Tim Bray has become one of the latest employees at Google and has shared his thoughts on starting there. He mentions growth, although he only skirts the issue in vague terms. I hope he says more on the issue in the future, because it would be insightful to see how Google is dealing with these issues internally. It would even be interesting to know what they perceive as the issues internally. I’ll be watching to see what else Tim he writes.

On Google

The one downside to Thrift at the moment is that it has very little documentation.  Everything I’ve learned has been through trial and error or by watching other developers at work.  If I get a chance I hope to provide a sample Thrift implementation, using Python, along with documentation.

Until then, Stuart has provided a nice overview of Thrift’s features by comparing it to Google’s recently released Protocol Buffers.  If you’re interested in either, read Stuart’s article.

Thrift

With that out of the way, I’m going to once again consider that last post of mine.  In that post I stated that publisher don’t get it, since they are charging the same for electronic books as they are for paper books.  Well, that comment was a bit short sighted on my part.  There are two sides to every coin.

It does appear to me that publishers must understand something, as many of them have fairly quickly digitized much of their content.  Instead of trying to hold on to paper books only, publishers have tentatively embraced ebooks and digital content.  While I may not agree with their business model, publishers should be given some credit for, at the very least, not shying away from a radical change to their industry.

Publishers have also embraced other ways of publishing.  Print on demand is growing rapidly (standard disclosure: I work for Lulu.com, a print on demand company).  Many publishers are accepting authors who started out with print on demand, even seeking them out at times.  If publishers aren’t entirely embracing print on demand, they are accepting it.

Kudos to publishers for not trying to sue new innovations to their industry into oblivion.

Still, there is more publishers could be doing.

Why don’t they lower the price on ebooks and digital content?  An ebook is almost entirely profit, since reproduction and storage costs are non-existent.  Why not make the barrier to reading an ebook low and then make profit in volume?  Consumers certainly know what a terrible deal ebooks currently are.  As ebook readers such as Amazon’s Kindle become more common place, consumers are going to demand that ebook prices drop, so why not avoid the pain of making consumers mad and just drop the price now?  Everyone will appreciate it in the end.  It will endear good will with consumers and give publishers a more realistic look at the ebook market than what currently exists.

Why aren’t publishers building an iTunes for ebooks?  I want a one stop destination where I can find nothing but ebooks.  Sure, places like Amazon can show me the physical book and ebook next to each other, but as ebooks become more the norm, there is an opportunity here for someone to capture the ebook market the way iTunes has captured the music market.  Consumers are going to want picking up an ebook to be as easy as downloading a song.  In the digital world, the two actions are actually the same thing – the only difference is the properties of the file I download.  An interface like the iPod’s album cover flip navigation would be a very powerful thing for ebooks – just replace the album covers with book covers.  Publishers would probably see sales increase if a tool like that existed, because it would increase the discovery of books.

Publishers should also be building content to draw in users.  The power of Google Adwords has been proven at this point.  Publishers have the opportunity to take advantage of that same power.  Publishing sites need to become destinations where consumers can go to find free stories, free ebooks, free author interviews, etc.  And right next to all the free, are links to the full length books to download or order, along with ads for related books.  Pull in the consumer with interesting content and they will be highly receptive to buying more.  The one caveat is that publishers need to make it dead simple to purchase and download ebooks or to order physical books.  The Internet is a place of instanct gratification and fast action.  Publishers need to provide for this so that consumers can buy without thinking, or else the consumer will be distracted by the next shiny object and never make the purchase.

I think there is significant first mover advantage in this space.  The first publisher to pull this off is likely to see very healthy returns.  For now, I suggest keeping an eye on Tor.  They so far seem to understand the internet better than most, with their site redesign in the works and the free ebooks they are currently giving away.  I’m looking forward to seeing what they are going to unvail and how closely it hews to the suggestions I just made.

Publisher Advice

Amazon sells most Kindle books for $9.99 or less. Publishers say that they generally sell electronic books to Amazon for the same price as physical books, or about 45 percent to 50 percent of the cover price. For a hardcover best seller like Scott McClellan’s “What Happened,” the former press secretary’s account of his years in the Bush White House, that would mean that Amazon appears to be selling the selling the book for about 25 percent below its cost.

– New York Times Article, Electronic Device Stirs Unease at Book Fair

It’s always hard to change your world view when the world changes underneath you.

Publishers Don’t Get It

Both make it very easy to follow someone, and once you do, if you’re like me, you end up following them forever just because you’re lazy and don’t get around to removing them. Eventually you probably sit down and do a purge, but that’s likely a once-a-year or a my brains are about to explode from too much information task.  Something has to push you over the edge to get you to that point.

