Related
face it, low level apis and languages are for hobbyists now adays. programming languages like C# are huge. .Net makes it so easy and quick to write applications. that's what professionals use. the amount of applications released will be explosive. go ahead, go to android, it's crappy java, and if you use the native development kit,it compiles against the old instruction set. android has hardly even made a dent in the market. it will never be mainstream,just like linux. whine all you want, but microsoft got a bad rap because of crappy coders who crashed the os all the time. it is no longer a hobbyist OS but a consumer OS.
no matter how easy C# is and everything but sometimes you still need access to low level functionality. I agree it's a good idea to recommend modern languages for development on WP7 or any future OS but I don't think it's alright to completely deny the privileged mode APIs, like they were previously called on WM. It's alright if you need to get special certificates if you want to do such risky things (you need that already on WM6) but it still should be possible after all.
I agree about the Android part of your post. On this board it seems to get hyped quited a lot but in real life sales are not really that groundbreaking like everyone thinks. WM is still a lot stronger. Still, I don't even see any advantages you'd get with a switch to Android!
I'm a .net dev but there's instances where you need the unmanaged APIs to do things. There's lots of gaps in the .net compact framework.
I got a nexus one for me and my wife. I was a WM fanboy but I have to say I will never use a WM device again. I have a zune hd and it sucks at everything. The new browser is worse then WM 6.5. I think WM7 is gonna fail big time. What are you talking about that android is not getting market share? Android will gain 20% more market share in two years.
http://online.wsj.com/article/BT-CO-20100209-717900.html
http://www.readwriteweb.com/archives/android_market_share_doubles_will_overtake_palm_soon.php
http://phandroid.com/2009/11/15/android-stealing-symbian-winmo-market-share/
shep211 said:
I got a nexus one for me and my wife. I was a WM fanboy but I have to say I will never use a WM device again. I have a zune hd and it sucks at everything. The new browser is worse then WM 6.5. I think WM7 is gonna fail big time. What are you talking about that android is not getting market share? Android will gain 20% more market share in two years.
http://online.wsj.com/article/BT-CO-20100209-717900.html
http://www.readwriteweb.com/archives/android_market_share_doubles_will_overtake_palm_soon.php
http://phandroid.com/2009/11/15/android-stealing-symbian-winmo-market-share/
Click to expand...
Click to collapse
Noone knows what's going to happen in two years. That prediction was also made before WP7 was announced so it's somewhat meaningless now. Who said the browser is the same one that's in the ZuneHD? The web browser was an afterthought for the ZuneHD and it's a prime feature of WP7. I expect it to be much improved.
RustyGrom said:
Noone knows what's going to happen in two years. That prediction was also made before WP7 was announced so it's somewhat meaningless now. Who said the browser is the same one that's in the ZuneHD? The web browser was an afterthought for the ZuneHD and it's a prime feature of WP7. I expect it to be much improved.
Click to expand...
Click to collapse
I have seen all the wm7 videos and its the same os. They said WM7 would not be based off of the zune hd os but it is. Watch the zune hd videos then WM7 videos. Same os and same browser. They are just adding phone support. I preordered my zune hd and was every unhappy with it. Nothing new or better then ipod. The browser has to be the worst browsing experience ever. Cant download podcast from the phone. You have to dock it and load the podcast on the phone. Cant listen to music without headphones. Tons and tons of stuff that makes you think WTF.
Interesting considering the browser IS different, and you other complaints wont be there on a phone.
Its not the same OS, its just they standardizing there UIs.
RAMMANN said:
no matter how easy C# is and everything but sometimes you still need access to low level functionality. I agree it's a good idea to recommend modern languages for development on WP7 or any future OS but I don't think it's alright to completely deny the privileged mode APIs, like they were previously called on WM. It's alright if you need to get special certificates if you want to do such risky things (you need that already on WM6) but it still should be possible after all.
Click to expand...
Click to collapse
RustyGrom said:
I'm a .net dev but there's instances where you need the unmanaged APIs to do things. There's lots of gaps in the .net compact framework.
Click to expand...
Click to collapse
yes i know we need certain APIs to do certain things,but nobody knows yet what the new framework will be like. who's to say what is limited and what isn't when we don't have any of this info yet.
I completely agree with o2neouzr. Whay are people whining about it when they have no idea what will and won't be included in .NET CF 4?
I've only done a small amount of phone development on WM6 and C# has been fine. My app responds to missed calls and texts when I'm on the motorbike. I recently discussed porting it to the iPhone with a friend. It turns out even really basic stuff like sending a text, knowing if you've missed a call, running with the screen off are all impossible on the iPhone - even with approval from Apple. I reckon we'll be allowed to do a reasonable amount. It'll still be better than the crappy iPhone.
Think of an Operating System like the rule of law in a country; now there is a rumor that the supreme leaders decide to enforce that the only dress code allowed is T-Shirt, of course, the rationals in breadth and depth as in any totalitarian country is only known to the supreme leaders.
"Death to fashion designers!" yelled the fanatics.
"They can't see the simplicity of T-Shirt!" chimed the supporters.
" It is easy to make!!" cried the guards.
" It takes only a minute to paint!!!" said the propagandist.
" It is ready for mass market!!!!" snapped the enforcers to anyone who dare to challenge the new establishment.
"Burn in hell for daring to wear any other dress! T-Shirt is the future! Any one wearing dress is living in the stone age !"
Of course it is hard for concern citizens like we, fashion designers - native developers, to have a decent conversation with this group of people. How can words explain the beauty of our love ones (customers), walking through the crowds of T-Shirts --- in designers' dress. How can we explain the search of excellence, as the way of life, to so many of us, perfecting the art, year after year, to bring out the best of us for the world to see the beauty of free expression.
It is not like that we can't make T-Shirt, it is no-brainer for most of us. The beauty of simplicity is simple but not simpler. A native developers try to achieve simplicity without making things simpler.
Fred23 said:
I completely agree with o2neouzr. Whay are people whining about it when they have no idea what will and won't be included in .NET CF 4?
Click to expand...
Click to collapse
It's not really about having the limited stuff of .NET CF in mind and doing PInvoke for really standard issues. It's about the whole architecture of the runtime. can you do DLL injection or hook into another process with .NET? Not at all. DLLs don't even exist there. You understand... we need the native API if we want to do all the funny things. Recent example: On the Omnia2 there's the taskmon service running which is closing applications if you open more than 3 or 4 (!!!!). chainfire guy wrote a tool that injects into the service to stop this nonsense. More examples are the File explorer extensions, Quick menu, XTask etc. You don't need this stuff? Fine! Go get an iphone!
RAMMANN said:
It's not really about having the limited stuff of .NET CF in mind and doing PInvoke for really standard issues. It's about the whole architecture of the runtime. can you do DLL injection or hook into another process with .NET? Not at all. DLLs don't even exist there. You understand... we need the native API if we want to do all the funny things. Recent example: On the Omnia2 there's the taskmon service running which is closing applications if you open more than 3 or 4 (!!!!). chainfire guy wrote a tool that injects into the service to stop this nonsense. More examples are the File explorer extensions, Quick menu, XTask etc. You don't need this stuff? Fine! Go get an iphone!
Click to expand...
