Archive - Oct 2009
October 31st
Google Wave Federation on Ubuntu – Openfire XMPP, StatusNet, Asterisk and Beyond
Submitted by Aldon Hynes on Sat, 10/31/2009 - 10:48Last night, I spent more time exploring Google Wave, and find there are so many more areas that I need to explore. I am especially interested in Google Wave Federation.
On the GoogleCode blog, they talk about “the Google Wave Federation Protocol, the underlying network protocol for sharing waves between wave providers.” They go on to explain, “Yes, that's between wave providers: anyone can build a wave server and interoperate, much like anyone can run their own SMTP server.”
With this in mind, I set up my own wave server. To do this, I followed the steps in the Google Wave Federation Prototype Server Installation Wiki Page.
At first glance, it looked nice and straight forward for Debian and Ubuntu users. Make sure you have Java 6 installed, grab the Openfire package, install it, configure it, and add the Wave software. Unfortunately, it turned out to be a bit more complicated.
If you are running on an older version of Ubuntu, you cannot simply install sun-java6-jdk. The package isn’t found. You can install openjdk-6-jdk, but then the Debian package can’t find sun-java6-jdk and it won’t install.
I did install sun-java6-jdk and Openfire 3.6.4 on one of my machines. However, I couldn’t get it to start. Then, when I finally got it to start, it wouldn’t recognize the administrative password. I went through several cycles of installing, uninstalling, tweaking, and couldn’t get things to work consistently. It was then that I discovered two important things:
First, when you start Openfire, depending on the machine you are using, it make take a minute or two before it will start responding to http requests. So, install Openfire. Wait a minute or two. Then, attempt to configure it. This led me to the second issue. After you configure it and set up the administrative userid and password, it appears that you need to restart the Openfire service. Once you’ve restarted it, and given it a minute or two to get up to speed, you can login to the administrative panel, and continue on with the configuration.
Openfire is an XMPP server. XMPP is a standard protocol that is used for chat programs like Google Talk. It is used by StatusNet for connecting with their micro blogging. StatusNet connects with Twitter, and the Openfire server has plugins that allow connections to other IM services like AOL and Yahoo. Openfire also has plugins to connect to Asterisk and SIP, so that you can use an Openfire session to connect to Voice over IP services. These are things I have yet to test.
My first test to make sure that Openfire was up and running was to run Pidgin, a chat service that runs nicely in Ubuntu. Not only did it connect nicely with my Openfire server, but I managed to connect over IPv6 for some of my testing.
Once I had Openfire running smoothly, the next step was to add the Wave component. In the instructions, they say to download the software using the ‘hg’ command. I did not have ‘hg’ as an available command. Hg is part of the mercurial package, so I did a
sudo apt-get install mercurial
To install Mercurial. Then, I retrieved the code and tried to build it. However, I had initially installed just the Java Runtime Environment, and I didn’t have ‘ant’ available as a command. I installed ant and the full Java Development Kit and the build worked. It took about a minute to build on my machine.
The next step was to build certificates. The Wiki page for Certificates provides a simple easy way to build your own certificates. However, it suggests getting a CA-issued certificate. They suggested going to http://xmpp.net/. However, as of yesterday, XMPP stopped operations. They have a link to https://www.startssl.com/. However, Google Chrome says that StartSSL’s certificate is not valid. On top of that, the StartSSL pages just aren’t working for me right now. So, I’m running with my own self-signed certificate.
The next place where I ran into an issue was installing my certificate. The run-config.sh file has a line
CERTIFICATE_FILENAME_LIST=${WAVE_SERVER_DOMAIN_NAME}.cert
If figured I’d just name my cert file to be my domain name. The problem is that the procedure for making cert files uses an extension of .crt instead of .cert So, my first attempt to run failed. Fortunately, I found that fairly easily and now have the server up and running.
The Openfire administration pages shows the component session running. The message from the wave server looks like it is initializing okay, but it isn’t discovering any other wave servers. It is getting a message that
INFO: Couldn't find wave on initech-corp.com
initech-corp.com does not appear to have wave
So, it looks like I’ve got a wave server up and running, but not connected to anything or doing anything. If anyone else is experimenting with Google Wave Federation, let me know. In addition, if anyone wants to do other things with my XMPP server, I’ve set it up so that you need to contact me to set up a userid, but I’ll set up userids for people that want to experiment.
