Double Feature: Sudoku Pseudo Updates and Facebook App Security

Two separate but slightly related thoughts today.

Updates to Sudoku Pseudo have been completed. Updated to the latest Facebook PHP API and Graph API. Does not mean Facebook won’t change things next Tuesday. If you did not know, Facebook typically deploys changes on Tuesdays. To accommodate the latest changes, had to upgrade hosting plan and upgrade to latest PHP version. As far as I can tell, I have caught everything that changed. I even found some bugs along the way. Of course, anyone that has worked with me professionally knows that I find bugs whenever I use any piece of software.

I also added a very brief tutorial on solving Sudoku Pseudo puzzles. I am not exactly pleased with the result, but it is something and can be a starting point for future refinements. Suggestions are welcome.

Bugs that were fixed include: Practice 2x2 puzzles showed invalid puzzles; time clock was non-functional and/or updated too frequently; undo/redo did not always work properly when reloading the puzzle or going to a second puzzle.

A while ago Facebook changed policies for apps getting a list of your friends. They currently restrict applications to only get friends who have also authorized the app. This means that the pool of friends’ pictures is made up only those friends that have also authorized Sudoku Pseudo. I made a change so that is has a preference for using your friends’ pictures over the stand-in pictures, i.e. yellow smiley, red square.

Facebook App Center submission guidelines apparently have changed. There used to be a requirement for a certain number of daily users. New requirements are for additional graphics to be developed - another future project.

It would be helpful though if you authorized Sudoku Pseudo. At least one person has commented to me that they did not even know it existed. You do not actually have to do the puzzles unless you want to. Authorizing the app increases the number of authorizations that Facebook sees and the number of pictures I see when I am doing puzzles. See below for further thoughts on authorizing unknown Facebook apps.

*** Second Feature: Facebook App Security ***

Every so often I post recommending that people check their currently authorized Facebook Applications and to remove any that you no longer have a use for or are unfamiliar with. For some games it is beneficial to your friends that play to still have you authorized as a user. Farmtown, Farmville, and Mafia Wars come to mind. So even though I don’t use those anymore, leaving the apps authorized benefits my friends that do actively play them.

This comes to the forefront because of the recent “This is…be like...” meme. I have seen reports that this is click bait and is collecting information about you for who knows what actual purpose. This is why I always suspect links that ask to be authorized to access my Facebook profile. I am sure many of them are benign. But some of them could be quite dangerous.  I personally tend to be more security phobic than most. I do know people that are worse than I am. But think of this: You should not make it so easy to steal from you that you actually encourage those that normally would not to actually do so. Simple example: Would you leave large sums of cash sitting in the backseat of your car fully visible to the casual passerby? This will get your car window broken. So at least do the simple things to protect yourself.

You can check your authorized Facebook apps by: 1) Logging in to Facebook. 2) Click the down arrow on the top blue bar on the far right. 3) Select settings from the menu that shows. 4) Click on "apps" on the menu on the left. 5) Click show more/all in the center portion. If you hover your mouse over apps, you can click the X to remove or the pencil to edit what permissions you give the app.

So if you do not want to authorize Sudoku Pseudo, I understand. But if you are willing to authorize every other Facebook app that comes along, I would hope that authorizing Sudoku Pseudo wouldn’t put you out. I promise that Mad Elf Software won’t intentionally abuse the access.

*exit rant soapbox*

You, Risk Management, and the Power of Two

Risk management is not a term heard by many. But it is important to most people and most people practice it in some form every day. Driving a car is risk management. Walking across a street is risk management. You have something to do or somewhere to go and there are potential problems in completing those tasks.

The power of two is guidance for when you should evaluate your past estimates. If something takes less than twice as long as you expect, this is probably acceptable. Think of your daily commute: If it normally takes 30 minutes and one day it takes 50 minutes, probably nothing to be concerned about as it was just a bad traffic day. your normal 30 minute commute took 20 minutes, then perhaps the lights were in your favor. The concern points should be around twice your estimate or half your estimate. This is the power of two: Both 1/2 and 2 are powers of 2.

If your commute took 2 hours, it probably wasn’t just a bad traffic day. Perhaps there is something more you need to do, i.e. investigate a new route, leave earlier, leave later. If your commute took 15 minutes, it probably wasn’t the lights in your favor, but that something else is amiss. Perhaps you are going to work on a holiday. Or perhaps you found a better route or worse everyone else found a better route. These are the points where you should think about what might be going wrong. It still may be nothing is actually wrong and nothing needs to change.

