Technology
#pcsw SEO Driving Traffic to your site with OpenID
Submitted by Aldon Hynes on Sat, 03/28/2009 - 11:39The first session at PodCamp Western Mass that I attended was focused on SEO. I tend to avoid SEO discussions because too many of them are snake oil, but this discussion had some real good points.
As a parenthetical comment, I realize that there are a lot of things that I could do to boost the SEO of my site, but that isn’t a major focus for me.
Tish brought up the importance of making sure that when you add a comment to a blog you add the URL pointing back to your own blog. I saw that and raised it one. When you are making a comment on a system that supports OpenID, be sure to use your OpenID to point back to your blog.
People wanted to know how to do this, and it is a little complicated to explain during the discussion, so I thought I would post a brief blog post and explain how to do it.
First, you need to have an OpenID. Many of you already do. If you don’t, there are plenty of places where you can get free OpenIDs. To get a good starting place, I like to recommend claimed.com. You can create your own OpenID there and if you have multiple OpenIDs, you can link them together there. Another site I like to recommend for setting up your own OpenID account is MyOpenId.com.
However, the odds are that you may already have OpenID. Livejournal, Wordpress, Vox, AOL, Yahoo and Technorati all provide OpenID. Some of these may be good ones to use. However, if you use one of these, you will set up your blog point to your OpenID.
To do this, you need to add OpenID delegation to the meta tags to the header of your blog. As an example, if you add
<link rel="openid.server" href=http://www.livejournal.com/openid/server.bml>
<link rel="openid.delegate" href=http://exampleuser.livejournal.com/>
This will make it so that when you use your Blog as your OpenID, it will go over to LiveJournal and check to make sure that you can log in as exampleuser.
With that, I would encourage everyone to set up OpenID delegation from their blog. Then, when you add comments one another blog, log to userid using your blog as your id. In most cases, this will leave a link back to your blog and help drive more traffic to you.
#DigiDay Recap, Part 1
Submitted by Aldon Hynes on Thu, 03/19/2009 - 10:51Last Thursday, executives interested in marketing and advertising in mobile and social media gathered at the W Hotel in New York City for a conference called DigiDay. The first half of the day focused on Mobile marketing and advertising and the second on marketing and advertising in Social Media.
The day started with a conference sponsored by ChaCha. ChaCha is a service where you can text any question and receive a response from one of 55,000 ChaCha guides. These guides, many of whom are work at home moms or college students, have an active community for finding ways to quickly answer any question that might come in. The answers are supposed to reflect information on the web, as opposed to their own personal opinions.
When you text a message to ChaCha, you receive a text message back which includes an advertisement. These advertisements can be targeted by location or topic. Currently, location targeting is done at the area code level. It has been used by over four million people and their surveys get over a twenty percent response rate. Currently, they are receiving over ten million questions a month.
It seems like a great service so I thought I would give it a try. While no registration is necessary, on their website, you can register your email address and phone number. I registered my email address and verified it successfully. However, I never received my verification code on my cellphone.
I thought this would provide an interesting question, and I sent a text message to ChaCha asking why I hadn’t gotten my verification code. The response was unhelpful, simply saying that I didn’t need to use the verification code to ask questions. I replied, acknowledging that it wasn’t required but that I wanted to anyway, and asking how to get it. The second response was as useless as the first.
On Saturday afternoon, while I explored the Hebron Maple Festival, I noticed a car with the State Representative license plates for Assembly District 55. I sent a text message to find out who the State Rep for the district was, and was informed that it is Rep. Pam Sawyer. The message included her phone number and an advertisement for H&R Block. So, currently ChaCha is batting .333 for useful answers. I’ll probably keep using it from time to time.
However, I was also disappointed to receive a text message at the same time informing me that I had used 3 of 5 questions during a 48 hour period and that I could only ask 2 more questions over the next 18 hours. Considering that one of the questions was an attempt to get an answer for the previous question that they failed to answer adequately, and even that answer was not adequate, I was disappointed. Have you used ChaCha? How well has it worked for you? If you haven’t, you can text to their short code, 242242.