Click to collapse
Ummmmmmmmmm DLLs do exist in .net. In fact, pretty much the entire .net framework is just a collection of DLLs. Have you ever even written a .net app or any software at all for that matter? It seems your hatred is extremely unfounded.
There likely won't be a file explorer to extend and all of the other things you list get into modifying the UI so those won't be allowed either. They're ugly hacks that shouldn't be needed on any platform. Yea, WinMo needed them to cover up it's many flaws.
You don't need native APIs to do those things necessarily. There just needs to be managed APIs to do them. They won't be allowing it though so it's a moot point.
If you want to completely replace the UI, go get an Anroid phone!
There's a middle ground that they need to find. I feel that's what they're shooting for but it doesn't sound like they're gonna hit it. I'm expecting only a subset of .net CF as even that would allow you to do 'too much' in their mind.
alright I admit I have written bull****. of course there are DLLs in .NET but I rather mean they are handled little different than usual DLLs. So I thought common sense is you call them assemblies instead of DLLs. oh! And yes I have written Windows applications in .NET but always refrained from using it on WM due to performance issues and because it's so incomplete and for all the interesting stuff you have to use PInvoke anyway. Also if you at least had checked the link in my signature before judging then you would have seen that I'm actually writing software for WM. besides that, it's even my daily job to write software for Windows/WM for like... ahm... 12 years now.
RustyGrom said:
There likely won't be a file explorer to extend and all of the other things you list get into modifying the UI so those won't be allowed either. They're ugly hacks that shouldn't be needed on any platform. Yea, WinMo needed them to cover up it's many flaws.
You don't need native APIs to do those things necessarily. There just needs to be managed APIs to do them. They won't be allowing it though so it's a moot point.
Click to expand...
Click to collapse
That's the point actually. What I love about WM is that you can easily (or not so easy) hack your way through it and if that's not allowed anymore in a new revision then a lot of fun is taken for me and I guess for "a few more folks" here on XDA. WM like it is with 6.5.x is REALLY comfortable with me and that's no joke! I'm trying to customize and modify it because it's a lot of fun to do so, not because the OS is so bad. Currently I'm running 6.5.x with Sense 2.5 on my Topaz and everything is running really smooth and responsive and I can't complain at all!!
RustyGrom said:
If you want to completely replace the UI, go get an Anroid phone!
Click to expand...
Click to collapse
I hate Android! I don't really want to code in Java. And their native API is a joke isn't it? If WP7 isn't suitable anymore (and right now we don't really know for sure) then I have to keep going with the old **** aka WM 6.5.x until I'm old and grey. why should I change OS if there's nothing that matches the customization possibilites of our present OS?
I could understand the frustration for not having full access to native API, full multitasking, etc.
It will be no joy ... no fun ... especially when you simply cannot customize the UI "the way you want" (You can still customize the WP7 start screen).
You cannot express your full creativity.
That is really valid concern.
But ....................................
Could this "new situation of WP7" trigger another kind of creativity?
To the extreme, who would think creating fart application? I know, probably this is a bad example of creativity, but still ... it is kind of creativity.
My point, in whatever situation, people will simply adapt and certain people will excel with their creativity!
Could that be YOU?
RAMMANN said:
I hate Android! I don't really want to code in Java. And their native API is a joke isn't it? If WP7 isn't suitable anymore (and right now we don't really know for sure) then I have to keep going with the old **** aka WM 6.5.x until I'm old and grey. why should I change OS if there's nothing that matches the customization possibilites of our present OS?
Click to expand...
Click to collapse
RAMMANN said:
If WP7 isn't suitable anymore (and right now we don't really know for sure) then I have to keep going with the old **** aka WM 6.5.x until I'm old and grey. why should I change OS if there's nothing that matches the customization possibilites of our present OS?
Click to expand...
Click to collapse
If I understand this Charlie Kindel (who supposedly is the guy in charge of all this API/SDK stuff), it's not their intention to lock the "fun stuff" out indefinitely. What he's saying is more in the lines of "we want a consistent and logical environment/ecosystem for developers and users, and in order to achieve that, we need to change from the "let's push everything we have out the door" mentality towards "let's make it really good", and this requires serious focused step-by-step approach". They want to start with "consumer application" and hobbyist developers (the latter probably means fart app makers and such), which is understandable because they want to build consumer appeal first. Other categories of developers will be addressed after that.
So, while I don't expect much "fun" you're talking about in WP7, I think this will evolve over time. It's unfortunate that we're not getting things to play with right now, but maybe you won't need to wait until you are old and grey haired. And also maybe we'll really get great APIs/SDK, which will be more useful, consistent and complete than the current incompletely and sometimes incredibly poorly documented mess. Just trying to be positive here.
RAMMANN said:
That's the point actually. What I love about WM is that you can easily (or not so easy) hack your way through it and if that's not allowed anymore in a new revision then a lot of fun is taken for me and I guess for "a few more folks" here on XDA. WM like it is with 6.5.x is REALLY comfortable with me and that's no joke! I'm trying to customize and modify it because it's a lot of fun to do so, not because the OS is so bad. Currently I'm running 6.5.x with Sense 2.5 on my Topaz and everything is running really smooth and responsive and I can't complain at all!!
I hate Android! I don't really want to code in Java. And their native API is a joke isn't it? If WP7 isn't suitable anymore (and right now we don't really know for sure) then I have to keep going with the old **** aka WM 6.5.x until I'm old and grey. why should I change OS if there's nothing that matches the customization possibilites of our present OS?
Click to expand...
Click to collapse
I agree with you. Many people hate WM6.5, I like it. I enjoy the customizability as many others on here do. However, I would also enjoy 'retiring' from the need to do that. It's kind of annoying knowing that the first thing I have to do when I get a new phone is go home and HardSPL it and flash a new ROM. I've heard some of the newer devices are much better out the door (HD2 for example) but haven't seen this for myself.
I don't think the members of this forum are part of Microsoft's target audience for end users. Devs, yea, but not end users. They are shooting for people disenchanted with their iPhone, Android, and more importantly the other 75% of the cell phone market that's still using "feature phones" (aka dumb phones).
It definitely looks like MS is going to lock down much more than WM6.5. Will we still be cooking custom roms? My guess is yes. But your common user will have more restrictions on what apps can and can't do.
I look at this as a new challenge more than anything else.
vangrieg said:
If I understand this Charlie Kindel (who supposedly is the guy in charge of all this API/SDK stuff), it's not their intention to lock the "fun stuff" out indefinitely. What he's saying is more in the lines of "we want a consistent and logical environment/ecosystem for developers and users, and in order to achieve that, we need to change from the "let's push everything we have out the door" mentality towards "let's make it really good", and this requires serious focused step-by-step approach". They want to start with "consumer application" and hobbyist developers (the latter probably means fart app makers and such), which is understandable because they want to build consumer appeal first. Other categories of developers will be addressed after that.
So, while I don't expect much "fun" you're talking about in WP7, I think this will evolve over time. It's unfortunate that we're not getting things to play with right now, but maybe you won't need to wait until you are old and grey haired. And also maybe we'll really get great APIs/SDK, which will be more useful, consistent and complete than the current incompletely and sometimes incredibly poorly documented mess. Just trying to be positive here.
Click to expand...
Click to collapse
That's a good way to look at it. I just hope that they don't get too restrictive from the start. I think the .net CF would be a good starting point that they should add onto but it almost sounds like they're only picking and choosing pieces from it. If we've got the whole .net CF (maybe minus a few things they don't want us doing) I would be fine with that.
