Starting June 29, 2005
Installing fiber into my home, "Professional" journalists are amateurs sometimes with podcasting, Podcast with Donna Dubinsky about Palm and more, What if VisiCalc had been patented?, New podcast-friendly version of ListGarden is released, Podcast with Tim O'Reilly, Beta of a new podcast-friendly upgrade to ListGarden, Where is Don Bulens?, Podcast with David Isenberg, Open Cellphone podcast with Tom Evslin, Podcast with MySQL's Marten Mickos, An interactive conference and thoughts on new media, More recordings and creating a wiki to point to them
Installing fiber into my home [link]
I now have the Verizon FIOS fiber-optic-based Internet service in my home. I've written up how this all works (what goes in your neighborhood on the poles, how it goes into your house, what they do, what you get inside, etc.) in an essay with lots of pictures. See "Installing Verizon FIOS fiber-optic Internet service to my house".
"Professional" journalists are amateurs sometimes with podcasting [link]
An observation: From when I first previewed my copyright training video at Demo earlier this year, I've had professional audio people (years in radio, TV, or public address sound) giving me pointers to help me with the sound on my video and podcasts. (I have produced at least three public series, Software Licensing, Open Cellphone, and Open Source SIG, and I've been interviewed on many more.) I've been relying on Doug Kaye of ITConversations more than you can imagine. I've watched a friend spend months and years learning the craft as he moved from being a print journalist to covering a similar beat for NPR. I constantly compare my work to what I hear on the airwaves, in movies and videos, and in other podcasts. I've even spent lots of my own money trying to make as pleasing sound as possible. I'm learning to hear the differences between different microphones, to understand why so much equipment is often used (de-essers, compressors, limiters, filters, etc.). I feel that sound is like layout, typography, writing, punctuation, print resolution, etc. -- it can get in the way when we don't take advantage of what has been learned over the years and when done well it helps get information across better. I'm getting sensitive to levels, clipping, and so much more. (Hopefully some of this is reflected in better podcasts from me, but I'm not always as successful as I'd like. Maybe if I buy just one more piece of equipment...)
With the rush to podcasting by just about every "content" provider, we're now seeing "journalists" who are "professionals" in one medium (usually print reporters or bloggers) trying to publish in another (audio) and they often sound like total newbies. I like the new outlet for information that podcasting provides. I like it that newspaper and other "text-based" reporters are now filing stories in audio form with actual interviews where you can hear the interviewee squirming. But, as I hear the sometimes poor audio quality and blog-like "I'm doing it myself" character of these podcasts I'm reminded of how "professional" journalists just a little while ago were deriding us bloggers as something of poor quality. As I hear them learning-by-doing in the audio sphere, just as we bloggers-turned-podcasters are, I hope they are getting a greater understanding of us and how the rough quality of our blog writing may not reflect on the quality of our thoughts and messages.
I am enjoying some of these "big media" podcasts, such as the Boston Globe Bizcasts with DC Denison, Scott Kirsner, and others. You can hear how rough the sound engineering is. (Scott did a wonderful podcast with me that was posted in early August but recorded last February -- you can hear the echoes in the room and the computer fans blowing and it wasn't just a bit during an intro for effect like the ambient sound NPR edits in. I talk a lot about innovation and inventing and VisiCalc.) Other newspapers are jumping into the fray, such as the San Francisco Chronicle with their "Chronicle Podcasts" (their recording quality varies, too). David Berlind and Dan Farber of ZDNet have the "Between the Lines" podcast series. David and I are frequently comparing notes about equipment and podcasting techniques and he posts what he's learning on his blog. Steve Gillmor's sound quality is all over the map as he experiments on the new Gillmor Gang. The original recordings, produced by Doug Kaye at ITConversations, were quite good and were the eye opener that got many of us sold on this medium.
Some of these people are very good interviewers and it really shows in the podcasts. Scott has done the MITX Fireside Chat series of live interviews that I frequently write about here on my blog (such as with Malcolm Gladwell and Clayton Christensen). Podcasting lets him use that talent for more of us. David Berlind often asks tough, knowledgeable follow up questions as he interviews vendors on his podcasts -- you can almost see the horror on the faces of the PR people in the room with the interviewee on speaker phone. Larry Magid has been a print columnist for many years, but he's also been doing radio shorts for NPR and then CBS for a long time. His interviews for ITConversations have been really interesting (like the ones with Gordon Moore and Buzz Aldrin), but he's been doing daily radio for years.