After the breakfast, the first panel was “The Mobile Marketer Roundtable: The Elephant in the Room: The Economy:“ Personally, I’m a bit tired of all the gloom and doom discussions about the economy. Yes, the economy sucks. However, there are still lots of people doing lots of interesting things. Tell me something I don’t know.
Fortunately, June Bower, VP of Marketing for Cisco-WebEx did tell me something interesting I didn’t know. There is a WebX app for the iPhone. Over 70,000 copies have been downloaded already and WebX will be coming to other smartphones soon. Another interesting idea from this panel was the cellphone as sales assistant. Someone is going to come up with an easy way for a user of a mobile device to find something he is looking for in a store. That will be a cool app.
There were discussions about ‘click to consume’ and the closest people have come up with so far have been buying ringtones, wallpapers and games. None of these are all that compelling, but they have been lucrative.
The biggest hurdle that members of the panel saw to mobile devices playing a bigger role was getting marketers to understand the role of mobile as part of their 360 marketing.
A final thought from this panel was that to television people, a mobile device looks like a small TV. To computer people, it looks like a small computer. More and more, simple telephony is playing a smaller and smaller part of mobile market.
This was brought home in the next panel, The State of Mobile Media by the Numbers, when Julia Resnick, VP Mobile Media Products for The Nielsen Company spoke about their research. The iPhone is drastically changing the data usage of mobile users and Android and Blackberry Storm are also making data a much larger part of the mobile platform. The other interesting tidbit that she revealed was that the average age for children getting their first cellphone is now 9.7 years. They also revealed that the average teenage sends 2300 text messages a month. That works out to around 75 text messages every day. I guess I’m not that heavy a texter after all.
The following panel, Keynote Panel: The Mobile Platform Implosion, spent time looking at appropriate metrics for mobile usage. Nothing particularly memorable came out of that panel except for the observation that cookies on mobile devices are a problematic stop gap measure. More interesting was a rant about metrics about how each decade has had it’s own ad science, but then about 2005, all that ad science went out the window simply for measurement without a lot of consideration of what was being measured and why.
It was an interesting observation. If you know what you are measuring and why you are measuring it, then you can determine if you are reaching your goals. Yet many people do not seem to have a clear idea of what they are measuring or why they are measuring it.
After this panel, a spokesperson for a company called Mojiva got up and made a sales pitch. It wasn’t all that compelling. What was compelling was the discussion afterwards. During the Q&A, he was asked about Twitter. He dismissed Twitter as diarrhea of people spending too much time online and having no mobile implications. The large community of participants at the channel who were having a great discussion about the conference on Twitter were merciless. They spoke about it as an epic fail, a credibility failure, a debacle, a shame, and some suggested that it is sometimes it is just better to get off the stage.
Experimenting with the Google FriendConnect API
Submitted by Aldon Hynes on Sat, 03/14/2009 - 23:16This evening, I started experimenting with the Google FriendConnect API. In particular, I wanted to see if I could get anywhere with the OpenSocial REST and RPC Protocol.
What I’m really interested in doing is writing a standalone PHP script that will go out to FriendConnect and gather information for me. Reading through the documents, it looks like I want to use the ‘standard two-legged OAuth’.
The example they gave said that I should be able to access
h ttp://www.google.com/friendconnect/api/people/@owner/@self?oauth_consumer_key=<your
consumer key>&oauth_signature_method=HMAC-SHA1&oauth_timestamp=<the time
right now in millis>&oauth_nonce=<some nonce>&oauth_signature=<signed using your secret>
By going to the FriendConnect page and selecting the ‘For Developers’ section and then going to the ‘REST API’ tab, I managed to find my Consumer Key and Consumer Secret. I attempted to retrieve various bits of information, mucking around with the ‘http://www.google.com/friendconnect/api’ base. With just about every variation, I received a 404 error and no useful information.
