the dnorton experiment

Since Amazon announced the Kindle, I’ve been weighing whether or not to pick one up. Over time as more reviews have come in from people I trust, I have decided that the pros of this e-book reader outweigh the cons. The general theme is that heavy book reader and frequent travelers love the device. People who are extremely hung up over the Amazon DRM or design asthetics are generally critical of it. I fall somewhere between the two camps. I love good design – evidenced by my obsession with Apple products – and generally avoid vendor lock-in strategies – for instance, I have recently made the switch to Amazon’s MP3 downloads instead of iTunes. However, I can overlook these minor flaws for a device that could change how and how much I read. Below is my good vs. evil list:
Good (Pros):

• Holds up to 200 books out of the box with the potential for many more with an SD card
• Will eliminate the crowding on my bookshelf… hey, I live in Manhattan, and we’re running out of room
• All content is fully searchable (great for technical books)
• Can convert non-DRMed PDFs into Kindle format by sending an email to your Kindle email address
• New books cost an average of $10. Based on my recent book purchases, I should save at least $12/book over purchasing new hard-cover books. At a rate of 3 books a month, I would cover the cost in just over a year. At 5 books a month, it would be less than half a year. Plus, if I tack on savings for the EVDO access of about $15 a month that I would spend getting internet access through Verizon, it will pay off in just over a year.
• built in Oxford English dictionary
• bookmarking and annotations – notes typed up on the Kindle are stored on amazon.com and are fully searchable
• instantly book purchases from Kindle bookstore
• EVDO wireless access (experimental but free for now.) So, I could check my Gmail occasionally while at work. How convenient.
• environmentally friendly (reduce dead trees) – read this article for a breakdown of the environmental impact of reading on the Kindle versus reading paper books and newspapers
• gain some early adopter street cred among a very elite group of tech nerds who read books

Evil (Cons):

• $400 is pretty expensive
• Content has an Amazon specific DRM so that you can’t transfer books to another user
• reported design flaws like the “Next” page button and the ergonomics
• less reliable than paper books (books don’t need batteries)
• there is not (yet) a broad adoption – lack of community
• I am generally unfamiliar with reading on e-Ink displays
• The Kindle is a single use device. Most mobile handsets are becoming general purpose – like the iPhone.

Helpful Reviews:

• Would you like that book in paper or plastic? | ES&T Online News
• NYTimes – David Pogue’s review of the Kindle
• Tyler Cowen’s Review
• Mainelife Review
• Amazon.com Review from a user who has owned a Kindle for months
• Mobileread.com’s Kindle Forum
• Colt Fan’s Amazon.com Review
• Ernie the Attorney’s Review
• Megan McArdle: Atlantic.com
• macintouch.com – Kindle Review
• Amazon’s Kindle Draws the Reader In

I picked up the domain dnorton.org recently. I am excited to have a nice, short domain name although I feel a little guilty registering a .org – I’m not planning a charity. There’s nothing there yet, but I do have some simple plans… stay tuned

I think it’s about time to scrap the antiquated design for nycnortons.com. I need to spend some time making things look nicer because the default Wordpress theme can only take you so far, right.

Also, I think I’m going to move this blog to d.nycnortons.com. It’s not really a dev blog because I don’t have time for dev things.

As of 4:14 PM EST, there are 14,244 articles with iphone in the title returned by Google News. I thought I would create this contentless blog post with “iPhone” in the title to become #14,245.

Besides actually getting some work done on the Runner Log app, my marathon training has been going pretty well. I have averaged over 20 miles a week for the last 3 weeks, and ran 8 miles on Saturday at my expected race pace – 8:00 min/mile.

This week I have the following runs:
Tuesday – 4 miles, fast (7:30 min/mile.) I usually run this around the Battery
Wednesday – 7 miles, easy. Possibly Central Park, but that depends on the weather and whether or not my usual running partner for Wednesday runs is around
Thursday – 4 miles, pace.
Saturday – 12 miles, probably slower than pace. I just picked up a water clip belt so I can take extra fluids on the run in addition to the pack or two of Powerbar Gel, cellphone, ipod, GPS watch, hat, and sunglasses that I usually run with.

