Anyway, here’s what I wanted to share: Since Buzz integrates heavily with Google Reader, you can use it’s “share” bookmarklet to forward links on to Buzz. Here’s a Bookmarklet I hacked together that works really well for this…

javascript:window.open('http://www.google.com/reader/link?url='+
encodeURIComponent(location.href)+'&title='+encodeURIComponent(document.title)+
'&srcURL='+encodeURIComponent(location.href));

Just drop it in a bookmark, load up a page you want to share and click. You can add a note too, if you like.

Enjoy.

2010 is upon us, and the air is full of excitement for a clean start and a fresh outlook on the opportunities ahead of us. This year, I plan to devote some time working towards revitalizing my blog. Like many, I’m notoriously bad at getting distracted and neglecting it for months at a time. It’s easy to lose track of time and fall behind on posts. That’s why this year I’m taking part in Project 52, the brainchild of Anton Peck (who was in turn inspired by a tweet from Chris Wallace.) It’s a project with a pretty simple objective: post stuff, and post it often.

Once a week, share something new. Whether it’s a review, or a technical walkthrough, or just what’s going in your life: blog it. Blog the ever loving sweet crap out of it.

I won’t call it a New Years resolution because, well, I never keep those. I’m just going to hope for the best and see where it takes me.

Happy 2010, everyone, and best wishes for a happy and successful new year.

Photo “happy new year 2010″ by mugley.

My variation of this Automator workflow will:

• Detect when a Kindle is mounted.
• Confirms if you want to sync your device with Instapaper.
• Deletes any old Instapaper files on the device.
• Downloads a fresh Instapaper archive and places it on your Kindle.
• Asks if you want to dismount the Kindle, and does so if chosen.

Once you have this workflow in place all you really need to do is plug your Kindle in and wait for it to do it’s thing. Couldn’t be easier.

One issue to be aware of: while you have your Kindle mounted, any insertion of USB drives (or anything that will show up in /Volumes for that matter) will start the script. I’m looking for a workaround on that one, but it’s a relatively minor issue. Plug it in, sync it, unplug it. That’s the point of it, anyway.

You must also be signed into your Instapaper account through Safari for the Instapaper archive to be downloaded successfully. Automator uses your Safari session to grab the file.

Finally, the script should work with any number of Kindle devices, but only one may be plugged in at a time for the sync to work properly.

Download

You can download my workflow and customize it for yourself. You’ll want to move the file to your ~/Library/Workflows/Applications/Folder Actions directory and work with it from there. Alternatively, you can build the workflow manually.

Install the Workflow

1. If you downloaded the pre-built workflow, ensure it’s been copied to the location I mentioned.
2. Open Finder and press Control+Shift+G. In the prompt, enter “/Volumes” and press Enter.
3. If you don’t have the path bar enabled in Finder, go to the View menu and select “Show Path Bar.”
4. Right click “Volumes” in the path bar and select “Folder Actions Setup…”
5. Attach the “Update Kindle with Instapaper.workflow”, ensure both checkmarks are checked.

Plug in your Kindle, and watch the magic. ;)

Build The Workflow Manually

1. Open Automator and create a new Folder Action. Set the “receives files and folders added to” drop down to “Volumes” (select Other, then press Control + Shift + G and enter “/Volumes”, press Go and then OK.)
    
2. Add a “Run AppleScript” action with the following code:

   on run {input, parameters}
   
     tell application "Finder"
       if exists disk "Kindle" then
         return input
       else
         error -128
       end if
     end tell
   
     return input
   end run

3. Add a “Ask for Confirmation” action. This will allow us to cancel our Instapaper update if we’re not ready for a fresh copy.
    
4. Add another “Run AppleScript” action to check for any existing Instapaper files on the Kindle, and move them to the trash for us.

   on run {input, parameters}
     
     tell application "Finder"
       set kindleFolder to "Kindle:documents:" as alias
       set instapaperFiles to every file in kindleFolder where
         name of it contains "Instapaper-ReadLater-"
       move every item of instapaperFiles to trash
   
       -- If you'd like to automatically empty your trash, uncomment:
       --empty the trash
     end tell
     
     return input
   end run

   Lines 5 and 6 should be on one line.
    