Larry's style brings up an interesting question. He's been trained to do the "2 minute spot" style of recording with a zesty tone of voice. When you do long discussions with someone, what is the most appropriate tone? What about 10 minute podcasts? Should we edit and add additional laughs like Car Talk sometimes does according to On The Media's "Pulling Back the Curtain"? (By the way, I've always loved NPR's "On The Media" and now that it's available as a podcast I get to listen to it much more -- that's spoiling me to expect all NPR shows to be like that.) Larry and I have been discussing this whole tone thing. As we all experiment with podcasting, what lengths are natural for what type of listening? What style of affect should we use when? Over the years Larry's done live radio, 15-30 second spots, 2 minute segments, and 20 minute longer stuff - what are some of the "right" (most pleasing or helpful to the listener) styles for each genre?
Larry reminds me this moving of many written reporters to audio and video is like the switch from silent films to talkies. Not all great motion and expression actors could sound the way that worked with audiences. New popular players emerged. Being a "multi" media person now might be a valuable thing.
This switch to audio will be tough for some people trained only in writing. With writing you needed to know how to type (a skill) and manipulate a word processor and email and maybe part of a content management system. Nothing that hard for regular people, and most stuff is taught in grade school now. So, the main unusual skills are writing ability and "journalism" training. With audio, most of the reporters who are doing new podcasts need to have and be proficient with lots of expensive fussy equipment, do a new type of editing they don't teach in most schools (for making editorial changes as well as sound changes, compression changes, etc., etc.), speak fluently and clearly (or know how to fit it with editing, just as they often do with print), and more -- lots of geeky skills and a bit of how to "act naturally" on top of it. On the flip side, reporters are finding whole new ways to do their craft of "reporting". They may need additional skills but they have more outlets for their work. Of course for the techies at heart, it gives you a great excuse to learn about a whole new area and get lots of new toys.
It's nice to all learn something new together. We are creating a new medium together, pros and "amateurs" alike. Video may have killed the radio star, but with the Internet the radio engineers are now finding that more people want to learn from or use their skills.
Podcast with Donna Dubinsky about Palm and more [link]
We've just posted another podcast in the DiamondCluster Wavelengths series. This one is an interview with Donna Dubinsky. Donna was the CEO of Palm in the early days when it came out with the Palm Pilot. She then co-founded Handspring and was very involved with the Visor and the Treo. Handspring was acquired by Palm in 2003. She's still on the board of Palm but not actively involved in day-to-day management. John Sviokla and I asked her questions related to her experiences with an eye to learning from history about creating an open platform. Not only do we talk about the "open cell phone" as we do on the other Wavelengths podcasts, but Donna talks about some of the history of developing the Palm as an open platform, how they learned about doing wireless devices at Handspring, what the wireless carriers' mindsets are, and more.
This podcast should be of interest to a wide range of people, including people interested in modern computer history, the Palm platform, open systems, handhelds, the Treo, cellular carriers, and more capable cell phones. If you've never heard Donna speak it's worthwhile to listen to this to hear the voice of a real business pioneer who helped blaze new classes of uses of computing.
One interesting observation related to the "Long Tail" discussion: Palm found out that most people used some of the core applications shipped with the device but that customers would also find one other application that was compelling for them and that was a "make it or break it thing" (listen at about 3:00 in the podcast). "There was always some additional compelling application for the Palm owner," she claims. A dedicated device without general programmability would do the core applications (like calendar) but not that one needed application (like connecting to a corporate database or tracking the stars). This is a step-function that relates to the Long Tail -- it's not that a large number of long tail apps added up to lots of "extra" money but rather they drove consideration of a product at all. This is a key aspect of understanding open systems and the Long Tail. The interview has additional important observations based on history, too.