Why don’t Twitter and Google Reader (and all feed readers) add a bit of friction to the beginning of the process, so that the mass purge becomes a thing of the past?  Or at least something that needs to happen less frequently.

There should be a quarantine function.  Each new person I follow or feed I add automatically goes in quarantine.  It then stays there for a certain amount of days which I can set, say 10, 20, or 30.  At the end of that period, a message shows up asking me if I want to keep following that person or feed or not.  If I say yes, then the person or feed is moved out of quarantine, and if I say no, then the person or feed is automatically deleted.  With this function I now get the opportunity to evaluate each new person or feed I think I might want to follow over a period of time, so I can therefore make an informed decision, instead of the snap decisions I make now.

So many feeds and people that seem promising after one post or message don’t hold up over the long haul.  I wonder why I followed them and eventually they get cleaned out in a mass purge.  A quarantine system solves this problem.  Also, if at any time I tire of a feed or person, I can throw it or them into quarantine and the system works as if I had just added them.  During quarantine, the person or feed can redeem themselves, or they get deleted.  It would be my own little slice of purgatory that I get to control.

Quarantine Tweets and Feeds

Here’s the link to my OpenId page, for your perusal:

http://openid.programmersparadox.com/mmzyk

I signed up with myOpenId, but that might not be obvious from the URL above because I opted to use their service and make my domain my OpenId.

I think it’s cool that my domain can be my OpenId.  That was really the thing that made me decide to try it out.  I didn’t want to have yet another login with another provider, but being able to tie a portable identity to my own domain?  Now that was worth trying.

I’m happy now that I figured it out, but it is much more difficult than it should be.

First, I had to modify my DNS records to get the openid sub-domain setup on programmersparadox.  I don’t have access to the DNS records for this site, but my awesome hosting provider, Laughing Squid, set that up for me and set it up quickly, less than half a day after I requested it be done.

So far so good, but then came the pain.  To set up OpenId for programmersparadox, I had to first sign up on myOpenId.  Then I associated my domain with my account.  That’s easy eough.  Then I created the first OpenId account under programmersparadox, which is, naturally, my own.

Guess what?  It’s a completely separate account from the original I set up with myOpenId.  I now had to create an account with myOpenId under the programmersparadox domain.  Creating the account isn’t hard, although it took me a good while to finally decipher what the login name was for the new account, since the setup process doesn’t make it clear.

Once that was figured out, all I wanted to do was replicate the information from my original myOpenId account to the new OpenId account associated with programmersparadox.  I never found a way to do that automatically.  I ended up re-entering and copying and pasting the information by hand.

Come on myOpenId.  You’ve got a good thing going, now you just need to make it easier to use.  If I, as a technology guy, am having this much trouble, I can’t imagine what the luddite is going to think when I’m trying to convince them to sign up for this.

OpenId

Recently this Gizmodo article brought to light some facts I didn’t know about the current crop of ebook readers (This applies to Sony and Amazon. As for others, I don’t know).  The main fact being that when you “purchase” an ebook for the reader, you are in fact, only licensing the ebook.

I generally prefer to own things rather than license them.  I like to know I have the full power of ownership.  The issue of licensing brings up a question in relation to the price of ebooks: since I don’t actually own the ebook, why am I still paying such a high premium for the right to read it?  Yet another reason why the pricing of ebooks is skewed.

In addition to only allowing the licensing of an ebook as opposed to owning it, both ebook readers also restrict you from copying your ebook to someone else.  Even if you were allowed to own the ebook, I still think it is likely a restriction on copying would be placed upon you.  This is a problem that desperately needs a solution.

Much has been said on the web about how people pass around books and use borrowed and second hand books to discover new authors, which can then lead to more sales of books.  It’s the whole premise of the library system.  Yet this generation of ebook readers is threatening to kill that system.  You could argue the system can still exist if someone is willing to pass around the hardware, but that has drawbacks.  If I lend you my Kindle, then I can’t read any ebooks on it while you have it.  This would be akin to me loaning you not just a single book, but my entire library of books.

So what options are there that can make both readers and producers of ebooks happy, as far as copying?

The first solution is social:  Find a way to pay authors/creators/producers outside of the existing system, so that readers are then allowed to produce as many copies of ebooks as they want.  How this would work, I’m not sure.  For an example of how this might play out and how the public will react to it, pay close attention to the moves of the record labels as they try to deal with the downloading of music.