Figuring that I was probably doing something wrong with the OAuth settings, I downloaded the opensocial-php-client. The sample in Version 0.2 worked quite nicely when I attempted to use the key and secret that came in the sample, and went to "server_rest_base" => “http://sandbox.orkut.com/social/rest/”. However, when I tried changing the server_rest_base to http://www.google.com/friendconnect/ I received the same 404 errors. My guess is that there is something wrong with the documentation and I need to find a different server_rest_base. So far, I haven’t had any luck.
So, I thought I would try a different tack. I looked at the and found a sample for Drupal. I downloaded the plugin and installed it on one of my test servers. It installed properly and the installation portion seemed to work okay. However, when I configured it, it asked for a ‘Unique site identifier provided by Google Friend Connect during initial registration.’ I don’t think I have any such unique identifier, or at least if I do, I’m not sure what it is.
Looking at the various URLs for reports and settings for my site, I do find a twenty digit number that might be the unique site identifier, but I’m not sure if that is it. When I try to use the FriendConnect module for adding a comment, I get a box that says ‘loading…’ which seems to just hang there. So, it doesn’t seem to be working properly yet. In addition, since I’m mostly using Disqus for my comments, right now, I don’t see any great advantage for throwing on this Drupal plugin for FriendConnect.
Have you had any luck with the FriendConnect API? Let me know your experiences.
OpenID, XRI, XRDS, and Portable Content
Submitted by Aldon Hynes on Sun, 03/08/2009 - 11:27I continue to dig into what it would take to build a good Social Network Contact Management System. Yesterday, I had a long discussion with Arron Kallenberg from DandyID. I had talked about my difficulties with the DandyID API and he wanted to address them. In a nutshell, their API is evolving and the documentation had not caught up with the current usage. Now that I know about the updates, I’ve got the API working nicely.
As part of the discussion, we talked about Portable Contacts as a standardized method of receiving contact information that people enter into their social networks. DandyID provides access to information in Portable Contacts format. I also learned that Plaxo is supporting the Portable Contacts API. Details can be found at Portable Contacts at Plaxo. The Plaxo API Endpoint requires OAuth or HTTP Basic authentication, so I started looking at doing OAuth authentication from PHP. I still need to dig deeper into this, especially for trying to call functions from a php script that isn’t part of a webpage.
As I read the Plaxo documentation, I also got interested in XRDS Simple discovery. With that, let me talk about OpenID, XRI and XRDS. OpenID is a method of sharing authentication. In other words, you can use an OpenID userid and password to login to any site that accepts OpenID. This makes it so you need to remember less userids and passwords.
Some people express concern about OpenID phishing. If someone manages to get your OpenID password, they can get to all of the sites you’ve used OpenID with. On the other hand, if someone gets your password, you only need to change it once, instead of either changing the password at a lot of sites which have the same password, or keeping a file somewhere which lists all your passwords for every site you use. I know that I can’t remember all of the passwords I have for all of the sites I’ve registered with, without a bit of help.
Also, using OpenID delegation, you can point back to your site when you use OpenID. For example, I always use OpenID pointing back to my blog when I place comments via Blogger.
This leads me to XRI. XRI is the extensible resource identifier. It is sort of like a URL on steroids. There have been major arguments over when to use URIs and XRIs, but being the geek I am, I have two XRI identifiers that I use regularly, =aldon.hynes and @ahynes1. In XRI, a personal name starts with an equal sign, and an organization name starts with an at sign. I registered @ahynes1 since that works out to be my identification on Twitter, Identi.ca and other services.
One of the things that is interesting about XRIs is that OpenID version 2.0 supports XRIs. So, I can log into services that support OpenID V 2.0 using my id @ahynes1
To support XRI, XRDS, the Extensible Resource Descriptor Sequence was created. It can be used to discover information about a resource. As an example, in OpenID V 2.0, you would use XRDS to find the appropriate OpenID server for a given XRI identity.
With this as groundwork, I am now back to Portable Contacts. My contact information and the other services that I’ve subscribed to is additional information about me and my resources. So, it would be great if we could use XRDS to find servers that provide Portable Contact information about me.
It looks like I can do this using my XRI accounts and DandyID or Plaxo. Both 2idi.com which I use for my =aldon.hynes identity and 1id.com which I use for my @ahynes1 identity provide the ability to edit the ‘service endpoints’ that are revealed through XRDS.