See "Show #4: Donna Dubinsky, ex-CEO of Palm and Handspring" for information on how to download this 41 minute podcast. You can also subscribe to the whole series with the RSS feed.
What if VisiCalc had been patented? [link]
My name has been brought up a few times in the last few weeks with regards to software patents. I publicly questioned a Microsoft representative in Zaragoza who was saying that software patents were good, asking him how patents played a role in Microsoft's early success (since they held very few patents until the early 1990's). You can read Jackie Danicki's write-up of the exchange (search for "bricklin" on the page) and you can listen to it at minute 24:25 of the MP3 of Simon Brown's session from the conference on the page that Jackie points to.
I was a source for Randall Stross' article in the July 31, 2005, Sunday New York Times which has been picked up in a variety of places. (I also helped him find out how to get the information for the chart there that shows Microsoft had only a single issued patent by 1987 with company revenue of $350 million, 5 by 1990 with revenue of $1.2 billion, etc.) I pointed out that the software industry has used copyright and trademark for IP protection very well (both are keystones of the proprietary and the Open Source software worlds). The main quote: "Isn't Microsoft the poster child of success without software patents?"
The issue comes up again in a recent piece in ZDNet/CNet "Open-source allies go on patent offensive" where my name is mentioned on the second page in discussing what the effect would have been of patenting VisiCalc. (See my old essay "Patenting VisiCalc" to understand why we didn't.)
This morning a Newsgator custom keyword search feed found my name mentioned on Russ Krojec's blog entry "What if VisiCalc was Patented?" Russ argues that since VisiCalc wasn't patented competitors found it "...safer to copy the currently winning formula and avoid having to innovate. In this case, the lack of patents brought innovation to a standstill and we are all running spreadsheet programs that still operate like 25 year old software."
I beg to differ with him. He seems to assume that people haven't tried really hard to find something better. I posted a comment to his blog (as of this writing it's awaiting anti-spam/troll verification):
There actually were many different "non-VisiCalc-like" calculating systems developed over those years. We even tried one at Software Arts with TK!Solver, and Lotus tried Improv. Then there was T-Maker, Javelin, etc., etc. Many of us tried to find new metaphors. They didn't catch on (there are reasons why, I believe, but it isn't for lack of trying). On the other hand, innovation in VisiCalc-like spreadsheets continued, with Lotus doing things we wouldn't, and then Microsoft moving things further ahead with Excel going in areas Lotus neglected.
When Mitch Kapor did 1-2-3, he copied the features of VisiCalc he thought were worthwhile and didn't (or changed) features he didn't think were appropriate. Microsoft added features (and Windows support) that Lotus held back on.
Innovation sure didn't "stand still". There was no "avoiding" of innovation.
I'm not against patents in general (they are good for some industries, I guess), but I do have real problems with how they are affecting the software industry which has other means of protection and incentive that have proven successful to society. Of course, as I've written, they are the current law of the land and I still apply for them at times. In fact just today I received one of those "It is my great pleasure to congratulate you on the granting of your patent!...Celebrate...by ordering a patent plaque or frame today" letters as another patent from the Trellix days finally came through this week (we applied in 1996). (I don't own the patent -- Interland does.)
The topic seems hot, so I figured I should post this, but I've been talking about software and patents for over 15 years so I am a bit burnt out about it. In my copyright video I even have a very short clip showing me testifying (along with Mitch Kapor) in front of a Congressional subcommittee in 1990 on the topic (which I'll get onto the Internet one of these days -- the complete tape of the hearing is hours long).
New podcast-friendly version of ListGarden is released [link]
I've just released the new ListGarden RSS feed generator Version 1.3.1. This is a major upgrade with lots of new features. It makes it very easy to create podcast RSS feeds, including automatically generating an HTML page with a "List of Shows" and links to the audio files like the one I use for my Software Licensing podcast. (There is a step-by-step tutorial on how to do this -- see "Setting Up an RSS Feed for Podcasting Using ListGarden".) The product is still free Open Source released under the GPL and runs on Windows, Mac OS X, and Linux, client-side or server-side.