The second solution is also social: Authors/creators/producers trust in the good of people and remove DRM or copy-prevention from ebooks and allow people to make as many copies as they want.  Perhaps what will be discovered is that everything will still work out in the end while those who want to get payed get payed and the buying public is happy because they don’t face restrictions.  Will there be a few bad apples who use this system to pirate ebooks?  Certainly. However, I also believe that many people will still pay for ebooks while giving their friends copies, who may then also become paying readers.  Of course, no one is going to make a change to the current system based only on my word.  I currently only dream of having that much power.

But neither of those solutions are what I really want to talk about.  I want to address the third option, which is the technical solution.

Let’s step back for a moment and examine a situation with a non-electronic paper book.  I happen to be reading Brave New World at the moment, so let’s use that.  I finish reading Brave New World.  I then discover that my coworker is an uncouth near illiterate Luddite.  In an effort to bring him into civilization, I loan him my copy of Brave New World.  While he has my copy, he can read it at his leisure.  However, because he has my only copy, I am unable to read the book until he returns it to me, I purchase another copy, or someone else loans me their copy.

It should be possible to recreate this exact scenario with ebooks.  I purchase an ebook.  I read it.  After having read it, I want to let you read it, so I copy it to your ebook reader.  You are now free to read the copy you have, while the copy I have on my machine stops working.  It only resumes working when you copy it back to me, I purchase a new copy, or someone else gives me their copy.

This an exchange that can be created technologically.  Is this a form of DRM?  Yes, in so much as it restricts you from taking advantage of some of what a digital format has to offer, especially the cheap copying ability.  However, this does recreate, in a small way, the world of physical books that people know and love.  I think that because this system mimics physical books most people would be accepting of it and would play by the rules laid out.  Would someone crack the system?  Of course.  All DRM eventually gets cracked.  Still, I think that is inconsequential here because this system would give people enough flexibility that they wouldn’t feel overly constrained and would be willing to abide by it.  It would also give the authors/producters/distributors some measure of control over the copies of their ebooks so they wouldn’t feel like they were losing all of their revenue to piracy (whether the piracy argument is true or not: remember, perception is a part of reality).  If an ebook doesn’t do well under this system, it is because the ebook failed, and it cannot be blamed on everyone obtaining a pirated copy.

I wonder what it would take to convince Amazon or Sony to give this system a try.  At the very least, it’s a step forward from the dark ages that ebooks are in now.  The current landscape of ebooks resembles the time in history when copying was done laboriously by hand.  Ebooks need to at least catch up to the time of Gutenberg’s printing press.  How can a medium become popular and profitable if it isn’t allowed to spread?  That’s the dilemma with ebooks today.

Ebooks: Failings and Solutions

A quick intro to garbage collection

A little more about garbage collection

Explaining garbage collection

These links provide a general overview of how garbage collection works, but are probably too technical for the non-computer science minded to fully comprehend or care about.  The link to explaining garbage collection is the best metaphor for garbage collection I’ve ever seen.

And yes, I opted to de-stress tonight by reading about garbage collection.  That has to make me a grade A geek.

Thanks goes to Daniel Ehrenberg for writing those links.

I Check Out By Taking Out The Garbage

In the test that is set up in Second Life, the bot is shown a gun that is in a red suitcase.  There are two observers with the bot.  One observer leaves.  The gun is transfered to a green suitcase.  The observer that left is then called back.  Upon returning, the bot is then asked which suitcase the observer who left would look for the gun in.  The bot correctly answers the red suitcase, since that was the last suitcase the observer saw the gun in, even though the bot knows the gun is in the green suitcase.

To achieve this reasoning, the bot has been seeded with a logic statement that if an observer sees something, they know it, and if they don’t see something, they don’t know it.

Apparently some people are calling this a great AI achievement.  Am I missing something here?  I’m not understanding what is so great about this.  It seems to me that it can be accomplished with a state machine.  Just what are research dollars being used for these days?

Please enlighten me if I’m missing something.  Or if you know more technical details about this that might fill in the gaps in my knowledge to make this more impressive, let me know.

Very Artificial Intelligence

Well, the NYTimes went out and implemented the idea before the ink was even dry on my post.  They’ve now proven it is possible, and their post makes it sound like it was all done by a single person.

Of course, the link to the service is down as I write this, but it’s still an amazing feat.

Of course, the other aspect of this is what opportunities are presented by having the NYTimes’ archive back to 1851 online?  What kind of applications can be developed with this wealth of information?  Maybe an interesting way of teaching history can be created using the archive?  Perhaps the archive could be searched for lost articles that in some way ring similar to events of today, to show if, in fact, history is repeating itself?  There are tons of possibilities.

And it should be noted all the articles in the archive are also in the public domain.

Thanks to boing boing for originally bringing this to my attention.

In The Wild: A Google Of One