I don’t know enough about Microsoft’s advertising strategy to comment on it, because I very rarely give Microsoft any notice (Especially when you consider that I work in an office of Apple Lovers and spend my time following Open Source fanatics on Twitter), so whether these videos are just advertising techniques or not, I can’t say.

But the Courier is grabbing my attention, so rather than trawl the web everyday for new information I set off a Google Alerts Feed to let me know of new Courier info, and to be honest, I wish I hadn’t, because it get’s filled with utter rubbish.

So here’s my take on everything Microsoft is currently doing, what I think the Courier is (vaporware or not?) and finally what happens next.

Microsoft Where Have You Been All My Life

I make it no secret when writing these posts that I am not a fan of Microsoft, it’s nothing to do with their ethics or anything like that, I just found Mac OS X and Linux were more than suitable for ANY person to avoid Windows, so I just wrote off Microsoft in terms of OS, their old mobile devices are ugly and while Windows 7 Phones seem to have a lot nicer interface, I’ll have to see how I feel after I get time to use their SDK (Past experience has been pretty bad in terms of customisation and I am all about the aesthetics).

But recently I’m seeing them in a different light, sure I think Windows 7 is still pretty bad and I hate it when my other half calls me up with the first sentence “My laptops broken”, but Microsoft seems to fixed up their attitude to mobile devices “to a degree”.

They’ve looked at mobile devices (Phones, Tablets, Other), considered them and said to themselves, “They aren’t computers”.

One team point to Microsoft right there! Yeah I admit, they are computers, but the UI needs to be different, you can’t just put a UI that works on a nice big screen straight onto a tiny screen and hope it will work just as well. I am in 2 minds about this point though, Microsoft have sat back and come up with this great Courier concept and their Windows 7 Phones (even the new Kin phones), but what the hell happened with the HP Slate?

Apple iPhone – Came with a totally mew UI

Android – Came with a new UI (although alot of similarities to iPhone)

Apple iPad – New UI (Sure the front screen is the same, but the apps have opened up into that screen real estate nicely, and I bet Apple have given Developers a new set of UI elements to take advantage of)

Windows Phone (Old) – Well Windows have scroll bars on them, Start button goes in the corner, nice and small, and no one cares about silly UI pictures to make things pretty, grey is pretty right? *Designer cries in the corner

So this new UI spree that Microsoft have gone on is a great thing for their products, but how real is it? That’s gotta be your call, I think other parts of Microsoft have locked into this new approach, but the core of the company . . . maybe not so much.

When, Where, How Much . . . The Courier Ain’t Real Mate

Straight off . . . . Is the Courier Vaporware? . . . . answer: Maybe

Microsoft have old videos of “Courier like” devices, the UI looks old, ugly and not too user friendly, but it has some big similarities to the original videos.

You can see the UI is very different, but it’s a thought provoker, has this been in the works for a long time, or did they forget it a long time ago and now figured this is the time to launch it?

I get the impression they asked a design team to take the Codex concept (above) and turn it into a viable, attractive product. This team produced the first video below.

Then the Codex team who probably created the prototype, took a look at this and figured, we can do this this and this, but probably not this, but maybe like this. Right, said the design team, so now we need to show you what you want the new UI to be and how it’s going to work.

That’s is where we are up to, plenty of blog’s are going nuts over a video Microsoft released with regards to using a stylus and fingers on a touch screen the size of a table (Microsoft Surface). Yes the Courier uses a stylus, I see the link, but come on, a table size screen compared to a small notebook, they aren’t directly related, take away that they are trying to prove that stylus’s mustn’t be banned (Especially after the Steve Jobs quote declaring war on stylus), but they shouldn’t be doing this to prove the product is good, just let the product do that itself.

Vaporware – maybe, maybe not, anyone declaring yes or no, ssssshhhhh . . . . you don’t actually know anything [Neither do I].

What Happens Next – Well Microsoft Launch It Surely?

So this section is more of a discussion about whether or not it’s feasible to make a Courier type device, regardless of who does it.

The answer is obcourse, Computer’s are powerful beasts, so I reckon it’s easily done.

