I favor Twitter Bootstrap personally – Foundation gave me some headaches when using it with Rails, Bootstrap’s syntax is a lot nicer (IMO), and Bootstrap just has more resources available for it (theme color generators, plugins, sample templates, etc.).  But at the end of the day, sites built with Bootstrap look like Bootstrap, and that’s a pretty major downside.

A few months back I stumbled upon WrapBootstrap.  They have hundreds of great themes (and they’re generally all HTML so you can adapt them to Rails, PHP, WordPress, whatever), and they’re CHEAP.  Most themes run around $10-$15, while some in the upper echelon run around $20.

To date, I’ve made several sites using themes from here, and I’m quite impressed.  It saves me tons of time and makes the client’s site look unique.  It’s a win-win.

Check them out today - WrapBootstrap.  It’s a great idea, and for rails projects, it’s just a matter of dropping in the CSS and JS into the pipeline, and building out your layouts.

A couple weeks ago I caught some slack from the creator of ManageWP for the following tweet:

@dougc84 Use a car? Check out walking – cleaner, easier and freer

— Vladimir Prelovac (@vprelovac) August 3, 2012

Oops.  I do try to walk everywhere though…

But the above statement is true.  I had thought about getting ManageWP for a while, but the cost structure really prohibited me.  I don’t manage hundreds or even tens of WordPress sites, and I can’t justify paying for any additional services every month.

I was on the phone one night with the General Manager of CMS Advertising Group, and she started ranting and raving over her new discovery – WP Remote.  I decided to try it out, and after a few initial hiccups setting up some GoDaddy sites I manage (oh, GoDaddy… grr), I had every site I manage available in one dashboard.

While not as feature-intensive as ManageWP, I didn’t need all the bells and whistles ManageWP offers.  ManageWP is a fantastic offering, but from communications with some people, I’ve heard that those “bells” don’t always work depending on your host, setup, etc.  I didn’t need features like posting directly from the dashboard or analytics – I can hop over to those sites in one click in WP Remote and get what I need.

Where WP Remote beats the competition is it’s simplicity and small feature set.  It’s design is based on Twitter Bootstrap’s responsive variant, so it looks great on my iPhone as well as my laptop.  I log in, it contacts each site (you do have to install a plugin on each site), and tells me if I need to upgrade WordPress, any of its plugins, or if the site is unreachable.  I can also make simple notes to tell me specifics about each install, such as those GoDaddy sites where I should not update WordPress from WordPress itself but through GoDaddy.  You can also group sites together, allowing easy access to subsets of sites you work on for different organizations.

And finally, the kicker – free backups.  And it backs up the wp-content directory as well – something ManageWP tends to have trouble with.  They’ve stated on their blog that because it is a free service, they may ask individuals to eventually provide their own CDN (Amazon is the one they mentioned specifically) for those backups, but at the moment, they’re complimentary.

So, it’s worth a look.  I’m not trying to take people away from Manage WP, but if you’re looking for a simple set of tools for sites you maintain (but maybe not post to that often), take it for a spin.

I have an upcoming project where data needs to be exchanged between groups of customers in realtime. As primarily a Rails developer, I know that the only real way to do this is with some sort of Pub/Sub system (that are often outrageously tedious to configure), or with one of these hot new JavaScript frameworks.

While I love Rails, I don’t feel like it’s the right tool for the job. I feel like it would provide a lot of unnecessary overhead, even if it’s just being used on the data layer, handling server-side requests and whatnot. Over on Railscasts.com, Ryan Bates had a recent video on Meteor.js, and a while back, on Backbone.js. Looking at what all was involved from a development standpoint (I’m big on rapid development – it doesn’t make sense to spend a day doing what someone else has done already), it seemed like Meteor was super easy, and I’ve been looking for an opportunity to play around with MongoDB on the backend. Sounded like a win-win.

Long story short: I feel betrayed.

Let me put this out there first: My learning of data structures and syntaxes was from Java. I get how everything’s written. I get what everything means. But I’ve been burned by other developers who have been heavy-handed with Javascript on their applications (including one I maintain that has nearly a meg of Javascript dependencies… A MEG!), and over the last few years, have tried to stay away except where it’s absolutely necessary and provides a benefit to the end user. I think that may have made my Javascript skills a bit on the lax side, but I can get around pretty well.

Meteor’s tutorials, screencasts, videos, etc. – they are all fantastic. And I understand Meteor is definitely in its infancy – it’s still in “preview,” as they call it. As I started playing around trying to figure out how this framework works, there’s simply a lot missing.

Routing is a manual process, and pulls in Backbone.js to handle it. That really irks me that they list that as one of the main packages you can add. If I wanted Backbone support, I would have gone with Backbone.

Controllers? Heh. Forget about it.

Models? What?

Semblance of order? None.

