Curiouser and Curiouser!

Curiouser and Curiouser! on radio http://matt.blogs.it/ RSS feed for topic radio Copyright 2006 Matt Mower Squib/0.1 self@mattmower.com self@mattmower.com en-gb Musing aloud about how to join two users together Sun, 16 Jun 2002 14:17:20 +0100 You are using Radio, blogging a posting from the Radio desktop website.  You post and as the page reloads BlogPlex floats a note on the screen telling you that there are three new people for you to explore.  You click in the note and are presented with an aggregation of some of their recent postings (with various filters and options for the advanced user) related to the interests in your profile. You decide to subscribe to one of them via the button provided.  One of the others you widen the filter to see everything they have written recently and decide that the signal/noise ratio is probably not high enough for you. In the background the BlogPlex service has summarized your new posting and used it to update your profile.  Maybe this will lead to new people being added to your watch list, or to you being added to someone else's list.   http://matt.blogs.it/entries/00000017.html Radio vs. Email Sat, 22 Jun 2002 01:32:39 +0100 <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> <A href="http://radio.weblogs.com/0100887/2002/06/18.html#a310">Steve Yost on ubiquitous collaboration tools</A>. Steve Yost, inventor and proprietor of <A href="http://www.quicktopic.com/">QuickTopic</A>, disagrees with <A href="http://www.hyperorg.com/blogger/archive/2002_06_01_archive.html#85176225">David Weinberger's</A> assertion that collaborative software fails to thrive because companies are afraid to "hyperlink the hierarchy." The real problem is more mundane, Steve says: ... [<A href="http://radio.weblogs.com/0100887/">Jon's Radio</A>]</BLOCKQUOTE> An intriguing hypothesis on the challenges of getting new technology ideas to take root in organizations [<A href="http://www.kellogg.nwu.edu/faculty/mcgee/htm/blog/">McGee's Musings</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » Interesting.  From a quick scan I'm not sure how QuickTopic differs from, say, using a Yahoo group where participants can either use it as a list (with single & digest options) or a web forum.   I have switched from reading radio-dev as a list to using it as a web forum.  But that's because I use Outlook for most of my email and it just sucks, sucks, sucks! But beyond that I would be looking for radio-dev to arrive as an RSS feed.  In fact much of the stuff I currently receive as email would be better arriving in my Radio news aggregator.  Of course, at that point my aggregator is going to have to become a lot smarter and work much harder for me. The other technology that could do for email is Instant Outlining (IO).  Radio is getting interesting in this regard.  It's going to take a lot of work to make it a killer app, but it's certainly on that track. As someone else has said, Radio: the best $40 I ever spent on software!  </BLOCKQUOTE> http://matt.blogs.it/entries/00000032.html liveTopics progress Mon, 24 Jun 2002 23:33:08 +0100 Once again I'm neglecting all other interests in the rush to make liveTopics a workable tool for Radio listeners.  I'm learning what a struggle it is to debug problems on an unfamiliar OS over a 4,000 mile and 5 hours gap.  Maybe I should just write better code, it'll be easier in the long run! The good news is that two people now seem able to use liveTopics and hopefully by the end of play tomorrow there will be four.  If we don't find any more serious bugs then I am on schedule to release version 1.0 at the end of the week.   http://matt.blogs.it/entries/00000037.html Foghorn Longhorn Tue, 25 Jun 2002 11:55:48 +0100 <A href="http://www.theregister.co.uk/content/4/25878.html">Windows Longhorn slips again, becomes megaproject</A>. Gates holds forth on Microsoft's next Big One [<A href="http://www.theregister.co.uk">The Register</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » I especially enjoyed the section: Gates's list of what's planned for Longhorn is largely user's eye view, classic eye-candy of the sort that gets bolted on to the company's interim releases, but given that we're currently talking about a major overhaul, these ought to be more integral to the finished product than has often been the case in the past. Gates alludes to the database angle by asking of current operating systems: "Why are my document files stored one way, my contacts another way, and my e-mail and instant-messaging buddy list still another, and why aren't they related to my calendar or to one another and easy to search en masse?" Sounds just like Radio.  I think UserLand should sue!  </BLOCKQUOTE> http://matt.blogs.it/entries/00000042.html TrackBack for Radio Sun, 30 Jun 2002 19:54:33 +0100 I'm really excited by the <A href="http://www.movabletype.org/docs/mtmanual_trackback.html">TrackBack</A> feature that has been implemented for MovableType.  I'd really love to see something similar for Radio. My first thoughts were to implement an RCS extension that uses XML-RPC calls.  The idea is that whenever you post to your blog, an upstream callback extracts the links from the text and sends them to the RCS server.  RCS would then keep a per-site record of all such pings.  Another XML-RPC call would allow Radio to get information about all pings to any pages relating to your site.  Working in the same ways as referrer information does now. The one major disadvantage I see to this approach is that it isn't how MT's system works.   http://matt.blogs.it/entries/00000069.html This post is part of the liveTopics demo. Tue, 02 Jul 2002 19:30:38 +0100 Radio is a really cool application.  You can do lots of interesting things with it.  Like publishing information in RSS format. http://matt.blogs.it/entries/00000085.html TrackBack for Radio Sat, 06 Jul 2002 00:19:26 +0100 <A href="http://davidwatson.org:8086/archives/000386.html">David Watson</A> has Radio talking to Movable Type's TrackBack feature. [<A href="http://www.scripting.com/">Scripting News</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » Cool.  I'll be looking to play with this ASAP.  </BLOCKQUOTE> http://matt.blogs.it/entries/00000098.html KMpings & Trackback Tue, 09 Jul 2002 20:12:09 +0100 <A href="http://www.highcontext.com/blarchive/2002_07_09.html">The KMpings Experiment</A>. I created a little blog called <A href="http://www.highcontext.com/kmpings/">KMpings</A> that allows any blogger writing about knowledge management to ping their post to a tracking page (if their software supports it). Think of it as a themed <A href="http://www.weblogs.com">www.weblogs.com</A> for the knowledge management community. I wanted to try out this experiment since I think the <A href="http://www.movabletype.org/trackback/">TrackBack function created by Movable Type</A> has a lot of potential for aggregating blog posts within communities of practice on the web or an intranet. Please post any feed back you have to this message or shoot me an <A href="mailto:david@highcontext.com">e-mail</A>. [<A href="http://www.highcontext.com/">High Context</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » I really want TrackBack for Radio. And KMpings sounds like a great idea.</BLOCKQUOTE>   http://matt.blogs.it/entries/00000130.html Spelling checker for IE (and Radio!) Wed, 10 Jul 2002 19:13:59 +0100 <A href="http://www.theshiftedlibrarian.com/2002/07/09.html#a2614">To Answer Kate Z's Question!</A>. <A href="http://www.iespell.com">ieSpell - F**kin A Man</A>. <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> "I urge any serious bloggers out there who haven't tried ieSpell yet to go over to <A href="http://www.iespell.com">www.iespell.com</A> and do so.  Immediately!  The newest version now provides support for rich edit tools (like Radio's for instance) as well as AOL and other IE based browsers.  Worthy of a micro-donation for sure. I'm going to suggest the possibility of user-defined short-cuts (or smart tags) to the author.  It seems like the perfect tool to provide all those cross-system, bloggers like myself a way to maintain a central list of shortcuts." [<A href="http://radio.weblogs.com/0100364/">...useless miscellany</A>]</BLOCKQUOTE> [<A href="http://www.theshiftedlibrarian.com/">The Shifted Librarian</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » Well spotted Jenny, downloading now....</BLOCKQUOTE> http://matt.blogs.it/entries/00000148.html Trackback for Radio inches closer Wed, 10 Jul 2002 22:08:05 +0100 <A href="http://www.highcontext.com/blarchive/2002_07_10.html">Radio TB Ping Development</A>. <A href="http://home.netcom.com/~luskr/weblog/radio/categories/kLogs/2002/07/10.html">Ron Lusk is working on a script for setting up TB pings in Radio:</A> <BLOCKQUOTE> I've posted a message to the <A href="http://radio.userland.com/discuss/msgReader$16648?mode=topic&y=2002&m=7&d=10">Radio Userland discussion group</A> asking where to hook into the system so I can ping KMPings once for each entry. </BLOCKQUOTE> [<A href="http://www.highcontext.com/">High Context</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » Fantastic news.</BLOCKQUOTE> http://matt.blogs.it/entries/00000163.html Why Radio? Mon, 15 Jul 2002 22:47:25 +0100 Why have I choosen Radio over MovableType?  It's a question I've asked myself recently. I think MT looks like an excellent blogging system.  In a few years time I think that MT (or son-of-MT) is likely to be the choice for bloggers who need a little more than Blogger (or son-of-Blogger) will provide.  I don't believe, as much as I love it, that Radio will be that choice. However I do believe that Radio could be the klogger tool of choice.  