Categories
Dev Ideas Tech Web

Kevin Rose’s Tiny Blog Idea

I like Kevin Rose‘s idea. I share the feeling that something is missing.

Sometime ago, I had this idea that I could start a project/startup all live via Google Hangouts and screencast. Every moment I was working on that project I would open Hangouts and share my thinking process, contacts, discussions, coding, etc. This would obviously be stored on YouTube and the user could go back and/or know every day what I was working on and why each feature was the way it was.

It is not the same idea but I believe that behind both ideas lives the same feeling/frustration. Sharing not only the end result (text, photo, code, feature) but also the feeling/process.

Categories
Dev Tech Web Works

Why you should NOT use GoDaddy has your hosting provider

GoDaddy is a great place to register domains. PERIOD

I have been running a dedicated server little over one month we can report 3 occasions of downtime. And many other issues when the server simply is acting erratic.

On a situation, I waited (over the weekend) 36 hours for a ticket reply.

Last week during a business meeting it stopped working, I went to the status report page of GoDaddy and it said that dedicated servers were with problems. Took them 5 hours to solve.

Today I started getting errors from crons running and I can’t ping ssh.github.com (unknown host) from the server and Just-ping results ok all over.

It is sunday (I work everyday, contrary to GoDaddy support). The chat support for dedicated servers is has good as not being there. Hours waiting and no answer. And that power cycle message that I just started and that can make the server go down for 30-60 minutes (again).

Went to the status page, all good. Then, went to dedicated server control panel:

GoDaddy status vs reality

I think this picture illustrate the question.

If I didn’t have so much on my to-do list. I would change it today. But I have deadlines and things already planned. The only thing I know is that I will move out of GoDaddy.

Categories
Android Dev Mac Tech Web

Dealing with hosts on Android SDK Emulator – MAC OS X 10.8

I was getting some odd bugs with responsive design on some Android phones and need to test it. After spending an afternoon at the Shopping Mall annoying the everyone to let me test every Android phone I could find I decided to install the Android SDK on my MB Air running OS X 10.8

After some tweaks I managed to set the emulador working, there is plenty of tutorial online on that I couldn’t test my dev sites because I always use hosts and your localhost on the emulator 10.0.2.2 (if using MAMP with port 80 or add the port after the ip eg 10.0.2.2:8080).

So my solution was:
As I use a define so I only change one file from dev to production I have a define for my WEB_URL the only question was to change that define:

if($_SERVER["SERVER_NAME"]=="10.0.2.2"){
	define("WEB_URL", "http://10.0.2.2/myfoldername/public_html/");	
}else{
	define("WEB_URL", "http://myhostname/");
}

BONUS TIP: Giving the Emulator localhost access

When you create a Avd you need to give it a name change myAvdNameHere to the name you gave Avd and this way you have access to localhost

emulator -avd myAvdNameHere -partition-size 128

More on this Emulator localhost acceess

Note to self:

cd ~/Development/android-sdk-mac_x86/sdk/tools/
./emulator -avd phone -partition-size 128

Categories
Dev

Mysql Delete with null left join

If you do:

Select * from table left join other_table on id=ext_id where ext_id is null

it works great. But if you want to convert that select to a delete you need to include ‘table.*‘:

Delete table.* left join other_table on id=ext_id where ext_id is null

It kind of makes sense, but MySql could be smart and get that we only want to delete from where there actually exists content.

Categories
Dev

Analysis of long tail keywords for SEO with Google Analytics

After the recent Mayday update to Google SERP I decided to have a look into how much was the long tail keywords influenced by this.

Long tail keywords are normally keywords with 3 or more words. To track it down I used the following regular expression: [^\s\+]+(\s|\+)+[^\s\+]+(\s|\+)+[^\s\+]+

Steps inside Google Analytics. Traffic Sources->Search Engines->Google->Non paid

On the filter keyword [containing] I input the regular expression: [^\s\+]+(\s|\+)+[^\s\+]+(\s|\+)+[^\s\+]+

After I just tracked down the last 3 month number of visits with and without filtering the long tail and calculated the % value of the longtail.

long tail keywords/total organic = % of the value for the long tail

5 minutes and I realised that mayday update did *not* influence my sites long tail keywords.

Hope that helps and if someone knows of a simpler or more effective way to do this I would love to know.