Later, I’ll reinstall statusnet. I had set up a very early instance of laconi.ca, the precursor to statusnet, but I haven’t done anything with it in ages. Now that I have a nice XMPP server, I will want to do an new installation and see how it runs with Openfire and ultimately, Google Wave Federation. If I really get motivated and have some free time, I’ll work on getting Asterisk up and running again. Also, still on the list is looking at Shindig and Drupal for Google Gadgets, but that is even further down the list.
So, are you doing anything interesting with Google Wave Federation? Got any cool ideas?
October 30th
Social Influence and Coffee
Submitted by Aldon Hynes on Fri, 10/30/2009 - 18:16Today, I received a review copy of Social Media Marketing For Dummies by Shiv Singh. I had heard Shiv speak at Digiday Social and was very impressed, so when his publisher asked if I wanted to review his book, I jumped at the chance.
I’m only part way into the book right now, so I will save my review for later. Instead, I’ll reflect a little bit on what he writes about and apply it to an event in my own life today. Several years ago, we got a Senseo Single-Serve Coffee Machine . It served us well over the years, but has not survived the hard water at our house all that well, even when we’ve tried decalcifying it.
The problems we’ve had with the Senseo is that it is out of stock, and the pods, which are sort of like coffee filled teabags are also sometimes hard to get. So, Kim wondered about getting a Bosch Tassimo Single-Serve Coffee Brewer , similar to what her folks have. To relate this back to Mr. Singh’s book, Kim’s parents acted as referent influencers on Kim’s initial considerations about buying a new coffee machine.
However, the coffee for those machines comes in little plastic cups that get thrown out; not very eco-friendly. We talked back and forth and decided that the Keurig Single-Cup Home-Brewing System might be more to our liking. Especially since you can get a Pod Holster
for the machine and a Reusable Coffee Filter
I read through some of the reviews on Amazon and they all sounded pretty good. However, I found that this is usually the case on Amazon. They only people who review are people that really like the product and a few that really hate the product. It is probably fair to consider these online referent influencers. However, I asked Kim to check some of the foodie sites she goes to and she reported a thumbs up from these sites. Referring back to Mr. Singh’s book, these would probably be considered by us the expert influencers, although, in my mind, the line between referent and expert influencers is a bit fuzzy online.
In his book, Mr. Singh references John R. P. French and Bertram Raven’s work on social power. People interested in Social Media Marketing for Academics, might want to start there.
All of this begs the question, how do you get a cadre of social influencers online to support your brand? I look forward to getting more of Mr. Singh’s views on this as I read more of his book.
Geek Day
Submitted by Aldon Hynes on Fri, 10/30/2009 - 12:15Yesterday, I spent much of my day playing with new technology in the pipeline as well as some older technology that is new to me, and it looks like today may be a bit similar. With that, I want to share some views from the geek trenches.
Google Wave
I am now on Goggle Wave. You can find me as aldon dot hynes at googlewave dot com. It is still in early testing and I’m only slowly finding friends and waves there. However, it does have a lot of interesting potential. I like the ability to have a real time discussion and still have it end up like an email thread. I like the idea of having these discussions publicly, like on a mailing list. I like the idea of being able to insert comments in the middle of someone else’s comment. I like the idea of having gadgets, and I’m glad they now have tags.
I’m disappointed that a good permissioning system isn’t in place yet, and I’ve been frustrated at times when a wave can get really slow with a lot of people on it.
On the usage side, I’m very interested in how people are using it for community gathering, political discussions and potentially journalism. It will be interesting to track usage and how usage changes, and I sure hope there are some good academics looking at this from various academic frameworks.
For the technology, I’m very interested in what can be done with gadgets, and I’m hoping to play with creating some gadgets. I’m interested in what can be done with gadgets and Shindig, and especially with the Drupal Shindig OpenSocial Integrator. Ultimately, I’m interested in how this will fit Open Handset, or Android.