I understand that many people create their podcasts pretty much by hand, uploading the audio files by FTP (easy) and then manually editing the RSS XML (tedious and error-prone). Some use external services to turn a blog feed into a clean podcast feed. The new version of ListGarden is a way to maintain the RSS feed more automatically yourself -- you just browse to a list of uploaded files and click on "Select" to fill in the enclosure information.
The new features include support for enclosures, support for extra XML such as that used by iTunes, and automatic backup to a local file and/or the server.
One special new feature, not in the beta version, is a way to download option settings from a web page. ListGarden has lots of option settings and some fields need HTML and other templates. This feature let me put sample settings on a web page and have you add those settings by just entering the URL. This made the tutorial much less tedious. It also lets anybody create such templates, much like templates for blogs. Since ListGarden can create HTML files (like the companion List of Shows file) this will make it easier for VARS and others to provide support.
See "Version 1.3" for a detailed list of the new features and where to download.
Podcast with Tim O'Reilly [link]
The fifth show in my Software Licensing podcast series is now available. I talked with Tim O'Reilly, founder and CEO of O'Reilly Media. The reason I first asked him to do a podcast was to discuss O'Reilly's policy for copying sample code from their computer books. They publish something like 500 books with a total of about 2 million lines of sample code and most programmers I know refer to at least some of those books periodically. Most of the books I checked don't tell you what the copying policy is. In 2001 Tim posted an informal description of their policy on the O'Reilly website which I assume most people have not read or don't find easily. I felt that it was of value to my listeners (which include corporate attorneys who like to know these things) to understand what the policy is and the philosophy behind it. While "fair use" law cover this, the ethical thing is to know the copyright holder's intent, too, in figuring out how to apply that law.
The discussion of the policy takes up the first 10 minutes or so of the podcast. The rest has discussions of O'Reilly Media's experience with copyable and online books and the effects of piracy, Tim's thoughts about the value of openness and the "architecture of participation", where there's value in the Open Source ecosystem (who will be the "Dell" of Open Source), the balance between what you own and what you give away, and more. If you've heard Tim speak recently some of this will be familiar, but the difference here is that it is a conversation (literally) so he probably goes into more depth on some things and covers a wider area of topics in response to my questions. If you are interested in Open Source or business models and haven't heard Tim speak, it's definitely worth the time. He's runs a for-profit business that is heavily involved in the Open Source world and has real data on what works and what doesn't.
See "Interview with Tim O'Reilly, O'Reilly Media" for the details of downloading the podcast or subscribing to it.
I've been really pleased with the reaction to my Software Licensing podcast series. People especially comment about the interviews with Marten Mickos of MySQL and Linda Hamel of the Commonwealth of Massachusetts ITD. I love how podcasting works where I just upload the MP3 file and update the RSS feed (using the hopefully final code I finished this morning of my podcast-friendly ListGarden upgrade, of course) and, voila!, dozens of people download it in the minutes before I can even let you know about it on this blog (thanks to podcatchers iTunes, iPodder, iPodderX, etc.). I love how I can see that listeners appreciate what I'm doing well enough to subscribe. A "page read" is one thing as positive feedback, but a "subscribe" is even better.
Beta of a new podcast-friendly upgrade to ListGarden [link]
It's been almost a year since I last upgraded my ListGarden RSS creation tool. ListGarden has gotten great reviews, has been downloaded thousands of times (the rate has not dropped off), and is apparently being used by many people. ListGarden is pretty specialized because most people who create something that uses RSS (a blog or a database-driven site) have the RSS creation built into their system. The program been most helpful to people like me who use non-blogging tools to create their blog (like Trellix Web or Notepad). Being released under the GPL has helped it stand out (and help me learn about the GPL firsthand in a more meaningful way). I've just added some features that may make it helpful to a new set of users.
When it came time for me to post my first podcast I realized that ListGarden couldn't help me create the type of RSS feed I needed to be a real, "podcatchable" podcast. It didn't support the <enclosure> element. Editing XML by hand is tedious and error-prone, so most people like using an automated tool. Phil Windley created a patch to make ListGarden work for him (that's what you can do with Open Source...) but I wanted a different UI and, anyway, there were other features and bug fixes I wanted to add, too. It was time to upgrade ListGarden. Before I started my podcast series I did some programming and added support for enclosures and a way to browse a directory on a server to choose the file and automatically fill in the file name and size (I assume that you can use a normal FTP program to upload the file). Since May I've been using that version for all the podcasts I've done but I didn't finish the coding so that I could release it to the public. (I did find and fix some bugs during that time, though.)