To start off with, the hand writing recognition (which I think is a pretty important part of the Courier), can be achieved with good accuracy. According to my colleague @mattcolliss (Twitter, look him up), who is doing a lot of work  in a different text recognition area, thinks text recognition of this kind is pretty much a done deal, it can be accomplished, and it can be accomplished well. I believe him, but if you dont, check out this video from Mark Cummins, a developer of Plink Art an image processing tool for photo’s, the accuracy is insane, so hand writing should be fine.

Then consider the actual task being performed. The first thing to notice is that each screen is essentially a different application. You could argue that it is a single application and by all means take that point of view, but I tell you what, designing a single app to perform the tasks that you want, would quickly become a pain to develop for. So from this point on, I’m going to take the view that they are 2 seperate applications running on each screen. Therefore the obvious is needed, multi-tasking. This never used to be a big deal, but iPhone changed it to be a big functionality. Phones don’t need it, the fact the iPad doesn’t have multitasking is a big shocker to me, but this thing will have to have it, but there are processors that can do this, and 2 screens means the casing could have the space to store and power the screens and applications that will run on them. (As a proof, think of your netbook running Linux, can run a browser and open office right?)

So multi-tasking is easy to get on this device (We assume). The applications can be developed independently (great for devs, or is it?) We see that nice flicking from one screen to another, this is great, but how does this actually happen from a developers point of view? The best way I can think of achieving this, is the platform (or OS to some), which the applications run on top of, must be able to take some image from the first application and performing the animation from one screen to the next (Taking the problem away from the individual application) then dropping the image back down in the second application.

This is something that you may take for granted, but I know that Ubuntu has a similar funcionality when you drag and drop an image from a folder into Gimp and it’s essentially the same functionality without the animation

But Courier is rumoured to run on Windows 7 Phone or Zune platform, now if they have added in a similar concept to Android’s Intents, then this should be perfect for the platform being able to take out and put in different images and data with a nice animation. If they don’t do this, then 3rd party applications don’t stand a chance because I can’t see how my application would take in an event of this image being dropped in my application and I REALLY wouldn’t want to have to handle the animation of the image once it reaches my animation.

In my mind, Android seems like a good idea to use for Courier, but I’m an Android Dev and I’m giving Microsoft the better of my doubt and assuming their new Windows 7 Phone is a vast improvement over the old Windows Phone SDK and therefore has a well thought out way to achieve this goal.

We’ve discussed Windows 7 Phone, Zune, Android, what about the rest?

iPad - I’m sorry but no go, the main reason for this, is that it would be a big pain to implement an app that would have a browser on one half of the screen and a note taking app on the other. Plus I don’t want to have to type in lots of text, I want to write it out, much more natural on this kind of scenario. Sure I could use a stylus, but will the screen work with it and be precise enough? Then let’s say you want to use the iPad Browser and a notepad application, you could probably do this, but the beauty of the Courier is taking the information (images, text, videos, pdf’s) from one source and editing, putting into new notes etc, so going in and out of different app’s will just become impractible. Don’t get me wrong, some awesome developer or team will produce some great utility apps and take ideas from the Courier but I doubt it would be as fluid.

Chrome OS - I’m not sure if this is a.) a good idea b.) possible

The reason I’m thinking it might be a bad idea is that you instantly limit it to HTML5, which is coming on leaps and bounds, but is it up to the stage of a customised UI with bells and whistles? It may well be and can do all this, but I just can’t help but think Android would be a better choice to better manage applications. I’m not convinced myself so this is a totally open ended question because I’m not convinced by Chrome OS but I’m also quite optimistic judging by some of the talks I’ve attended on the topic.

The reason I’m not sure it’s possible is after asking a Google engineer about whether HTML5 supported multi touch gestures, which is pretty essential to any touch screen device, the response was that it isn’t standard in HTML5, but it is in implemented in Webkit, so it’s probably working it’s way into Chrome and Chrome OS, but a long way to go to be as smooth as Android’s widget elements? (I have no idea just for the record)

(Finally) Linux - Yes, No, Maybe, Linux can do some great things, but it needs a lot of work to make applications that look as nice, work as nice and be as well integrated to work like the video’s shown above. Linux does support tablets, and we have seen other versions of the desktop Linux to work better on net-books, but will Linux be able to specialise a separate version of itself for these devices?