RustyGrom said:
I agree with you. Many people hate WM6.5, I like it. I enjoy the customizability as many others on here do. However, I would also enjoy 'retiring' from the need to do that. It's kind of annoying knowing that the first thing I have to do when I get a new phone is go home and HardSPL it and flash a new ROM. I've heard some of the newer devices are much better out the door (HD2 for example) but haven't seen this for myself.
Click to expand...
Click to collapse
I have used my Topaz with the 6.1 factory ROM for at least 2-3 months. At this time I couldn't complain. I didn't face any issues. I only started to flash new ROMs when 6.5 and later issues were available...
About retiring... I don't think you will. The need to immediately flash a new WP7 device is always given. Just imagine you got a new phone from a Telecom contract and you got all your hubs in pink...
RustyGrom said:
I think the .net CF would be a good starting point that they should add onto but it almost sounds like they're only picking and choosing pieces from it. If we've got the whole .net CF (maybe minus a few things they don't want us doing) I would be fine with that.
Click to expand...
Click to collapse
I'm not much a developer (it's not my profession, I just did a couple of things for fun), but, apart from performance issues which can more or less be improved in the new OS, I've seen two types of limitations with .Net CF on WM 6.x: it's been nearly impossible to make nice UI without awful performance, and there are big gaps in terms of accessing "low-level" and sometimes not so low-level stuff and hence need for PInvoke.
The first issue will probably be addressed by Silverlight and, possibly, better built-in UI controls and such. The second will probably be partly addressed by OS-level push, "live feed" APIs, which sometimes might help overcome the limitations of SNAPI. The obvious question marks are things like non-SNAPI event handling such as hardware key processing, power management, device IO etc. Theoretically, this could be done in an environment such as .Net, why not, but not in its current incarnation of CF 3.5. If they added functionality to .Net, it could make life easier for devs.
Obviously though, some things just aren't realistic in managed code, so no alternative browsers/video players for us apart from shells over MS engines (which means no .mkv support if it's not built into the OS already). It could be possible, though, if they're serious in trying to do things right this time, that they want to create those APIs/SDK together with devs. Say, it would be an awesome, albeit slow, path to sit down with CoreCodec and build a piece of the SDK on a specific case, understanding and overcoming the challenges step by step. Don't know if we can hope to get that kind of attitude though...
It's good to use .NET languages, but it is limited, at least on CompactFramework, so it's impossible to use ONLY .NET: you have to use unmanaged code for creating some things that .NET doesn't allow to.
vangrieg said:
I'm not much a developer (it's not my profession, I just did a couple of things for fun), but, apart from performance issues which can more or less be improved in the new OS, I've seen two types of limitations with .Net CF on WM 6.x: it's been nearly impossible to make nice UI without awful performance, and there are big gaps in terms of accessing "low-level" and sometimes not so low-level stuff and hence need for PInvoke.
The first issue will probably be addressed by Silverlight and, possibly, better built-in UI controls and such. The second will probably be partly addressed by OS-level push, "live feed" APIs, which sometimes might help overcome the limitations of SNAPI. The obvious question marks are things like non-SNAPI event handling such as hardware key processing, power management, device IO etc. Theoretically, this could be done in an environment such as .Net, why not, but not in its current incarnation of CF 3.5. If they added functionality to .Net, it could make life easier for devs.
Obviously though, some things just aren't realistic in managed code, so no alternative browsers/video players for us apart from shells over MS engines (which means no .mkv support if it's not built into the OS already). It could be possible, though, if they're serious in trying to do things right this time, that they want to create those APIs/SDK together with devs. Say, it would be an awesome, albeit slow, path to sit down with CoreCodec and build a piece of the SDK on a specific case, understanding and overcoming the challenges step by step. Don't know if we can hope to get that kind of attitude though...
Click to expand...
Click to collapse
Yup. I'm expecting a 'revolution' of sorts in the UI standpoint. Like I've stated previously, it sucks trying to make a good, finger-friendly UI. XNA will make game/3d creation a breeze. Silverlight will be awesome for more traditional style UIs. I'm just wondering how high level and low level they'll get. For example, if I wanted to make an app that looked exactly like the main UI of the phone would I be able to just create a panel object, give it the text for the title, text and controls on that panel, and it will take care of the placement, input control, etc? Or what if I want to go the opposite route and create a UI of my own, will I be able to do that? I'm expecting things like button inputs to be provided. It would be crazy for them not to. That's part of the benefit of standardizing the buttons, they can easily bubble those up to devs.
The leaked docs show that native APIs and unmanaged code will be available to an extent but only to OEMs and carriers. I'm sure if people make enough noise they'll back down and approve apps that make use of that for others as well if they show the need. Microsoft's corporate culture traditionally hasn't been as 'religious' as Apple's and is more willing to back down on things if pushed.
Hey there, thanks for reading.
I decided that I want to learn how to program apps and possibly games for WP7.
Because my holidays are almost starting, now's the time for me to do so.
I already have Visual Studio for Windows Phone up and running, and I encountered the getting started pages on the official silverlight website.
However, it kinda feels like being thrown into the deep and having to do everything in one big leap, instead of being able to start with basics and then move on in smaller steps.
Does any of you have advice for me on what would be good ways to get started? Any advice is appreciated.
I do have vast experience with PHP, but I believe that the syntax is completely different from SL.
Thanks in advance!
My advice would be to learn the basics of c# first. Any previous online tutorials for c# apply to windows phone 7. There are some minor differences but the basics are the same.
Check out http://social.msdn.microsoft.com/Forums/en-US/windowsphone7series/threads which is the official forums for Windows Phone 7 and there's a lot of good info there.
The code samples at http://msdn.microsoft.com/en-us/library/ff431744(VS.92).aspx are also worth playing around with.
Ideally you want to know C# for a starters. There's plenty of tutorials on MSDN or the web, but I can also recommend buying the book: Pro C# 2010 and the .NET Platform from Apress.
It's learns you the language, and also goes over XAML related technologies, and even though it's mostly WPF (Windows Presentation Foundation) the core principles apply for Silverlight as well.
After you learned C#, it's time to learn Sivlerlight in dept, and by this I mean XAML, and the limitations in the .NET subset used for Silverlight.
For tools you wish to get Visual Studio 2010 Express, and Expression Blend 4. You can get the Release Candidate free for now, to try out development.
Make a note of that your computer need to support hardware-visualization to run the emulator without a incredible amount of lag!
If you're used to programming you might quickly learn Visual Studio, but Expression Blend requires a good amount of time to get used to. I'll recommend setting off a weekend where you use 2x8 hours to learn it.
You'll be using both Visual Studio and Expression Blend at the same time, C# for application-logic (ie. C# code) and Blend for design.
Designing in Blend is really simple, see for example expressionblend.png.
One of the articles I want to recommend you to read once you got into this entire XAML thing, is this article explaining how to do paging, as it's one of the core concepts, but also provides a good exercise in using Expression Blend.
Also, to get a handle of the basics, Microsoft has put together a site to encourage people to get involved with WP7 and Silverlight. The website is: http://www.microsoft.com/design/toolbox/
Its a real beginners course to all the programming one will be involved in on the WP7 platform. This won't get you very far in programming, but will teach you the absolute basics. From there you can always get assistance from the great members of XDA (me excluded)
Pretty cool site. Now we just need them to release a free-WP7 version of Expression Blend.