Why? Because Radio has such potential in both a networked (social) and standalone (personal) context.  Because Radio is a general computing platform that has been specialized to handle blogging but could also be specialized for a thousand other applications. I, along with others, are looking to take it to the next stage with k-log ready tools.  Userland are doing their part with things like Instant Outlining and RCS. So, that's why Radio.   http://matt.blogs.it/entries/00000214.html Why Radio? Tue, 16 Jul 2002 22:13:04 +0100 <A href="http://radio.weblogs.com/0108194/2002/07/16.html#a132">Hmm... Ya think?</A>. See I think that all the cute little features that are being added to Radio are going to "in a few years time" bog it down terribly. However MT is a wad of perl code, which means it can be a point of departure for whatever anyone wants to do with it, without having to learn a proprietary "less than impeccably documented" scripting language (grumble grumble.) However, initial ramp-up with radio can't be beat (to say nothing of the 40m of cloud space), so here I am. I like radio alot, but I have to say the more compelling reason to use it over other blogging software is ease of setup & the year of service that comes with using radio. As a developer, I've given up on trying to do really neat things with radio. It's just too hard to track down the documentation. [<A href="http://radio.weblogs.com/0108194/">The Universal Church Of Cosmic Uncertainty</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » I think there are two different issues at play here: language and platform. Is MT a better blogging system than Radio because it is a wad of Perl code rather than a wad of Usertalk code?  I'm not sure but I guess so if only because of Perls popularity.  Neither would be my first choice for writing a complex application but both are adequate to the task. But the important issue for me is the platform.  I would class MT as an application and Radio (along with Frontier) as a platform.  In particular an ideal platform for delivering groupware applications.  Here's an example from my experience:  About 5 years ago att UNL we were looking for a learning mangament system for lecturers to use to construct & deliver on-line courses.  A question which stymied most vendors we spoke to was: <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> "How do you handle a lecturer who wants to update his module whilst on a cycling holiday in the south of france?"</BLOCKQUOTE> For the most part they had no answer to the disconnected scenario and had to bluff or fall back to legacy software "Oh they can edit stuff in Word and then C&P when they get back."  Right... I can imagine constructing some very novel solutions to this kind of scenario with a combination of Radio & Frontier.  I can imagine some novel applications for k-logging too. It's the flexibility and power of the platform that I'm betting on.  </BLOCKQUOTE> http://matt.blogs.it/entries/00000224.html Amazon webservice Tue, 16 Jul 2002 23:40:39 +0100 <A href="http://biz.yahoo.com/bw/020716/160329_1.html">Press release</A>: "Today Amazon.com launched its first version of Amazon.com Web Services, a platform for creating innovative Web solutions and services designed specifically for developers and web site owners." [<A href="http://www.scripting.com/">Scripting News</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » Cool. I wrote a "bookcase" tool for Radio showing what I was reading & linking back to Amazon via my associates ID.  But it was such a pain getting things right that I never get around to updating my current book!    </BLOCKQUOTE> http://matt.blogs.it/entries/00000235.html Faster damnit! Faster! Wed, 17 Jul 2002 00:52:58 +0100 I've had complaints about how long my weblog takes to load. Looking at it there are a lot of extraneous bits that seemed like a good idea at the time but I don't want any more.  They're gone now.  Also I've cut down how many days posts live on the home page from 7 to 3.  It seems faster now, although most of my posting is done using a 56K modem.  Nothing seems fast with that bandwidth!   http://matt.blogs.it/entries/00000237.html Let's see what's on the slab... Thu, 08 Aug 2002 20:52:14 +0100 Well in the last couple of days I have been working hard on the liveTopics 1.0 release.  It's so close I can almost feel it.  We're testing and hopefully will have the kinks worked out in the next couple of days then I can finally get this sucka out the door. Also I'm really besotted with TrackBack but haven't seen it work the way I would like yet.  So I've rolled a TrackBack server in Frontier that comes with a Radio client.  The two communicate with a simple XML-RPC interface that would allow any klogging system to join in. At the moment the Radio client automatically harvests each posting for links (when you submit it) and automagically pings each one.  The ping contains the permalink for the post, the Url of your weblog, the title of the post, your name & email address.  But you can drop most of this information you don't want to pass it.  I guess some people will also want fine-grained control over what they ping.  That shouldn't be too hard. Along with this are some macros to show your TrackBack information against each item. At the moment the server is hosted on my laptop which isn't ideal but is good enough for testing.  The next job is to find a better host and then look at adding a simple federation mechanism.  That would allow lots of different people to provide TrackBack servers and share the results. More on this later. http://matt.blogs.it/entries/00000266.html Integrating klogs with Big-KM Thu, 08 Aug 2002 23:08:31 +0100 In order for klogging to be successfully I think it is going to have to come to an understanding with Big-KM. Example:  BigCo has invested half a million dollars in a big knowledge management system for their world-wide operations.  This kind of investment can become a lode-stone around any other systems neck.  For klogging to thrive here it is going to have to integrate. Here's one idea I have for how this could work. <OL> <LI>Extend Big-KM System-X so that it can aggregate RSS feeds like Radio, MT and others do now.</LI> <LI>Extend your klogging software to allow per-post meta data.  (liveTopics does this for Radio)</LI> <LI>For each project in System-X define a set of topics that will act as trigger phrases for that project</LI> <LI>Get the kloggers to use those topics when they want to involve a post in a particular project</LI> <LI>Now subscribe System-X to every klog in the organization and watch as it indexes and archives all that information.  Each project grabbing only those postings that are appropriate (by use of the trigger phrases)</LI> <LI>This means that the klogs add value to the big-KM system.  Suddenly it doesn't just have the dry dusty project documention, but all the live vibrant stuff that people are really doing!</LI> <LI>Now extend System-X to generate a per-project RSS feed.</LI> <LI>If I am on the project I can subscribe to this feed.  Now instead of receiving email from System-X or having to go to an arbitrary web page, I get all the "official" project stuff (new documents, forms etc...) delivered in my RSS stream.</LI></OL> Closing the loop between the big-KM and the klog so that they both add value to each other. Just an idea....   http://matt.blogs.it/entries/00000278.html More on TrackBack Fri, 09 Aug 2002 17:01:45 +0100 <A href="http://scriptingnews.userland.com/backissues/2002/08/09#whatAboutTrackback">What about TrackBack?</A>. While I was in the hospital in June, the Movable Type folks implemented a <A href="http://www.movabletype.org/docs/mtmanual_trackback.html">feature</A> <A href="http://www.google.com/search?hl=en&ie=UTF-8&oe=UTF-8&q=trackback">called</A> <A href="http://www.movabletype.org/trackback/">TrackBack</A>. I'm not exactly sure all that it can do, but here's at least part of the story. (I'm posting this so I can get corrected if I don't understand the feature. It occurs to me that this post could use the feature, heh.) Anyone, anywhere can send a message to any Movable Type server to associate a URL with a weblog post. That URL will be shown in the list of TrackBack links for the post. Further, based on an email from <A href="http://radio.weblogs.com/0107808/2002/08/08.html#a251">Matt Mower</A>, for some reason that I don't understand, this can only work with Movable Type servers. I doubt this, because from all outward appearances it is using HTTP, which could be emulated by any program capable of doing HTTP. Matt thinks this feature should be implemented with XML-RPC. I'm not sure it'll take off no matter what it's implemented in. Here's the problem. By design it seems to assume that everyone plays fair. But eventually we all attract a relatively small number of people who would mark up every post with trash talk, if given the chance to. It's a predictable process. That's why I don't have a discussion group here (I used to), or a comments feature. It's why MSNBC is moving to weblogs over discussion software. It's basically why weblogs have a future for thoughtful discourse where mail-list-like collaboration tools are dead-ends. When I think about evolving weblogs, I try to avoid features that turn them into discussion groups. [<A href="http://www.scripting.com/">Scripting News</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » I think there has been a misunderstanding between Dave and I.  Maybe I misphrased something or it was misinterpreted.  Either way: I'm not suggesting that TrackBack can only be implemented in MT.  