Isaac Asimov - The Relativity of Wrong

The young specialist in English Lit, having quoted me, went on to lecture me severely on the fact that in every century people have thought they understood the universe at last, and in every century they were proved to be wrong. It follows that the one thing we can say about our modern "knowledge" is that it is wrong. The young man then quoted with approval what Socrates had said on learning that the Delphic oracle had proclaimed him the wisest man in Greece. "If I am the wisest man," said Socrates, "it is because I alone know that I know nothing." the implication was that I was very foolish because I was under the impression I knew a great deal.

My answer to him was, "John, when people thought the earth was flat, they were wrong. When people thought the earth was spherical, they were wrong. But if you think that thinking the earth is spherical is just as wrong as thinking the earth is flat, then your view is wronger than both of them put together."

The basic trouble, you see, is that people think that "right" and "wrong" are absolute; that everything that isn't perfectly and completely right is totally and equally wrong.

However, I don't think that's so. It seems to me that right and wrong are fuzzy concepts, and I will devote this essay to an explanation of why I think so. ....

The entire essay by Isaac Asimov is really worth a read, as it provides a very nice look at what wrong means in science, which is often different than how it is used in common language.  It might be another case where part of the confusion between science speak and common language is the words don't quite mean what we think they mean.

Maybe it's citizen anchors, not citizen journalists

I found this really insightful.

We like to say new media is allowing us all to be journalists. But it’s
probably more accurate to say it lets us all be anchors. Sure, the
Internet also allows people with local knowledge or serious expertise
to speak directly and be picked up by a wider audience, but it doesn’t
fundamentally do a whole lot to increase the population of those people.

And I love his proof point of Palin as the main example

That makes Palin the perfect post-postmodern politician, in a way: A
totally self-contained text, a signifier with no referent. You don’t
really need to know anything to love her or to hate her, because she’s not about
anything except… Sarah Palin. Obligingly, she places no demands on
either her supporters or her detractors, because what they decide to
think of her is all they need to know to decide what to think of her.
At the center of her media narrative is… the media’s narrative about
her, bouncing down an infinite corridor of mirrors. If Jorge Luis
Borges had a talk show on a cable channel run by M.C. Escher, it would
look like CNN right now. Welcome aboard the Goodship Palin, now sailing
from the desert of the real.

What I'm listening to

After I got my HTC Hero and purchased DoggCatcher, I've now got this entirely seemless podcast experience as my phone grabs the newest stuff any time I'm on wireless.  It caused me to do some trimming and tuning on what is being pulled down as I work towards having approximately the same number of hours of podcasts a week as I've got car rides and yard work.  For those interested, here is what I'm listening to now.