The current process to get started is way to messy, but I guess they already know that
Allright, thanks for these replies
My holidays are (unofficialy) starting in three days, so I'll have plenty of time available.
Luckily my rig runs the emulator buttersmooth (I ran the unlocked WP7 rom with it, to try the OS in it's current state) so that won't be an issue.
These links should get me going in the right direction (H)
Another must-have link is the unofficial Pivot and Panorama controls: http://phone.codeplex.com/
im ready for this!!!
Wow this looks like an interesting journey. Holidays? If you are a student, you can get a waiver on the $99 startup fee to post 5 applications on the windows marketplace. Check out www.dreamspark.com for more student resources.
Thanks for the information provided! I was stepping into WP7 app programming too.
Yea, thanks for telling me but I already knew about dreamspark
I've checked and I can enter by signing up with the e-mail my school provided.
However, since it's valid just one year im not using the key untill I actually own a WP7-device.
enadiz said:
Wow this looks like an interesting journey. Holidays? If you are a student, you can get a waiver on the $99 startup fee to post 5 applications on the windows marketplace. Check out www.dreamspark.com for more student resources.
Thanks for the information provided! I was stepping into WP7 app programming too.
Click to expand...
Click to collapse
Sounds quite awesome. I only had MSDNAA access so far, but I believe I should have Dreamspark access when I start on my bachelor of engineering, after the summer holidays.
Btw. I made a compilation of tools and blogs that'll help developers getting started, http://forum.xda-developers.com/showthread.php?t=711629
Windcape said:
Sounds quite awesome. I only had MSDNAA access so far, but I believe I should have Dreamspark access when I start on my bachelor of engineering, after the summer holidays.
Btw. I made a compilation of tools and blogs that'll help developers getting started, http://forum.xda-developers.com/showthread.php?t=711629
Click to expand...
Click to collapse
Yumm. I'll get down to learning.
The key needs to be renewed every year .. as long as you have your student email, you'll be fine
My first post on xda-developers... +_+!
Yes, as you were told C sharp learning is important so try hard to learn it then go on learning VS 2010 express and expression blend 4. I am new to WP7 dev too and like you ... am dreaming of building a good application indeed - Games are so tough and complex to build and needs teamwork I think-. I just strated leaning C# and have found a lot of good websites trough google search (Windcape compiled most of them in a useful thread you may bookmark)
However, there are a lot of ressources and books about C# and WP7 programming there in torrents world. you can search them by yourself.
Here is a small cd from lynda.com I just uploaded to demonoid.me.
Lynda Windows Phone 7 App Development
I can post links but you can search demonoid website for it.
If anyone needs invitation to demonoid he may PM me here
Also, for a good intorduction to WP7DT just search google for this series : Windows Phone 7 Jump Start
NB: I wish if we can found or make a small teams of biginners wm7 programmers;
hexham said:
NB: I wish if we can found or make a small teams of biginners wm7 programmers;
Click to expand...
Click to collapse
I'm currently studying Computer Science and i am very interested in developing to WP7. I don't have vast knowledge of C# but in other languages i'm good. Let me know if there is an opening for me . Besti wishes...
Do think about the Trial Version - my version was to short - so people were complaining about it...
I haven't read through the replies yet so apologies if I'm repeating anything, but I've been a .Net Developer for around five years now so I feel like this is something that I can actually comment on.
I didn't train in anything to do with computers I just fell into this programming lark, I actually trained as an Actor believe it or not. Through needing to design a database in order to record my Acting Accounts I eventually got into V.B.A. and after ending up with a part time job using V.B.A. was finally forced to learn .Net through requirement changes at work.
I only mention this, so as you know that I wasn't born to this sort of thing and had to struggle to get my head round it.
I started with vb.net as that was almost the same as using v.b.a. but last year decided to teach myself C#. I've not used PHP so can't comment on any similarities or otherwise but I do know this.
C# was a huge shock to my system, absolutely huge. I almost gave up after a week of trying to translate my existing code, but managed to persevere and I'd say that after about two months of forcing myself to do C# only, I felt comfortable with it. A year later and I'm completely fluent and happy with C#.
It's a bugger to look at, I don't find it elegant at all, but once you get your head round the syntax then it all starts to make sense.
Get a good C# book, I'm not sure of any good ones, I've not used any, and go with the lessons in there, but don't let that stop you trying to develop your own experimental stuff in the meantime.
Syntax basics
Curly Brackets {} tend to indicate a routine block
E.g. A Function/Method/Void called Test would look like this
public void Test()
{
... Your code goes here
}
This is also the case with an If statement
E.g.
if(A==B)
{
... Your code goes here
}
Scoping/Dimming/Creating an object requires the Object type first followed by it's name
E.g. TextBox _textbox
The object being created is a TextBox and it's name is _textbox
A semi colon ; indicates the end of a block of code. If you have a particularly long line of code then you can use multiple lines but will
only need to insert the ; at the end of that particularly block of code
E.g. A=1; (This is the standard way of doing this)
A=
1; (This is not the way that I would actually do this but is just an example of what is permissable.)
I always found it useful to find C# code on the web and try to read and understand it.
The most important thing is though, don't give up. Visual Studio is widely considered one of the best coding tools in the industry and C# itself, once you get past the first impressions is a really easy language to pick up.
Thanks for the detailed explanation - the use of accolades ('{}'), multiple lines and the semicolon and the syntaxt for statements looks pretty similar to PHP.
That might just be enough motivation for me to finally get started.
yes, even though I asked about advice over halve a year ago, I never got quite to it... Maybe these similarities plus the fact that I have a Microsoft Dreamspark developer account plus the fact that in March I might have a device will be of help. If that's not enough motivation, I don't know what will be :#
Go for it!
You can always ask questions here or on somewhere like StackOverflow.com if you get stuck.
Object Orientation takes some getting used to, but atleast the garbage collector should look after your memory for you
Dave
Interesting, I hope you will get something.
Just adding my two cents here...
I thoroughly recommend that anyone getting into WP7 programming takes a very good look at the MVVM pattern (see here - it's WPF based but the difference is really only in the objects available, most silverlight code can be directly ported to WPF and vice versa). Using this pattern not only enables you to easily swp the UI for your app, it also makes debugging a whole lot easier due to the lack of UI code in the classes that actually do the work (the view models).
Another thing to try and get your head around is asynchronous programming with services, but that is not really something you want to start out with Sorry, forgot we were dealing with silverlight for a minute this is something you're going to have to learn right from the start, along with using background threads for processing data.
For anyone that needs any specific advice feel free to PM me. I'm new to the WP7 thing, but looking at the classes available which aren't available in silverlight it shouldn't be too hard to get into. And the design patterns will be the same.
Can some some one point me to the right way and what do I need to know in term of computer language.?
Sent from my HTC-HD2 using Board Express
Check out http://create.msdn.com - it has everything you'll need to get started.
As for knowledge, it really depends on the app you're wanting to develop. If you can get away with using data already available on the net, with little to no processing needed, you can really build the UI in Blend without any programming experience what so ever. Sure, you'll need to wire up some events for buttons and what not, but even this is almost automatic.