An open source project like this would need the money, time and attention, which is sometimes Linux’s biggest fall, and anyone considering doing this task has to ask whether a Linux implementation is better than opting for a custom Android version (I know Android is a platform on top of Linux, but in my mind they are targetted at different types of devices).

Conclusion

I’ve discussed a number of different topics, and mainly the result is, I still want one, I think Microsoft is toying with the idea of releasing and might be seriously considering it, I am certain it can be done, and if Microsoft don’t do it, someone else will.

Things that will kill this for me though, requiring me to use Microsoft Stuff, I don’t use hotmail, sorry no export to thunderbird and poor spam filtering killed that for me, I don’t want to use an online Microsoft Office, Google Docs is what I use at work, it’s what I use for personal projects and everything else has latex of Open Office and if Microsoft has an online sharing tool, that doesn’t fit into any of these categories, just please let it work, and please let it work well.

Microsoft phones and this have made me consider there is hope for Microsoft yet in terms of products, but it’s still way up in the air.

Take Away Point – Tablet, Netbook, Desktop Computer, Laptop, Courier are ALL different types of devices with different use cases. Treat them that way where-ever appropriate.

http://community.research.microsoft.com/blogs/alpineinker/archive/2008/10/01/microsoft-research-codex.aspx

This phased me for quite sometime, how do I use a widget like a button, or a view as an overlay?

I asked on the Google Groups, got a great reply from Mark Murphy, and he explained about how you can use relative layouts to add views on top of the map view and set x, y coordinates and use z ordering.

I started to think about what the MapView was, then a quick look at the API and bam, addView.

So here’s how you add a view to a MapView and I’m sure you can add buttons to it etc

The LayoutParams is the key to defining where the marker view is kept in relation to the map.

Enjoy!

I’ve spent probably the best part of a couple of days attempting to get Proximity Alerts working in Android. There are a large number of Google Group threads, all leading to essentially someone saying it still doesn’t work. There was one or two that had working solutions, or some who managed to get it working to a certain degree.

I managed to get a working implementation of proximity alerts by combining the information from a number of threads, this is just a guide through what I learnt and should get you up and running with your own alerts.

Just for the record, I need to learn a great deal more about Intent Filters and Broadcast Receivers as my knowledge of them is fairly vague, so please correct me in the comments if I say anything wrong, or should be implementing things differently.

Lets start of with the basic intent action:

This is simply a string I have made up to signify the action of an intent is a proximity alert action (i.e. this identifies the intent as a proximity alert). The usual convention for naming the intent actions is com.some.package.action.ACTION_NAME, notice the action before the action name. This then needs to be applied to an intent filter, which is then applied to a broadcast receiver. This can be done in the AndroidManifest.xml file, however I didn’t have any luck this way (told you I needed to do some learning) so for now I can only show how to achieve this in code.

In your Activity:

ProximityAlert Class:

It’s worth noting at this point that you only need one broadcast receiver for a number of proximity alerts, since one receiver will pick up any number of intents fired off defined by the intent filter. We’ve got the broadcast receiver with the intent filter applied to it. Now lets define out proximity alerts function.

Breaking this down, you obviously set radius and expiration to anything you wish, expiration = -1 will mean the alert will never expire. We get the LocationManager as shown, the intent is just an intent using your action string I previously mentioned. In my example I added an extra item, now in the Google Groups there was quite a bit of talk about needing to apply a different extra to the intent, this isn’t necessary, however the VERY IMPORTANT THING to getting multiple proximity alerts working is in the request code in the PendingIntent.getBroadcast. In the java docs for pending intents, it says “Private request code for the sender (currently not used).” So one would assume it doesn’t matter, Wrong, it’s extremely important. You need to send a different request code for each proximity alert so the location manager can identify each of the pending intents as different, otherwise (I think) it caches them and treats them the same, meaning all proximity alerts will have the same extra data (or at least that’s what I found).

So I hope that helps someone, if I have missed anything out let me know, I know my naming for these classes is poor, but I was working on some coursework and refactoring was bottom of the list I’m afraid.

Do as I say, not as I do.