Just that, as it is implemented in MT it can only be served by MT and is most useful to MT users.  It doesn't suit me very much.   I also don't like the way you have to TrackBack enable things, use special URL's, have bookmarklets etc..  all that gets in the way to me. I envisage an open XML-RPC based system.  The TrackBack data should be available to & from any system and can track arbitrary URLs (no requirement to TrackBack enable anything). Also with the prototype Radio client all the work of pinging is done for you automatically.  As part of the publishing action Radio will figure out all the posts being referenced and ping them automatically.   That's how I want it to work, you might want it different which is why I say it's a prototype. As to the inherent design problem in trackback, well, I agree with the comments made.  From a certain viewpoint. However I see TrackBack not so much as a weblogging tool but as a k-logging tool.  It gives you the ability to know what someone else is contributing to projects you are working on and that could be vital.  As are discussion forums and all the other collaborative tools that help people do useful work. Will TrackBack be absused?  Sure.  But so can any technology.  If the abuse becomes a problem we can evolve strategies for addressing it.  For me this is a time for experimentation, it's too early to abandon a potentially useful idea like TrackBack because it has a potential for abuse. Example (and shooting from the hip) : Problem: "nusiance pings appearing on my TrackBack report."  This seems a lot like the problem of spam email to me.  Collaborative spam filtering looks set to deliver good results here, maybe it could do the same for TrackBack? [Disclaimer: TrackBack - I am a believer!]  </BLOCKQUOTE> http://matt.blogs.it/entries/00000284.html TrackBack questions Fri, 09 Aug 2002 20:16:32 +0100 I've had a couple of comments to the effect of "why implement TrackBack in XML-RPC when its perfectly fine in HTTP/XML?" I guess the answer is that Frontier & Radio make it so easy to implement XML-RPC clients & servers that it was the easiest way to glue together a TrackBack solution that worked the way I had in mind.  I guess there is no no reason why it couldn't also support the same method's as MT as well.  I have no technopolitcal axe to grind here. Something that really bothers me about TrackBack in MT though is the idea of enabling.  Have I got it right?  You have to enable a post for TrackBack before it can receive pings?  And the trackback url isn't the same as the link to the posting?  This seems clumsy to me.  Can someone explain what's going on with that?  Am I confused?   http://matt.blogs.it/entries/00000287.html Tracking back Fri, 09 Aug 2002 21:22:44 +0100 <A href="http://tig.nareau.com/2002/08/09.html#a203">Rahul</A> has the solution. Trackback and comments suffer from the same problem:  they aren't seemlessly integrated into the act of posting to a weblog.  Rahul's points out that a combination of RSS UIDs and blogrolls would seemlessly pull this off.  It would automatically thread the conversation back from your post.  This is a solution that doesn't detract from the weblogging experience.  All you need to do is post, and the rest is taken care of.  The requirements to participate would be that you need to have a weblog and be on someones blogroll (in someones neighborhood).  To screen out inconsiderate particpants, just put their blogs on your block list.   It gets really simple if this is an orgnizational setting: a list of all the people that run weblogs could provide a definitive blogroll, a one-stop-shop for all participants.  Nice. [<A href="http://jrobb.userland.com/">John Robb's Radio Weblog</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » Rahul's idea is interesting, but it isn't TrackBack.  It still assumes a near-line level of connectivity between two blogs.  Maybe that's okay in some situations but I don't think it should be a requirement. As to seamless integration.  The Radio plugin I have devised provides that.  Any url you referencein a post gets pinged when that item is published.  No need to do a thing. Of course some people will want more control.  That's okay too...  </BLOCKQUOTE> http://matt.blogs.it/entries/00000289.html TrackBack and RSS Fri, 09 Aug 2002 21:58:47 +0100 Staying on theme... I see MT's TrackBack has the ability to generate some RSS output, but it doesn't actually look like RSS (as in, I don't think you could just subscribe your aggregator to it). Here's what I have in mind.  I want to know whenever someone pings any of my pages.  I register with the TrackBack server a pattern like <A href="http://radio.weblogs.com/0107808/">http://radio.weblogs.com/0107808/</A>* Whenever someone pings a page under that URL it gets added to a dynamic RSS feed with a URL generated for me.  I can subscribe to that feed in Radio and see all the trackback pings appear as news items. If I'm interested in who is writing about things Dave is saying I might add a pattern <A href="http://www.scripting.com/">http://www.scripting.com/</A>* and get his TrackBack results as well. Does MT's TrackBack do this already?  I couldn't really tell what it's RSS was doing... http://matt.blogs.it/entries/00000292.html liveTopics and categories Sat, 10 Aug 2002 13:57:09 +0100 The the real use of categories to my mind is to allow upstreaming to multiple locations.  That way I can have my public weblog (and maybe a salon blog too like <A href="http://blogs.salon.com/0001048/">Marc Barrot</A>) as well as a number of private klogs. I would still like to be able to use liveTopics when posting to a private klog, but the topics perhaps should not be shared or, if they are shared, it should be done in an intelligent way.  Specifically each category needs it's own Table of Contents where the URL's to posts are not cross-referenced. I'll have to figure out how to do this and it probably won't be easy. Of course a better solution would be for Radio to natively support the idea of mutli-site operations. http://matt.blogs.it/entries/00000300.html Radio news handling Sun, 11 Aug 2002 16:30:07 +0100 <A href="http://radio.weblogs.com/0109150/2002/08/08.html#a424">More Flexible News Scanning Needed</A>. This really hits the mark -- I've been traveling for two weeks with very limited connectivity. [<A href="http://radio.weblogs.com/0109150/">Blunt Force Trauma</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » I too wish that Radio would handle news more flexibly.  The idea of a "poll now" button would be genuinely useful as would a way to adjust the frequency with which different RSS feeds are polled. However when you get to the point where you want to poll every 3 minutes I think you may be reaching the breaking point for the medium.  I would be thinking about moving to Groove, Instant Messenger, Shared/Instant outlining or something like that to handle a real-time interaction.  The results of that interaction could then be published klog style for everyone to share.  </BLOCKQUOTE> http://matt.blogs.it/entries/00000308.html Getting referrers via RSS. Thu, 15 Aug 2002 10:52:12 +0100 <A href="http://radio.userland.com/directory/6742/community">Radio Wishlist - RCS Referers: RSS feed and rolling 24 hours.</A>. Can I get my referer lists as "RSS" feeds from the "Radio Community Server"? Can we make the list a rolling 24 or 25 hours instead of a clean sweep at midnight? [aka <A class=navigatorLink href="http://dijest.com/aka/categories/blueSkyRadio/">Blue Sky Radio</A>] [<A href="http://dijest.com/aka/">a klog apart</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » Now this would be cool. TrackBack information should also appear this way (that's how I'm implementing it).    </BLOCKQUOTE> http://matt.blogs.it/entries/00000324.html Plugging leaks in Radio security Thu, 15 Aug 2002 11:59:11 +0100 <A href="http://radio.weblogs.com/0109150/2002/08/14.html#a462">Protecting Radio Folders</A>. A simple Meta tag to keep prying eyes from browsing weblog folders you don't want people to see. [<A href="http://radio.weblogs.com/0109150/">Blunt Force Trauma</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » Another example of the security theme that is developing.  This is obviously going to become more important as Radio seeks to be the de facto <A href="http://radio.weblogs.com/0107808/2002/08/15.html#a275">PKP</A> tool.</BLOCKQUOTE> http://matt.blogs.it/entries/00000329.html Aggregating IA/ is pretty sucky Tue, 20 Aug 2002 13:18:45 +0100 What is it about Radio's new aggregator and the <A href="http://www.iaslash.org/module.php?mod=node&op=feed">IA/ feed</A>? I seem to get all their articles over and over, no matter how many times I delete them.  I've even got three copies of the whole set in the aggregator at the moment. This sucks! http://matt.blogs.it/entries/00000368.html liveTopics to create virtual weblog channels Tue, 20 Aug 2002 16:57:14 +0100 Here's an idea I've been thinking about for the use of liveTopics. At the moment as author's we categorize our posts for our readers.  If using default Radio by explicitly putting them into categories (or, by default, not doing so).  With liveTopics I can add some granularity to that.  But basically it doesn't have too much impact on my reader.  It also doesn't give the reader much choice. What I'd like to do is offer the reader is the chance to create their own categories and here's how I think it would work: <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> We add a "customise this site" button that pop's up a list of all the topics available on the site. The reader can then group topics they are interested in together to create "virtual channels."  