jennicam and the more adult voyeuristic versions that came later. I also think that there might be a market for security operators using mobile streaming video. The market would probably be for the small to medium business's that don't already have a 24/7 security presence on site. I know a few people who either get paged or phoned when the security system is set off. Generally it's because some punk kid threw a rock or something through a window, but you never know. More then once i've heard their wives comment that it is a bit scary when their husbands have to go on site to check things out at 2am.

So I can see a future where a small business owner gets a text message to their phone when the security system gets triggered. Maybe with a guid or something embedded in the link so they don't have to login to view the feed. Then right away the owner can hopefully tell if it's a situation where the police need to be called, and if police are needed forward the feed to them so they can ascertain the amount of presence they need at the scene. If it's just a rock through a window they can bring some wood and a hammer to board it up until it can be fixed.

I guess there might also be a market for residential use. I'm just sure how i'd feel if someone at ADT could view the inside of my house at anytime.

Now I have to clean out my history, so that my searches to see if there are already mobile streaming video enabled porn sites doesn't get me in trouble with my wife.

2006-06-03T23:07:00.000-04:00

Thoughts About : Near Field Communication

Wikipedia: Near Field Communication

In the jumble of wireless protocols like RFID, Bluetooth and Ultra-Wideband is Near Field Communication. It's designed to connect devices through magnetic field induction by bringing them within centimeters of each other. It's designed for making payments with your phone, but there's got to be more fun that can be had with this. I like the fact that it's range is so short and from my limited knowledge i'm guessing it would be hard to build something like the BlueSniper Rifle for magnetic field induction.

Microformats combined with NFC will be big, you could carry around a bunch of personalization information with you. Before I get in my car I can pass my phone over the door and when I get in, the seat is how I like it and the radio has my presets. You could probably extend that to any car like rentals. Forget using a usb drive to carry my bookmarks and stuff. I can just place my phone on the computer and there they are, so is that proposal doc I have to finish writing.

I would really like to see this in action. If you have to hold the two things together for more then a second and wait for a green light, well it's going to be less then remarkable. If you can literally just wave it in front of the other device and it will reliably transfer the data, then it'll be something special.

This might kill things like QRCodes, if passive NFC is cheap and easily deployed. Why take a picture of something when you can just pass your phone over it and it transfers information or sends you to a mobile website. Plus it can be indirectly two way, so what ever is passed to you, you can pass on to something else. Say you pass a poster for a new subway sandwich, there's a little NFC enabled graphic on it. You pass your phone over it and it's a coupon for a half off sandwich. So you goto the nearest subway and pass your phone over the reader there to get your half off sandwich. If your wearing a tinfoil hat please stop reading. Now when the marketing group at subway bought these posters they embedded a unique key in each one and they know where each one is placed. While the campaign is going on they can see where people are coming from to redeem the coupon at each store. After a bit of A/B testing they could figure out that one area only needs a 30% off coupon while another requires 50% to drive traffic to the subway.

My last thought on this is a system for connecting to friends. On your way into a bar you swip your phone over a NFC tag that gives you a unique id for the bar that is then used to update your whereabouts on your mobile myspace/xanga/[insert site here]. At the same time the bar's system lets you know who on your list of friends is at the same place. Basically a automated interface into a site like dodgeball.com.

2006-06-02T23:04:00.000-04:00

Idea [#2] - Event Specific Social Experience

I was thinking that it would be cool if you could use your mobile to make group based decisions at a concert or bar. Say your at a concert and at the front of the building is a QR Code. The code takes you to a mobile site that has a list of all the songs the band may perform, say 1.5 times the number that will actually get played. Then on the site you and everyone else will move your favorites to the front of the list, or maybe just pick 5 songs. Before the band gets on stage a "magic process" runs and produces the playlist based on the whims of the crowd. I'm not a musician so i'm not sure if playlist chioces are written in stone for some reason, or if they are mutable.

Another nifty thing would be to have something on your mobile where you define your mood and it's locally queryable. So then when you walk into a Starbucks, the store checks your mood and matches it up with everyone elses. So if everyone in starbucks is depressed then something slow and mellow plays on the stereo and if everyones happy something a little more upbeat is played.