I want to build a webapp to track my run progress for my marathon training.  So, being a geek, I thought what better way to track my time than to build a webapp.  Because I've started working with Flex at work, I decided that I wanted to build a Flex UI running on top of a REST based Rails webapp because I want to gain some experience with both of those technologies.  I'm going to give a little back by documenting the process.   I don't know if this will help anyone else, but I certainly hope so.

I'm also going to be pulling data off of a ipod nano with the nike+ipod kit and possibly my Garmin Forerunner 305 for my actual miles.

Architecture

These are the architectural requirements for the project.  Pretty simple really.

Requirements

• Ruby on Rails 1.2

• the Flex SDK
• Textmate

Table model

• user - simple user table that has login/pwd for user sessions
• runs - table that has individual run data - time, distance, pace, expected distance
  
• device - optional information for device that supplied run data (like ipod)
• session - it's better to use an active_record_session store than a file store in my opinion because it's much easier to support.
  

Rails plugins

• REST based authentication - http://svn.techno-weenie.net/projects/plugins/restful_authentication/
• This is another brillant plugin by Rick Olson, the Rails guru.  I am constantly amazed at the quality of the code he writes... just check out Mephisto and Beast.
  

Why REST and not WebORB?

So if you are like me and new to these technologies, you want to know what exactly is WebOrb... does it have anything to do with spiders? (no) From the WebOrb for Rails site:

“WebORB for Rails is server-side technology enabling connectivity between Flex and Flash Remoting clients and Ruby on Rails applications. WebORB for Rails can be installed as a plugin into any Rails application to expose Ruby classes as remote services. The product provides a complete implementation of the Adobe’s AMF0 and AMF3 messaging protocols and thus supports any Flash Remoting or Flex client.

Flex Builder projects can be configured to use Flex Data Services and point to a WebORB for Rails installation, however, the current release supports only the RPC featureset of the Flex Data Services technology. “

Another explanation for weborb from this Google Groups discussion states:

What WebORB allows you to do is to transfer Rails objects directly from
the server to the client, using a protocol called AMF, which is a binary
protocol (unlike REST, which is XML / text based...)

So, after a little research, I'm sticking with REST.  I don't have any need for passing large, complex objects like byte arrays or for using [RemoteObject] to send typed classes to Flex (although I am doing that at work.)

Flex UI

I'll save this information for Part II to be continued...

Joost is the new streaming video venture from Niklas Zennstrom and Janus Friis of Kazaa and Skype fame. It’s the hottest new beta product on the Internets right now, and in my opinion is pretty impressive. I happen to have a token (invite) to pass out, so if anyone is dying to try it out, leave me a comment.

There has been a lot of buzz recently about OpenID. In the last couple weeks, we’ve seen announcements from Microsoft about adding OpenID support to their CardSpace identification technology and AOL adding OpenID support for all of their user accounts.

So, what is OpenID? From openidenabled.com

OpenID is a URL based identity system. An OpenID identity is simply a URL, and all the OpenID specification does is provide a way to securely prove that you own that URL.

Unlike most identity systems, OpenID is completely decentralized. The end-user may choose whichever server they are most comfortable with, and may even run their own if they choose.

What this means to me is that sites that support OpenID save you the hassle of managing your identity on every site that you visit. Wouldn’t it be great if you could keep all of your personal information in one place (or more) with one password to remember. When you wanted to change information, you only have to do it once. OpenID is an attempt to address this concern.

My next step is to add OpenID support to taskhack using the new Rails plugin.
add OpenID to a Rails app
The No Shit Guide To Supporting OpenID In Your Applications—Dan Webb does a great job explaining how to set up a Rails app using the ruby-openid gem

update 02/28
I have added OpenID support to taskhack. I used code from East Media:
Consumer Plugin for OpenID

Mephisto is a blogging engine (small CMS system) built on Ruby on Rails by two of the Rails greats - Rick Olson and Justin Palmer. I really wanted something other than Wordpress to maintain this site. I debated between installing a Radiant CMS or mephisto and decided that mephisto offers all that I need.
More info:
Mephisto Wiki

This new part of nycnortons.com has a specific purpose. This is my development site where I can write about tech stuff, science, etc. I really needed somewhere to do a brain dump when I’m working on personal projects. And this is it. It won’t be fancy or even well written, but it may have some interesting information.