Greetings,
It has been a long few weeks since my website host performed an "upgrade". Unfortunately this upgrade broke the scripts which run on my website to make the CryptoQuote Daily Challenge work at all.
I fixed this problem... then ran into another where I could not access the database. Before the upgrade, I used a web-tool to update the database. After the upgrade, this tool is not available. I raised a ticket with my host, but have seen no updates or progress.
Instead, I wrote my own tool to be able to update the database, and just now pushed the first quote, an apology to my users.
I truly appreciate your patience, and especially the kind tone of the emails about this problem.
I'm glad that this has become a part of the routine for many of you, and I have some exciting features planned for CryptoQuote in the next few months.
Thanks again for your patience and support.
Wednesday, December 9, 2009
Wednesday, December 2, 2009
Flashlight 2.0
Flashlight 2.0 is live on the App Store today, and I'm super excited. This is a great update with full customization, something I've been hoping to do for many months.
Unfortunately, there is a problem with the app. I left in a "debug flag" in the app which resets the screens every time you open the app to their standard set. This means any customization you do will be gone the next time you open Flashlight. I'm very sorry this got out into the wild. I'm embarrassed and I will make no excuses. I do, however, wish that I could test the "Release" build that I send to Apple on my own phone before it goes out on the App Store. That could have been the one last test where I realized I'd made this mistake...
As a second stroke of misfortune, I have lost all of my source code for this version. Somehow my computer and Dropbox got confused and my whole directory of this code was never added to Dropbox, and when I renamed folders, I got an old version from when I first started thinking about Flashlight 2.0.
UPDATE: Due to some serious confusion, the source code was tucked away in a different folder on my MacBook! I have submitted the update (2.0.1) to Apple and hope it will be on the store in about a week.
So, to fix what should have been a one-line, five-minute change, I need to redo all of the work for the entire 2.0. I hope that you will bear with me as it may take a few weeks to finish (again), and then a week or two for Apple to approve it for the App Store.
Thank you for your patience.
Unfortunately, there is a problem with the app. I left in a "debug flag" in the app which resets the screens every time you open the app to their standard set. This means any customization you do will be gone the next time you open Flashlight. I'm very sorry this got out into the wild. I'm embarrassed and I will make no excuses. I do, however, wish that I could test the "Release" build that I send to Apple on my own phone before it goes out on the App Store. That could have been the one last test where I realized I'd made this mistake...
UPDATE: Due to some serious confusion, the source code was tucked away in a different folder on my MacBook! I have submitted the update (2.0.1) to Apple and hope it will be on the store in about a week.
So, to fix what should have been a one-line, five-minute change, I need to redo all of the work for the entire 2.0. I hope that you will bear with me as it may take a few weeks to finish (again), and then a week or two for Apple to approve it for the App Store.
Thank you for your patience.
Thursday, November 19, 2009
Website stats, iPhone & iPod adoption rates
Greetings, these are percentages of my top 15 devices hitting my website. I filtered out and rolled up percentages so this adds to 100%. iPhone 2.2.1 is hidden because it did not make my top 15 devices...
Notable summary: 70% of my traffic is iPhone. Of the 3.x iPhones, 77% have upgraded to 3.1 (or 3.1.2). Of iPod touches seen here, 50% have upgraded to 3.x.
40.70% iPhone 3.1.2
13.48% iPhone 3.1
7.81% iPhone 3.0.1
8.35% iPhone 3.0
11.89% iPod touch 3.1.2
3.16% iPod touch 3.1.1
14.62% iPod touch 2.2.1
Notable summary: 70% of my traffic is iPhone. Of the 3.x iPhones, 77% have upgraded to 3.1 (or 3.1.2). Of iPod touches seen here, 50% have upgraded to 3.x.
40.70% iPhone 3.1.2
13.48% iPhone 3.1
7.81% iPhone 3.0.1
8.35% iPhone 3.0
11.89% iPod touch 3.1.2
3.16% iPod touch 3.1.1
14.62% iPod touch 2.2.1
Monday, August 3, 2009
CryptoQuote Daily Challenge Woes
There are a few things that seem predictable, one of which is that software problems will usually occur when the programmer is least able to remedy them in a timely manner. I had such a problem this weekend.
The day before my mother's wedding, while I was out running errands, I checked my email on my iPhone to find this message:
The iPhone version of my word puzzle game, CryptoQuote, has a feature where you can compete in a timed challenge each day. The app will download the puzzle of the day from my server. When I first released CryptoQuote for the iPhone last year, I had filled the server with plenty of quotes to last. I was certain that I had 2 years worth of quotes, but I also knew I couldn't easily address this problem until I had some time to sit down and debug what was happening on the server. My quick reply:
Once I get to my mom's house for the evening, I disappear to her den and use her iMac to investigate. What?! The quotes did run out. By now it was the evening, so I put an informational quote online for today and began rewriting a tool to scramble up a quote (and talking with my family). My temporary quote was:
I was still so certain that I had loaded multiple years of quotes, some 900+ quotes. I spent the better part of the next hour writing a simple script to "scramble" a quote for use by CryptoQuote. I finished this, found a new quote, scrambled it and added it to the database for tomorrow, my mother's wedding day. I saved the scramble script to my server so I could download and use it later that night to fill in a few more quotes to get thru the rest of my trip.