Over on StatusNet there has been a discussion about Google Wave Federation and how Google Wave could interact with a system like StatusNet. This is another area I want to explore.
So, as an early adopter geek, I’m really enjoying Google Wave, and I’m hoping to explore it much more.
Sidewiki
On the Thursday #socialmaketing discussion, @ckieff asked people what they thought about Google Sidewiki. I hadn’t played with Sidewiki, but I’ve been hearing a lot of buzz about it, so I figured it would be a good idea to kick it around a little. My first comment was about my surprise that you couldn’t use it with Google Chrome. Several people shared this comment, and then @googlesidewiki responded that Google Sidewiki has just come to Chrome. Actually, it only partially came and there is more work to do, but it is a step in the right direction.
Over on Firefox, I installed the Sidewiki addon. So far, it doesn’t seem to do a lot that I couldn’t already do with StumbleUpon and a host of other sites. In addition, for some reason, I often have problems with Google authenticating from Firefox. It seems like Google frequently gags on my Firefox cookies. Maybe when Sidebar gets connected up to a Google Gadget and from their into a Google Wave, it will be more interesting. If you’re doing anything interesting with Google Sidewiki, let me know.
FlashBlock
One of the things that I’ve found more and more annoying are Flash programs in browsers. They suck up CPU time, play annoying music, and generally don’t add much of value. So, I was please to find FlashBlock as a Firefox addon yesterday. I’ve added it and things seem to run much more quickly and nicely without Flash. Unfortunately, there isn’t anything similar for Chrome.
This does present interesting new issues for advertisers using Flash.
Twitter Lists
I’ve also started playing with Twitter Lists. I’ve set up a few different lists and I look forward to when they can be accessible via the API. It seems like group functionality is one of the next great frontiers and I’m especially interested in how the group information can get shared. For example, some of the information that I entered in my Twitter Lists has already been entered in FriendFeed, PeopleBrowser, Facebook, and probably half a dozen other areas. Perhaps group information is something that needs to be added to Portable Contacts and sites like DandyID
Joomla
As I believe I’ve mentioned earlier, I’m starting to play a little bit with Joomla. Don’t worry, I’m not abandoning Drupal. Meanwhile, I’m listening to a friend on Facebook talk about a Drupal versus Wordpress smackdown going on at her office. You need to use whichever tool meets the needs.
Ubuntu 9.10
Next on the list is upgrading some machines to Ubuntu 9.10 I'm hearing good stuff, but just don't have the time yet.
Anyway, I keep getting distracted by a shiny new tool, so I’d probably better post this entry and get back to everything else that has piled up.
October 29th
Debunking Digital Publishing and Advertising Myths
Submitted by Aldon Hynes on Thu, 10/29/2009 - 09:24Many great speakers presented many great ideas at the Digital Publishing and Advertising Conference in New York City earlier this week. However, some of these great ideas are myths that need to be debunked.
Leading off the list was a comment by Walker Jacobs, Senior Vice President of New Media Ad Sales for Turner Entertainment. In the Keynote Panel, The Media Moguls Address the Digital Content and Advertising Economy, he suggested that all of their content is paid content, the question is who is paying for it. This exposes an interesting myth that takes several different forms about the divide between paid content and truly free, user generated content.
One form of the myth is that the only good content is content that people get paid for. Other forms include the idea that free content is some new creation of the digital world and that people who share free content, perhaps even making money off of this are bad.
So, let’s look at some forms of free content. I don’t know if there were many academics in the audience that have published significant articles in peer reviewed journals. If so, I suspect most of them did not get paid for their content. I know that I have not been paid for articles I’ve written for peer reviewed scholarly journals. Related to this are the chapters that people write for books edited by others. Many of these chapters add significantly to the literature on one topic area or another, but again, it is typically the editors that get paid, and not the people creating the content. Again, I’ve had chapters published in a couple different books which I’ve not been paid for.
There is nothing new about free content being produced, and sold at a profit by book companies, editors, and journal publishers, and much of this content is very high value content.