As best as I can tell, what I need to do is add
<service>
<type>http://portablecontacts.net/spec/1.0</type>
<URI>http://www.plaxo.com/pdata/contacts</URI>
</service>
to my XRDS document, and people should be able to find my portable contact information via Plaxo. If I use http://www.dandyid.org/api/poco/ as my URI, they should be able to get my portable contact information from DandyID.
1id.com uses a webform to create this information, and I need to choose options like select, append and priority. I took the default values and we’ll see if they work
You can see my XRDS documents here and .
The next step is to add pointers to them in the header of my blog using the format
<meta http=equiv=”X-XRDS-Location” content=”http://xri.net/=aldon.hynes?_xrd_r=application/xrd%2bxml;sep=false” />
I need to decide if I should use =aldon.hynes or @ahynes1 and if I should have those point to Plaxo or DandyID. Then, I need to find good ways of testing and taking advantage of this. It seems as if Portable Contacts, discovered through XRDS may be a great building block for OpenSocial and other efforts to share information across platforms.
If you’re a non technical blogger, you probably don’t want to start playing with this yet, but you should keep it in the back of your mind as part of the direction social networks may be heading. If you’re more of a geek, kick it around. Share your ideas. Let’s see what we can do with these tools.
#FollowFriday - @JoeCascio @sara6633 @wkossen @SHHHE
Submitted by Aldon Hynes on Fri, 03/06/2009 - 09:52A popular meme on Twitter is to post a list of interesting people that you are following on Twitter on Fridays. Since my blog posts are fed into Twitter, I thought I’d post it here and get some double coverage as well as use this as an opportunity to talk more about my ideas for a Social Network Contact Management System.
The idea is to have a system where you can keep track of all your friends on various social networks, as well as the contacts you’ve made with them. #FollowFriday is a great example. Currently, I’m following nearly 1500 people on Twitter. As I pick out people for #FollowFriday, it would be nice to pick out people that I’ve been talking a bit with recently, but haven’t posted on #FollowFriday before, or at least for a long time.
I chose these four people because they have helped shape, either directly or indirectly, some of my thoughts about a Social Network Contact Management System. Joe and Willem have both spoken with me about the programming issues. Sara is co-founder of DandyID which is a great source for keeping track of your social networks. Priscilla is a key person behind PeopleBrowsr, which does some interesting cross social network stuff.
The thing that I’ve always liked best about PeopleBrowsr is the ability to group people by tags. With that, I can look at various streams, such as the stream of people in Connecticut, people in a Tweetup group in Connecticut, Newspapers from Connecticut, bloggers that are on EntreCard, bloggers that are on CMF, and so on. I’m not sure how people can follow more than a couple hundred people on Twitter without a tool like PeopleBrowsr. I used FriendFeed in a similar manner before PeopleBrowsr came along, and I’ve heard you can do some really nice similar stuff in TweetDeck, but I haven’t tried it.
Also, I’ve just started playing with another feature on PeopleBrowsr. You can bring up a stream or grid of Facebook, LinkedIn, or Plaxo friends on Twitter. Unfortunately, it seems to try to do a match based on name, instead of their actual userids, so since one of my friends is named Tim O’Brien, it is bringing up lots of different Tim O’Briens on Twitter. Not quite what I want.
Beyond that, I would like to be able to extract the information in PeopleBrowsr concerning the services I’ve subscribed to as well as the services my friends have subscribed to, much like I can with DandyID, MyBlogLog, FriendFeed and others. Even more importantly, I would love to be able to extract the tag information.
Using my first very early pass of SNCM, I’ve found about 140 people that I follow on other social networks that I don’t follow on Twitter. I need to start following some of them. I need to explore the tool in PeopleBrowsr to see whom else I ought to be following, and then, there is always Mr.Tweet as another source of good people to follow.
So, if I’ve recently added you on Twitter, that could be the reason. I may do similar adds for other networks as well, once Twitter is under better control. Hopefully, I’ll even come up with a better way of tracking how I contact people.