This one took me a while to figure out. I have my PATH environment variable set up to include other paths that are also defined by an environment variable. For example, I have Maven set up in the following way:
M2_HOME = C:\Program Files\Apache Software Foundation\apache-maven-2.0.9
M2 = %M2_HOME%\bin
PATH = %M2%;[a whole lot of other dirs]
Today, I tried to run a Maven build, and it wasn’t recognizing the maven\bin in my path. When I did an echo on my path, the path had the following:
I got the full path C:\Program Files\Apache Software Foundation\apache-maven-2.0.9\bin
What was going on?
Well, apparently, Windows has a limit on the size of your path and when it hits that limit, it just stops interpreting your %VARIABLES% set in the path.
So, the solution is to either reorganize all of your files into more succinctly named directories, or remove some of the dirs from your path.
I’m working on a long post about Merb, but I figured I’d post a link to MerbCamp videos from the event in October. I was unable to attend, so these videos are really, really nice to be able to watch.
It also looks like they were able to stream some of the event live, and people participated in the conversations over IRC. Way cool.
I’ve now watched almost every video posted on the site. My favorite videos include:
Keynote by Ezra Zygmuntowicz - Ezra reviews the core tenets of Merb and introduces Nanite. Nanite is a really cool library for scaling the backends of web apps. Basically, when you build a web app, you want to be able to respond to the browser quickly. However, some tasks take some power to process, so you can kick these tasks out to Nanite Agents. You can have as many machines running as nanite agents as you want. They can be located anywhere, and can be added as needed. Very cool.
MerbSlices by Daniel Neighman - This is something I’ve wanted to see in Rails and other frameworks for a long time. Basically, it provides a way to build reusable applications. So, I have frequently had clients ask if I could just add a forum, a blog, or some kind of CMS to their site. Integrating authentication and user accounts across a custom site to a forum (like vBulletin) or a blog (like WordPress) is a real pain. If a forum slice or a blog slice would be a great mixin to a custom site.
Using Multiple Databases with Merb by Wesley Beary - Wesley talks about an experimental site that he’s got on Github that uses CouchDB and a RDBMS together. As I mentioned in a previous post, CouchDB is one of the newer technologies that I’m following. Pretty cool. I wouldn’t recommend sharing images though like he talks about in his presentation just for the reason that the port may be blocked on some networks. I’m sure there are other security implications as well.
jQuery on Merb by Yehuda Katz - Yehuda shows some good examples of how to get by in life without JavaScript helpers, like the ones we often rely on in Rails. This was also a good introduction to jQuery.
Keynote by Yehuda Katz - Yehuda talks about future goals for Merb and why Rubyists don’t need to be ashamed about Ruby.
I have to admit that I was disappointed in the “From Rails to Merb” video. While the discussion was interesting, I was really hoping someone would address good strategies for migrating a site to Merb. Oh well.
Overall, there is some great content at the MerbCamp site, and this would have been a very fun event to attend.
Although I consider myself a pretty tech savvy guy, I’m not as up to date on Internet social life. Today, I updated my status on my Facebook account for the first time ever, and then I thought, Oh, I bet that is what Twitter is really all about.
I visited Twitter and lo and behold, a new video there that explains Twitter in plain English. I never really grokked Twitter, even though I knew what the Fail Whale was. I guess I’m starting to get it now.
Now I just need to actually use it.
(video provided by a cool company named Common Craft)
You may have noticed that I haven’t updated my blog for a while. I’ve decided that I’m going to try to write more during my lunch breaks. So, what have I been up to?
I’m still working for FamilySearch as the Developer Support guy for all public APIs. This has been a fun and rewarding experience. I get to write documentation and code samples, present at conferences and webinars, and meet with a lot of companies in the genealogy space. I’m really excited about what is coming in the genealogy industry. Lots of advancements are being made. I’ll update more of what I see coming in a later post.
I’m still doing Ruby programming on the side for USFamilyTree.com. I get to do most of this work during my commute to and from Salt Lake City. UTA has free WiFi on their express buses, which makes it possible for me to still stay productive.
Family life has been really good. Our little boy is almost 2 years old now, and he keeps us laughing and having lots of fun. We’re excited to announce that we are in the application process again to adopt another baby. The second time has been a lot faster for us and we’re getting really close to being approved and on the waiting list.
This blog has typically been about web technology and such, and I plan to keep it on that subject. Some projects that I’ve been watching closely include: Merb, CouchDB, and Rubinius. These will likely be the topics of my next posts.
I’d like to continue with my Wordpress/Hostmonster screencasts, but I haven’t been able to find the time or place to make and record new episodes. Hopefully I can find a Saturday to crank out another one soon.
Don’t be overwhelmed by the 99 slides. It moves really fast, and walks you through the history and concept of BDD.
Perhaps my favorite concept from the slide show is “Design is a Process, not a Phase.” As I have embraced Behavior Driven Development, I have seen improved software design come from the process. I believe this is true because writing the specs help create a clean public interface for your objects and classes. By having the tests that verify the behavior of your code, you can easily refactor without the worry of unknowingly breaking desired behavior.
If you find yourself falling behind in getting all of your emails read, like I have in the past week, you may find it useful to put the following in your Gmail search:
is:unread in:inbox
This will filter your messages to just those that are unread and in your inbox. Pretty cool. This help a lot for me because I have a lot of filters on my mail that mark some mailing list emails as ‘read’ and put others under different labels without marking them as read.