  • Drunk and Retired.  I honestly can't even remember how I originally found this podcast, it must have been through something ruby related at one point.  This is Michael Cote's original podcast with Charles.  They talk about random tech topics, zombie topics, now some fatherhood topics, but it's all around interesting, and lots of fun to listen to.  Charles has some really great insights on development, and I've loved listening to his various journeys through ruby, java, and javascript.  I enjoyed D&R enough that I went back to the beginning and listened to the entire backlog.  I even really enjoyed the 4 hour scotch tasting episode... and yes, your read that right.
  • IT Management Podcast.  Once Cote got hired by Redmonk, he started producing more podcasts covering a range of subjects Redmonk was doing analysis on.  I tried most of them, but IT Management Podcast is the only one that stuck.  John Willis has a real wealth of knowledge in the industry, and very recently got hired by Canonical on cloud strategy.  The podcast is honestly more cloud than "good old fashion IT management", as the boys would say, but that suits me just fine.  If you are looking for a good source on what's new in cloud computing, this is one of the best out there.
  • AstronomyCast.  Once I got my telescope this was suggested to me by a friend of mine.  AstronomyCast is incredible, just flat out incredible.  Each of the 162 30 minute episodes to date takes on a specific topic in Astronomy and gives it a really wonderful treatment.  The back and forth being Pamela and Frasier is really fun to listen to, and the do a great job of making the information very accessible.  Honestly, anyone above the age of 12 could probably get into this quite easily.  After listening to the entire AstronomyCast back catalog I feel like my knowledge of Astronomy is now at quite a reasonable level.  This podcast also has the added benefit of being very wife friendly.  It is one of the few science or technology podcasts that Susan is eager to listen to when we are doing a long drive in the car.
  • Slacker Astronomy.  Before I found AstronomyCast I found Slacker Astronomy.  The slackers publish far less often then AstronomyCast, and it is a more free form model.  To get into Slacker Astronomy you need to know a few more things about physics and astronomy out of the gate, as there is less explanation there.  I appreciate that though, and am always psyched when I find out there has been another slacker podcast posted.
  • 365 Days of Astronomy.  In case you didn't realize, this is the International Year of Astronomy.  This podcast was an interesting experiment to get volunteers from all over the world to put together 10 minute podcasts and have a new one every day.  They aren't all great, but there are a lot of great ones in there.  And the breadth of volunteers and subjects is quite nice.  Even if you aren't into astronomy, check out this one epsiode on Roswell that really debunks a lot of the story and timelines associated with that myth.
  • Geologic Podcast.  The theme song in front of the 365 days of Astronomy was done by George Hrab, and eventually I decided to check out his podcast.  It's not about geology.  George is a musician, a skeptic, and all around great story teller.  He also has a sense of production values for this podcast that rival some of the greatest radio being done today.  The Geologic podcast airs once a week, and the moment I get my new episode I jump from whatever I'm listening to and flip over to it.  The last time Nick and Heather were around I put in on in the car, and it happened to be the episode that starts with worst gig ever.  I'm not really sure how we managed to stay on the road as we were laughing so hard during that story.
  • Radio Lab.  Radio Lab is This American Life meets science reporting.  It tells some really compelling stories that all have some element of science in them along the way.  I'm also really happy that it gives Robert Krulwich an outlet again, as I've always really appreciated him as a popular story teller.  I find him far more compelling that Brian Green or Neil deGrasse Tyson as a popular voice for science.  I do realize that he doesn't have the same credentials, but his ability to do outreach is far greater.  Radio Lab is a WNYC production that is done somewhere irregularly, but great none the less.
  • This American Life.  It is probably hands down the best radio being produced today.  A bad This American Life episode is still quite good.  A good one... will make you weep, laugh out loud, and totally rethink some opinion you've held, all in the same hour.  I'm really happy that the folks at This American Life let the podcast out there, because many other shows of that quality on NPR have not.
  • Planet Money.  Spawned out of the great This American Life reporting on the financial crisis last year, Planet Money was an attempt to bring a 3 times a week podcast out there that tried to explain economics for mere mortals.  Planet Money was strong out of the gate, but had some issues finding a voice after 4 months of the financial crisis.  Fortunately, they eventually did, and really broadened their approaches to exploring economics as applied to many fields.  The many episodes they've done over the last 4 months on the economics of health care in the US have been incredible, and really informative, and often surprising when you start to understand the complexities in the current health care system.  It also gives you a much more nuanced understanding to what a "free market" means, because every market is just actors inside of constraints.  Economics is really about understanding how those constraints (be they incentives, regulations, taboos) change how the actors interact.  The Pirates Have Timesheets episode gives you a really nice example of that.
  • Wait wait don't tell me.  This is the NPR news quiz show.  It's sort of daily show light and airs weekly.  I get it as a podcast mostly because 11am on Saturday is a dubious time for me to be near a radio.  If we're working around the house, I typically listen to it live, if not, it's podcasted for me to enjoy it later.
  • The Media Project.  While On the Media is probably the more popular national show on this subject, I really like the take our local public radio station does with this.  Sometimes I'll catch the Sunday rebroadcast live, if not, it's on the player and I listen to it that way.
  • This Week in Science.  This was a suggestion in DoggCatcher, and the first episode I listened to seemed quite good.  I think this one is sticking around.