If you do need to process the data before using it you should have at least a minimum knowledge of programming though. It really doesn't matter what language, but C# is what's used for WP7 (you can also use VB, but I don't think many do).
Just to point out that C# is almost identical to Java.
You can download WP7 SDK and free Visual Studio editions from Microsoft to start making WP7 apps. Normal apps are written using .NET (you can pick C#, VB.NET or Managed C++) as SilverLight app. Games are also written in .NET but on XNA framework. There maybe additional charges for XNA framework.
The SDK should come with a WP7 emulator software. That's where you spend your time debug and play with your app. In order to load the app to your phone, you will need to pay Microsoft $99/year to open a dev account so that you can pushlish the app to market or setup your phone as a dev phone so that you can load your own apps directly to the phone.
Or you can go jailbreak route and sideload apps that way.
Thanks, m a rookie when it comes to this. i just started to take c++. anyways everyone start with no knowledge right.
brummiesteven said:
Just to point out that C# is almost identical to Java.
Click to expand...
Click to collapse
I maybe should have elaborated on this, at least I intended to
Anyways, Java is generally a learners language, you'll find it quite easy to learn as there are lots of tutorials etc on the internet. If you were to ever study computer science you'd also do java
So yeah, Java might be a good thing to start learning then when you get good enough C# will be a breese.
brummiesteven said:
So yeah, Java might be a good thing to start learning then when you get good enough C# will be a breese.
Click to expand...
Click to collapse
Not sure I agree, as a complete novice you're better off going directly for C# (if that's the end goal) rather than using java as a stepping stone. It may end up confusing the hell out of you if you've done things in java a certain way only to find that C# does them completely different.
That said, I still think the choice of language is secondary - first and foremost you have to get into the right mindset, i.e. you have to learn programming, not a programming language. Of course, for WP7 development (at least for now) it very much depends on the type of app you're doing as a lot (too much?) can be done without ever touching code.
programming has a lot of similarities once you've done enough (same as learning languages in general), however, the beginning learn curve is HUGE. Understanding things from variables to classes to object orientation... and that's just the starters. I wouldn't worry about what language is better/easier, both java and C# are some of the easiest and most powerful around, but it's wrapping your head around the basics in general that is a real challenge.
i asked what language because i started my c++ programming this semester and i wanted to know if this the right path on making apps for the phones. as a right now im doing allot of read and watching allot of videos before i try of making a app, since i never have made one.
to be honest, focus on uni first. learn the principles of programming which you'll definitely cover in C++. get them down pat. once your semester is up and you ace the course, then move to another platform. early on it's not good to jump between many languages. it stuffs up your understanding pretty quickly. if you focus on 1, get it solid, it makes it easier to move to the next, and following.
eventually they'll just feel like 1 language... but focus on 1 to begin with, and that should be your studies, not a phone app.
@emigrating great post. That site has started my journey on App Development for WP7. For any other absolute n00b such as me, check out this series of short videos. It explains everything as clear as day and seems to go at a reasonable pace for a beginner like me. THere are 64 vid's in total, i'm currently on the 5th and am loving it
Thanks , I'm going to check them out right now.
Audio said:
@emigrating great post. That site has started my journey on App Development for WP7. For any other absolute n00b such as me, check out this series of short videos. It explains everything as clear as day and seems to go at a reasonable pace for a beginner like me. THere are 64 vid's in total, i'm currently on the 5th and am loving it
Click to expand...
Click to collapse
For anyone that can help.
I made I calculator app like many have started with, but I have a problem, it only calculate 2 terms. How can i fix this?. Plus I'm trying to to a tip calculator now, I have don't a simple one in c++ can I use the same code?
Sent from my HTC-HD2 using Board Express
I've written an article that I believe will be beneficial to all software developers, specifically app developers. It's about how to architect your application code for simplicity, reusability, and testability.
So be warned, this topic is going to be geared more towards a high-level code architecture discussion.
Snippet:
I won't get into the details (for that visit the links above), but at the high level, the MVP philosophy is to separate your code into 3 logical areas:
Model - business objects, business logic, services
View - interfaces which each represent a single screen of the application
Presenter - presentation logic for Views that coordinate between them and the Model
If I had to make an analogy, I'd compare the MVP pattern to a car - where the Model is the engine, the Presenter is the chassis and linkages, and View is the body panels, paint, stereo knobs, pedals, and door handles.
What you gain by partitioning the code along these divisions is
* simplicity - by separation of concerns
* portability - via UI-agnostic abstractions and interfaces
* testability - removing UI dependencies and extracting out as much high-level logic as possible means you can now write more automated unit tests and have more coverage
Traditionally, if you wanted to write a mobile app and deliver it to several platforms (say iOS / MonoTouch, Android / MonoDroid, Windows Mobile, and Windows Phone 7), you'd probably have to rewrite significant portions of the app for each platform. Even if the UI worked exactly the same across all platforms, you'd end up rewriting the same presentation logic for each. MVP's advantage is that you can separate the high-level presentation logic from the view implementation - so you'd be able to reuse the Model and Presenter between all platforms, and only have to rewrite the View. Another major benefit of MVP is that it is also very easy to write tests against the Model and Presenters. The only manual testing that needs to be done is at the View (and even then there's some automated testing available, at least for WP7).
Click to expand...
Click to collapse
read the full article
UPDATE 5/2/11
I've released my PMVP framework as open source at codeplex.
http://pmvp.codeplex.com/
This includes the source code for the framework (abstract classes to extend), utilities, WP7 implementations, as well as a complete and functional example application which demonstrates the PMVP framework, data serialization, Location Services (GPS) usage, and Bing maps integration.
Hmm... Do you have any "high-level code architecture" really big and complex WP7 app to prove your concept? Your QuickText is too simple and small, can be written without that kind of theorization.
sensboston said:
Hmm... Do you have any "high-level code architecture" really big and complex WP7 app to prove your concept? Your QuickText is too simple and small, can be written without that kind of theorization.
Click to expand...
Click to collapse
You don't need a really big and complex app to benefit from using the MVP pattern. Even a simple app can benefit just because when you separate the concerns by MVP.
Sure, with MVP there are more classes, and all you're doing is really shifting the location of the complexity - BUT every class is way smaller and very tightly focused. My XAML code-behinds are dead small & simple. My ViewFacades are simple. My Presenters are simple.
Would you rather have a single large bloated code-behind, or several smaller code files each with a single purpose? Are you interested in porting your app to other platforms? Do you want automated code testing coverage? That is where the advantage of the MVP pattern comes in.
It requires a little more up-front planning, but you reap the benefits quickly after that.
I'm working on an (admittedly trivial) example app, but it should feature most of the common scenarios that developers will encounter. It'll be posted to codeplex and i'll link to it here when it's ready.
zaijian said:
You don't need a really big and complex app to benefit from using the MVP pattern.
Click to expand...
Click to collapse
You need it, not me. Telling very trivial and banal things on the developers forum without real experience in the large multiplatform projects it's just a blah-blah-blah. Sorry, but for me your article is not even worth the time spent on reading.
not that MVP isn't the samething but most people are more use to MVVM now over MVP in general terms since that has been the buzz abbreviation with WPF/Silverlight for a few years now.