By the time we got home that evening, and I got my 3-year-old daughter to bed, I was falling asleep myself and populating quotes slipped my mind. The next day, I saw this email just as I was about to head out to begin the wedding day:
One mistake on top of another, I had a bug in my quickly rewritten "scramble" code. Since nearly one third of the usual number of people had "given up" on this puzzle, I opted not to fix it to be fair to those early birds. I'll have a good quote and test the scramble for tomorrow, I told myself as I rushed out the door after sending this response:
My mother's wedding day was great and again I was exhausted and sleep beckoned. The next day we woke up late for brunch and spent the day with family before packing up and finally making the trip back home. I realized after arriving back home at 11:30pm that I hadn't posted a new quote. Finally after changing the first diaper this morning at 6:00am, I posted a new quote and populated the database. I can proudly say that I have verified, there are over 1900 quotes in the online database, and this will take the CryptoQuote iPhone daily challenges thru the year 2014, including February 29, 2012.
Problem solved.
So what happened? When I first uploaded the quotes, I neglected to account for how sql handles single quotes. So I created a giant file of SQL INSERT commands, one for each quote. The first quote that happened to have this parsing problem wasn't until July 31, 2009, so only about one-third of my quotes were entered into the database. Everything up until the syntax error.
How could I have prevented this problem?
The day before my mother's wedding, while I was out running errands, I checked my email on my iPhone to find this message:
Hi. I play this puzzle every day & for yesterday & today I can't download the daily time challenge. Any suggestions?
The iPhone version of my word puzzle game, CryptoQuote, has a feature where you can compete in a timed challenge each day. The app will download the puzzle of the day from my server. When I first released CryptoQuote for the iPhone last year, I had filled the server with plenty of quotes to last. I was certain that I had 2 years worth of quotes, but I also knew I couldn't easily address this problem until I had some time to sit down and debug what was happening on the server. My quick reply:
Thank you for buying CryptoQuote and for this bug report.
I am currently on vacation but I can hopefully fix this problem later tonight.
I'm sorry for the trouble this has caused. Rest assured the server has quotes for 2 years already.
Once I get to my mom's house for the evening, I disappear to her den and use her iMac to investigate. What?! The quotes did run out. By now it was the evening, so I put an informational quote online for today and began rewriting a tool to scramble up a quote (and talking with my family). My temporary quote was:
My apologies. The daily quotes were erased from yesterday into the future. A real quote will be ready tomorrow and on into the future. Thank you for your patience.
I was still so certain that I had loaded multiple years of quotes, some 900+ quotes. I spent the better part of the next hour writing a simple script to "scramble" a quote for use by CryptoQuote. I finished this, found a new quote, scrambled it and added it to the database for tomorrow, my mother's wedding day. I saved the scramble script to my server so I could download and use it later that night to fill in a few more quotes to get thru the rest of my trip.
By the time we got home that evening, and I got my 3-year-old daughter to bed, I was falling asleep myself and populating quotes slipped my mind. The next day, I saw this email just as I was about to head out to begin the wedding day:
I just tried to complete the cryptoquote for 8/1. Finally gave up.
When I went to look at the solution, n represented n so it is
impossible to solve. What's the deal?
One mistake on top of another, I had a bug in my quickly rewritten "scramble" code. Since nearly one third of the usual number of people had "given up" on this puzzle, I opted not to fix it to be fair to those early birds. I'll have a good quote and test the scramble for tomorrow, I told myself as I rushed out the door after sending this response:
You are right. I had a problem with the server and am on vacation, so I had to do re-write some software to scramble a quote. Apparently there was a bug because it should not have mapped N to N.
I won't change it for today since then some would be able to solve the puzzle while others would have given up.
Should be better tomorrow. Thank you for your patience.
My mother's wedding day was great and again I was exhausted and sleep beckoned. The next day we woke up late for brunch and spent the day with family before packing up and finally making the trip back home. I realized after arriving back home at 11:30pm that I hadn't posted a new quote. Finally after changing the first diaper this morning at 6:00am, I posted a new quote and populated the database. I can proudly say that I have verified, there are over 1900 quotes in the online database, and this will take the CryptoQuote iPhone daily challenges thru the year 2014, including February 29, 2012.
Problem solved.
So what happened? When I first uploaded the quotes, I neglected to account for how sql handles single quotes. So I created a giant file of SQL INSERT commands, one for each quote. The first quote that happened to have this parsing problem wasn't until July 31, 2009, so only about one-third of my quotes were entered into the database. Everything up until the syntax error.
How could I have prevented this problem?
- Run the sql command as a transaction. This way, everything would have to work or else the database would not accept any of them, which would have made the syntax error more disastrous and caused me to seek out the source of the problem right away.