These along with the default selection are bundled up into a cookie that is stored in the readers browser (with their permission). The next time the reader visits the page they only get posts that match the selected "virtual channel." along with a drop-down to change channel and the customise button.</BLOCKQUOTE> Anybody else think this is an interesting idea?   http://matt.blogs.it/entries/00000382.html Entropy, big-KM, klogging and the wheel Tue, 20 Aug 2002 19:57:17 +0100 <A href="http://www.rolandtanglao.com/2002/08/18.html#a2630">Roland's Natural Klog Progression.</A>. I spoke of <A href="http://dijest.com/2002/08/13.html#a1927">four klogging roles</A> last week: catalyst, coach, armorer, practice leader. <A href="http://radio.weblogs.com/0107808/2002/08/19.html#a287">Matt Mower</A> advocates the the role of "Intranet Editor:" <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> Much as the users of a Wiki should occasionally re-factor pages that are becoming "busy" I think that a good intranet editor should be grooming the klogs in their organization and drawing together useful strangs to form part (or all) of the static intranet.</BLOCKQUOTE> <A href="http://www.rolandtanglao.com/">Roland Tanglao</A> builds on this: <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> I think <A href="http://www.rolandtanglao.com/categories/radiouserland/2002/06/02.html#a2066">a natural progression for knowledge</A> is: <OL> <LI>blog breaking news <LI>harvest it periodically (say weekly) into an FAQ and/or other knowledge base type of documents <LI>Put the link into a a directory that supports transclusion like Manila style directories. </LI></OL> K-Log => (FAQ or other knowlegebase article) => directory.   K-Logs need to be periodically (at least once a month) harvested for content that should go into an FAQ or other knowledgebase document and links that that should go into a directory. This is the job of a K-Log editor :-)! I have been trying to do this with VanEats but after a klog gets to a certain size, it really needs to have some time set aside for it.</BLOCKQUOTE> Practice Leader is probably the closest to a dedicated multi-author editor. Summarizing work in a field, showing the aggregate progress and useful threads. Structuring knowledge into FAQs or other KM systems may be a natural progression, especially as klogging tools and KM tools build bridges. Entropy, bad. Fighting entropy, expensive, slow.  Self-review is a powerful tool for learning. Going over my own posts for the past week, month, and quarter has shown patterns I missed, ideas I was skirting but never wrote outright. It reinforced brief social connections, blogs to which I linked to and people with whom I briefly corresponded. It takes concentrated time and effort. It helps me to print out all the pages on my blog for that period; something about shuffling through paper. Folks are trying hard to automate this work. Summarizers. Cluster analysis. Text to Structure converters. Taxonomy systems. But the expert author of the original content is often the best judge of relevance. [<A href="http://dijest.com/aka/">a klog apart</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » I think one of the things about klogs is that are no better than any other KM system when it comes to entropy.  In fact they are likely to be a hell of a lot worse -- it's just the entropy matters less.  Any information system that isn't properly maintained has the potential to quickly deteriorate into chaos. The fact is that most people don't want to have to find just the right place to put something.  Most people aren't going to review what they have done.   You can force this behaviour, you can encourage it.  But is it really necessary that everyone has to become a librarian in order to function in a knowledge environment? My alternative is that we recognize and promote the value of good editing (and, hence, good editors).  Have an editor/practice leader to head each area whose responsibility it is to aggregate good knowledge.  Then reward them when they do it well.   Example:  Look at the number of search engine queries for specific keywords.  Tie those keywords to projects/areas.  If the number of searches trends downwards something is working.  Okay, too simplistic? Then suggest something better! An area I have been thinking about is how I would integrate the idea of uploading files to a KM system when klogging.  One approach would be to provide some kind of clever dialogue to allow the user to specify where they want the file to end up.  That sounds like hard work for me & for the user. Alternative strategy:  Allow the user to put a file in an enclosure.  Radio will upstream it to the KM server as part of the RSS feed.  The KM server will toss the file into an upload bucket in an area based upon the metadata of the post (ala liveTopics).  It's then up to the practice leader for that area to decide where the document actually belongs and move it there (or indeed if it belongs at all). Is this less efficient?  Maybe so.  Is it more effective?  I think so. Agree? Disagree?  Ideas?</BLOCKQUOTE> http://matt.blogs.it/entries/00000390.html Something worth talking about. Tue, 20 Aug 2002 21:15:08 +0100 <A href="http://radio.weblogs.com/0103807/2002/08/20.html#a468">Something Unexpected: Scotts Radio</A>. <H3>Something Unexpected: Scott's Radio</H3> <TABLE> <TBODY> <TR> <TD vAlign=top><IMG src="http://radio.weblogs.com/0103807/images/scottsradiobookcover02_sm_blogsidebar.gif"> </TD> <TD>   </TD> <TD> As the author of the <A href="http://www.fuzzygroup.net/go/?oreillyfromradio">O'Reilly Essential Blogging</A> chapters on Radio, I clearly have a commercial interest in Radio.  You'd think that I'd want people to just buy the Essential Blogging book and NOT give content about Radio for free.  You'd think that but you'd be wrong.  I really want to see Radio do well along with great people like Jake and Lawrence.  And more documentation is pretty much always a frothy good thing for products. So... Inspiration struck me yesterday when I was digging through the 240 gigabytes of digital bile that I call a hard drive(s): <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> O'Reilly cut a lot of my text on the Essential Blogging book.  (these are all labeled as "Missing")  Why not aggregate that content along with my previous writings on Radio and release it as a free book under the GNU Free Documentation License?  This content still gets tons of hits from Google so it's clearly useful.  </BLOCKQUOTE> A quick demand (ok gentle request) to my partner, Gretchen, for "A really cool cover" and within about an hour, she IM'd me the graphic at left.  And I've been in hard core content massage since 3:37 am on this oh so soggy Boston day.  I won't tell you that this content is perfect -- there are clearly some broken links and other editing style things that need to get done.  But there is a lot of content and it's useful.  It'll get improved more over time but following the Open Source mantra of "Release Early and Release Often", I give you: Scott's Radio ==> <A href="http://radio.weblogs.com/0103807/stories/2002/08/20/scottsRadioCover.html">Read Stories</A> <==</TD></TR></TBODY></TABLE> [<A href="http://radio.weblogs.com/0103807/">The FuzzyBlog!</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » Thanks Scott for publishing this.</BLOCKQUOTE> http://matt.blogs.it/entries/00000392.html You know when... Wed, 21 Aug 2002 08:51:06 +0100 Okay I'm a convert to Radio's outliner now.  I just caught myself, in a Word outline, pressing F2 to navigate the outline nodes!! http://matt.blogs.it/entries/00000404.html Implementing trackback services Wed, 21 Aug 2002 10:52:04 +0100 Radio Tip: <A href="http://radio.weblogs.com/0107233/categories/radioquestions/2002/08/20.html#a645">TrackBack tutorial</A> ~ <A href="http://www.whump.com/moreLikeThis/link/03015">Homebrew TrackBack Tutorial</A>  Got this from Whump.com. If you have TrackBack Envy (I do), here's your revenge: <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> "It's easy for a non-Movable Type weblog to send <A title="Moveable Type Manual" href="http://www.movabletype.org/docs/mtmanual_trackback.html">trackback pings</A>. They use a <ACRONYM title="REpresentational State Transfer">REST</ACRONYM> interface: <CODE><A href="http://foo.com/mt/mt-tb.cgi?tb_id=ID&"><A href="http://foo.com/mt/mt-tb.cgi?tb_id=ID&">http://foo.com/mt/mt-tb.cgi?tb_id=ID&</A></A>;title=TITLE&url=URL</CODE> So you can write a bookmarklet that 'does the right thing' if you know the site's base URL. Trackback supports a few more parameters in the URL so you can send a summary of the reply, and the name of the site sending the ping." Over at Hit-or-Miss, they've been building <A title="Homebrew TrackBack Tutorial" href="http://www.hit-or-miss.org/projects/trackback/">their own Trackback server in PHP and MySQL</A>." </BLOCKQUOTE> [<A href="http://radio.weblogs.com/0107233/categories/radioquestions/">Dog News: radio questions</A>]  [From <A href="http://dws.us/weblog/categories/radiofaq/2002/08/20.html#a2187">DWS Radio FAQ</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » Can someone explain to me what the tb_id part of the URL is? I'm still not convinced by MT TrackBack.  But I shall certainly try and support the current interface in my own TrackBack server implementation.</BLOCKQUOTE> http://matt.blogs.it/entries/00000406.html An unstable Radio is an unhappy Radio Thu, 22 Aug 2002 11:08:18 +0100 Well, thrice is the charm. Again, radio crashed and I lost posts. I was able to reconstruct them from local cache, but enough is enough. [<A href="http://radio.weblogs.com/0108194/">The Universal Church Of Cosmic Uncertainty</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » I'm intrigued about why Radio is crashing so often for you and why you are losing information.   