Today I finally released the upgrade, including a variety of other features I've added in the past weeks. It is a beta version and there is barely any documentation for the new features, though for people who know podcasting RSS already that shouldn't be much of a problem. Over the next few days I'll upgrade the existing documentation and add new documentation about podcasting so you won't need to know any of that.
The special podcasting-friendly features include: Support for enclosures, the ability to use namespaces and add additional sub-elements to both the channel and the item elements (which lets you, for example, set the new values that iTunes takes advantage of), and the ability to maintain an HTML-format list of shows on the website along with the XML-format RSS feed (for listeners who don't use podcatching software or when you want to link to an entry on a webpage for a single show).
Other features: The ability to automatically save backups of all the data both locally and/or on the server, and a simple non-HTML way to markup the descriptions (this lets you podcast with formatted entries without needing to create a companion blog or know HTML).
For more information on this beta version, see the ListGarden News page on the Software Garden website. I'd appreciate any feedback.
Where is Don Bulens? [link]
Regular readers of this blog have heard me refer many times to Don Bulens who was CEO of Trellix for many years. After I left Interland/Trellix last year I haven't mentioned him. Periodically, though, people ask me where he is (he has lots of friends in the industry). I've kept up with Don at least every few months since I stopped working with him and here's the info:
Soon after I left, Interland scaled back the Massachusetts office where I had worked (I guess independent of me leaving). It's still there (at a slightly new location a block from the old one) and the complete service Don led that uses our website creation system is alive and well and contributing to their bottom line (as I read their latest PR about their relationship with Dex). Don left a while after I did and then did some work with various VCs and considered a variety of jobs (and finally stopped spending 60% of his time in Atlanta and being away from his dear family).
In March of this year Don joined EqualLogic, Inc. as president and CEO. EqualLogic makes intelligent, all-inclusive iSCSI storage area network (SAN) solutions. That may sound quite different than doing website development software, but Don has always been a whiz at OEM and VAR relationships -- he got the Lotus Notes VAR channel going, for example, and got the old Trellix Web Windows client onto millions of OEM products and the server-based tool used by many well known brands. He likes being a CEO building a company and its sales channels. He also likes it that the company makes products that try to simplify things in a complex area, just like Trellix did. As I see it, EqualLogic is also in a much more family-friendly location, just over the border in Nashua, New Hampshire (Don lives north of Boston), so it's a win all around.
So, for those of you who have been asking, now you know where to find him. Good luck, Don!
Podcast with David Isenberg [link]
The third "Wavelengths" Open Cellphone podcast is now available. This time John Sviokla and I talk with David Isenberg about his "Rise of the Stupid Network" essay that he wrote while at AT&T, the democratic principle of the "Freedom to Connect", cities putting in their own connectivity infrastructure, the turnover in buying new cellphones, and more.
See "Show #3: David Isenberg".
Open Cellphone podcast with Tom Evslin [link]
We've just posted the second show in the "Wavelengths" Open Cellphone podcast series. DiamondCluster's John Sviokla and I talk with Tom Evslin about getting a big company to do new things, what he learned at AT&T starting WorldNet and at a successful Voice Over IP company, about the problems of vertical integration, and much more.
John really like a metaphor that Tom mentioned about why parts of a vertically integrated company often have a hard time competing with those parts as used by (and sold to) a non-vertically integrated company: It's about "surface area", meaning how exposed the managers are to customers and markets. As Tom wrote on his blog last February in "AT&T: Lesson From the Crypt #3: Vertical Integration Doesn't Work Anymore":
Vertically integrated companies can't compete! The oxymoron of “internal customers” is poison to a competitive culture. That is the lesson of the computer industry and it is a lesson the telecommunications industry apparently has not learned yet...