- Made an iCal event to remind me in every possible way (email, popup notifications, etc.) that the quotes were going to run out.
- Kept a copy of the "scramble" tool available on my server so I could add an emergency quote very quickly. Even better, make the tool web-based so I could scramble a quote from my iPhone and solve these types of problems via any machine.
- Active monitoring. I'll be setting up a script to run on my iMac now, to probe the next few day's puzzles, and send me an email if there is any problem.
Wednesday, July 8, 2009
iPhone/iPod usage stats for July 1 to July 7
I get a significant number of hits from iPhone users, and thought I'd share my raw breakdown of the user agents I'm seeing. As you can see (and might expect), far fewer iPod touch users are upgrading to 3.0 -- possibly due to the price of the update, and also possibly due to the fact that users just don't plug their iPods into their computers as often.
19.86% Mozilla/5.0 (iPhone; U; CPU iPhone OS 3_0 like Mac OS X; en-u
13.11% Mozilla/5.0 (iPod; U; CPU iPhone OS 2_2_1 like Mac OS X; en-u
6.83% Mozilla/5.0 (iPhone; U; CPU iPhone OS 2_2_1 like Mac OS X; en
3.52% Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10_5_7; en-us) Appl
2.74% Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9
2.55% Mozilla/5.0 (iPod; U; CPU iPhone OS 3_0 like Mac OS X; en-us)
1.51% Mozilla/5.0 (iPhone; U; CPU iPhone OS 3_0 like Mac OS X; fr-f
1.32% Mozilla/5.0 (iPhone; U; CPU iPhone OS 2_2 like Mac OS X; en-u
1.30% Mozilla/5.0 (iPhone; U; CPU iPhone OS 3_0 like Mac OS X; de-d
1.08% Mozilla/5.0 (iPod; U; CPU iPhone OS 2_2 like Mac OS X; en-us)
1.06% Mozilla/5.0 (iPod; U; CPU iPhone OS 2_2_1 like Mac OS X; es-e
0.90% Mozilla/5.0 (iPhone; U; CPU iPhone OS 3_0 like Mac OS X; es-e
0.83% Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.11)
19.86% Mozilla/5.0 (iPhone; U; CPU iPhone OS 3_0 like Mac OS X; en-u
13.11% Mozilla/5.0 (iPod; U; CPU iPhone OS 2_2_1 like Mac OS X; en-u
6.83% Mozilla/5.0 (iPhone; U; CPU iPhone OS 2_2_1 like Mac OS X; en
3.52% Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10_5_7; en-us) Appl
2.74% Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9
2.55% Mozilla/5.0 (iPod; U; CPU iPhone OS 3_0 like Mac OS X; en-us)
1.51% Mozilla/5.0 (iPhone; U; CPU iPhone OS 3_0 like Mac OS X; fr-f
1.32% Mozilla/5.0 (iPhone; U; CPU iPhone OS 2_2 like Mac OS X; en-u
1.30% Mozilla/5.0 (iPhone; U; CPU iPhone OS 3_0 like Mac OS X; de-d
1.08% Mozilla/5.0 (iPod; U; CPU iPhone OS 2_2 like Mac OS X; en-us)
1.06% Mozilla/5.0 (iPod; U; CPU iPhone OS 2_2_1 like Mac OS X; es-e
0.90% Mozilla/5.0 (iPhone; U; CPU iPhone OS 3_0 like Mac OS X; es-e
0.83% Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.11)
Thursday, February 26, 2009
when 2D is better
My first video game machine was an Atari 800, and every game I grew up loving on that system was in 2D: River Raid, PAC-MAN, Joust, Q-Bert, and so many more.
I'm a social gamer. My preference in games is an experience that is super easy to pickup and play with simple controls. I'd rather play a game that anyone else is willing to try.
This is why my newest iPhone game, LightRunner, is set from a top-down 2D style. That, and this view gives clarity to the whole situation. You can clearly see where the other players are, and where the opportunities for escape are. The 3D view is flashy, and makes for better screenshots, but detracts from the strategic play. Without the HUD on a 3D game, you're knee-jerk reacting to what's in front of you.
I'm a social gamer. My preference in games is an experience that is super easy to pickup and play with simple controls. I'd rather play a game that anyone else is willing to try.
This is why my newest iPhone game, LightRunner, is set from a top-down 2D style. That, and this view gives clarity to the whole situation. You can clearly see where the other players are, and where the opportunities for escape are. The 3D view is flashy, and makes for better screenshots, but detracts from the strategic play. Without the HUD on a 3D game, you're knee-jerk reacting to what's in front of you.
Saturday, January 3, 2009
A few simple things to tune your iPhone apps
Here are some simple things to make sure you are doing in your iPhone app for maximum performance:
* Opaque views
* No [UIColor clearColor] background on text views
* Minimize use of nested views
* One view per .xib file
* Opaque views
* No [UIColor clearColor] background on text views
* Minimize use of nested views
* One view per .xib file
Subscribe to:
Posts (Atom)