I am looking forward to what you write up.
sensboston said:
You need it, not me. Telling very trivial and banal things on the developers forum without real experience in the large multiplatform projects it's just a blah-blah-blah. Sorry, but for me your article is not even worth the time spent on reading.
Click to expand...
Click to collapse
Well I'm sorry it doesn't provide any value for you. The MVP pattern has been very beneficial for me in my job as a software engineer developing enterprise systems. It has provided real value for me and my coworkers and we use it in shipped, in-production, systems.
Besides, the multi-platform aspect of MVP is trivial compared to the benefits of separation of concerns (single responsibility principle) and inherently automatable testability.
So if you're already a top-notch developer, well then I have nothing to offer you. For everybody else who's interested in making themselves better developers, then I remain available for discussion in the hopes of furthering the craft.
mvermef said:
not that MVP isn't the samething but most people are more use to MVVM now over MVP in general terms since that has been the buzz abbreviation with WPF/Silverlight for a few years now.
I am looking forward to what you write up.
Click to expand...
Click to collapse
The standard MVVM pattern has its benefits, mostly being that it's very quick to develop. However, it's not as consistent in its separation of concerns as MVP is, but if short-term speed is more important than purity of design, then stick with MVVM.
I'd like to point out that with my MVP implementation, i've managed to integrate MVVM into MVP, not replace it. The way I did that is to simply extract some of the MVVM functionality into the Presenter tier, while keeping the powerful databinding capabilities of MVVM in the View tier.
MVP pays major dividends when considering the long-term picture. Better separation means easier debugging, maintenance, testing, and extensibility.
zaijian said:
The standard MVVM pattern has its benefits, mostly being that it's very quick to develop. However, it's not as consistent in its separation of concerns as MVP is, but if short-term speed is more important than purity of design, then stick with MVVM.
Click to expand...
Click to collapse
It's quicker to develop because you end up writing less code.
Less code == less bugs == better code IMO
zaijian said:
MVP pays major dividends when considering the long-term picture. Better separation means easier debugging, maintenance, testing, and extensibility.
Click to expand...
Click to collapse
This is interesting. Could you please elaborate on how MVP is a more (or better) separated pattern than MVVM? I get the feeling you're doing MVVM wrong if this is your conclusion, but I could very well be wrong.
If you have actually read the white papers and examples from Martin Fowler then you would actually equate MVP and MVVM. They are nearly identical. Only some minor differences. As for separation of view from model you actually have less code with mvvm. As the other poster has indicted it might be that your implementation of MVVM might be slightly off.
emigrating said:
It's quicker to develop because you end up writing less code.
Less code == less bugs == better code IMO
Click to expand...
Click to collapse
That is undoubtedly true, to some extent. However, less code also implies less control and less flexibility. While it takes more code to implement MVP, a lot of it is simple plumbing code - code that could, perhaps, be eliminated in the future using something like PostSharp or T4 to generate code.
But this is where as a developer you have to weigh the costs and the benefits. The benefits are separation of concerns, portability, testability.
emigrating said:
This is interesting. Could you please elaborate on how MVP is a more (or better) separated pattern than MVVM? I get the feeling you're doing MVVM wrong if this is your conclusion, but I could very well be wrong.
Click to expand...
Click to collapse
The MVP pattern encourages a separation between presentation logic and UI-specific logic, whereas MVVM does not.
Presentation logic is behavioral - for instance, say you have a collection of products, which is displayed in a ListBox on a PhoneApplicationPage. There is a button on the page which will delete the selected product.
In MVVM, the ViewModel holds the view state (products collection), and also the ICommand which contain the logic to execute: 1) delete the selected product from the data layer, and 2) update the view state. The MVVM databinding handles the last step of synchronization between the products collection and the listbox on the page.
Now in there is some high-level logic which could be abstracted - the rule that when the delete event occurs, that the data layer is updated, and the view state data is updated. If you are interested in portability, then you'd want to be able to reuse this presentation logic in another app platform, even on a web page version, so that's why you'd extract that code and put it in a presenter. But you keep the view state in the ViewModel to take advantage of the databinding.
Another benefit of removing presentation logic from the ViewModel is that it no longer couples your View tier to your Model tier (other than using the data objects). Removing dependencies makes your View tier much simpler and cleaner.
mvermef said:
If you have actually read the white papers and examples from Martin Fowler then you would actually equate MVP and MVVM. They are nearly identical. Only some minor differences. As for separation of view from model you actually have less code with mvvm. As the other poster has indicted it might be that your implementation of MVVM might be slightly off.
Click to expand...
Click to collapse
Fowler's white papers make a distinction between MVP and MVVM - he actually calls it Presentation Model, and his definition predates MVVM.
So yes, i've read his white papers, i'm familiar with them, and I reference them extensively. There is a significant difference between Presentation Model and MVP, namely that the Presentation Model (aka MVVM) contains presentation behavior logic, whereas in MVP that logic is contained in the Presenter tier.
emigrating said:
It's quicker to develop because you end up writing less code.
Less code == less bugs == better code IMO
Click to expand...
Click to collapse
Martin Fowler himself says:
All three of these patterns (MVP/Supervising Controller, MVP/Passive View, Presentation Model), by introducing an extra class, produce a arguably more complex design - but splitting a class that does too much into separate classes to fulfill each responsibility is a good practice.
Click to expand...
Click to collapse
So MVVM is "less code" than MVP, but MVP is just MVVM taken one step further to separate different responsibilities into difference classes - which he notes, is a good practice.
I'm having a great time discussing this, but I think my intent in my posting of this is being misunderstood, [obama]so let me be clear[/obama]:
I'm not advocating the use of MVP+MVVM over standard MVVM - that is something for the developer to decide, after weighing the pros and cons of each against their requirements.
I'm merely offering a way to enhance MVVM applications to provide the additional benefits an MVP implementation provides.
I developed my implementation of MVP because I want to work with an Android developer and see the real, tangible benefits of the pattern for cross-platform application development. So in this, the argument of MVVM vs MVP is moot, as Android does not have the luxury of an MVVM framework! So besides the other benefits of MVP, if cross-platform development is of an interest to you, then MVP is almost a must.
I've released my PMVP framework as open source at codeplex.
http://pmvp.codeplex.com/
This includes the source code for the framework (abstract classes to extend), utilities, WP7 implementations, as well as a complete and functional example application which demonstrates the PMVP framework, data serialization, Location Services (GPS) usage, and Bing maps integration.
I was wondering what resources are out there to become a DEV and make my own ROM's. I have googled it a little bit and not really found anything worth while so far. I did run across a good book but its not going to be published for another 2 months. Does anyone know of any tutorials? Books? How to guides?
There are plenty of kitchens right here on xda. And remember when you search on google add on xda to whatever you are searching. I had a bunch of bookmarks for this but never backed them up. Sorry buddy.
Sent from my BAD A$$ EPIC TOUCH 4G
I didn't think about using the term kitchen when searching. I will give that a shot. Thanks.
Sent from my SPH-D710 using Tapatalk
bdpatch said:
I was wondering what resources are out there to become a DEV and make my own ROM's. I have googled it a little bit and not really found anything worth while so far. I did run across a good book but its not going to be published for another 2 months. Does anyone know of any tutorials? Books? How to guides?
Click to expand...