Things I used to listen to but gave up on for one reason or another

  • LUG Radio.  This show was great, but it ended.  Damn you!
  • Linux Outlaws.  While it's still probably the best Linux show out there, it's far too dry for me.  I know a lot of folks that like this more than LUG Radio because it's PG language instead of a hard R, but I don't mind vulgarity at a certain level in my podcasts.
  • Security Now.  I really felt like the information density was too low in this.  While I do like Leo... the other guy got on my nerves from time to time.  I know lots of folks that like this podcast, I'm just not one of them.  Perhaps if you were more of a Microsoft user it would be more relevant.
  • Floss Weekly.  I was pretty frustrated by the treatment of Justin on the OpenSim episode.  I just don't think it's right to put someone into apology mode on an interview show about the platform their project is written in, especially a show that gives smalltalk a pass.
  • TWiT.  John Dvorak is always wrong.  I'm now convinced that is an axiom of the universe.
  • Google Developer Podcast.  This had similar issues to a lot of the Google developer bits on youtube.  Informative but bland and too much reading scripts.
  • The Gaurdian's Tech Weekly Podcast.  This had exactly the opposite issue, it was too over produced.  You could hear them watching time codes the entire episodes.  One of the beauties of podcast medium is that if it's 28 minutes or 34 minutes, it doesn't matter.  You don't need to rush something, or cut it off, to fit into a standard time block.  They seem to have missed this memo.
  • Some Ruby Podcast.  Honestly, I don't even remember which one it was, but it had the same issue as Linux Outlaws.  High on dry facts, low on interesting stories or banter.  You really need to be a multi level black belt at podcasting before you are allowed near a sound effects board, and these guys broke that rule.

I'm sure I've forgotten some along the way, and I'm sure my listen list will be a bit different in a year, it always is.  I'll actually be interested in revisiting this post in a year and see what's changed.  Perhaps this might help you discover some new things to listen to.  If so, or if you have other suggestions for your favorite podcasts, or this inspires your own write up of what you are listening to, I'd love to get a comment from you.

P.S.  Hope all you Americans out there have a happy Turkey Day tomorrow.  Turkey and Cranberries are about the best things in the world as far as I'm concerned. 🙂

Updated (11/28/2009) to be more specific on my criticism of FLOSS weekly, as Randal correctly called me out for having just a vague slap down.

I have a theory... no you don't, you have a guess

One of the reasons that many people have a misunderstanding of evidence in science has to do with the differences in terminology used by scientists and by the general public.  The biggest misunderstanding is around the word theory.

How many times have you, or someone you known, come upon a situation they didn't fully understand and then propose an explanation.  They'll typically start that statement with "I have a theory...".  But they don't have a theory, they have a guess.  Out of some random facts that happened to be around they guessed at an explanation.  It might be true, it might not, there might be a lot of missing facts they used in the process.  They may have cherry picked facts, either intentionally, or just because of some other built in bias.  It's just a guess.

The truly diligent folks might come up with with a way to test their guess, that makes it a theory, right?  No, it doesn't.  If they come up with a set of tests that would either prove or disprove their guess, that makes it a hypothesis.  This is where science begins, an explanation to a phenomenon that has a way to prove or disprove if it is true, and that could predict other behavior in the future.

If you have a hypothesis, and rigorously test it over and over again, creating multiple experiments with control groups, publish a number of scientific papers reviewed by your peers, get feedback from a great deal of other people proposing counter hypothesis, and yet it remains widely believed by experts in the field that your hypothesis best fits all the data at hand... then, and only then, you get to call it a theory.

My own thoughts Google Chrome OS

When google decided to create their own web browser a year ago, I was both sceptical and annoyed.  Did the world really need another web browser?  Couldn't they just put that effort into Firefox?  What I hadn't realized at the time was that Google was about to push the boundaries of what a web application is, and their is no way they could pull that off without some control of the browser.

Google has this really cool tech called GWT.  It basically lets you write a rich web application in Java and compile it into a AJAX application.  Javascript as machine code.  The cross browser stuff is in the compiler, so you don't need to think about it when you write you application in Java.  If your whole business is writing really complex applications for the web, something like GWT is really a necessity.

If you want to see how far GWT can take you, check out Google Wave.  Real time key by key synchronization across the web to multiple users at once.  This isn't the web we are used to, it's something a bit different.  Once you start trying to use Wave for real, with a couple of people, you find out that your browser is totally overwhelmed by that much javascript.  Run it in Chrome, and life is a lot better.  Things load faster and are responsive in the way you expect a desktop application to be.  Seeing Google Wave in Chrome makes you start to realize the the idea of desktop on the web isn't such a crazy idea.

