Blogging Roller

Sun has given every possible indication that Open Solaris will be run as a true open source project. The latest indication is the make-up of the board of directors: Casper Dik, Roy Fielding, Al Hopper, Simon Phipps, and Rich Teer. (via Simon Phipps - congrats Simon!)

IBM Adopts Open Development Internally: "Following on the success of its Eclipse open-source development platform, IBM has quietly been using a form of open-source development internally to create technology the company will sell commercially.

IBM calls its model Community Source, which it defines as a collaborative, internal, open-source-style environment for developing and testing new technology.

Danny Sabbah, vice president of strategy and technology for the IBM Software Group, in Armonk, N.Y., said IBM is using its Community Source model across 100 projects and 2,000 developers in the company. These projects span the IBM Software Group, Systems Group, Research and Global Services, he said."

In case you missed it, Google has added satellite images to maps.google.com. They've got coverage for the entire US and low resolution images for the Caribbean and South America.

Over the weekend I finished writing Chapter 16: Distribute Files Podcast style. The code for the chapter is a simple JSP-based web application called FileCaster, which you can use to manage a Podcast newsfeed. You can add a new Podcast by uploading a file or by referencing a file somewhere on the web. The app was easy to write, thanks to Commmons File Upload and good old JSP. I need to review it once more tonight and submit it to Manning tomorrow.

I've got to admit, I've never really paid all that much attention to Podcasting. I've reviewed a couple of Podcast book proposals, I've implemented Podcast newsfeeds and I've got the required software, but I've just never gotten into it. To make up for this horrible deficiency in my geek credentials, I spent most of the weekend listening to Podcasts playing in the background as I worked. I couldn't make it through many of the Podcasts I downloaded because they were either too mumbly and under-produced or too slick and over-produced (like a drive-time radio show). I did find some things to like. Here are the subscriptions that I didn't delete from my iPodderX Lite subscription list:

• IT Conversions [blog|feed]: Some of these interviews and conference sessions are simply amazing, fascinating stuff.
• Evil Genius Chronicles [blog|feed]: Interesting pop-tech talk and some good tunes.
  
• Caribbean Free Radio [blog|feed]: Good music and an interesting Caribbean perspective (recommended by Evil Genius).
  
• The Chris Pirillo Show [blog|feed]: Interesting talk about selling content, organizing conferences, pop-tech, etc.
• Geek News Central [blog|feed]: Nice review of tech news and nice overall experience, but I didn't really learn anything new here.

Claire Giordano has written an excellent article on the business benefits of blogging, pulling together a bunch of nice quotes, links, and useful advice.

are the most believable ones (and I mean that in the nicest way possible).

Currently, a Roller user can ask Roller to "ping" weblogs.com when a new post is made. That was a nice feature back in the day, but in modern times there are many different ping targets. Some users might prefer to ping Javablogs.com, while others might prefer to ping Technorati. Some might even want to configure a different ping target for each weblog category.

Anil Gangolli has written a proposal for configurable pings in Roller and has asked for comments. If you're interested in joining the discussion you can join the Roller dev mailing list, comment in the wiki (write me for a login), or if you're really lazy leave a comment here.

We deployed a beta of Roller 1.1 at blogs.sun.com today, featuring the new built-in "Planet Roller" blogs.sun.com aggregator, which includes both BSC hosted and externally hosted Sun bloggers.

On the planet page, we substitute Technorati rankings (via Technorati.java) for the normal hot-blogs listing because we don't track hit counts for external blogs.

And, if you're looking at the Technorati rankings and wondering what a bush-league blogger like me is doing up there with MaryMary and Tim Bray? Well, I'm sorry to say, it's all due to bogus whuffie. At the time, I actually wrote to the Technorati folks to point out to them that my place in the Technorati 100 was undeserved. They have since rejiggered their formulas (sorry to get so technical) and I'm out of the list, but still highly ranked.

I meant to blog about this earlier, but last week was a very busy one. Simon Phipps spent a day working with the command-line version of Planet Roller last weekend and built something very cool: an aggregated blog (at www.webmink.net) that brings together posts from all of his sites: his Sun blog, personal blog, Java.Net blog, del.icio.us links, and Flickr photo stream -- all aggregateed together to form one blog with it's own feed. Subscribed!

Even when you're doing work you love. In other news: today I finally finished Chapter 14: Blog Your Build Process and Chapter 15: Blog From a Chat Room.

Above we see the blogbot entering a chatroom. Snoopdavey greets the bot and the bot, being quite professional and not sassy in the least, responds with "Unknown command." Next, snoopdavey instructs the bot to post a blog message titled "Hello World" with some content in JSPWiki format. Blogbot converts the content to HTML, posts it to a blog server, and responds with the new post ID. Next, snoopdavey decides to unblog the message he just blogged and the bot complies.

Pretty cool and all, but probably useful only to the terminally geeky.

Alexis Moussine-Pouchkine: In his "FeedReader" tutorial (version française ici), Rich Unger explains how to develop an RSS reader on top of the NetBeans platform (using P@'s and friends' Rome Atom/RSS tools).

It's sad to see Danese leave Sun. I talked to her at length before I joined Sun and she's given me lots of good advice along they way. She helped to convinced me that taking Roller to Apache is the best way to ensure the continued success and long-term viability of Roller as an independent open source project and helped me find my way to ApacheCon 2004 so that I could learn about the Apache way. I started at Sun about five months ago, I work remotely and I never really got to know Danese all that well, so it's very interesting to read the farewell messages and retrospectives on the important work she has done at Sun (e.g by Jim Grizanzio, Claire Giordano, and Danese herself). Thanks Danese and good luck at Intel and with your new blog.