They state on their docs that you can order your files any way you want and specify where things can go and can’t go. The flexibility is fantastic, and that’s why I choose CodeIgniter as my preferred PHP framework when I need to go that route. MVC is a fantastic concept and an awesome way to separate and organize your files in a really common-sense manner (after you understand it, that is), but sometimes the rules need to be broken. In a Rails app, for instance, sometimes you just need access to current_user in a model. For some business cases, it makes sense. To do it otherwise, you end up writing more code, tests become trickier, and you introduce code duplication. CodeIgniter lets you be more free in how you do things, and I appreciate that.

But… Meteor… there’s simply too much flexibility. Your files don’t make sense. You have a method in one area but you can’t call it because you’re using backbone for something else and they don’t talk to one another.

On the other hand, there’s not enough flexibility. You have one (that’s right, just one) layout file. I don’t know a single application I’ve ever worked on that has had the same layout across the board. And sure, you could do some inline logic – I mean, it’s all Javascript anyway – but it just ends up being messy and dirty and before you know it, you have a 400 line layout view.

What if you want your app to be XHTML compliant, instead of HTML 5? I don’t know why you would anymore, but maybe your end user demographic are those that use outdated browsers and old computers, and you want to stay compliant for them (I still deal with that with one client). You can’t change the doctype.

Finally, the documentation leaves a lot to be desired. Sometimes it just states something that may be blatantly obvious to the developers on where a bit of code goes, but it’s simply not so when you’re learning the language.

What I’ve determined is this – If you have a single-page, non-mission-critical application that you want to get live pretty quickly, Meteor’s an awesome tool. If you want to do something else, it’s time to look elsewhere. As for me, I’ll be looking back to Backbone, Ember, or Angular to accomplish what I need to do. But I will certainly be keeping an eye on Meteor for what’s to come.

Before entering VCU, I had a basic knowledge of some programming. I could write some HTML, I knew how to make a layout with tables (uuuggghhh), I could fiddle around with JavaScript, and I had an OK knowledge of QBASIC and Visual Basic. I am eternally grateful to VCU for teaching me databases (with a focus on Oracle, but we did use some MySQL), PHP, Java, and C++.

Now, their languages of choice are the languages bulked with Visual Studio. Java isn’t taught any longer (from everyone I’ve talked to), and PHP is being phased out.

I have a buddy of mine that goes to a local community college, and is working on transferring over to VCU. They’re making him use Dreamweaver for all of his web coding, connected to a hosted Microsoft Access database.

I completely understand why they chose the languages they did when I attended. It was to learn structure, formatting, and flow.

However, over the last few years, as I’ve started to work on more freelance projects, I realized how far out of reality they are when it comes to real world projects.

For starters, anyone looking to do programming on their own needs to know how to set up a server. We touched on some basic linux commands and had some basic SSH access, but I feel like I spent my life in Midnight Commander because it was a GUI that let me do everything I needed to do. Never once did we talk about find or grep. We never talked about configuring a MySQL server, installing Apache, understanding that there were other servers besides Apache, or learning what “dependencies” were.

While in the PHP class, we dealt with some interactive forms with some rudimentary database connections, we never talked about frameworks such as CakePHP or CodeIgniter, or even the MVC structure.

In the database class, we learned Oracle inside and out, but were required to make an application using the tremendously horrible Oracle application builder software. I felt like we spent more time figuring out how to connect the application to the database than actually working with database structures, like joins, unions, etc.

The application development classes were focused on Java. I can’t complain about that – JavaScript is a direct relative, and when I worked with Flash for a little while, ActionScript was a derivative of it as well. I feel Java helped me understand control structures. The C++ class further honed that ability, but I feel I got little out of it. I still can’t understand the difference between a .m and a .h file, except that one’s a template for the other (which, arguably, doesn’t make sense).

While I’ve learned Ruby on Rails, MVC, CodeIgniter, MySQL, and how to set up a server on my own, I feel that some of the essentials were missed in lieu of continuing on a comfortable path instead of an innovative one. The people that write these courses must understand the technology inside and out. I get that. There also was no iPhone, and Ruby was in its infancy in 2005. Frameworks were just coming into play.

However, seeing where schools are taking things today genuinely worries me. I believe my education was well worth every dollar, but… there needs to be more.

Where’s the focus on testing? That’s still a concept I struggle with daily. I learned the old-school way of simply trying the form out with dummy data. Over and over and over and over and over again. Breaking it in every scenario. We had to write reports on how we would test the application, but never wrote a dime’s worth of testing code. So, I still write applications in that manner as that was the way I was taught, and arguably, I can put my applications up to anyone’s TDD applications.

I believe some of the Visual Studio languages do incorporate a MVVM structure, but it’s still not the same as MVC, which is (again, arguably) a more popular framework.