5. Add a “Get Specified URLs” action. Delete the default URL and add http://www.instapaper.com/mobi.
    
6. Add a “Download URLs” action. Point the Where value to your Kindle/documents folder..
    
7. Add another “Ask for Confirmation” action. This time we’ll be asking whether we want to dismount our Kindle or not.
    
8. Add one more “Run AppleScript” action to dismount our Kindle from the file system, allowing us to just unplug the device without any further steps.

   on run {input, parameters}
     
     tell application "Finder"
       if exists disk "Kindle" then
         try
           eject "Kindle"
         end try
       end if
     end tell
     
     return input
   end run

    

First, here’s what the related portion of my lighttpd.conf file looks like. We’re just replicating what Lessn’s own .htaccess file tells Apache to do.

$HTTP["url"] =~ "^/g/-/" {
  server.error-handler-404 = "/g/-/index.php"
}
$HTTP["url"] =~ "^/g/" {
  server.error-handler-404 = "/g/index.php"
}

Important – I have my Lessn installation located at /g, so you’ll want to tweak these rules to match wherever you installed yours.

Now open up Lessn’s /index.php. Beneath the includes we’ll be adding a few lines to grab the token from the request uri, like so:

include('-/config.php');
include('-/db.php');

if(!isset($_GET['token'])) {
  if(isset($_SERVER['REQUEST_URI'])) {
    $_GET['token'] = substr($_SERVER['REQUEST_URI'],
          strrpos($_SERVER['REQUEST_URI'], '/') + 1);
  }
}

There you go: a functioning instance of Lessn on lighttpd. Enjoy.

• Live Search – Check out that text box to your left. Try searching for something and behold it’s AJAXy goodness.
• Dynamic Archives – Also powered by AJAX. No pages to have to link through; the list grows as you scroll.
• Twitter Mentions – Each blog post (and a few of the pages) include a “Twitter” tab. The tab lists any mention of the page’s long or short permalink structure. It uses the excellent Backtype API to accomplish this. I developed it originally against the Twitter Search API, only to discover how awful they are at retaining data like this. Backtype works flawlessly and has a very respectable retention history.

Of course, the most obvious changes are the revised color scheme and layout. I got sick of the dark look, so I replaced it with a more eye-friendly, lighter appearance. There’s also a few new pages hidden around, if you take a look.

So, I hope you enjoy the new look. I’ll hopefully be releasing the theme in a month or two after I give it the old dog food treatment and work out the kinks.

I want my computer to work. All of the time. No exceptions. (Who doesn’t?)

I don’t care whether you use Windows, a Mac, or Linux; you’ll encounter show-stopping problems, and probably a lot more frequently than you care to admit. That isn’t to say the web isn’t without it’s problems. Datacenters go down, tables get corrupted, and worse of all things get hacked. The difference is, if a site goes down I can come back to it later. If my desktop OS goes down, I’m screwed… and potentially my data is screwed along with it. A reliable web service won’t suffer the same fate.

An all too common sight.

Now, Google unveiled quite a bit of new information on Chrome OS today, and officially opened the code for the first time to developers. I have a mostly positive impression coming away from the webcast, but I do see a couple of problems that Google hasn’t really addressed.

The Good

#1, The OS is self-repairing. That is, if a byte goes bad, or someone develops a piece of malware that circumvents the sandbox and eats your OS, the machine will revert to a good copy patch itself to a fresh state. No lasting harm done.

#2, Dead simple. I could see slapping this on an old laptop and giving it to a grandparent (see Bad #1.) No file system confusion, or worries about viruses, or needing to ensure they have the latest updates. It’s a set it and forget solution for the tech-clueless.