A horizontal company has a high surface to volume ratio. It sells all of its outputs in a competitive market. It is free to buy all its inputs in a competitive market. Its managers are not isolated from the markets they compete in. A vertical company spends most of its time and energy dealing with itself rather than the external market. Meetings are dominated by esoterica like transfer pricing between divisions because this is what determines internal success. The real marketplace is distant from most of the managers trapped inside the vertical structure. Buy vs. build and capital allocation decisions inside a vertical company are made by office politics in a vain attempt to optimize across the whole vertical organization. Horizontal competitors optimize only for the layer they are competing in and so end up being superior to the vertically integrated company layer by layer.
Podcast with MySQL's Marten Mickos [link]
I've posted a new podcast in the Software Licensing series. This one is an interview with Marten Mickos, CEO of MySQL AB. MySQL is a very popular database released under the GPL license that is also available under a "dual license" with a more traditional proprietary license that doesn't have the GPL "redistribute derivative works under GPL, too" requirements.
I ask people about pronunciation before I start recording. In the USA, Marten will accept most reasonable pronunciations for his name -- like the "Mahr'-ten Mee'-kohs" I think I used -- and the original pronunciation for MySQL is to spell it out -- My-Ess-Que-Elle -- but the customer is always right, so "My-See'-quill" is OK, too, if that's what you want.
The interview starts with about 15 minutes of talking about the dual licensing they have and the question of how to tell when you can't use the GPL license and need the commercial one. I hope this is helpful to the lawyers in my audience since the company does not like to say much about this. We then talked for the rest of the almost 40 minutes about business models, his feelings about Open Source, ensuring ownership of copyrights (in an Open Source project and in general), software patents, the fact that about 40% of MySQL's active installations are on Windows computers, various segmentations of their market with respect to different licenses, and whether an Open Source-based ISV can make a reasonable amount of money in the eyes of investors.
An interactive conference and thoughts on new media [link]
The response to Friday's conference has been phenomenal. There was real energy in the room. When asked, at least half of the people in the room said they'd written code in the last 6 months. Almost everybody reads Groklaw. Dozens of people participated. It was this participatory, interactive nature that keeps coming up in descriptions. Pamela Jones of Groklaw, who did a long write-up of the meeting after listening to the 4 hours of recordings, mentions that multiple times.
I tried hard (with the help of the other organizers) to make the conference work this way. I had gone to other events where there was little time left for the audience, and even that was for "questions" for the "esteemed panel" to answer. I've also gone to two of Dave Winer's BloggerCons which are mainly audience give and take. Dave's been pushing that format very strongly recently (thank you, Dave! Yet again you help the rest of us and make a big mark on the world), and I decided to get some flavor of that here, while trying to respect the fact that some of the invited panelists came great distances and should get a reasonable amount of time to put in their comments and that we only had 1 hour for each part.
We posted in advance that "It is expected that all attendees will actively engage in collaborative discussions with panel members." We also told the panel members to encourage participation and only gave the big panel (with 5 speakers) a few minutes each to do an introduction, not the normal 10 or 15 that would have taken up all the time. Attendees paid attention to this and constantly had things to say. I'd run over with a handheld mike (which kept things under control so basically only one person was talking at a time) which made it obvious to whoever was speaking that it was time for someone else. Body language of a moderator can help communicate without needing to break in while someone's speaking, and keeping the mike in the moderator's hand let's you fade one person out and fade in another.
The other thing about this meeting, I think, is that the attendees were mainly all used to the Open Source world which has a lot of participatory activity. Also, the area is so broad that for almost any question a different person was the local "expert". So the style fit the topic of the SIG.
In today's world, the fact that everybody can participate is now a given. I view myself as a toolmaker and look to tools. The tools of the old world were for "broadcasters" where one small group (but not too small -- it needed "production" people, expensive equipment, and capital) tries to provide something that the general population "shares" by listening. Think: newspaper, radio, TV, books, magazines. The tools of today are for individual expression where anybody can decide what they want to communicate or listen to and there is no requirement to try to make it "broad" and interaction has a much lower threshold. The tools are designed for individuals with almost no money, but often scale to large groups. Think: the web, blogs, email, IM, podcasting, cellphones, search, wikis. Google searches are all "long-tail". Cellphone calls are all long-tail, not like top-40 radio. Even listening on music players and the Internet is long-tail: In an interview with Larry Magid on ITConversations, RealNetwork's Rob Glaser said [at minute 2:45] that in a given month over 90% of Rhapsody's one million songs are played at least once and the top 100 songs make up only 1% of the listens. I wonder how identical people's music libraries are if you look at their high-capacity music player. Probably many quite different mixes and that personal part is what makes an iPod so popular and better than radio.