Click to collapse
I would read google's documentation on Android, learn Linux commands, and also learn java. This is a good start. If you use kitchens, you are not dev'ing. After this, reading the cyanogen wiki will teach you some of the less "official" deving methods.
Edit:
Links
First go here and learn all of it. (weeks of hard work and dedication)
http://linuxcommand.org/index.php
Then go to a local library and pick up a recently published book on java programming and learn it. (weeks of hard work)
Then learn all of this (more weeks of hard work)
http://developer.android.com/guide/index.html
Then learn all of this (more weeks of hardwork)
http://www.freeyourandroid.com/index.php
Finally get involved with Cyanogenmod
Becoming a dev is a huge commitment, it is why professional devs are paid so well.
Sent from my SPH-D710 using Tapatalk
kingsway8605 said:
I Becoming a dev is a huge commitment, it is why professional devs are paid so well.
Sent from my SPH-D710 using Tapatalk
Click to expand...
Click to collapse
I'm just curious, how much do devs usually make and who actually pays them?
Sent from my SPH-D710 using Tapatalk
Sn1per 117 said:
I'm just curious, how much do devs usually make and who actually pays them?
Sent from my SPH-D710 using Tapatalk
Click to expand...
Click to collapse
I said professional devs. The hobbyist devs on xda usually are doing it because they enjoy doing it and picked up the skills over the years. But it isn't any easier and that is why the behavior on the forums sometimes annoys them. As far as professional devs who work for a company like Samsung, they make between 50,000-100,000 a year.
As far as independent professional devs, Paper camera has sold over 1,000,000 apks at 2 dollars a pop, do the math.
Sent from my SPH-D710 using Tapatalk
I say make a couple of apps first and learn how Android works. Then move to themeing, then make kernels, then roms
Sent from my Epic™ 4G Touch
kingsway8605 said:
I said professional devs. The hobbyist devs on xda usually are doing it because they enjoy doing it and picked up the skills over the years. But it isn't any easier and that is why the behavior on the forums sometimes annoys them. As far as professional devs who work for a company like Samsung, they make between 50,000-100,000 a year.
As far as independent professional devs, Paper camera has sold over 1,000,000 apks at 2 dollars a pop, do the math.
Sent from my SPH-D710 using Tapatalk
Click to expand...
Click to collapse
Devs dont get 100% of their sales
Sent from my Epic™ 4G Touch
Here's where I learned my skills, this determines on what you're trying to develop. Most roms here, with the exception of source builds don't require much actual programming experience since they are closed source, primarily they have small tweaks like de-odexing, XML mods, etc.
Hands-On | 90% - 90% of my skills have come from hands-on experience and learning as I go along. For example, say I'm writing an app for the very first time. (Don't start with an app unless you've practiced with the platforms language first) As I went along on my first app I learned an incredible amount about android developing by reading developer.android.com and SO as I went along and came across something I wasn't sure about. Ditto kernel development, years ago I started with no knowledge and then bam I'm half a pro . Also, I knew nothing about ASP.NET/MVC/C# about a month ago, now it's kinda fluid to me.
School(Computer Science) | 8% - Yeah, it costs all that money for 8% or less. There can be some unique challenges in a software program, but almost everything you do in school is either so off the wall simple or has no real world application. I also think they spend too much time teaching by the syntax of the language, when really you should group them. Syntax is *mostly* trivial, it's better to group languages. Example: learn the Object Oriented Style (Perl, PHP, C++, Java, C#) which should give you a decent foundation in all those languages. Then learn the Procedure-Oriented Style (C, PHP, Perl, Cobol, BASIC (eww)). Then learn the rising in popularity Functional-Style (F#, SQL (mostly, this ones an exception since you often won't write many SQL scripts save for basic queries), Erlang, Haskell, C#). Notice: I listed some languages in multiple categories as the styles of programming are concepts, and some languages support either concept.
Co-Workers | 2% - Sometimes you learn by asking a co-worker or another developer. It's not how I did it primarily, but I know others who work best and learn best working in a team.
So to recap:
If you're trying to learn software programming itself I suggest starting with common scripting languages which are very easy to learn (HTML, XML (this will help for android), PHP). Then move on to some OOP languages (Java (Android*), C++). Then if you plan on working with kernel source code you should work with C and familiarize yourself with the procedure oriented style. Java is by far the easiest first language to learn OOP because of the wide-range of libraries available (to do almost anything) and the JRE, which gives you a nice garbage collector so you don't have the worry of memory management and freeing objects manually like you do in C/C++, etc.
If you just want to bake a rom with minor adjustments or a closed-source rom, I suggest you learn how to create themes first, as many of those concepts will help in baking a rom. And if you have questions this is a development community last time I checked, although sometimes it doesn't seem like it. So ask questions, ask other developers (I used to get questions ), ask on the board (if you follow the rules), ask on StackOverflow, etc. Don't be a nuisance to other developers though, I had someone that constantly blew up my phone nonstop all hours of the day asking silly silly questions. I don't mind questions, but i'd be spending time with my GF and my phone would explode with 20 questions that I already answered 20 times each
Also, some books may not be bad if you have absolutely no idea about programming whatsoever. But nowadays, you can find anything in a book on the internet. The benefits of a book is they include all their source code (usually) and can motivate you more since they give you projects and challenges. I had one Java book, and it drove me crazy because nothing was standardized. Instead of using the Java HashMap methods, or the Java Stack methods they wrote their own APIs for everything it drove me crazy :O
bbedward said:
Here's where I learned my skills, this determines on what you're trying to develop. Most roms here, with the exception of source builds don't require much actual programming experience since they are closed source, primarily they have small tweaks like de-odexing, XML mods, etc.
Hands-On | 90% - 90% of my skills have come from hands-on experience and learning as I go along. For example, say I'm writing an app for the very first time. (Don't start with an app unless you've practiced with the platforms language first) As I went along on my first app I learned an incredible amount about android developing by reading developer.android.com and SO as I went along and came across something I wasn't sure about. Ditto kernel development, years ago I started with no knowledge and then bam I'm half a pro . Also, I knew nothing about ASP.NET/MVC/C# about a month ago, now it's kinda fluid to me.
School(Computer Science) | 8% - Yeah, it costs all that money for 8% or less. There can be some unique challenges in a software program, but almost everything you do in school is either so off the wall simple or has no real world application. I also think they spend too much time teaching by the syntax of the language, when really you should group them. Syntax is *mostly* trivial, it's better to group languages. Example: learn the Object Oriented Style (Perl, PHP, C++, Java, C#) which should give you a decent foundation in all those languages. Then learn the Procedure-Oriented Style (C, PHP, Perl, Cobol, BASIC (eww)). Then learn the rising in popularity Functional-Style (F#, SQL (mostly, this ones an exception since you often won't write many SQL scripts save for basic queries), Erlang, Haskell, C#). Notice: I listed some languages in multiple categories as the styles of programming are concepts, and some languages support either concept.
Co-Workers | 2% - Sometimes you learn by asking a co-worker or another developer. It's not how I did it primarily, but I know others who work best and learn best working in a team.
So to recap:
If you're trying to learn software programming itself I suggest starting with common scripting languages which are very easy to learn (HTML, XML (this will help for android), PHP). Then move on to some OOP languages (Java (Android*), C++). Then if you plan on working with kernel source code you should work with C and familiarize yourself with the procedure oriented style. Java is by far the easiest first language to learn OOP because of the wide-range of libraries available (to do almost anything) and the JRE, which gives you a nice garbage collector so you don't have the worry of memory management and freeing objects manually like you do in C/C++, etc.