I do have Radio crash here (for example I cannot compact the weblogData.root file) sometimes but basically it's quite sound. If you're not 100% committed to abandoning it, can you blog some more details of your situation?</BLOCKQUOTE> http://matt.blogs.it/entries/00000425.html XTM export of a weblog Mon, 26 Aug 2002 21:25:56 +0100 Anyone with an interest in XTM want to check <A href="http://radio.weblogs.com/0107808/outlines/topics/topics.xtm">this</A> out? It's my weblog + topics exported as a topic map via liveTopics.  I'd be interested in any opinions as to the correctness of my XTM implementation, use of tags etc...   http://matt.blogs.it/entries/00000459.html Klogs can improve the value of what you write Wed, 28 Aug 2002 15:27:46 +0100 <A href="http://radio.weblogs.com/0107808/">Curiouser and curiouser!</A> in <A href="http://radio.weblogs.com/0107808/2002/08/27.html#a339">There's a hole in my bucket...</A>. <BLOCKQUOTE>As a klogger, over the past 3 months or so, I have recorded & published tens if not hundreds of thoughts.  I doubt if I shared one quarter of output during the last 6 years I worked at various companies.  Oh I would probably have emailed here and there, spoken up during meetings.  But I wonder just how much knowledge is being lost, second by second, in most companies by each employee.  Then multiply up...</BLOCKQUOTE> But even if they would catch those thoughts, it's going to be very difficult to find something relevant and to understand it our of the context. More or less like forum discussion: you have to follow for some time to make sense of it. Going through blog archives is not easy... So far I benefit more from the distributed dialog and from the collective filtering. So, blogs is more for sharing, rather than capturing... [<A href="http://radio.weblogs.com/0109961/">Mathemagenic</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » I don't think this problem is necessarily inherent in blogging/klogging as practiced, more a problem in the simple calendar based access method most weblogs provide by default. But there are other options, for example a Radio weblog with liveTopics adds another dimension for relating posts together to create a train of thought.  You can follow a topic from a post into a table of contents where you can see other posts referencing that topic.  You can also see, for each post, other topics that were associated with it allowing you to hop from one subject of conversation to another. The next level is based upon XML topic maps (in XTM format) which I am experimenting with generating right now.  This will allow you to reconstruct the weblog to serve different purposes and, by merging topic maps from different weblogs together, to analyse a larger conversational "space." All in all I think klogs will, ultimately, vastly improve the ability for people to find things that are relevant and meaningful among the discourse of themselves and others.  </BLOCKQUOTE> http://matt.blogs.it/entries/00000470.html Keep your tinder dry Thu, 05 Sep 2002 12:44:21 +0100 When I came across it again today I remembered that <A href="http://www.eastgate.com/Tinderbox/">Tinderbox</A> is the application that got me into weblogging. Although I couldn't use it (it's Mac only) I consumed the website with great interest and a certain amount of jealousy.  It was also the first reference I had seen to weblogging and from there I got to Radio Userland. I'm really hoping they port Tinderbox to Windows soon. http://matt.blogs.it/entries/00000491.html Find tool for Radio developers Mon, 09 Sep 2002 20:47:20 +0100 Although Radio includes a Find & Replace facility which lets you search the database I often find it more annoying than useful.  This is because it too much power with too few controls.  Most of the time I want to search for a specific table or script.  But with Radio's built-in find any occurrence of my search term in any object, I also get to iterate through these interactively in little windows that popup in random spots on the desktop, and I'm never quite sure whether that beeps means my search is finished or not. Tonight I snapped and and wrote a findObject() routine to search the database for object names, and then added a mechanism to allow specifying the object type: table, text, outline and so on. Drop <A href="http://radio.weblogs.com/0107808/gems/tools/utils.root">utils.root</A> into your Radio tools folder & restart Radio.  You will have a new menu Utils in which will be a find sub-menu.  It takes the current target as the start point for the search, or the root if you have no start point.  Take a look at utilsSuite.findObject() if you want to see the guts. Happy searching.     http://matt.blogs.it/entries/00000516.html Here's hoping that there isn't an RSS3.0 Tue, 10 Sep 2002 08:12:14 +0100 <A href="http://rss.benhammersley.com/archives/001321.html">Questions for Dave</A>. In the comments to Entry below, Dave Winer says: "To everyone, I believe the new spec is leaps and bounds... [<A href="http://rss.benhammersley.com/">Content Syndication with XML and RSS</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » Depressing, depressing, depressing. One of the 7 habits (of Stephen Covey fame) is "<A href="http://www.leaderu.com/cl-institute/habits/habit5.html">Seek first to understand, then to be understood</A>."  The tension in this discussion is rising fast and empathic listening seems to be out of the window.  It's all sounding more and more like a debate in the house. I think <A href="http://groups.yahoo.com/group/syndication/message/3299">Joe Gregorio</A> makes an interesting point.  If the RSS1.0 folks could stand to walk away from the name RSS then much of the tension disappears.  You can't stop Dave issuing his own 2.0 and trumping him with an RDF 3.0 is just going to escalate things.   Sounds a bit like the middle east doesn't it? <A href="http://groups.yahoo.com/group/syndication/message/3251">Bill Kearney</A> has already proved that any new standard, even one that Userland doesn't support, can introduced into Radio. I vote for taking the creative energy in this discussion into a new TLA and a new forum.   The worst that suggests itself to me about this route is having to support two specifications.  But isn't that the case already with RSS1.0?</BLOCKQUOTE> http://matt.blogs.it/entries/00000524.html Updated find tool for Radio Thu, 12 Sep 2002 20:19:00 +0100 I've updated the <A href="http://radio.weblogs.com/0107808/gems/tools/utils.root">find tool</A> for Radio to fix a subtle bug that would cause it to skip some tables.  I've also added menubars as a search type and an option to use an exact name match rather than a substring match (the tool is still case insensitive, but then Radio table names are case insensitive anyway). To install just drop the .root file into your Tools directory.  The Find menu will appear until Utils on your menubar. http://matt.blogs.it/entries/00000539.html Radio RCS needs better referrer tracking Tue, 17 Sep 2002 18:19:44 +0100 <A href="http://radio.weblogs.com/0110772/2002/09/17.html#a299">Radio wishlist: automated referers harvesting</A>. Does anyone know of a simple way of automatically storing (or e-mailing) my list of referers every day just before they are reset, so I can look at them when I have the time? [<A href="http://radio.weblogs.com/0110772/">Seb's Open Research</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » So many people want this I would hope that Userland would add it when the dust settles on Frontier 9.0.  I added a free <A href="http://sm6.sitemeter.com/stats.asp?site=sm6curiouser">site meter</A> to my blog just to get better referrer tracking.</BLOCKQUOTE> http://matt.blogs.it/entries/00000552.html Autotrackback Tue, 24 Sep 2002 18:46:51 +0100 <A href="http://radio.weblogs.com/0101679/2002/09/24.html#a843">Automating Trackback</A>. <A href="http://www.decafbad.com/news_archives/000292.phtml#000292">0xDECAFBAD</A>: I'd also like something that scans a blog entry I post for links, then investigate those links for Pingback/TrackBack availability - all to make the system even more automatic.  I'm game.  I've <A href="http://www.intertwingly.net/code/autoping.py">implemented it</A> in Python.  And written a Perl <A href="http://www.intertwingly.net/code/excerpt.pl">extracter or excerpts and pinger</A> which uses takes advantage of this.  I'll be testing it on this blog entry.  Once debugged, I'll connect this to my publishing process (though a simple cron job would do) and forget about it.  As Ben <A href="http://www.stupidfool.org/archives/2002/09/000211.shtml">says</A>, There is nothing inherent in TrackBack that makes it any less transparent to either of the users involved. The previous sentence is in there just so I can get two trackback entries out of this. <IMG alt=smiley src="http://static.userland.com/shortcuts/images/qbullets/sidesmiley.gif"> [<A href="http://radio.weblogs.com/0101679/">Sam Ruby</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » This is exactly how the TrackBackish server I wrote worked.  The Radio piece automatically scanned the post for links and sent pings for each of them to the server. Why to a server?  Because that way everyone knows where to find them (and you can federate to spread the load). The idea was that you could obtain information about pings simply by specifying patterns to match, such as <A href="http://radio.weblogs.com/0107808/">http://radio.weblogs.com/0107808/</A>* to see who had pinged any page on your weblog.  