And using Access for a database? They must be absolutely insane. Access is a joke. I wrote a database system (frontend and backend) for a small company with about 5 users using Access before I knew better. Eventually we transitioned the backend to MySQL, and the speed of the system increased exponentially, but we were still using Access for the frontend, and it was prone to crashing. All. The. Time.

But what really scares me is how they’re teaching people technologies that are platform dependent. To code in .NET, you MUST have a Windows machine, or at least a virtual machine with Windows loaded up. You can teach the same fundamentals in PHP/MySQL, and not worry whether the developer is using Dreamweaver (which for real developers is a useless, bloated application), or TextMate, or Coda, or Notepad++. You can teach fundamentals of data structures using something like Ruby or Python just as well. Pointers and Garbage Collection are important topics, but are becoming less and less relevant as we’re getting away from using compiled languages, and more toward interpreted ones.

The only real exception is writing actual software applications, but it’s still difficult to find classes that teach application development for the iPhone or Android. And realistically, why aren’t schools teaching iPhone development or Android development instead of C#?

And while WordPress is a rather simple platform to use, why is that not something that’s ever taught? It’s a great cumulative overview of combining both PHP and design – one thing a lot of developers overlook. If your design sucks, no one’s going to use your code, no matter how stellar it is. There could easily be a class on CMS applications – making students develop sites in WordPress, ExpressionEngine, Joomla, and Drupal. Not every web site needs the overhead of Ruby, or a hundred custom built methods when you could load 2 plugins in WordPress and get everything you need.

Lastly, what about NoSQL databases? I can’t say I’ve ever heard of a college or university talking about MongoDB or CouchDB. While I don’t think every project needs a schema-less database, as much as they are gaining in popularity, it’s hard to ignore the fact they’re out there.

I just worry about what’s happening with the schools and how they’re evolving in their IT programs. I feel the focus drastically needs to shift from “tried and true” to “upcoming and new,” while teaching the essentials more thoroughly. And if they’re not going to teach them, at least provide those other options as additional resources for students to research on their own. Again, I’m grateful for what I learned at VCU, and I wouldn’t take it back if I could, but I make my living off MVC framework-based languages, and on things I never learned in school.

What do you think?

Typically, in ruby, a method is declared as such:

def somemethod(arg)
  # do something
end

However, today, on complete accident, I wrote the following:

def somemethod arg
  # do something
end

No parenthesis! And it still works fine. I ended up rewriting it with the parenthesis for readability sake, but I thought it was interesting and I thought I’d share. I am using Rails 3.2.5 and Ruby 1.9.3-p194 at the moment – I’m not sure if this applied to previous versions of Ruby or Rails.

A week or so ago, I pushed some changes to carrierwave_securefile. It’s probably still a bit buggy, but that’s why it’s not a 1.0 version!

Additionally… well, this comes with a story:

If you use git to push changes to a repository, you know how tedious it can be to run this every time you want to push some changes:

git add .
git commit -m "my commit message"
git push

So, in about 5 minutes, I whipped up a super-handy script that I call Gitify. It compresses all of those commands into one, making the “push” an optional parameter. Check out the Gitify GitHub repo to check out the usage and download it.

After a bit of further inspection, I discovered the glorious Linux app known as htop. It’s like top, but just a heck of a lot more useful. With htop, you can scroll the list, kill processes with one keyboard tap, and see awesome graphs of your current CPU’s status. It’s pretty awesome, and I would encourage anyone that uses top often to give it a spin.

On CentOS, you can type yum install htop and it’ll be on your machine in no time flat. Simply type htop to start, and you’re off to the races.

If you want more information, you can visit their project page over on sourceforge.

You’re welcome.

This weekend, CodeSchool.com offered a free communication weekend to meet your next divorcee free access to their site.  As this weekend was an important weekend for myself and my girlfriend of a year (yay anniversary!), I didn’t know if I’d get around to messing with it.  However, I was graced with some free time, so tonight, I’ve spent a good chunk of my evening playing around with their Rails Testing For Zombies course, and got all the way through without a single answer wrong – sweet!

Disclaimer: I technically did get the last answer correct, it just didn’t like the manner in which I wrote it – it didn’t like my use of assert_select, and wanted me to use Capybara’s “within” instead…

So with that, I’ve got a chunk of badges earned from CodeSchool.com.  Check ‘em out here:

http://www.codeschool.com/users/dougc84

After doing a bit of researching, I stumbled across an app called PhoneView. It’s something like $20, but it’ll archive and backup all over your texts, phone logs, etc. It also lets you download your voicemails and manage your application files. Pretty powerful little tool. I figured I’d give the trial a spin. Now, my girlfriend is not the only person I text, but she is the only person that I never delete a message for. So I knew when the total text message count came up, I knew it was all us.

22,000+ messages.

