I’m a little biased here, being a ruby on rails user, although not professionally anymore. This still makes a great point that most programming purists fail to see.

The point is that the cost per request is plummeting, but the cost of programming is not. Thus, we have to find ways to trade efficiency in the runtime for efficiency in the “thought time” in order to make the development of applications cheaper. I believed we’ve long since entered an age where simplicity of development and maintenance is where the real value lies.
David Heinemeier Hansson
Tuesday, July 12, 2005

I like programming and doing it well, but I really enjoy what it enables us to do. I think it’s just the means to the end, not the end itself.

I am constantly using Expose` on my MacBook Pro. When I started working on Ubuntu at work I couldn’t remember what the setting was called that emulated expose` (at least a little bit).

It’s called scale. Alt+Shift+up is the default key combo.

Sorry about that. I got a little over zealous and moved my website without enabling mod_rewrite. Everything should be back to normal now. Maybe a little quicker.

I finally upgraded to a VPS. No more slugishness. Bought a slice at linode.com. Now I can finally have a decent solution for my rails and merb apps.

The new hellaphone at github can now monitor the queue and updates the progress. More to come soon.

What’s on the immediate horizon:

• Look and feel updates
• Slimming down the code for less to download
• Ability to reorder, pause, and delete from the queue

New release of hellaphone. I scrapped the old code, got rid of cakephp, rolled a lighter, custom framework. Got rid of the installation troublesome mod_rewrite dependancy that cakephp required. I may even end up with a super light MVC after it’s all done. Grab a copy at Github here.

I was getting a new install of hellanzb up and running and noticed that it wouldn’t download the nzb’s from newzbin. I found that the problem was an old uri path in the python scripts. My file was located at /usr/share/python-support/hellanzb/Hellanzb/NewzbinDownloader.py.

Right after the class definition “url” is defined. Just search for v3.newzbin.com/dnzb. Change http://v3.newzbin.com/dnzb to http://v3.newzbin.com/api/dnzb and save. 

I hope this helps out others until hellanzb gets a fix up. I haven’t checked, but this may already be in HEAD.

Isn’t it funny how everyone has a good idea for a website/service that will make money?

Does this ever happen to you?

Every now and then one of my friends or colleagues gives me a call with a new idea for a website or service to offer. They call with excitement and a tone of voice that sounds like they believe they’ll be rich soon. They never do any due diligence to see if the same exact service or niche is already being filled. Even when their idea is unique that seems to be all they want to do, tell me the idea. Tell me the idea and collect some sort of paycheck for basically doing nothing but talking to me like a lush would to a buddy during a long night out.

It’s a real shame too, some of these ideas are genuinely good. Unfortunately, I only have so much time. I usually require them to do some leg work.

• Something like research what other services are out there that compete or are related. Make a list. 
• Define an initial feature set.
• Brainstorm our business model.
• Define costs
• Brainstorm marketing ideas

Most people never get past step one. This weeds out the majority. Some will actually define a feature set, but then tend to drop off past the business model. Defining costs almost always loses everyone else.

I may start giving quotes to these people with the ideas, but that would require more questions that they probably wouldn’t ever get back to me on.

I started using git full time for my personal or side projects. Git is so much easier to get running than svn, but still doesn’t have a way to easily sync sandboxes (branches in git speak) after getting up and running like svn. I needed to setup a remote repo to push and pull from. It’s very easy.

Here is the code to do it.

If you don’t have git, then get it.

• Ubuntu: sudo apt-get install git
• OS X: http://code.google.com/p/git-osx-installer/

 

• ssh your_remote_server.com
• cd /var
• mkdir git && mkdir git/your_repo_name.git
• cd git/your_repo_name.git
• git –bare init

On your local machine:

• cd your_local_git_repo
• git remote add origin ssh://your_remote_server.com/var/git/your_repo_name.git
• git push origin master

Can a coder have too high of standards?

I overheard someone say they dislike someone with too high of coding standards. As if this is a major problem!

The reasoning behind this thought process was this: They usually aren’t pragmatic enough to be goal oriented to meet deadlines and are more concerned with technology.

I say, “Bullshit”! These are not mutually exclusive.

• Well written code can help meet deadlines.
• Code that is easier to read always takes less time to add features to.
• Code that is well designed saves time modifying and/or adding features
• Fixing bugs is always easier in well designed code

I’ll concede that the first time something is developed, writing it to be modular and not one off takes a good programmer a little longer. The second time around though that time is made up and then some. Of course this is not an issue if you never have any changes, scope creep, or a boss/client that knows exactly what he wants and specs it out perfectly the first time.

I’ve been in many projects that are rushed and shortcuts were taken. The code is unbearable by all but the original coder. Even he takes awhile to get things done. Not that it’s ever my job, but I always spend the time rewriting reusable, elegant code then documenting it. I always spend more time than I would have if I had just patched it, but I save time for me next time and anyone who must maintain it.

Sometimes the rewriting is pieces at a time. This means making everything backward compatible with what is currently written and being used at the time of being written. As I replace parts of the code, I go back and remove the code that was made backward compatible for the pieces being replaced. Eventually the old unmaintainable code is replaced by lean, easy to read, documented code.