The "media" of yesterday were producers of broad content. The "media" of today are really tools for individuals to communicate with few or many people -- the tools scale and the thresholds are very low. The tools assume a long tail and are not area-specific for "broad interest content". So Google (and others) with search (a "newspaper" or "book" with very narrow grained material that has more or less whatever you want to read, not just what someone decided you want to read today), email, and blogging (a newspaper that's as easy to maintain as sending email to one person so it's worth it for one or ten or a thousand readers) is a new "media" company in that media means a way to read or hear what we want from others. Wikis are "books" that are easy for a group to write, either very small or large.
Back to Open Source, Doug Heintzman of IBM said [at minute 43:45 of the "Business Models" session and on BetaNews] that they're spreading the collaborative, "bottoms up" development process, culture, and tools to their other development projects because it works better than the siloed ways of the past. On my trip to Europe, I was reminded how many cultures and subcultures there are and how rich each is. I saw people interacting with others in so many ways. Nobody was listening to a radio -- they were all talking to others, texting, immersed in their personal worlds composed of so many diverse parts. Broadcast, one-way tools and thinking don't work there. Personal expression tools do and, like at IBM, they will change the world for the better to be more from us and, hopefully, of what's best in us.
More recordings and creating a wiki to point to them [link]
I've finished processing the three morning sessions of the Mass Software Council Open Source SIG meeting and posted them online. They are really good and the reviews so far have been quite positive. The main recording left to do is the lunch keynote, but I'm not sure yet if that one will be posted. To see the full list of what we have, go to the "Kickoff Meeting" page. To follow additions, check out the blog.
The constant addition of recordings, associated slide files, links to reports in blogs, etc., fit in well with the blog I set up to notify those that are following along. A blog isn't, though, a good fit for people who come in later. It would be much better to have an organized webpage with everything neatly formatted. This is a volunteer-run organization with many contributors, so a centrally-controlled website would put the burden on the webmaster (who at present is me). At the meeting, a common refrain was the move to using wikis for organizing stuff for a group. The combination of blog and wiki looks really good.
Until now I've never created a wiki of my own, but I really had no choice but to use one here. I searched a bit on the web and found good reviews of UseModWiki. It's written in Perl, is released under the GPL (good for an Open Source SIG) and is easy to install on a web server (no need to connect to databases, it's basically one Perl file, etc.). I just downloaded the zip file (which assumes you will be on a Windows server, which is not what I was going to do), edited the first line of the program to point to Perl, edited the line with the location of the data file, uploaded it to the server (using "ASCII" mode!) which is a normal inexpensive shared-hosting account, set a few permissions and created a data directory using an FTP program, and Voila!, a wiki. I then edited the sample config file to change the wiki name, added some passwords, and turned on calculating file differences, and then uploaded that to the data directory. Finally, I started creating the actual pages, a home page and one for the Kickoff Meeting, using my browser. Pretty easy. Of course now I'm itching to make a custom CSS file for the site, but I'm resisting the best I can.
Quick and dirty direct edit web authoring is a major win for collaboration. Successful examples are blogs and wikis. In the Business Models session of the event, Doug Heintzman from IBM told us how IBM has been learning from the Open Source development model and that they're implementing it for other projects, even those that aren't with an Open Source license. The Novell people told how they've moved to wikis for project coordination.
If you are interested in Open Source, I think listening to the recordings is very worthwhile. They can be listened to in any order. If you are wondering how an Open Source project is really organized, or what it's like to turn a proprietary software project into an Open Source one, at least listen to the third session. I found it fascinating.
© Copyright 1999-2014 by Daniel Bricklin
All Rights Reserved.
See disclaimer on home page.