Blogging Roller

• Microsoft's Dare Obasanjo on IE 7 beta 2 preview release
  "The general RSS reading experience is rather unsatisfactory"
• Microsoft ships RSS platform
  Winer: "I have not installed the software myself [...] Even so, I am confident that they have done an excellent job of supporting RSS"
• Proposal - Atom Admin Publishing Protocol
  Atom-like protocol for administering blogs, users and group blogs
• NeoOffice tipping point
  "consider tossing a few notes in Patrick's tip-jar"
• Google Toolbar API
  Niall hacks the Google Toolbar
• Project app-hosting on Java.net
  A demo server for Java.Net projects... bravo!

Ian Kallen: Google is pretty much operating an open-relay, the blogosphere's equivalent of an SMTP spam-mill, because they lack the imagination to watch their own numbers and their spam rolls out unabated. This has been a ballooning problem for at least a year and a half. It's actually kinda inconvenient. Don't think they haven't been advised. Long before maverick-man coined the term "splog" I'd been sending my friends at the big G data on the extent of their problems. They know.

Roders Caldenhead: A new era begins today for the RSS Advisory Board, an independent organization formed in 2003 that publishes the Really Simple Syndication (RSS) specification, helps developers create RSS applications and broadens public understanding of the format.

The board is taking on eight new members: Meg Hourihan, Loïc Le Meur, Eric Lunt, Ross Mayfield, Jenny Levine, Randy Charles Morin, Greg Reinacker and Dave Sifry. I'm serving as chairman this year unless they kick me to the curb.

And they released draft 1 of a new RSS specification that looks like... gasp... a real specification! They're are not trying to change RSS. They're just trying to create a better spec and that's a good thing.

Ted Husted: The OpenOffice suite provides an interesting opportunity for open source products. Since the suite is free, open source, and multiplatform, using this tool with our projects is little different than using Subversion or Ant.

Problem is, the format is not change-log friendly. By design, all changes made to a ASF product are logged to one of the mailing lists, where they become part of our "communal memory". When a change is made to an OpenOffice document and checked into the repository, it is logged as a change to a binary file. No one watching the project knows what changed unless they spend several minutes opening the document and reviewing the internal change log.

Albeit, Roller is deliberating whether to use the OpenOffice to maintain it's user documentation. The vote is pending now. Since OpenOffice can save to multiple formats, my suggestion is that we also checkin a companion HTML document, so that everyone can see what changes in real time. We'd contnue to edit the ODF file, and just Save As to HTML before checking in both files. Film at 11.

Yes that's right: Ted Husted (of Struts fame) is blogging!

Plus, Ted has been participating on the Roller dev list and most recently raising some issues with the use of OpenOffice.org for our user and install guides. As you can see, Ted's biggest concern is that, when docs are in a binary format it is hard to monitor doc changes by watching the diffs in SVN commit notifications. Does your open source project use OpenOffice.org and if so, how did you deal with this issue?

Rocked the Cradle.



So go see 'em if you get a chance. Actually, Son Volt played at Raleigh's Disco Rodeo (good god I hate that name), but  it was a Cats Cradle production. NC's own Tift Merritt opened and put on quite a good show too.

Matt Raible: It's interesting to hear someone from Sun say that "JDO is dead". What are your thoughts? Should Roller change their persistence backend just to satisfy Apache?

I've been following the new bubble hype-masters and it seems to me that, for a lot of folks, Web 2.0 is about enabling companies and investors to profit off of user created content, simpler programming models so that built-to-flip companies can throw up new sites fast and slicky-slick AJAX interfaces to lure investors and unsuspecting users to the shiny new services.

So Henri, you're right, Web 2.0 is not about having cool software to install on your own personal web server, it's about getting locked into services provided by and trusting your data to Web sites that you do not control. It doesn't have to be that way, of course, and perhaps I'm exaggerating a bit just for fun (and hits). But maybe we need a list of "the best open source Web 2.0 software" -- and it would include things like open source blog servers, wiki servers, photo galleries, content management systems, social bookmarking clones and etc. Where would you start looking for such software? Google for "open source web 2.0" and where do you end up? Lucky you Henri, you're already there.

Chris Adamson: Pixel-accuracy has led to a miserable fool's errand for both of the major Java GUI toolkits, Swing and SWT. Being rather coupled to the native Windows API's, SWT should enjoy a huge advantage here, but its obsessive fidelity to Windows' GUI shows no signs of ushering in a new era of Java Desktop development... just a lot of very loud evangelists, and complaints about its performance on other platforms. Swing, despite its bulk and sometimes ruefully ornate design, is probably better suited for the long run, because its goals are ultimately based in functionality, not appearance.

Rich Sharples has written up some of his thoughts about group blogging and Roller's relatively new group blogging features -- including some good RFE ideas.

O'Reilly's new rough cuts service gives you access to books as they are being written. I especially like the fact that they "welcome feedback on both the book's content and the service itself." It's definitely a good idea to get the readers into the development process. Kinda reminds me of the SourceBeat concept. Via Joho the Blog.

We're getting ready to release Roller 2.1 (incubating). This release turned out to be a pretty big one. We focused on performance, comment management and spam prevention, but we also added a host of other small changes, fixes and improvements. Here's a summary of the most significant changes: What's new in Roller 2.1.