Chrome OS UI (Mockup)

#3, You can login to your Google account on any Chrome OS-based netbook and instantly have access to an identical looking “desktop” as you have on your own machines. Your bookmarks, browser history and even theme are synced and displayed. Fantastic. This also means, of course, if you drop your netbook and kill it, you can buy a new one and be right back where you left off in no time.

#4, It’s free. Netbooks running the OS won’t have to worry about Windows licenses, and thus keep the cost of their hardware cheap. Granted, Linux has always been an option for the same reason, but this is no standard Linux OS. I don’t care if it’s Ubuntu or Fedora or Gentoo; there is no such thing as a user friendly desktop Linux environment. It appears Chrome will be the answer to that problem.

The Bad

Solid State Drives. I understand why Google is going that route– the performance advantage is huge over traditional drives– but requiring them to use the OS is absurd. I get requiring a SSD on new devices that ship with Chrome OS. I think if someone walks into Walmart and buys a Chrome OS-loaded netbook, they would expect tip-top performance.

But let’s say I’m not going to walk into Walmart and buy one. Let’s say I have a perfectly fine laptop sitting next to me that I want to give to Grandpa. Chrome OS is an easy choice, because all he does it check email and browse the web, and I don’t want to have to worry about him getting viruses. Alas, unless Google changes it’s mind on this, that isn’t going to be an option- because that old laptop doesn’t have a Solid State Drive. Who cares if Grandpa’s OS takes 7 seconds or 7 minutes to load? It’s a free PC, and it’s running your OS, Google.

If I’m savvy enough to install Chrome OS my own hardware, then I should be free to do so. Don’t lock me into your ideas of what a netbook should be.

The Ugly, and questions left unanswered

Privacy. I’ve always been a proponent of the “privacy is dead” argument when it comes to the Internet… or in any context, really. There isn’t such a thing. Still, having every aspect of my OS… my bookmarks, my customizations, my browser history… housed on Google’s servers is a little concerning even to me. I’d like to see us able to define a different sync-point than Google’s servers. Like Mozilla Weave or Xmarks, let us use our own FTP server as a place to store this cloud data as an option. At the very least, give us an indication of what sort of encryption you’ll be using to secure OS data.

Is there room for gaming on a Chrome OS device? Of course we aren’t going to be playing World of Warcraft on these things, but Google hasn’t mentioned if they’re going to support hardware acceleration at all, or whether OpenGL support will be bundled. NVIDIA’s Ion netbook platform is growing in popularity; will there be support for that? They demoed a Flash game; considering the sandboxed environment, will they be open to partnering with companies like Unity and OnLive to deliver cloud-friendly plugins for gaming?

Likewise, considering Chrome is all about HTML5, and a major aspect of HTML5 is it’s local storage capabilities; how is Chrome OS going to handle this? Will there be a finite amount of local storage for these things, or is Google imagining a way of syncing this data to their cloud and somehow still maintaining offline capacity? Perhaps there will there be no such thing as offline capacity, considering Google’s buyout of the 700MHz frequency spectrum, or are these devices going to be sold by cell carriers with built-in modems?

There’s a lot of interesting uses for this Lists system. You could setup your list of favorite tweeters, or build a list of all the employees in your company who are tweeting. There’s a lot of neat ways clients could implement this feature going forward too, and Twitter is opening up an API specifically so they can.

In any event, check it out, and if you’re into game design or the video game industry at large, check out my @game-industry list. It’s got 200 (and growing) Twitter users from the game industry, and pop me a @reply if you’d like to be added or know someone who should be.

P.S. If you’re looking for a way to easily browse some of the Lists being built on Twitter, check out Listorious.

Today I came across a web service called Aardvark, which one might define as a user-generated KGB; that is, it’s a community of users connected via web browsers, cell phones and instant messengers that define their own areas of expertise, and are automatically sent questions from other users that target those criteria. For example, earlier today I asked “What is a good, lightweight and free (open source preferred) FTP client for Windows?” and received 2 answers within a few minutes; one user recommended Mozilla FTP, and the other FileZilla. When the answers came in, I was notified via Google Talk on my Blackberry and PCs. Alternatively, I could have received notifications via SMS and email. Twitter notifications from the @vark account would be a nice touch, too.

