Archive for August, 2007

Monitoring the buzz

Published by Stuart Yeates on August 29, 2007 in Discussion and Community. 1 Comment

At OSS Watch we like to know who how people react to our events, our documents and our blog. Sometimes this feedback comes to us directly, via our personal email accounts, our catch-all info@oss-watch.ac.uk account, our mailing lists or even by phone.

Sometimes, people take about us on their own mailing lists, blogs and websites. Initially we kept track of such feedback by checking inbound links reported by Google. This, however, has become less and less reliable, partly because Google now have separate indexes for much of the web that doesn’t show up in most searches. The total number of inbound links to the OSS Watch website, as reported by Google, seems to be static, even over periods when we know that new links were added, so this number appears unreliable.

Currently we keep track of such feedback using Google Alerts These regularly run a given search across a number of the Google indexes and email you of any new hits. Any Google query string can be used. I have alerts, for example, for “Stuart Yeates” and “OSS Watch,” but they can get very complex. I recommend using as tight a query as possible over as wide a range of media as possible.

Personally I find Google Alerts strangely reminiscent of alerting services in the bibliographic world, which have been running customised queries against newly published journal articles for at least ten years.

As mentioned recently on the mailing list, we’re also currently playing with Google Analytics.

So how do other people monitor the buzz they’re creating?

Review of Sun’s weblog publisher

Published by Stuart Yeates on August 24, 2007 in Software. 1 Comment

Sometime ago, I took a look at StarOffice, the commercial edition of OpenOffice. An educational version of StarOffice is available as a free (if large) download. I found that while it integrated nicely into my Ubuntu desktop, it didn’t have my personal holy grail, a grammar checker. (I’m still looking for a text editor or word processor with a grammar checker that runs on Ubuntu without emulation/virtualisation).

Presumably as a result of this, I was contacted by the marketing people at Sun, inviting me to try their new blogging plugin. Being the slow learner that I am, I said yes.

Before I knew it, I had a 1.3 MB .oxt file in my inbox and shortly afterwards a CDR in the post. The first thing I did was to install the version on the CDR and check to see whether they’d got a grammar checker (they hadn’t). The second thing I did was click on the little icon inviting me to download the latest patch. The “patch” turns out to be 369.0M and I couldn’t download it anyway, since I don’t believe I’ve got a login to “SunSolve” which is apparently necessary. I’m a little bewildered as to why they shipped me review copy of an out of date base system, particularly since it was a CDR rather than a mass-produced CD.

Installing the “sun weblog publisher” turns out to be an exercise in frustration.

  • All the dialog boxes pop up behind the current windows rather than in front of them
  • The install failed saying it can’t deploy a java launcher
  • Having worked out that it can’t find the version of java installed on the system and that it doesn’t look for $JAVA_HOME like every other piece of Sun software I know of, merely browsing to the appropriate tab in the options dialog seems to make Java work.
  • The extension installs into my home directory, rather than into a system-wide location.

Having set the sun weblog publisher, it seems to do exactly what it says on the tin.

I’m sure I’d find it more exciting (not to mention useful) if I used an office application all day and had content already in office formats that I what to publish, but I don’t. My postings on the OSS Watch blog are mainly spur-of-the-moment reflections and musings, and the shortest distance between my brain and the blog is still via Emacs rather than an office suite. My postings to my photography blog are largely built around posting photos, and I store my photos on flickr, so I post blogs using flickr’s built-in blogging tool.

I’m guessing those people already using OOo, StarOffice or StarSuite and with volumes of content they want publish will find in practical enough, and the integration with the rest of the suite seems pretty smooth (though I’m not an advanced user of the suite, so maybe there are kludges I’m not noticing).

Weblog publisher isn’t open source software, it’s a propriety (10 USD) plug-in that Sun have created that runs in both open source (OpenOffice) and propriety (StarOffice, StarSuite) frameworks across multiple platforms. I think it’s a great idea for a business model and a great way for Sun to leverage the huge install base of OpenOffice. I just can’t see myself using the software day-to-day.

Another review of the software can be found at Linix.com.

Mailing lists and RSS feeds

Published by Stuart Yeates on August 23, 2007 in Discussion and Community. 0 Comments

Recently there was a discussion on the managing community led projects mailing list about whether an RSS feed of the mailing list should be added to the OSS Watch planet. A number of issues came out of this that drove home to me:

  • People have access information in very different ways (don’t assume how readers are going to access your content)
  • People subscribe to an information source for a particular set of reasons, and if source changes and no longer interests them (or they no longer have time to consume the source), they unsubscribe (be consistent)
  • Information sources have different capabilities for reuse and repurposing (don’t assume that content on third party systems is readily repurposable)
  • People like to see information exactly once (don’t SPAM people)

My original suggestion, went down in flames, but I learnt from it.

I’ve just heard that the next upgrade of JISCMail “before Christmas this year” will have better RSS support.

Communities can’t flourish in walled gardens

Published by Ross Gardler on August 21, 2007 in Information Environment, Software, Standards and Community. 3 Comments

I recently posted on the dangers of using closed Social Networking sites to develop community. Since writing that post I have entered into numerous discussion about my position. These discussions have taken place in both “walled garden” tools as well as within open communities such as mailing lists, direct emails and the blogsphere. In that time I have listened to, and learnt from, many different views and I’ve started to come to the conclusion that, well, I’m only partially right to when I say:

I predict only one or two of the current Social Networking sites will survive, and they will be the ones that share their network data first.

Randy Metcalfe pointed me at a BBC story in which Michael Geist agrees with my “sharing data” point:

 The better approach - for users and the sites themselves - would be to work towards a world of interoperable social networking.

However, Micheal disagrees, like many others, with my claim that only one or two Social Networking sites will survive:

Some services may believe that it is in their economic interest to stick to a walled garden approach; however, given the global divisions within the social networking world, the mix of language, user preferences, and network effects, it is unlikely that one or two services will capture the global marketplace.

I agree with Michael, and others. There is lots of room out there for niche players. What we need is open standards for creating interoperable networks.

I recently started reviewing open source social networking tools and the standards they adopt. This will be published as an OSS Watch briefing paper sometime in the next couple of months. Please let me know, via your comments, of any social networking/news/bookmark tools you think I should look at.

IBM learns from open source

Published by Ross Gardler on August 20, 2007 in Discussion, Strategy and Policy and Community. 0 Comments

I’ve always claimed that I have learnt far more from my involvement with open source than through many years of education, employed work, contracting and continuing education. Open source exposes me to people from a much wider domain and skill set than any other aspect of my life.

In a recent interview Bob Sutor, IBM’s Vice President of Open Source and Standards made a similar observation about IBM and their involvement with open source:

It taught us how to better collaborate with others who don’t work for IBM; it demonstrated that business models can evolve; it showed us that a good intellectual property strategy balances both “open” and “closed”; and it taught us that software that grew up in a non-corporate setting can be excellent, wildly successful, and meet customer needs. Linux, along with other open source software and open standards, showed us that being flexible in our thinking and business models is lower risk than adamantly clinging to past practices that might have worked once but now aren’t solely what customers really need.

So, it’s not just individuals like you and I that can learn from open source, it’s also huge organisations like IBM.