I've been exchanging some very productive emails this weekend with Jed Brown, who's working on the new Performcaning for Firefox release. We've finally got Roller and Performancing playing together nicely and I think we're going to have some good news for Roller and Performancing users coming in our next releases, that's Roller 2.1 and Performancing for Firefox 1.1 both coming soon to a browser near you.

Project Teepee. Great news. I love the new CVS client in Netbeans 5 and now Subversion is getting the same treatment, possibly as soon as May. Via Roumen.

I'm very happy to announce that my proposal Java and REST: Implementing the Atom protocol was accepted for JavaOne 2006. Tags: topic:[JavaOne 2006], topic:[Atom protocol]

So says Nick Wilson. Any chance that MetaWeblog API will working in the new release? (sorry, I'm not registering just to leave a comment). Tags: topic:[Performancing], topic:[MetaWeblog API]

John Clingan: More group blogs are popping up on blogs.sun.com.

We're using Mark James' wonderful Silk icons in Roller now. You can get all 700 of them at Mark's site FamFamFam.com and he's got pretty much everything covered whether you're developing an IDE, an ecommerce site or a blog server. The price is $0.00 and the license is Creative Commons (Attribution).  Here's a small sample:



We were using the old Java Look and Feel Repository icons in Roller. You know, they're really in need of an update. Maybe the icons from Open Office and the various Netbeans incarnations could be bundled into a nice Java icon pack.

A couple of bloggers at BSC have asked about Performancing for Firefox, so I've been trying to get it working with Roller. Since Roller is not on Performancing's  list of Blogging Service options, I tried "custom blog," specified "MetaWeblog API," and entered my blog server settings. No dice. So, I turned on Roller debugging and watched what Performancing is sending over the wire. Here's what Performancing sends for the MetaWeblog API newPost() method:

<?xml version="1.0"?>
<methodCall>
  <methodName>metaWeblog.newPost</methodName>
  <params>
    <param> <value> <string>t</string> </value> </param>
    <param> <value> <string>t</string> </value> </param>
    <param> <value> <string>p</string> </value> </param>
    <param> <value> <string>:</string> </value> </param>
    <param> <value> <string>/</string> </value> </param>
    <param> <value> <string>/</string> </value> </param>
    <param> <value> <string>l</string> </value> </param>
    <param> <value> <string>o</string> </value> </param>
    <param> <value> <string>c</string> </value> </param>
    <param> <value> <string>a</string> </value> </param>
    <param> <value> <string>l</string> </value> </param>
    <param> <value> <string>h</string> </value> </param>
    <param> <value> <string>o</string> </value> </param>
    <param> <value> <string>s</string> </value> </param>
    <param> <value> <string>t</string> </value> </param>
    <param> <value> <string>:</string> </value> </param>
    <param> <value> <string>8</string> </value> </param>
    <param> <value> <string>0</string> </value> </param>
    <param> <value> <string>8</string> </value> </param>
    <param> <value> <string>0</string> </value> </param>
    <param> <value> <string>/</string> </value> </param>
    <param> <value> <string>r</string> </value> </param>
    <param> <value> <string>o</string> </value> </param>
    <param> <value> <string>l</string> </value> </param>
    <param> <value> <string>l</string> </value> </param>
    <param> <value> <string>e</string> </value> </param>
    <param> <value> <string>r</string> </value> </param>
    <param> <value> <string>/</string> </value> </param>
    <param> <value> <string>x</string> </value> </param>
    <param> <value> <string>m</string> </value> </param>
    <param> <value> <string>l</string> </value> </param>
    <param> <value> <string>r</string> </value> </param>
    <param> <value> <string>p</string> </value> </param>
    <param> <value> <string>c</string> </value> </param>
  </params>
</methodCall>

My post had the title "test" and the content "test" and as you can see, neither of those words appears in the XML above. Obviously something is horribly broken. I've offered to help the Peformancing folks get Roller support going, but I'm not sure I can really do anything to help. Looks like the problem is on their side.

Assuming that Performancing's other options (Blogger.com, WordPress.com, TypePad.com, LiveJournal.com and MSN Spaces) work fine, this is pretty depressing. Plain old unadorned MetaWeblog API is the one that's broken, the one that's tested last and the one nobody cares about. Is that really true? Did the Performancing developers really have to write custom code for each and every blog server?

Tags: topic:[Performancing], topic:[MetaWeblog API], topic:[XML-RPC], topic:[Blog client]

I spent Saturday rewriting the second half of my search/monitoring chapter and then Sunday and Monday I spent updating my C# examples. I also built, ran and tested all of the example code for the book and committed a number of fixes to the Blogapps CVS. And finally, I sent the whole Part II bundle back to Manning for typesetting and proofreading. The end is near!

I had been using Visual Studio C# 2003 but since that's almost three years old now, I moved everything over to Visual Studio 2005 and .Net 2.0. Moving to .Net 2.0 was pretty easy. I downloaded Visual Studio C# 2005 Express Edition and that worked fine for almost all of my examples, but not for my one ASP.NET example. I had to download Visual Web Designer 2005 Express Edition to get that working. Both of those IDEs are free at least for the next year. You can get them here: http://msdn.microsoft.com/vstudio/express. Nobody pays for IDEs anymore, not even Micro$ofties.

Lots of good feedback on my blog search service comparison. Everybody except Google and Yahoo wrote in with corrections and additions and in response I've updated the matrix.