The pundits have largely been missing the point.  While browser as operating system has been a concept since the late 90s (which was skillfully retarded by Microsoft by bundling a web browser that could never support that level of complexity), things are different now.  Over the last 5 years google has been slowly rolling out a set of applications, free to use, that already moved a lot of people fully to the web.  It's hard to find someone without a gmail account at this point, and most of them are using it as their primary and only email.  Google docs is really nice, and I found that far more useful to collaborate on all my interactions with the outside world than the old "email a word doc" model. 

I get that a lot of people fear the cloud, and point to Microsoft's fiasco with with Danger as a reason to distrust the cloud.  But Google isn't Microsoft.  And more importantly, you know what happened before, people lost data.  Never in the history of computing up until the Danger computers crashed did people loose irreplaceable data on a computer.  Much like the fear of flying overwhelming people in a way that the much great risk of driving to the local store to get Milk doesn't.  People with just the wrong amount of knowledge make very odd risk assessments.

The thing I'm most thrilled about is that Chrome OS is going to help us keep an open web, as least based on everything I've seen now.  It's Linux, and the entire stack is going to be open.  That means it's not going to support all the rich internet application alternatives being pushed by Microsoft and others.  Chrome OS is going to drive a lot more open standards on the web than any set of committee meetings ever would.  And that's good for all us, regardless on whether or not we're using Chrome OS to access the web.

Staying connected while you are away, using an IRC Proxy

Internet Relay Chat (IRC) has been around for just about ever.  In the 90s it was used for chat rooms and warez sites mostly.  For the past decade it's become one of the key pillars of communication for the Open Source Community.  It has the advantage that the client and server are free and open, and there is an inherent redundancy system built in.

One of the challenges of IRC, over say email, is that you need to be online to see the discussion.  On a really global project this is a problem, because of the pesky fact that daytime is determined by facing the Sun, and living on an orb, only 1/2 the earth gets to do that at a time.  Life would be so much simpler if the flat earthers were actually right.  But there is a way to stay connected, even when you are not, which is using an IRC proxy.

Using an IRC Proxy

The get started you will need a Linux machine, that you can run code on, that is always on.  It doesn't need to be on your network, but you need shell access.  If you are a Linux geek level 4 or higher, this is probably not an issue.  You probably either have a Linode or a home server that's always on.  If not... well sorry... your journey ends here.  There is not, as of yet, a cloud service to provide this for you.  Please come back once you level up.

The next step is the actual IRC proxy.  IRC is a simple enough protocol, which goes over clear text, that many people have written a man in the middle server for it.  You connect the proxy to the IRC server as you, then you point your IRC client at the proxy.  When you are connected to the proxy, everything works as normal.  Your messages are sent back and forth in real time.  When you disconnect from the proxy, the proxy keeps you logged into irc and logs everything that goes on.  The moment you reconnect to the proxy all those messages are replayed to your client.  You now have a full offline ability.

My favorite IRC Proxy

There are many out there.  A few years back I spent some time trying to get one that I didn't hate, and I landed on miau.  I've even packaged it for ubuntu, so if you are on that platform, it should be easy to install.  Once installed, read the same miaurc on configuration, it's really well documented, and should be easy enough to get rolling.

Although miau supports a password to connect to it, I don't really trust running another service connected to the internet that just has a password in clear text.  My solution here is to have miau only listen to localhost (127.0.0.1), and ssh proxy to the machine.  Pick a port (like 4098) on your local machine and have that forwarded whenever you connect to that server.  In linux this would look like the following in you .ssh/config.

Host your.server.name.com
LocalForward 4098 localhost:4098

The have your IRC client (like XChat) connect to localhost:4098.  This will mean that you will only be connect to IRC when you have an ssh link to your proxy server.  It works quite well, and is about as secure as you'll get.

Why bother?

If you made it this far, you probably already know why.  When development conversations happen at 4am your time on IRC, you are probably never going to participate directly.  But, having access to the conversation when you connect in the morning is a very good thing, and I've walked people through this setup enough times in the past, writing it down for posterity seemed like a good thing.