Sure, some of you might think “that’s it?”  But, unlike most techies in the world, I have no problem with picking up the phone and/or not responding to every single notification that comes up on my phone.

PhoneView offers not only their archiving, but also allows you to export an entire conversation, with attachments, to a folder of your choice.  You can get a TXT, CSV, XML, or PDF version of your conversation too.  Pretty nifty.  So I exported it to a PDF (which, by the way, was a 950+ page, 60+ MB file…), crossed my fingers, and hoped deleting all of her previous texts wouldn’t be the end of the world later on.  So I swiped across and hit delete on her conversation.  It took a couple minutes for it to actually delete everything (and… well… if it takes that much time to delete something… yikes), but when it did, I noticed texting felt instantly more responsive.  Swiping a text message notification responds instantly, instead of a 2 or 3 second pause each time.

As I’ve been using my phone for the last day or so, I’ve noticed my phone as a whole is significantly more responsive.  Usually after my morning check of Facebook, Twitter, my RSS feeds, text messages, emails, Groupon, Livingsocial, catching up on Scrabble, Words With Friends, Scramble With Friends, and, my new addiction, Snoopy’s Fair, my battery life is down below 80%.  Granted, that’s how I start off my day – I get all of that out of the way and I don’t worry about anything until later on.  But it’s a lot of content to take in.  Between being groggy and actually getting through all of that, it’s probably about an hour out of my day. Today, at 2:30, my battery is at 85%.  Normally by now, it’s in the 60′s.

So was it worth it? The $20 to pull everything off my phone into my DropBox?  Definitely.  I’ve got an awesome account of meeting my girlfriend via text messages, and all of our conversations, in one nice, neat folder.  And my phone’s running much, much better.

There’s been a lot of buzz in the Ruby world about RubyMotion this week. After looking at their getting started tutorials, documentation (what little there is), and more, and reading up on what it is and where is comes from, i’ve come to the conclusion that it’s a great idea, but it’s truly poorly executed.

As a developer that primarily works with Ruby on Rails, I was very excited when I first heard about the prospect of being able to write my apps in Ruby. But after looking at the implementation, I’m thoroughly disappointed.

1. I know some basics of Objective C, but I feel like you have to really know Objective C in order to use this. Which begs the following question: “Why would I even want to learn something that does something that’s already been done?” Sure, I prefer Ruby’s beautiful syntax over Objective C any day of the week (and twice on Sunday). I also prefer TextMate over XCode. But interface aside, you’re learning how to write Objective C in another syntax.

2. XIBs and NIBs. For those that don’t know, these are the files generated by XCode’s Interface Builder – it’s your interface into the application, much like a Rails (or Sinatra, or, or, or…) app has layouts and views. This is one of those “make or break” points, which the developer of RubyMotion states that it makes more sense to write the interface in code.

   Wrong.  Wrongy wrong wrong wrong wrong.

   There are far too many Ruby programmers so focused on writing beautiful, elegant, and expressive code that’s DRY that they forget about what’s most important – the user experience. And when you really sit down and think about it, are you more likely to design an app in a language where you have to code every. single. detail. of every. single. element, or in one where you can set up the interface in a few minutes per page (obviously more if you’re using custom graphics, but there’s always exceptions)? This also leads people down the path of throwing buttons up that aren’t placed correctly, interface elements that don’t belong, or custom interface elements that look… quite frankly… bad. Ever opened a free app only to feel like it didn’t really look right? I see every single RubyMotion app like this.

3. Cost. While I would love to start screwing around with this, if only as an experiment in another language, I can’t. Well, unless I shell out $150, that is. For a language that’s purely open source, with frameworks that are purely open source, what would make me want to pay a team of 1 person $150 for an application that’s closed and doesn’t guarantee sustainability, reliability, and fast updates? If this were built by a team (implying more than 1) of ex-Apple employees dedicated to this, then I’d be more than happy to pick it up. But what’s the guarantee that this won’t sit on a shelf untouched for months?

4. Lack of MVC. I mean, come. on. Looking at the file structure makes me want to cry all the way back to my days of learning Java classes and basic PHP code sans frameworks. It’s simply archaic. Heck, you can’t even write a view if you wanted. Isn’t that the point of ObjC, is to have an OO implementation of C, complete with MVC? What good is that if you can’t define models and views separate from your controllers?

5. Poor documentation. What’s provided is a great “how to get started” guide, and a few tips about configuration settings. When it gets down to turning your ObjC code into Ruby… good luck. It seems pretty intuitive if you know the API like the back of your hand, but if you don’t, I feel like you’re going to be doing a lot of guess work. And guess what – increased development time.

So, all that said, it really does look like a promising product, but due to the issues above, I would rather take a class and learn ObjC well, and invest my time in the tools Apple uses to create beautiful and powerful iOS apps. And in the end, XCode’s that that bad, Mr. Sansonetti.