Travis Berthelot

Don't use empty exception blocks.
Don't create strings for each request.
Don't load the driver more than once.
I would start with that first.

Why tell him to use AJAX on a server problem?

Session is the server object in question and not javascript or a request from a client.

Tim Holloway wrote:Ah, yes. Forgot about Blue-Ray! Since that's a platform that strives to control what users can and cannot do, not only does J2ME save the chip budget for flashier things, the more restrictive nature of J2ME makes it preferable to Android. The same can be said, of course for almost any sort of process controller environment. Just ask the Iranian Atomic Energy Commission.

True, but Iran does have the material needed for a bomb.

I would like to say that when I first used J2ME I had some of the same feelings. Now days I just ask people what can't you draw on a Canvas for J2ME that you can for a View in Android or vice versa. You could replace HTML5 or J2SE as well in either order I still have the same answer. What else do you really need? Sure you have sensors and other crap, but that is really device specific and you have optional APIs for that other stuff.

Rahul Sudip Bose wrote:

Travis Berthelot wrote:Location aware self replicating robot armies. You may laugh at it now, but just wait a few years.

Robots replicating...sounds impossible,fantasy. Anyway how are the two connected ?

So, they will know where they are.

Well I have some glorious news for you. Blue-Ray supports J2ME (well almost) and it is selling like wild fire.

So I don't agree with doom and gloom. Maybe I should open source my game development kit to prove the point, but I am still looking for a real funding before I do it.

On another note, I support Android and J2SE as well, but I will never drop J2ME. Although, my 3D support varies by platform.

Location aware self replicating robot armies. You may laugh at it now, but just wait a few years.

I use J2ME for my AllBinary Platform and there are around 1.2 billion devices around that use it. So hardly dead. Just not a lot of money being spent by business for J2ME.

Actually the module stuff is pretty much done and part of the new version yet to be released.

http://openjdk.java.net/projects/modules/

Since it is not all OSGi like I was wondering what you thought about it.

What do you think is the best free implementation of OSGi?

I used Knoplerfish years ago and it was pretty great.

Also I would like to know what the book author thinks about the new Java Module JSR vs OSGI?

It does so by dynamically loading Java Archives. That way you can use jars that are not in the class path.

Devender Thareja wrote:Travis, you surly are the most money making app developer here.
Did you consider IPhone as your games platforms? There is a larger customer base for Iphone. You must be deterred by Objective C language or the requirement of buying MAC. Right?

I answered this question before, but my post is missing. Hmm.

Well I will give a short answer this time.

I do like Apple and the products they produce. It was not a technical issue it was a closed market issue. Now that an more open competitor to AppStore has come out I will think about it again when I have a hit game. I don't have that game yet.

IPhone doesn't have enough future market to convince me to develop for it as a main platform. However I expect PC/Laptops and Android phones to be over 1 Billion devices. J2ME devices also fit well with a large market, but they have closed access controlled by the network providers. As such I will keep MIDP in mind since carriers may get sane and see the light. This could hurt Android domination, but I would still welcome it.

I would also think about putting them on a game console as the quality and sales improve.

So, I will hold off on IPhone until I have a hit game since the market is not large enough. Yes I know it is 20 million strong, but that is nothing in comparison to the 1 billion PCs and open handsets to come.

You already have 3 applications on market. It's been only a month or so since the Android Market started accepting paid apps. Very Impressive I must say.
You are determined to do more, that tells me that there is a sustainable business model.
Do you develop games for other platforms too.. such as iPhone etc. The 3 games you mentioned, did you port them on Android or you wrote them from start.

