Pete Carroll, why do you vex me so?

Sorry, I know you have all been clamoring for some more SMMRs and I have not been delivering, because football. And because school. And because servers. And because Comic Sans.

Well, the Seahawks are out of the playoffs, so I can spend more of my free time finding great music, instead of yelling myself hoarse at a coach that makes the wrong call at every opportunity.

GRRR! Pete Carroll

School is, well… school. But it’s my last class, and then I am free of this insatiable monster that devours my time and willpower. #SENIORITIS

I have migrated loudifier.com to a new server. Can you tell? No? Good. Before, I was using a generic LAMP stack on a dual-core laptop with 3GB of RAM, running Linux Mint, and being used for web browsing and office documents. Now I’m running Ubuntu Server on a dedicated single-core, 2GB laptop from 2005. I switched to NGINX, APC, and Varnish to make up for the decreased processing power, and it took me a long time to get (almost) all of the kinks worked out. The whole thing performs surprisingly well, and cost me exactly zero dollars. Here, you can see that the new loudifier data center is literally held together with zip ties.

loudserv

Comic Sans, you ask? Well, I read about a charity auction called Cosmic Sans, that sold off space-themed paintings of each letter in the alphabet to raise money for kids who don’t read good. It’s a great project and all, but I really clicked the link hoping to find a font that was a modern replacement for the ubiquitous Comic Sans. When I didn’t find one, I decided to just go ahead and make one myself. Little did I know, making fonts is hard, and extraordinarily time consuming. After sinking about 60 hours into my font, Comic Spans, it is nearly complete. It will get it’s own post soon, but here is a preview:

Comic Spans preview

Apple is half-assing the iPad mini

Without the manic, price-be-damned, attitude of Steve Jobs, Apple just doesn’t know how to make or launch a new product.

The new iPad mini is a great idea. Finally, you can do all the cool, iPaddy stuff you have always wanted to do, only in a more convenient, 7 inch, form-factor. Except it is going to look awful, because Apple only put half of its behind into its oft-extolled Retina display technology.

Don’t get me wrong, the jump from the iPhone 3GS to the iPhone 4 was amazing. It looked great, and the hardware that came together to give you 326 pixels per inch revolutionized the industry. But the software was another story. All they did was double the size of the images in apps, and use simple 2x scaling, which works, as long as you scale your pixel sizes accordingly.

Enter the iPad, then the Retina iPad, then the Retina MacBook, and then the iPhone 5. By this point, you would think that Apple would have ways for developers to make apps that would intelligently scale based on the display they are shown on, because none of these devices has the same resolution, or even the same aspect ratio. You would think that by the sixth revision of iOS, which now accommodates a 16:9 display on the iPhone 5, apps would be coded with some sort of getDispResX(), getDispResY() system so there would only be ONE version of the app for ALL iOS devices. You would think so, but you would be wrong. Now, if the developer hasn’t updated an app for the iPhone 5, you get ugly black bars.

On Android, you can play Robo Defense on nearly any screen, nearly any phone, nearly any tablet. One beautiful app.

Which brings me to the iPad mini. 1024×768, at a whopping 163 ppi. Pretty lackluster when compared to the rest of Apple’s lineup, and even worse when compared to the rest of the 7 inch tablet market. For example, the Nexus 7 comes in at 216 ppi and costs $130 less than the iPad mini.

Apple could have stuck a 1600×1200 display in the same space, and bumped the iPad mini up to 253 ppi, but then they would have had to come up with a whole new range of apps to accommodate a device that they didn’t want to make in the first place. With this lame display, they can recycle all of the apps designed for the original iPad. Using apps for an older device also means they can cheap out on the processor, mitigating the risk of releasing a new device with none of their usual hype and fanfare. 

What they should have done, way back in iOS 4 with the introduction of the first Retina display, was create a responsive framework that allows apps to look nice on any reasonable display. Instead, the developers of three quarters of a million apps have to scramble to fix their apps every time Apple makes a new device.

It’s kind of funny. First, the original iPad was just a giant iPod touch, and now, the iPad mini is just a small iPad 1 made out of glue.

Comcast sucks, is shady

Comcast sucks. I know, it’s not a very original idea, but today I encountered one more example of just how much they suck, and how shady some of their business practices are.

Where I work, we have Centurylink high speed internet. 50 megs down, 50 megs up. I run a speed test occasionally, just to see. There are several websites that offer this service, and speedtest.net is my go to, but they all pretty much give me the same result: 30-35Mb down, 20-25Mb up. We have a huge, complex, network, including live mirroring to a secondary site. I’m in operations, so I leave QoS issues to the network guys. For me, 30/20 is close enough to 50/50 that I won’t complain.

I was investigating pricing for Comcast Business Class, because I operate this site in the “personal-use” category, which is colored decidedly gray in Comcast’s TOS for home internet. They have a friendly comparison page designed to show you how much faster they are than whoever you use now.

This is what happened when I ran their speed test from work:

Comcast Business Class speedtest

It wasn’t a fluke. I tried it several more times, checking other sites to confirm that I was getting the usual 30/20.

Speedtest.net:Speedtest,.net

I understand that Comcast’s reputation is about on par with the likes of GoDaddy or The People’s Republic of China, but this is a pretty bold skewing of information. To me, there are two things that really stand out about just how shady this really is.

Comcast’s business speed test and Speakeasy both use the Ookla speed test software, but the Speakeasy test provides multiple servers distributed around the country.

Speakeasy:Speakeast

I can connect to the Seattle server and get a best-case scenario measurement. Comcast doesn’t give you a choice of servers. My guess is that they have a single server located in a remote part of Nebraska somewhere that will give you a terrible result no matter where you are.

The biggest problem I have with their credibility is a blurb on the business speedtest site that tells you to use the Xfinity speedtest if you are already a Comcast customer.

Xfinity:Xfinity

Note the ability to connect to a local server. My guess is that the business speedtest only has one server, and that it is located in the middle of nowhere. While this isn’t necessarily an untruth, it is certainly dishonest, and is just one more reason to not trust Comcast.

If Comcast wasn’t the only high speed internet provider in my area, I would chew my own arm off to get away from them. They turn over customer data to the MAFIAA at the first sign of conflict, and are fibbing, if not outright lying, about how they compare to the competition. If you can get your high speed internet elsewhere, do.