Review: Memories of the Future

I just finished Wil Wheaton's Memories of the Future Volume 1, and hope that Wil get's writting fast to get Vol 2 out there.  This book is just too damn fun.

The book is an episode by episode look at Star Trek The Next Generation, wherein Wil provides a 6 - 8 page synopsis of the episode in the way that the Mystery Science 3000 folks would do one.  It's incredibly funny, and has lines like "well as long as we're not advancing the plot, why don't we do a pod race?".  For anyone that watched ST:TNG growing up, this book is a really amusing look back, especially on all the uneveness of the first season.  Each episode also then has Wil's favorite quote from it, the obligatory technobabble, and his personal memories of shooting the episode. 

Vol 1 covers up through Datalore (the first 1/2 of season 1), and is constantly making comments about the disaster which is Angel One.  I can't wait for Vol 2 which is going to start us there take us to the end of season 1.

Wil is really an incredible writer, and his great sense of humor comes through in this book in spades.  After reading Just a Geek in the spring, I was happy to pick this book up.  Honestly, I found it hard to put down.  It's just too damn funny.  If you had any opinion at all on ST:TNG (loved or hated), do yourself a favor and get this book.  You will not be disappointed.

Mediawiki vs Drupal for a community site

After our android hack-a-thon, Frank put a page in the MHVLUG wiki to start to stub out Android information that we're all finding useful.  It's small, and largely a stub, but it's a start.  When I went back to shift a couple things around, I started to really wish this whole things was Drupal instead.

Drupal gets a lot of hate in the tech world, so I'm sure that I'll get complaints or at least scoffs.  I'll get into why I wish this was drupal in a bit, but first, some theories on why people hate drupal.

It's popular.  There are over a million drupal sites out there, and you'd loose your tech street cred if you didn't scoff at that which is popular.  I suspect that our high school experiences dealing with being unpopular probably shape that point of view.

It's a big php application.  PHP has a low barrier of entry.  This is a double edge sword, because your average php developer is probably less skilled than your average developer in another language (Visual Basic has this even worse, as some time with the daily wtf will show.)  Being both php and big means that drupal security issues come up relatively regularly.  Like with wordpress, you need to keep on top of updates.  A 2 year old drupal install, like a 2 year old wordpress install, is basically a rusty door waiting to fall off.

People learn one tool, then try to make the world out of it.  I'll brush off the first 2 complaints, as I think they are both addressable.  When talking to O'Connor recently, I found a much more reasonable complaint, which he was in the middle of.  Drupal is a content management system, and if you use it for semi structured content storage and display (i.e. a ghetto database), it's great.  But because it has a module structure, a lot of people turn Drupal into a web application development framework.... which it really is not.  If you want that, check out rails, django, cake-php or something that actually gives you that level of control.  This is a standard issue - if you have a hammer, everything looks like nails.

If I was doing mhvlug.org over from scratch, it would be with Drupal.

What I've found over the last almost 7 years with MHVLUG is that being a wiki was better than being a static website, but that the number of edittors for the site is still in the single digits.  We're a group of 150 people on the mailing list, 20 - 30 at a monthly meeting, 10 - 12 at our monthly lunch.  It's a solid community, but it's one with a pretty well defined reach, that's been more or less constant now for at least the last 4 years.  People are fading in at basically exactly the same rate as people are moving away or fading out.

The bulk of what gets updated is information around the meetings.  If that was stored in a semi-structured way, it would be a whole lot easier to update in one place, and have it viewed different ways in different places.  If I didn't have a project list queued out the door, around the block, down the hill, and... well you get the point, I'd seriously think about this one.

I'm pretty happy with how things are coming together in exactly this way when it comes to the farm website build on drupal.  Now that I understand what the right "types" structure is for the farm, the rest of it is just falling into place.

So for people looking to put up a website for a community in this day and age, I'd highly suggest that drupal is a good place to start.  Yes, you'll catch grief from your other tech friends, but such is life.  Long term, I think it's a pretty good call.

Update: I ended up redoing the MHVLUG website as a drupal site, with an extensive writeup on how.