2006-05-26T12:21:00.000-04:00

Enterprise Social Bookmarking

I've been thinking that social bookmarking would be great in the enterprise. There are a few things that I would like to see in an enterprise level social bookmarking application. One of the main paridigm shifts in bringing a social bookmarking tool into the enterprise is to keep a smaller more focused tag cloud and provide tools to help with that. This helps to direct the users into areas they should be focused in and hopefully curtail random browsing that could effect job performance.

Some points i think would make the Web 2.0 social bookmarking tools useful in an Enterprise 2.0 setup.

Administration

Off limts urls - This kind of goes with proxy blocking rules. If your company is blocking Youtube then Youtube links should not be able to make it into the system or at the least kept out of the public list. Sometimes you also have a situation where one user has access to something other users don't. I can see it necessary when a third party app is deployed that has less then ideal security, you may not want links showing up in the public list.
Automatic tag assignment - It would be a benifit to be able to apply certain tags automatically based on a url. So if I bookmark a cnn.com article it would be flagged as a news, or a bookmark to a competior's site would have a competitor tag. This way in the enterprise you can have a bit of control and have a chance at creating standards that everyone knows and can quickly look up the links they need.
Tag owners - Being able to remove a link from a tag would be nice to keep junk out of key tags the company uses for specific topics. It should be distributed, so someone is sales could be allowed to remove the competitor tag from any link tagged with competitor. This keeps key areas junk free and allows the administration to be spread out to Subject Matter Experts and keep a flood of removal requests out of the system admins inbox. Moderation might be appropriate but you should be able to turn it on and off per keyword.
Terminated Employees - I can't think of the best way to handle links by employees who leave the company. Maybe multiple options is best. It could drop all links from these employees to a default account or append something like "-gone" to the name. I would think the app would be tied to your directory system and it could be a problem unless you never give out the same ID more then once.

User Experience

Public, Private and Group links? - In the enterprise there are somethings you just don't want to share with the whole company, but inside your group it needs to be shared. An example could be someone in management flagging an outside companies home page with the tag acquire. Sharing that with other management would be fine but it could cause problems if the link with that tag made it to the public list. Groups could also be used for the for: ability to send a link to an entire group.
Consistansy check - If a tag is new an validation screen should be presented so that the user can see that they are creating a new tag and allow them to fix it quickly. Basically to keep spelling errors out like the one in the title of this bullet.
Internal/External Visual Queues - Some way to easily identify if a link is internal or not would be nice. It could be based on a certain tag generated based on the url or a manual checkbox.
Easy Submittion - This is a key feature that needs to be there and standardized through out the enterprise. Just installing a tool like this isn't going to have a huge impact or grow outside of a core group unless subittion is easy. A corporate wide toolbar for IE and firefox would be the best setup and probably help with other issues. Keeping it incredibly easy and without any cutting and pasting is key.

I'm all for the openness that is makes del.icio.us such a cool tool online, and maybe for small business's. I just think that for a large company where you want this type of thing to take off, having it wide open could present more problems in the long run.

2006-05-22T09:48:00.000-04:00

Idea [#1] - Mobile Ad Campaign

I was thinking about a campaign that would use posters with unique codes printed on them and a persons cellphone. Ideally the codes would be 2d barcodes like QRCodes but that might not work in the US but maybe Japan. Otherwise you would need to use unique alphanumeric codes or unique keywords. You could then place them around a city or event.

When someone finds the poster they would take a picture of the 2d code or enter the code string into a url shown on the poster. If your the first person to enter the code then you could win a prize and all subsequent code submittions would receive a coupon.

From the marketing perspective it keeps your brand at the front of a persons mind as they hunt for your poster.

Another spin on this could done at events, expos... Have a group of people with codes on their shirts walk around, each different like a poster. The winners need to goto the companies both to redeem their prize. Last spin would really only work for 2d barcodes. Have a group of people with all the same code on their shirt and then one person with the winning code. Hold that person back for a bit of time and then have them wear a jacket to the middle of the expo and then reveal their shirt with the code.