The service is brilliant, and I really hope it gains popularity. The only real problem right now is the limited number of users participating (think Twitter back when it was “twttr”). On the other hand, as my buddy Tom pointed out, one has to wonder how scalable the platform behind it all is, and whether they’ll be able to meet the demand as the community grows. I wish them the best of luck with it.

If you’d like to try Aardvark, you can sign up using my referral link or from their site.

Opera 10 Beta 2 (Unite Build)

For the last 6 months or so I’ve been using Google Chrome, and really enjoying it. It’s blazingly fast, has a tremendously elegant user interface, uses my favorite rendering engine (WebKit), is open source, and is updated very frequently (I run off the dev builds.) But it was far from perfect, especially in comparison to what I had left behind with Firefox. Namely, it lacks extensions, a robust web developer toolset (the Inspector is broken and falls way behind tools like Firebug) and suffers from a lack of customization options.

The last time I had tried Opera was with the release of v9. I had tracked the improvements the company had made to browser for many years, and tried each major release. Although I had always found it’s performance to be top notch, the user experience really bugged me. The interface was… bizarre, in my opinion. Although it could be customized I never managed to come up with a layout that rivaled the simplicity of Firefox.

Like many, I was intrigued when I heard about Opera’s “Unite” system. If you aren’t familiar with it, it’s an experiment (of sorts) in expanding upon the concept of the browser from just a viewer of the Internet into a participant; it embeds functionality (exposed as modular “services”) that transform it into a server. There are a variety of services, from a file browser and web server to a streaming media player and social note sharing hub. Heck, it can even serve as a personal OpenID provider. The concept is pretty exciting, and I’m already in love with the file sharing and media playing functionality, but after playing with the beta I realized that Unite isn’t really the big seller for me: the browser itself is.

In one release, Opera managed to fix everything that bugged me about the browser. The tabs, the navigation controls, the panels/sidebar; they repackaged the chrome in a way that is minimal (which I love), flexible/customizable (important for many) and easily transitioned to from other browsers. With the user interface problem resolved, I was able to sample the features old and new that Opera has to offer;

• Unite — As I said, it turns your browser into a server, offering media streaming, file sharing, etc.
• Turbo — I’ve already fallen in love with this on my laptop. When you’re on a low bandwidth connection (like when I’m tethered to my Blackberry) it dynamically compresses assets on a page to reduce download time. Image quality is dramatically reduced, but it gets the page to you dramatically faster than it would normally. LOVE this feature.
• Link — Sort of like Xmarks or Mozilla Weave on steroids. Syncs your bookmarks, notes and other data between browser installations on different PCs and on Opera Mini, which I am quickly growing to love on my Blackberry.

If you haven’t given Opera a try lately, please do. I think you’ll be pleasantly surprised with what you find. The latest beta version can be found here, the Unite-enabled beta builds can be found here, and the general release version is available from their homepage.

I spent much of yesterday getting my Slicehost slice setup, and today I went live with it. Slicehost isn’t your usual web host; they build you a custom, bare-bones server image that you can then configure to your heart’s content using SSH. Being a Linux newbie, it’s been a lot of fun working my way through it; from updating the essentials to getting Apache, PHP, MySQL and XCache running. Postfix has been a particular painto get working, but I think I’m almost there.

I’m super impressed with the performance I’m seeing right now, but I’m nowhere near the load I’ll be once I get my guild’s site running on here. Still, I’m very optimistic that we’ll be happier with this setup than our previous host. I do regret not getting this setup with a bit more memory from the get-go, but I’ll see where I’m at when I start moving more sites over. I could always setup another slice and distribute the domains across them.

FYI, I still need to make a few tweaks and ensure email is getting moved about properly, so I apologize in advance if you get an email bouncing from my @evansims.com addresses for the next day or two. You can reach me at evan sims [at] gmail [dot] com in the meantime. Thanks!