Is of course, coming back. We just got back from a short trip down to Stone Mountain, Georgia for a family re-union centered about my grandmother's 90th birthday celebrations. It was great spending time with the small family. When we gather together for a re-union, it almost feels like a big family. The kids loved the Stone Mountain train, the cable-car up to the top, and jumping from rock to rock all the way down. I wasn't expecting much from "redneck Rushmore" but the Marriot resort there is very nice as is the rest of the park. I might actually go back there someday.

Anyhow, upon my return I find that my blog has been down all day (due to a misconfiguration on my part) and, as usual, one of my office computers is freaking out. When I leave town, I power down the computers in my office (a Sun box, a low-end Linux box, and a Windows machine. No matter what, everytime I return, one computer won't start up correctly. This time it is the Windows box. The symptom: on start-up it boots Redhat Linux and I didn't even know I had Redhat on that box. Oh well.

In other news... I was supposed to finish Blogs, Wikis, and Feeds chapters 14 and 15 this weekend (you can probably guess how that went). Back to work...

Lars Trieloff has created a very nice bookmarklet that makes it easy to add bookmarks to Roller's bookmark manager. I've tried it and works. It and captures both the URL of the page you are currently on, as well as the newsfeed URL (if the page includes the autodiscovery tags). Read more here: Roller bookmark bookmarklet

Dare Obasanjo posted some criticism of the Atom Format and Protocol today, based on Ben Hammersley's talk at E-Tech. I'm not at the conference; I picked up on them from James Robertson, who also shares a rather cynical view of Atom. Dare's a smart guy and I really wish he was participating in the Atom Format and Protocol discussions, but he dropped out and seems to have lost touch with Atom. Even a dumb guy like me can see that. So, I'll try to correct some of his (and James') misconceptions about Atom.

Dare Obasanjo: During the break, Marc Canter and I talked about the fact that both the Atom syndication format and Atom publishing protocol are simply not rich enough to support existing blogging tools let alone future advances in blogging technologies. For example, in MSN Spaces we already have data types such as music lists and photo albums which don't fit in the traditional blog entry syndication paradigm that Atom is based upon. More importantly it is unclear how one would even extend to do this in an acceptable way.

I'm not sure I understand what the specific problem is here, but if there is a problem then that problem is shared by all of the nine ten incompatible forms of RSS as well. They all use essentially the same model and offer the same mechanisms for extensibility (allowing the addition of new elements at feed or entry level). Atom is not supposed to be a new invention, it is a rigorously specified standardization of existing technologies -- carefully designed to ensure inter-op.

Dare Obasanjo: Similar issues exist with the API. The API already has less functionality existing APIs such as the MetaWeblog API. It is unclear how one would perform the basic act of querying one's blog for a list of categories to populate the drop down list used by a rich client which is a commonly used feature by such tools.

Atom's support for categories is unclear, of course, because that section of the specification has not been written yet. When Atom Protocol is complete, it will offer more, not less, functionality than the MetaWeblog API. Contrast that with the MetaWeblog API, where the spec is "complete" yet categories are still unclear.

James Robertson: Sounds like events moved past the Atom API while it was being put together. Podcasting walked right past it. Their idea seems to be that you should just slap Base64 encoded content into the feed - uh huh - I sure want to download a 10MB screencast (more like 15 after being encoded, but never mind) each time I look at the feed. Looks like the Atom API is worth ignoring as something to actually use...

No, that is not the Atom Protocol solution for Podcasting. To reference a Podcast from an Atom entry, you'd use the Atom Link element.

Come on back guys. Read the latest drafts and get back into the process. We need you and we all need Atom.

Congratulations! You have been chosen to speak at the upcoming 2005 JavaOne(sm) Conference in San Francisco, California, June 27-30, 2005. Your submission entitled' Beyond Blogging: Feed Syndication and Publishing with Java ' has been accepted by the JavaOne Conference Program Committee.

The session proposal that Alejandro Abdelur, Kevin Burton, Patrick Chanezon, and I put together has been accepted for JavaOne 2005.

Josh Staiger: Join us tomorrow (Tuesday March 15) for the third weekly Raleigh/Cary Bloggers meetup.

What: An open meeting to talk about blogging, podcasting & whatever's on your mind
When: Tues @ 6:30 p.m.
Where: Cafe Cyclo, in Cameron Village

2020 Cameron St
Raleigh, NC 27605 (map)
(919) 829-3773

Who: Bloggers & people who want to blog

Optionally RSVP at the Meetup.com Raleigh/Cary Bloggers Meetup page.

Amongst other things we will talk about this week is the frequency of future meetups.

See notes on our last meetup.

Hope to see you there!

A device for keeping your eyeglasses from sliding down your nose. Hmm... where have I seen that before?

For Chapter 15 of Blogs, Wikis, and Feeds In Action I'm combining my Relay-IRC Chatengine, the JSPWiki rendering engine, and the BlogClient library together to create a wiki blogging chatbot, but why? Why would anybody want to use wiki syntax (or any other syntax) to blog from an IRC chatroom? Are chatbots just pure geek-fun or is there an actual business case for them? Anybody using chatbots, or more specifically blog-chatbots for in a business application?