This applies to any task that you estimate for yourself: knitting that scarf, reading that book, completing homework. So if you find yourself done in half the time, spend a couple minutes thinking about why it was so quick. Could just be you overestimated. But it could be that you missed something important. If you find yourself done in twice or more time, you should also spend a few minutes thinking about what went wrong.  Perhaps it was outside forces that you had no control over, i.e. it rained. Or there could have been something that you should have accounted for but didn’t.

Two tasks on my recent list were catching up on business accounting stuff and giving some TLC to Sudoku Pseudo. I had estimated the accounting stuff to take me a day or so. It ended up taking me a day and a half. No reason for concern as the reasons for the delay were obvious and it was still under the twice the estimate mark. Sudoku Pseudo has turned from what I expected to be a day project to be something that might take as much as a week. This is more than twice. So what went wrong? Facebook changed the API significantly. I stopped receiving updates from Facebook about those changes. The changes require changes to my web host. It is not Facebook’s fault. I know they change things constantly. Since I did not spend the time to keep track of what changes were being made and making changes to keep up or even notice that I wasn’t receiving updates anymore, I am forced to make a whole bunch of destabilizing changes at once and taking more time to do it.

New Beginnings

Today marks the beginning of a new chapter of my life. Friday was my last day as an employee of Everyday Health. You did not even know I was employed there and now I’m gone? You are not alone.

More than 5 years ago, I started what would become a long-term contract with TechSafari. After working on various projects with them, I started on consistently work on the Tea Leaves Health contract for them. In August, Everyday Health acquired Tea Leaves Health. Acquiring companies generally want acquired workers to be employees instead of contractors. They feel that this somehow gives them more control over the domain knowledge. So I became an employee.

Also during the past couple of years, I have been working on my own projects. Being an employee instead of a contract curtailed a lot that effort. Now is the time for me to go back to putting my passion into those projects. I hope that something becomes of them. Devoting full time to them will help them progress.

Didn’t I do this before? Yes and no. I stopped being a full time employee twice before. Once there was a plan, but the finances and time just didn’t work. The second didn’t have a plan, but I ended up being Mr. Mom, good son-in-law, and parent caregiver. This time around, there are no kids at home, parents are healthy and not moving, and finances are in line.

I will miss those that I left behind. Our paths will probably cross in the future, as the IT world may be big, but it is not that big.

There is a general plan with goals. There are time limits. This is the time for me to be focused and passionate on what I want. The plan is not set in stone. First task is to get organized. Accounting, bill paying, setting up work environment, blogging (this), and further refining the plans for projects that I can work on. I plan on blogging to keep myself in line. I probably do not do this every day, but goal is to at least do one a week.

The first project was going to be an integration project, but that has already been derailed. I usually do not like talking specifics about projects I am working on. This is for two good reasons: Generally the ideas are so vague or hard to communicate that I just end up waving hands when describing them. The second is that plans change and it seems that whenever I try to give details, those projects tend to fail. But I’ll do a whole blog post on that project as I think it is important to understand why it failed.

The second project is doing some TLC on Sudoku Pseudo. Nothing really big: updating the Facebook integration code and perhaps finally writing the tutorial.

The third project is trying to get some exposure and feedback for Maze Pseudo. Yes, I have a iOS game released. It happened right before WWDC and E3 and I haven’t really had time to focus on following that up more than getting everything setup for it: web site, forums, Facebook page. This won’t be to involved and is setup for future changes to this project.

The fourth project is what most would consider the first big project. I have an Apple TV for development purposes. Plan is to integrate Maze Pseudo with it. This will require supporting controllers for tvOS and iOS. I also plan on looking at the networking code, as this is the current bottle next for making the game play faster.

Beyond that there are projects for a cross platform tablet game, a multi-console game, another Apple TV app, two iOS productivity apps, and even a couple non-IT projects. I am not planning on doing all of this at once. I don’t really expect that these will all be done this year. It is possible that they never get done. The order has not been determined either. I have also been asked about doing a couple of other projects for/with others.

Right now the focus is to do enough on general planning, Sudoku Pseudo, and Maze Pseudo, so that I can get to working on the Apple TV Maze Pseudo project. The other projects are just what could be next.