You could also query for just about anything. I also had in mind a Wiki'ish syntax to add to the URL which the Radio code would strip out before publishing but would pass through to the backend server.  Something along the lines of <UL> <LI>*url - approve</LI> <LI>!url - disapprove</LI></UL> The server could track then track these "votes" but I hadn't got as far as thinking about what you might do with them.  </BLOCKQUOTE> http://matt.blogs.it/entries/00000578.html JIRA does RSS, so cool! Tue, 24 Sep 2002 23:05:56 +0100 Oh man <A href="http://opensource.atlassian.com/novissio/secure/IssueNavigator.jspa?pid=10000&tempMax=25&view=rss&reset=true">this</A> so rocks! I'm now have Radio subscribed to a <A href="http://opensource.atlassian.com/novissio/secure/IssueNavigator.jspa?pid=10000&tempMax=25&view=rss&reset=true">feed</A> coming out of my liveTopics <A href="http://opensource.atlassian.com/novissio/secure/Dashboard.jspa">JIRA project</A>.  I get an RSS item for each change that happens, i.e. someone adds a new issue, someone adds a comment, it's all there.  This is *so great* for project visibility. Unfortunately it does kinda point out the weaknesses in the Radio aggregator interface.  It was probably fine when you only had a dozen or so feeds but I'm up to 58 and some of them are really big.  I need a tabbed interface that lets me organize feeds the way I want.  I need keyword prioritzation, I need the ability to delete all items from a specific feed at once.  I'd like them outlined to save screen real-estate.  [Actually John Robb's just pointed out that I can delete all the items at once, using the magnifying glass icon - must check that out] Damnit I want aggregator Mark 2.   http://matt.blogs.it/entries/00000585.html Sharing zones of control Tue, 24 Sep 2002 23:48:52 +0100 <A href="http://www.smartmobs.com/archives/000171.html">IBM turns to social network analysis</A>. A critical resource embedded within organizations is the knowledge that highly skilled workers bring to work on a day-to-day basis. However, aside from human resource policies targeted at the attraction, development and retention of skilled knowledge workers, there has been little effort put into systematic ways of leveraging knowledge that is embedded in people and relationships. Given the extent to which people rely on their own knowledge and the knowledge of their contacts to solve problems, this is a significant shortcoming. Social network analysis allows us to understand how a given network of people create and share knowledge, helping us to move beyond this approach. [<A href="http://www.smartmobs.com/">Smart Mobs</A>] An important issue that would arise for me, if I were to work in an enterprise, would be to restrict my sharing to the organization. This would require a degree of corporate loyalty that I just might have some trouble with. From a personal standpoint, it would be more useful for me to share all my knowledge publicly: it would enable me to build more relationships with outsiders, and establish a reputation that is not limited to my organization. When the time comes to move on, I'd probably be in a better position. [<A href="http://radio.weblogs.com/0110772/">Seb's Open Research</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » As in most human endeavours I think there's going to have to be a compromise.  I can imagine increasingly relaxed zones of control over blogged information.  Sharing layers if you like: <UL> <LI>My personal private blog (backup brain)</LI> <LI>Team / Project Group / Community blog (private sharing)</LI> <LI>Intranet blog (corporate sharing)</LI> <LI>Public blog (real sharing)</LI></UL> All of which could be done now using Radio category routing.  A simpler interface could be introduced in Radio so that people can specify how wide they want that post shared and Radio selects the right routing category itself. I'd add that liveTopics (plug, plug) will soon support categories and less soon the idea of a corporate weblog directory.  This will group posts from different weblogs around shared topics.  Add theme support and you can cluster related topics to create a real navigable knowledge structure for each layer. I haven't forgotten about BlogPlexes either... [I do seem to have gone italic mad lately though]</BLOCKQUOTE> http://matt.blogs.it/entries/00000590.html For a well baked blog, add topics Sat, 28 Sep 2002 12:59:52 +0100 <A href="http://www.networkcomputing.com/1320/1320buzz2.html">Michael DeMaria</A> over at Network Computing wants weblogs to have topical lists of posts.  He points out that the time-based format isn't the easiest thing to use when looking for specific posts on selected topics.  There are obviously two ways find posts contain a specific topic: 1) Use a search engine.  This is the best approach to use when people are resistant to entering metadata.  2) Use a metadata tool like LiveTopics by <A href="http://radio.weblogs.com/0107808/">Matt Mower</A>.  Matt has built a tool for Radio that makes it easy for authors to enter in metadata with each post.  This makes it easy to provide directories that list post by topic (through use of the outliner).   Basically, Livetopics can create a simple list of topical links to posts, or a complex hierarchy of topical links.  Matt has a complex hierarchy on his site. [<A href="http://jrobb.userland.com/">John Robb's Radio Weblog</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » With thanks to John for the link. Clearly I think Mike makes a very valid point.  Weblogs make great diaries, but the by-date navigation structure sucks for locating topical information.  More information about liveTopics can be had by either clicking the liveTopics see-also reference under this post, or going to the <A href="http://www.novissio.com/products/liveTopics/liveTopics.html">liveTopics</A> page on the <A href="http://www.novissio.com/">Novissio</A> website.</BLOCKQUOTE> http://matt.blogs.it/entries/00000620.html Permalinks should be! Sun, 29 Sep 2002 21:41:47 +0100 Like a lot of people I'd like to move my weblog off of the weblogs.com server.  In a few short months I've racked up 24mb worth of "stuff" so I imagine I'll have used up the 40mb before too long.  Also as other people have commented the URL isn't terribly attractive. So I have my Novissio.com site and that would be fine for hosting, I can still ping weblog.com.  But what do I do about this site? What I would like to see is for Userland/weblogs.com to offer a smart redirection service that forwards requests to another site for you.  I can implement this myself I suppose, but they could do it at the server level which would be a good thing. Also, as I've suggested before, I think that Userland should look seriously at handling permalinks via a <A href="http://www.purl.org/">PURL</A> service.  That way permalinks really would be permanent. http://matt.blogs.it/entries/00000629.html Congratulazioni! Fri, 04 Oct 2002 22:51:37 +0100 Radio UserLand in Italian <IMG hspace=5 src="http://paolo.evectors.it/myImages/RadioBox.gif" align=right>Starting today, we are distributing the <A href="http://www.evectors.it/">Italian version of Radio</A>. It took longer than expected to localize (have you ever noticed how much text there is in the <A href="http://127.0.0.1:5335/system/pages/help">help pages</A> of Radio???), but now it's done. The great thing is that since the localization is done on a tool, we'll be able to keep it up-to-date with new Radio features (and to fix translation errors that we'll find). We are also providing hosting service under the blogs.it domain, running our own RCS. If you are an Italian blogger, you can <A href="http://www.evectors.it/itideatools/story$num=137&sec=8&data=ideatools">download Radio in Italian</A> now and try it for 30 days. You'll be up'n'bloggin in just a few minutes. Special thanks to <A href="http://jake.userland.com">Jake</A> who helped us a lot trough all the process. [<A href="http://paolo.evectors.it/">Paolo Valdemarin: Paolo's Weblog</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » Congratulations eVectors!  I hope they do brisk business. I'll have a lot more to say about the good folks @ eVectors in a couple of days.</BLOCKQUOTE> http://matt.blogs.it/entries/00000638.html Language logging Fri, 18 Oct 2002 10:28:02 +0100 Blogs for developing language skills. One of my friends has commented on my not perfect English and suggested her help for improving it. She is going to start Radio blog, so I think about the following: <UL> <LI>She makes a special category, which is not visible in her blog, but has an RSS feed. <LI>Then she uses this category to comment on my posts pointing to errors and suggesting improvements. <LI>I subscribe to this RSS, get my personal feedback and correct posts. </LI></UL> Implications: <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> For me it should be more effective than any course or private lesson: no stupid exercises, but just-in-time feedback to improve my writing. This could be a service that someone can provide for bloggers. Personally, I wouldnt mind to pay a bit for it. For multinational companies this could be a solution to help their employees developing language skills and overcoming fears of writing in foreign language.</BLOCKQUOTE> [<A href="http://radio.weblogs.com/0109961/">Mathemagenic</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » Excellent idea.  This would be a good adjunct to any language teaching course too.  Although managing the 300 or so categories required for each student could be a challenge...</BLOCKQUOTE> http://matt.blogs.it/entries/00000748.html Blogplex in WebWork Fri, 18 Oct 2002 10:39:12 +0100 <A href="http://radio.weblogs.com/0107789/2002/10/18.html#a987">Web-excellent-work!.