If you just want to bake a rom with minor adjustments or a closed-source rom, I suggest you learn how to create themes first, as many of those concepts will help in baking a rom. And if you have questions this is a development community last time I checked, although sometimes it doesn't seem like it. So ask questions, ask other developers (I used to get questions ), ask on the board (if you follow the rules), ask on StackOverflow, etc. Don't be a nuisance to other developers though, I had someone that constantly blew up my phone nonstop all hours of the day asking silly silly questions. I don't mind questions, but i'd be spending time with my GF and my phone would explode with 20 questions that I already answered 20 times each
Also, some books may not be bad if you have absolutely no idea about programming whatsoever. But nowadays, you can find anything in a book on the internet. The benefits of a book is they include all their source code (usually) and can motivate you more since they give you projects and challenges. I had one Java book, and it drove me crazy because nothing was standardized. Instead of using the Java HashMap methods, or the Java Stack methods they wrote their own APIs for everything it drove me crazy :O
Click to expand...
Click to collapse
My school must have a great compsci program because they focus on good programming styles and not syntax. There is no reason to start with any language besides java if your goal is to develop Android. And while I agree with your learning as you go approach, it would be a mistake and overwhelming imo for someone to jump into Android development without having a basic understanding of Linux, java, and Android. My advice given was for helping op to become an Android dev, not someone who posts roms. They are not always the same thing.
Sent from my SPH-D710 using Tapatalk
bbedward said:
Here's where I learned my skills, this determines on what you're trying to develop. Most roms here, with the exception of source builds don't require much actual programming experience since they are closed source, primarily they have small tweaks like de-odexing, XML mods, etc.
Hands-On | 90% - 90% of my skills have come from hands-on experience and learning as I go along. For example, say I'm writing an app for the very first time. (Don't start with an app unless you've practiced with the platforms language first) As I went along on my first app I learned an incredible amount about android developing by reading developer.android.com and SO as I went along and came across something I wasn't sure about. Ditto kernel development, years ago I started with no knowledge and then bam I'm half a pro . Also, I knew nothing about ASP.NET/MVC/C# about a month ago, now it's kinda fluid to me.
School(Computer Science) | 8% - Yeah, it costs all that money for 8% or less. There can be some unique challenges in a software program, but almost everything you do in school is either so off the wall simple or has no real world application. I also think they spend too much time teaching by the syntax of the language, when really you should group them. Syntax is *mostly* trivial, it's better to group languages. Example: learn the Object Oriented Style (Perl, PHP, C++, Java, C#) which should give you a decent foundation in all those languages. Then learn the Procedure-Oriented Style (C, PHP, Perl, Cobol, BASIC (eww)). Then learn the rising in popularity Functional-Style (F#, SQL (mostly, this ones an exception since you often won't write many SQL scripts save for basic queries), Erlang, Haskell, C#). Notice: I listed some languages in multiple categories as the styles of programming are concepts, and some languages support either concept.
Co-Workers | 2% - Sometimes you learn by asking a co-worker or another developer. It's not how I did it primarily, but I know others who work best and learn best working in a team.
So to recap:
If you're trying to learn software programming itself I suggest starting with common scripting languages which are very easy to learn (HTML, XML (this will help for android), PHP). Then move on to some OOP languages (Java (Android*), C++). Then if you plan on working with kernel source code you should work with C and familiarize yourself with the procedure oriented style. Java is by far the easiest first language to learn OOP because of the wide-range of libraries available (to do almost anything) and the JRE, which gives you a nice garbage collector so you don't have the worry of memory management and freeing objects manually like you do in C/C++, etc.
If you just want to bake a rom with minor adjustments or a closed-source rom, I suggest you learn how to create themes first, as many of those concepts will help in baking a rom. And if you have questions this is a development community last time I checked, although sometimes it doesn't seem like it. So ask questions, ask other developers (I used to get questions ), ask on the board (if you follow the rules), ask on StackOverflow, etc. Don't be a nuisance to other developers though, I had someone that constantly blew up my phone nonstop all hours of the day asking silly silly questions. I don't mind questions, but i'd be spending time with my GF and my phone would explode with 20 questions that I already answered 20 times each
Also, some books may not be bad if you have absolutely no idea about programming whatsoever. But nowadays, you can find anything in a book on the internet. The benefits of a book is they include all their source code (usually) and can motivate you more since they give you projects and challenges. I had one Java book, and it drove me crazy because nothing was standardized. Instead of using the Java HashMap methods, or the Java Stack methods they wrote their own APIs for everything it drove me crazy :O
Click to expand...
Click to collapse
This is great info thank you. I have linux experience and I have made a couple of hello world apps but nothing extensive. It looks like maybe i should start with a little theming or something like that. I know html and can read through XML. I think i will try and find some Java tutorials or books to get me going there. Know of any guides on theme's?
kingsway8605 said:
My school must have a great compsci program because they focus on good programming styles and not syntax. There is no reason to start with any language besides java if your goal is to develop Android. And while I agree with your learning as you go approach, it would be a mistake and overwhelming imo for someone to jump into Android development without having a basic understanding of Linux, java, and Android. My advice given was for helping op to become an Android dev, not someone who posts roms. They are not always the same thing.
Sent from my SPH-D710 using Tapatalk
Click to expand...
Click to collapse
My schools good too I'm not saying its all syntax but its a lot of write your code readable in the professors style. I've learned more in 3 months on the job than 2 years of school. I've learned way more from self teaching tho.
And for app development Linux knowledge isn't really necessary since you can develop android apps on any platform, but Java and OOP styles is necessary knowledge.
Sent from my SPH-D710 using Tapatalk
bdpatch said:
This is great info thank you. I have linux experience and I have made a couple of hello world apps but nothing extensive. It looks like maybe i should start with a little theming or something like that. I know html and can read through XML. I think i will try and find some Java tutorials or books to get me going there. Know of any guides on theme's?
Click to expand...
Click to collapse
What you should start with greatly varies with what you want to develop
Sent from my SPH-D710 using Tapatalk
bdpatch said:
I was wondering what resources are out there to become a DEV and make my own ROM's. I have googled it a little bit and not really found anything worth while so far. I did run across a good book but its not going to be published for another 2 months. Does anyone know of any tutorials? Books? How to guides?
Click to expand...
Click to collapse
Read and follow this information from Cyanogen himself
http://forum.xda-developers.com/showthread.php?t=667298
jerdog said:
Read and follow this information from Cyanogen himself
http://forum.xda-developers.com/showthread.php?t=667298
Click to expand...
Click to collapse
Thank you for the link. I got lots of stuff to start reading through now. I also found this book: http://www.barnesandnoble.com/w/xda...7854?ean=9781119951384&itm=1&usri=android+rom
It should hopefully be a good one as well when its published.
bdpatch said:
Thank you for the link. I got lots of stuff to start reading through now. I also found this book: http://www.barnesandnoble.com/w/xda...7854?ean=9781119951384&itm=1&usri=android+rom
It should hopefully be a good one as well when its published.
Click to expand...
Click to collapse
It will be - I know the guys who've put it together.