I wrote my first game in 1992-1993 using Turbo Pascal a corner pong like game. From 1995-2000 I developed a space conquest PC game using directx and C/C++ called Mortification. Despite working on it off and on for 5 years I never published it, and realized that I would never have the funding to make it what I wanted. During that time I developed some test games with direct x and some basic game testing with opengl. I stopped making games for about 18 months when I was working at a telecom then I decided to make some mobile games. So, I made simple versions of MiniSpaceWar and ZeptoRacer for J2ME back in 2002 and did some testing on my J2ME LG phone. Although, the FCC decided to make it so Carriers controlled the market entry points. I didn't have the 1k to spend getting on a mass market phone so I went back to developing other software (mainly Freeblisket an E-Commerce program and an InputAutomation testing tool). I heard about Android in 2006 and downloaded the non working SDK they did not apparently intend to release. I looked it over and realized that I would port my games over if it ever came out. When I found out about the FCC allowing a more open market I decided to try again. So in October-November of 2008 I created a plug-in to run games on Android for my game libraries. Since then I have mainly added features and made bug fixes to my now 60+ small game libraries.

So I have developed games for the PC in and out of the browser and for mobile devices with MIDP 1.0/2.0 and Android.

Note my Android games also run as a PC standalone app, as an Applet, and J2ME.

Does the fact that there will be too many different type of devices fro Android, bothers or excite you? How and how much do you plan to test you apps for each platform?

It excites my brain, and it plays well with my multi platform game libs. Although, practically I don't really care. I just want to make some games that make me money, and that is all I really care about right now. So all the great things about Android don't have much meaning to me anymore. While gadgets will always excite me it is no a real factor unless it can make me more money.

Since I don't have the actual phone I only have the Emulator. If I made enough to get each and every phone I would. I still may end up getting the G1 as my own personal phone since I don't have a contract cell phone.

Why don't you do what Ed did for his ReTranslate app. He mentions on his blog that he discontinue the demo model and put the full version for free for some time. When it reach about 500 downloads, he increased the price. He writes that after 500 downloads, the applications visibility improves significantly. Are you close to that magic number yet?

LOL, my license servers show >500,000 plays of my games since October. The download stats according to Google are >23,000 for ZeptoRacer which is not correct. So the only magic number I know of is ching $ ching $. I don't know about ReTranslate, but pulling my free versions would be a big mistake. I can eat up big players revenue while they try to kill me off with advert money that I don't have.

My games have an Update and Subscribe notification for those looking for improvements or more levels. So, if they like them enough they will click on the subscribe notification that should take them to a listing with both the free and paid versions. Honestly, since I don't have a phone I wouldn't know if the market url even works.

My sales strategy is probably going to be pretty static for the first 3 game titles that are out. I will probably try one ad game and one 99 cent game to see what happens. The only change I made for sales is a price change for all 4 games in one apk from 9.9x to 6.9x.

While my sales strategy probably could improve I will leave things mainly as they are.

I do however plan some things like when I release my next update and game I will probably have some kind of deal for free registered versions of my games to a person that can come up with the best control profile for a given game. I have plans for other similar customer cooperation for better game play.

So I will try to keep treading water while I make improvements and additions, but game sales philosophy is already down pretty good.

The reality is that demos reduce sales, but without them I can't get people to buy since I don't have advertisement money. So I will crank out free versions with a few levels to entice people until some deep pockets feel I am worthy or until I die.

I wrote MiniSpaceWar, ZeptoRacer, and SantasWorldWar. I have more games and updates on the way.

The free versions are in the top ten percent of 2300+ applications. My pay for versions have slow but continued sales. My sales are close to the worst while that was not expected I did not expect great sales either. So I am below expectations when I started, but given my situation it is not a surprise.

This is a several fold problem:

1. I don't have large business/government controls - No special deals with Google or T-Mobile - Not politically correct in the eyes of many - No comment snipers - No control over false Google popularity data i.e. my downloads are actually a bunch higher
2. I have no advertising budget
3. I have no phone so my game play and controls are probably never great

Possible solutions:

1. Support National Socialism
2. Win the Lottery of Life - get 165 million like Bill Gates did from the government through IBM in 1978, commit fraud like Warren Buffet, or literally win a lotto.
3. Eat up the market with more games

Conclusion:

I will not support the new Nazi regime, and I will probably not win the lottery of life and get free money from some stimulus. So that leaves producing a bunch of games with more simplified game play for the masses.

I have used NetBeans since 2002, and I have found it to be the most useful IDE for GUI development.

Although, I don't know where they are headed with the Thick Client stuff.

Do you?

NetBeans is the best, and most user friendly IDE around. From mobile to enterprise.