</A>. <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> <A href="http://roller.anthonyeden.com/page/rsal/20021016">Web-excellent-work!</A>. I know, bad title. Just wanted to add another note to the large list of pluses about <A href="http://www.opensymphony.com/webwork">WebWork</A>. It's so easy to protype new use cases (stories for you XP'ers') with webwork. I managed to knock off quite a few today to meet a delivery time for our project. Once you understand ww, throwing something together is a piece of cake! Throw in SiteMesh with a decorator and you can prototype an entire site in a day. Gotta love that! [<A href="http://roller.anthonyeden.com/page/rsal">Rick Salsa</A>]</BLOCKQUOTE> 'nuf said really? (<A title="Click to read comments about this item." href="http://radiocomments.userland.com/comments?u=107789&p=987&link=http%3A%2F%2Fradio.weblogs.com%2F0107789%2F2002%2F10%2F18.html%23a987">comments</A>) [<A href="http://radio.weblogs.com/0107789/">rebelutionary</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » I don't post on Java stuff too often.  It's not my day-to-day anymore and I don't often have a lot to add to what I see (although I file a good deal of it away for later).  However two factors make this look interesting: 1. I'm going to have to do a lot more Java consultancy work than I expected if I want to stay financially afloat.  Most Java jobs say J2EE (whether they mean it or not) and this looks like a great J2EE project to get to grips with.  (I have a lot of Java experience but absolutely no J2EE). 2. I was going to implement the BlogPlex server part of liveTopics as a Radio/Frontier application.  However my experiences developing on the platform make me very leery of attempting an even larger project on it. If I was a PHP expert like "!fuzzygroup" I would probably be thinking of Apache/PHP, but I'm not, so I'm considering implementing it as a J2EE service.</BLOCKQUOTE> http://matt.blogs.it/entries/00000754.html Radio Make Tue, 22 Oct 2002 12:27:28 +0100 Quick note: I have updated my <A href="http://radio.weblogs.com/0107808/gems/tools/utils.root">utils</A> tool for Radio (and Frontier I guess) to add a "Make Root" function.  When invoked this will make sure that every script in the targetted root is compiled.  Actually make is a misnomer really since I don't know how to detect whether a script is already compiled, so I just compile 'em all. I can now confirm that all 175 scripts in liveTopics compile properly.  They may not work, but at least they compile! http://matt.blogs.it/entries/00000760.html Fetch delivers Enterprise Streaming Wed, 23 Oct 2002 10:13:47 +0100 <A href="http://www.fetchserver.com/">Have you seen Fetchserver.com</A>. <A href="http://www.fetchserver.com/">Fetch</A> delivers on the <A href="http://dijest.com/aka/categories/blueSkyRadio/2002/09/04.html#a1981">enterprise streaming</A> side of <A href="http://dijest.editthispage.com/klogs">klogging</A>.  <UL> <LI>Server pulls SQL data from sources, on schedule, outputs RSS feeds.  <LI>Clients grab feeds, scroll news in a task bar UI. </LI></UL> Note the variable query frequency: pulled more often for rapidly changing data, presented more prominently for more important data. Why isn't this database bridge part of Radio or Manila? [a klog apart <A href="http://dijest.com/aka/categories/klogs/">klogs</A>] [<A href="http://dijest.com/aka/">a klog apart</A>] <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> » An interesting light weight RSS client that scrolls headlines and gives a click through to more info.  Reminds me of PointCast (who now seem to be InfoGate, "enabling leading media companies to offer turnkey premium subscription services to their clients"). As Phil mentions there is no reason why Radio/Frontier couldn't act as the FetchServer part of a Radio based enterprise RSS streaming network.  Although the Radio client would need a lot of work to be as functional in that context. The strength (and weakness) of the Fetch client is that it is always visible, docked to the taskbar.  With headlines scrolling continuously.  However I can imagine that it also becomes a distraction, or an annoyance, especially if you are subscribed to a lot of channels.</BLOCKQUOTE> http://matt.blogs.it/entries/00000767.html Some good bits from the x-log Wed, 27 Nov 2002 18:12:55 +0000 From <A href="http://radio.weblogs.com/0001134/">x-log</A> I found some useful stuff for advanced Radio users and Radio developers: <UL> <LI>the <A href="http://radio.weblogs.com/0001134/stories/2002/02/16/rframeMacro.html">rFrame macro</A> for creating inline frames</LI> <LI>the <A href="http://radio.weblogs.com/0001134/stories/2002/02/18/eventViewer.html">desktop event viewer</A> displaying almost real-time the radio event log in your desktop website (very cool)</LI></UL>   http://matt.blogs.it/entries/00000880.html liveTopics in RSS2.0 #2 Thu, 28 Nov 2002 15:55:44 +0000 <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> <A href="http://radio.weblogs.com/0107808/2002/11/27.html#a566">Matt Mower's Knowledge Log - ( liveTopics, k-log, radio, blogging, RSS )</A>: "This will enable a smart aggregator to use the topic's for filtering & combining feeds together." + filtering is in the pipeline for myRadio, on dates, keywords, and now topics. will be tricky to devise a UI. + filter a single feed, or multiple feeds. multiple feeds would require agreement on a common pool of topics, i think. + Syndication, with meta-data, gathered by smart aggregators, has a lot of possibilities. It would be cool to hear more about usage scenarios. [<A href="http://radio.weblogs.com/0100875/">Brain Off</A>]</BLOCKQUOTE> Mikel picks up on my post yesterday regarding adding topics to Radio RSS.  I've got a few things in mind for this, but I'm sure others will really lead the way.  Let's just address one point first. When trying to handle feeds from multiple blogs, inevitably, as Mikel points out, we will reach the situation where people using different words to mean the same topic.  This will be a problem, but hopefully not as a big of a problem as it could be.  It is for this reason that I have been tracking <A href="http://xfml.org/">XFML</A> so carefully.  With XFML we have the ability to say "A's topic X is the same as B's topic Y".  liveTopics already does XFML. So I think the first and simplest usage scenario will be within the type of aggregators that we have now as a way of filtering a feed to get rid of posts we deem irrelevant.  This will allow us to subscribe to many, many more feeds since we don't have to weed out so much chaff.  Although I think we'll need to be careful as it may make it more difficult to experience serendipitous moments. The next scenario I can imagine is as a way of producing a consolidated "on-topic" feed from a number of other feeds.  Combined with technology to scrape RSS from sites and databases and with a little automagic to add topics where they don't exist this could be very powerful. My imagination runs out here, maybe someone else..? http://matt.blogs.it/entries/00000888.html RssDistiller: vital tool for Radio Users Sun, 08 Dec 2002 10:43:53 +0000 Last time I checked out <A href="http://store.evectors.com/itproducts/story$num=1&sec=3">RssDistiller</A> from evectors I wasn't really into RSS very much.  It was just after I started using Radio and, frankly, I was more interested in messing around with it and what it could do.  What did I care about feeds?  Also creating patterns to distill sites is a bit of an art, who has the time? Well of course I'm a little older and wiser now.  RSS has grown to be very important to my thinking and to how I think business should be done.  So important that tools to get non-RSS delivered content into feeds are vitally important.  Of course, this is exactly what RssDistiller does. To create a feed from a website you point RssDistiller at the site and specify patterns marking the start and end of the areas RssDistiller should look at, and then the start and end of each "item" it should create.  RssDistiller will then turn that into a valid feed. For example the following patterns are how i configure a feed for a website that I use: <UL> <LI>ignore text before: <body <LI>ignore text after: </body <LI>start pattern: <LI>end pattern: <LI>item template: ##text##</LI></UL> fiddly, but worth it. RssDistiller is definitely worth <A href="http://store.evectors.com/itproducts/story$num=1&sec=3">checking out</A>.  http://matt.blogs.it/entries/00000963.html Looking for my next aggregator Wed, 08 Jan 2003 23:15:36 +0000 Okay this is a real problem with the Radio news aggregator.  There is an RSS item that I know is in there somewhere, I saw it this afternoon, but I can't find it. It has been buried by a mountain of new items.  As far as I can tell the only way to find it again is to delete enough items that Radio will show it on the news page again.  Oh I guess that I can find the feed URL and fake a URL for the Zoom feature to display all items from that feed... (since the feed doesn't appear on the news page at all right now). As I come to depend more and more upon RSS to keep track of the life & work of the people I am becoming friends and colleagues with I am increasingly finding that the Radio aggregator doesn't cut it.  This is just one of a number of items that are bugging me and I don't have time to address, even if I thought it was a worthwhile exercise. I'd really like to try out some of the alternatives.  Can anyone recommend any aggregator software?  