Of course, all of this is based on a fairly narrow view of what it means to be paid. Free content producers are often paid in social capital; strokes or accolades. I remember years ago, when a friend of mine who wrote for the Wall Street Journal had her first article front paged on that newspaper. She had a party to celebrate. Accolades for financially remunerated articles and for articles where there is no financial remuneration can be a significant payment. I still get a little thrill when one of my blog posts gets front paged on a site where I had not been expecting such an honor.
This takes me to “The Ultimate Digital Content Debate: Paid or Free?” The first piece of free content about the debate at DPAC was from @scanlon_pittPG who tweeted, “Wow what an ego! ‘the debate about pay walls and journalism over’ because of myself and my partner Steven Brill #dpac4” Brill went on to defend his statement with the assertion that “People are realizing that advertising alone cannot support news”.
While Mr. Brill might be ‘realizing’ this, or at least attempting to realize a profit based on this assumption, it is unsupportable in many ways. First, it is based on a “but we’ve never done it that way before” assumption. I’m assuming you know those corporate meetings. Some young creative innovator comes in with a great idea and one of the large old egos sitting at the end of the table says, “But we’ve never done it that way before”. If the innovator is smart and lucky, he leaves the meeting, finds an angel investor and sets up a company to champion the new disruptive approach, taking down the large old egos.
Yes, it is true that in the past, newspapers have had to rely on both subscriptions and advertising to cover their costs. As an example, based on a 2008 10-Q filed by the New York Times corporation, 60% of their revenue came from advertising and 40% came from subscriptions. However, with print versions, there are costs of raw material, printing and distribution that are much more expensive than the cost of web hosting today. In addition, much of the cost of newspapers include significant interest payments covering the debt servicing of leveraged buyouts from corporate consolidation that wracked the market years ago as well as the payout needed to investors and the large salaries demanded by top executives. For a good example of this, look at the bankruptcy filing of the Journal Register Corporation which included a $1.7 million dollar incentive pay plan for 31 officers and key employees.
One needs to question whether a properly set up online news organization without the cost of printing, excessive corporate debt, and excessive executive compensation really does need subscription revenues to survive. Early successes of some online only local news sites give reason to believe that the assumption must be questioned.
Ignoring for a moment the issue above about whether there can be quality free content, @scanlon_pittPG tweets that Steve Brill believes “You should pay for content because reporters have families.” @scanlon_pittPG goes on to observe, “Nice, but that is NOT a business model.” This gets to a very important point. Just because it might cost money to produce something, doesn’t mean that it is valuable or that people are willing to pay, a point that @bjornjeffery made on Twitter.
Yet the place where Brill seems to cling most firmly to outdated notions is the idea that the role of a paid editor to organize and make sense is required. As I walked home from the DPAC conference, I saw news scrolling freely on the side of a building. It announced a report that the Coast Guard’s exercise on September 11th did not violate any policies, but may not have been a good idea anyway. The reason it may not have been a good idea is perhaps because of these paid editors that Brill lauds that decided to run unsubstantiated stories on September 11th, the same editors that decided the Balloon Boy hoax was more newsworthy that the general strike in Puerto Rico on the same day. Just because it costs money to produce something, doesn’t mean what is produced is valuable.
Now this is not to say that there is not some valuable journalism being done where the journalists deserve to get paid. Here in Connecticut, Ted Mann of the New London Day has recently been doing investigative reporting into the Governor’s use of polling help from a UConn professor that is outstanding. This sort of reporting does provide a value; a value that people who care about our state should be willing to support in one way or another. National Public Radio has long produced reporting that people are willing to support and Spot.Us is creating a new model for funding investigative reporting that needs to be considered.
There is a lot of valuable content that is being produced. Some gets paid for, some doesn’t. We need to explore new models to make sure that the most valuable content does get properly paid for. We need to change models so that the overhead in producing and distributing valuable content gets reduced. More efficient ways of selling and purchasing online advertising and data could be a great help. Ways of making this available to smaller publishers would be a great help. However, setting up more paywalls, and supporting overhead like Mr. Brill may just be a step in the wrong direction.
(Originally published at DigidayDaily.)
October 28th







