Wednesday, September 07, 2016

Demise of ITT Tech

Looks like The Obama Administration was successful in shutting  down ITT Tech. I wander if they will be able to shut down the University of Phoenix this year?

The closure of ITT Tech makes me sad. A decade ago the school had a curriculum that was better than most local community colleges. ITT Tech students were actually getting a higher return for the amount of money they invested in education than other schools.

ITT Tech was briefly owned by ITT and was truly focused on giving students the skill sets needed for the technology environment of the day.

Businesses tend to align with their funding source. The primary funding source was the Federal Government and guaranteed student loans.

So, instead of focusing on developing a better curriculum for less. It appears that ITT Tech took the low road and turned into a mill for processing student loans.

Even worse, ITT Tech appears to have devolved into a student loan mill during a weak job market. The poor students would pop out of the company into a soft job market and simply be umemployed and deeply in debt.

I had the misfortune of wasting two years of my life as an instructor for a student loan mill in Utah. The company was very good at getting students to sign the dotted line on the loan application, but was providing a subpar product.

Many public universities appear to be little more than student loan mills. Since the universities don't have private investors, they don't get raise the ire of the public. When public universities go bad, Democrats simply yell for higher taxes.

Anyway, now that the Obama Administration has the formula for closing down private for profit schools, I wonder how many schools he will be able to close before the end of his administration.

Tuesday, September 06, 2016

Replacing Affiliate Ads with Direct Ads

I am upgrading the Community Color Web Sites to PHP 7. I am also replacing MySQL with SQLite3 and upgrading to HTML5. It's a big project that I hope to complete this week.

As part of the upgrade I will replace the Adsense Ads and Affiliate Ads on the site with direct local ads.

I like the basic concept of affiliate advertising. A site with a product to sell offers a commission on sales. Web sites run their ads and get an occasional commission.

There are tens of thousands of merchants with affiliate programs. These merchants offer millions or products.

I believe that affiliate marketing is a good thing. Because there are so many products, web sites can run ads for things they support without compromising their message.

The problem I have with affiliate marketing is that the huge networks that offer affiliate ads aggressively track user activity. These use the information they gather to manipulate user behavior. They often engage in activities that undermine the sites that run their ads.

I believe that my sites are well enough established to break from the clutches of the big marketing firms and that I will be able to replace the ads. Currently, I have over 10,000 page views a day. My 10 year old server freezes up when I have more. I believe the new new server and improved code will be able to handle a hundred times more traffic.

I decided that the best way to handle local advertising is to sell the ads in blocks of 100,000.  I am currently getting about $30 for every 100,000 page views. Since I will bill in advance, I decided to charge $25 for every 100,000 page views. This is a super low rate of 25¢ CPM.

The new site will go live in a week or so. I would love to have a few people buy ads before the site goes live so that I can tweak the program.

The ad is a 728x90 banner that is in the upper fold of the page (between the page title and contents). I put text ads in a stylized SVG block. You can pre-order ads on my test server.

Use this form to order text ads.
Use this form to order a banner ad (you need a link to a 728x90 banner).

The ads won't go live until I switch to the new server which I hope to do soon.

I only have 10,000 page views per day (which I've split into targets). I only want ads related to the target. I don't want to show ads for Denver Lawyers on the Moab Real Estate page. The flow of the program is as follows. A person orders an ad. If I have a relevant target, I will approve and place the ad.

Some time after the new site is live, I will send the user a payment request using Paypal (Paypal accepts Visa, Mastercard, etc). The request will show the ad placement. If the person is happy, they can pay the bill. If not I will remove the ad. I am using a delayed payment instead of a money back guarantee.

Monday, September 05, 2016

Still No Date Picker in Firefox

I am bummed out. Firefox still hasn't implemented the date picker that was in the HTML 5 specs that came out a few years back.

I hold that the date picker should be provided by the browser because it gives the client a unified experience across sites. Here is a sample form with the new HTML date types. Chrome and the other browsers I tested have date pickers. Firefox does not.

This means I can't depend on the feature. Arrrggggggghhhhhhhh!

Friday, September 02, 2016

New Privacy Policy

I created a new privacy policy page for the Communtiy Color sites.

The privacy policy tells the world that [gasp] I track activity on the site.

I track the activity to see what parts of the site people visit. I use this data to figure out how to expand the site. If people are interested in sports, I add more info about sports. If they are interested in history, I add more pages about history.

Our leaders and media vilify small sites for using cookies and tracking activity. The EU requires that web sites apologize if they use cookies.

A cookie is simply a device that allows web sites to carry information between pages. I do not use cookies for casual visitors to the site. However, I do drop a session cookie when a user logs in to the site. I need the cookie so that they can modify the information on their account.


Now, I find the idea that every site provides a privacy policy to be absurd. The only privacy risk of small sites is that they might send spam if they require a user's email address. This risk is adverted by being careful in giving out one's email address. Personally, I use multiple emails addresses and never give up my most private addresses.



As for this new policy:

I've been hoping to replace my large national advertisers with small local advertisers; So, I am using the privacy policy to sell my new local advertising product.

My advertising product seeks to replace advertisements from large national advertisers that track user activity with small local advertisers that do not.

So, while I think the demand that every site includes a privacy statement to be foolish, I am not adverse to using the page to sell a product. If advertisers bought their ads directly from sites, it would help replace ads from big advertisers that create user profiles with small sites that do not.

Thursday, September 01, 2016

Online Advertising

I've been working on upgrading the Community Color sites. The process is taking longer than I had hoped.

The new design will offer targeted local advertising.

The way this works is I assign each page a target. The Design an Ad page allows advertisers to select a target and create an ad related to the target. The program will allow both text and image ads. I create the text ads with SVG.

The cost for the ads will be $25 for 100,000 ad views. That is an ad rate of 25¢ per thousand ads. This is an extremely low rate.

The catch is that it might take several years for me to display 100,000 ads on some of the smaller targets.

The new design is not yet live, but my test server is up.

The process will work as follows. Advertisers can design ads on my test server. If I approve the ad; I will send a Paypal invoice to the advertiser after the new site goes live. Advertisers can then accept the placement by paying the invoice or reject it by not paying the invoice.

You can order an ad. See what it looks like on the site and then cancel the ad just by ignoring the Paypal invoice.

It would be fun to have a few ad orders before going live. That way I could test the program with live data.