It must run on Windows (so don't suggest iNews) and ideally it should be written in Java (but that's just a nice to have). http://matt.blogs.it/entries/00001052.html The Topic Exchange Tue, 14 Jan 2003 09:57:10 +0000 <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> <A href="http://blogs.salon.com/0000002/2003/1/14/#200301141">New and improved</A>. Thought I'd mention something new I've been hacking on for the last few evenings. It's not all done yet, but people are e-mailing me about it so here's a bit of an introduction:     <A href="http://topicexchange.com/">The Internet Topic Exchange</A> It's the first (as far as I know) real-life implementation of <A href="http://www.myelin.co.nz/cgi-bin/wcswiki.pl?RidiculouslyEasyGroupForming">Ridiculously Easy Group Forming</A>. Basically, it lets you create sites like <A href="http://www.highcontext.com/kmpings/">KMPings</A> just by filling out a <A href="http://topicexchange.com/new">form</A>. Once you've created one, you can send <A href="http://www.movabletype.org/trackback/">TrackBack</A> pings to it, and see them <A href="http://topicexchange.com/t/test/">like so</A>. There's also <A href="http://topicexchange.com/t/test/rss">RSS</A> for the aggregator junkies. With any luck <A href="http://radio.weblogs.com/0107808/">Matt Mower</A> will be supporting it with his <A href="http://www.novissio.com/products/liveTopics/liveTopics.html">LiveTopics</A> tool, so it'll be trivial to use from <A href="http://radio.userland.com/">Radio</A> as well. Any suggestions / feature requests? <A title="Click here to comment on this post." href="http://rcs.salon.com/rcsComments/comments?u=2&p=200301141&link=http://blogs.salon.com/0000002/2003/1/14/#200301141">Comment</A> [<A href="http://blogs.salon.com/0000002/">Second p0st</A>]</BLOCKQUOTE> Indeed I will be.  I hacked in the basic support for the configuration of this feature last night (since I was working on preferences code anyway).  Adding the ping code as another publishing activity should be trivial. What Phil has done is to implement a very simple, elegant, solution along the path of the BlogPlex idea I've been working towards.  With the Topic Exchange, it will be simple for users to cluster around topics simply by using them. What might be interesting is to combine this with the idea of synonyms (from XFML) so that even when people don't use exactly the same topic name, if they are talking about the same thing, they can still cluster with everyone else!   http://matt.blogs.it/entries/00001066.html RSS topics in the mix Tue, 21 Jan 2003 18:17:40 +0000 <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> <A href="http://groups.yahoo.com/group/radio-dev/message/7369">Dave Winer</A>. Callbacks for RSS-writing released [<A href="http://groups.yahoo.com/group/radio-dev">radio-dev</A>]</BLOCKQUOTE> This is good stuff.  liveTopics was using my own witches brew of code to inject topic information into the RSS feed.  Now it can do it with a proper set of callbacks.  And this will work across all categories as well.  Neat, thank you Dave.   http://matt.blogs.it/entries/00001084.html Hit-and-run weblogging Fri, 31 Jan 2003 10:52:32 +0000 <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> <A href="http://diveintomark.org/archives/2003/01/28/autocontent.html">This item</A> on <A href="http://diveintomark.org/archives/2003/01/28/autocontent.html">hit-and-run weblogging</A> coupled with <A href="http://radio.weblogs.com/0111261/2003/01/28.html#a47">this other item</A> on <A href="http://radio.weblogs.com/0111261/2003/01/28.html#a47">discerning weblog usage</A> gave me an idea: Is there a way, in Radio, to specify which posts are forwarded to your aggregator feed? Often I do a quick post (some smug webloggers might call these hit-and-run posts) for my own reference. Other posts have an audience in mind. I'd like to post both to my weblog, while forwarding only the latter to the aggregator feed. Has this been done? [<A href="http://radio.weblogs.com/0111198/">Blogfish</A>]</BLOCKQUOTE> For this kind of stuff I use a trick I picked up from Spike Hall.  I have a category Seeds which represents those things.  Any post I don't want to appear goes into Seeds and not the Home Page. Then I copied the #upstream.xml file from /www/system into /www/categories/seeds/ to prevent anything getting upstreamed to the cloud. http://matt.blogs.it/entries/00001123.html Birthday present for Paolo's blog Wed, 26 Mar 2003 16:20:18 +0000 <%googlami( "Paolo Valdemarin", "Paolo")%> wanted a macro that would link directly from a term like Paolo Valdemarin to the first hit returned by Google for that term. <A href="http://radio.weblogs.com/0107808/gems/googlami.txt">Here</A> is a macro that will do that. (To install, right click the link and select "Save As" then save it as googlami.txt in your Radio Userland macros folder). To use it you would use the source view in the post editor and type in: <%googlami( "what i want to search for")%> and it will generate a complete HTML link for that search.  If you want the text for the link to be different to the term, pass a second parameter with the text of the link, e.g. <%googlami( "Paolo Valdemarin", "Paolo" )%> Consider it a 1st birthday present for your blog :) http://matt.blogs.it/entries/00001324.html Pepping up liveTopics performance Mon, 07 Apr 2003 20:40:01 +0100 I knew that Radio's outline were slow but... Internally liveTopics uses Radio's outlines quite a lot.  When you are outputting outlines for activeRenderer it seems quite convenient to use Radio's built-in support them as a type.  But as my topics database has grown, liveTopics has begun to groan.  And this on a pretty zippy P4 1.6GHz, I can only imagine what it must be like for some people out there with somewhat slower machines (Marc?) So I took a first cut a few minutes ago at re-writing the liveTopics outline generator and did a little benchmarking.  At the moment it takes the Radio outline based generator 64 seconds to generate the XML for my topic liveTopics which is the largest topic in my database at 82 posts. By contrast it takes my hand-rolled OPML generator just 2 seconds to do the same work! It will take a little bit of time to replace the current generator properly but I anticipate the performance difference will be startling. http://matt.blogs.it/entries/00001366.html Manilla gets trackback, what about Radio? Wed, 23 Apr 2003 09:20:07 +0100 <BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"> Next set of features for Manila -- optional human-authored summaries for long weblog posts, along with the option, on a per-post basis to include the summary or the full text in the RSS feed. This is the solution for the conundrum that's been circulating in RSS-Land: should feeds include the full posts, or summaries? The answer is to let the author make the decision, and it's not a global, it's made on a per-post basis. The default is full text so people who do short posts never have to think about it. Along with this we'll have both sides of Trackback. We're <A href="http://static3.userland.com/mtweblog/">testing</A> with Moveable Type, but both sides will work without MT. It's not my #1 feature, but it's clear many users want it, so it's going in. [<A href="http://www.scripting.com/">Scripting News</A>]</BLOCKQUOTE> Sounds to me like both of these are features that Radio users would like too. http://matt.blogs.it/entries/00001428.html Bring back the voices Wed, 11 May 2005 13:53:47 +0100 <blockquote>"Once, on the Merritt Parkway heading for New York, I came upon The American Atheist Hour, the sheer tedium of which was wildly entertaining--there's nobody so humorless as a devout atheist."</blockquote> Via <a href="http://doc.weblogs.com/2005/05/10#hotTalk">Doc</a> I came across a <a href="http://www.thenation.com/docprint.mhtml?i=20050523&s=keillor">piece by Garrison Keillor</a> on the fortunes of Radio and, in particular, good-neighbour radio. <blockquote>"The deregulation of radio was tough on good-neighbor radio because Clear Channel and other conglomerates were anxious to vacuum up every station in sight for fabulous sums of cash and turn them into robot repeaters."</blockquote> Keillor thinks that the days of ClearChannel style Robot Radio are numbered and that the iPod owning, satellite radio listening, generation will tune out - robbing them of their reason to be. Keillor wants to hear voices, all kinds of voices. <blockquote>"I enjoy, in small doses, the over-the-top right-wingers who have leaked into AM radio on all sides in the past twenty years. They are evil, lying, cynical bastards who are out to destroy the country I love and turn it into a banana republic, but hey, nobody's perfect."</blockquote> Keillor doesn't worry about the rise of right-wing radio. He attributes it's growing popularity to a need for constant reassurance and the road rage induced in listeners by traffic congestion. Putting up with it is a small price to pay for all other voices, all the other worlds, radio can tap into. <blockquote>"Here at the low end of the FM dial is a show in which three college boys are sitting in a studio, whooping and laughing, sneering at singer-songwriters they despise, playing Eminem and a bunch of bands I've never heard of, and they're having so much fun they achieve weightlessness--utter unself-consciousness--and then one of them tosses out the f-word and suddenly they get scared, wondering if anybody heard. Wonderful."</blockquote> For me Keillor evokes happy memories of a time I never new in a place i've never lived. Where Radio was something the community used to reach out to each other. At the end of which I knew that my Radio station is right here and I'm talking into the microphone right now. http://matt.blogs.it/entries/00002575.html