UPDATE: I’ve just made a full example of the code here – http://www.gauntface.co.uk/blog/2010/12/28/how-to-multiple-proximity-alerts-in-android/

Word of warning before we start, this goes against the standard Android UI and should only be done when you are certain it is appropriate.

To handle screen rotations such that your activity doesn’t get destroyed and restarted do the following:

1. In your AndroidManifest.xml file go to the Application tab (Assuming you are in Eclipse, Manual programmers can still use this info, but will need to work out the corresponding xml), then under Application Nodes select the Activity you want to stop the rotations on, on the right hand side under “Attributes for <Class Name> (Activity)” scroll down to  Screen Orientation and set it to portrait or landscape.
2. Then in the same section set config changes to “orientation|keyboardHidden”
3. Now back in your class add the following function and you’ll be good to go.

// Handle Screen Orientation (Stop Activity being killed and re-started)

@Override

public void onConfigurationChanged(Configuration newConfig)

{

super.onConfigurationChanged(newConfig);

}

Your making an Android app and want to take advantage of Google Maps, you’ve been bashing your head against the desk attempting to work out what is going wrong.

So welcome to this little checklist of things I’ve forgotten to do when doing this task:

1. Right click on your project in eclipse and go to properties. Then under Android select Google APIs with the correct platform number, click apply then Ok
2. Open up your Android Manifest file, under the application tab scroll to the bottom and under the Application Nodes click on Add, then double click on “Uses Library”, now for the name enter the following – com.google.android.maps
3. Your activity which is going to be displaying the map isn’t a plain old activity, it needs to be a MapActivity like: public class MapExample extends MapActivity
4. Finally your map view in the layout file should look something like <com.google.android.maps.MapView android:id=”@+id/mapview” android:layout_width=”fill_parent” android:layout_height=”fill_parent” android:clickable=”true” android:apiKey=”YOUR_API_KEY”/> where your api key can be found by doing as mentioned on the Google Maps site -http://code.google.com/android/maps-api-signup.html

This should get you up and running with lots of mappy goodness.

After attending one of the Google Developer days in London, I couldn’t help but think that Android is going down a bit of a bad road, taking away some of the simplistic things that made it such a perfect platform to develop for.

So in the good old days of Android, you downloaded a single SDK and everything worked, now you have to download it, and then run an application inside the SDK to install the required versions of code (which actually wasn’t all that obvious to me when I needed a new install of the SDK). But as the day progressed and I got to play with a load of devices, you quickly see that the OS versions on each device become a big problem.

Say I want to develop for HTC Tattoo, but want to also let my app run on the HTC Hero (HTC Hero = V 1.5, Tattoo = V 1.6), I should develop for 1.5 right? Wrong. I should be developing for 1.6, then setting the min SDK to 3 (or the equivalent for 1.5) and then set target version as 4 (or the equivalent for 1.6). Is it just me or is that a little bit backward? I appreciate the reasoning behind this, since you need to let the phone know the app is suitable for devices with smaller screens. But is this really the best way to do it? Maybe it’s just me but this seems really counter intuitive.

But this problem falls back to the simple fact that we are developing for a number of different Android versions and hence a number of different SDK’s. So my one app should be developed in 1.6, but set the min sdk version to 1.5 to cover those devices, then test on 3 different screens, then test on 2.0 and perhaps 2.0 with 3 different screens? 2.1 anyone? get the picture?

And no you can’t assume backwards compatability, so for example, I developed Wheres My Contacts for Android 1.6 however in 2.0 the contacts API changed, hence my app doesn’t work on the 2.0 platform. Now I would of really liked it if the market saw I require access to read and write to the phonebook contacts, but also saw that it was built for 1.6 and not targeted at 2.0, as what happens with the HTC Tattoo because of the screen size.

So we’re already seeing a little bit of inconsistency within Android, let alone the differences in the mobile devices themselves etc.

I think the whole point of this ramble is that the beauty of Android was it’s an open platform that phones would use and we could just develop for the latest version end of, Google release new update, we get said update, good times. By handset manufacturers chopping and changing the versions between handsets, an app developer has his work tripled (atm for 1.5, 1.6, 2.0) but this is only going to get worse as time goes by. The Android engineers acknowledged there is some fragmentation and are trying to prevent but there is alot to overcome.

Phone manufacturers are introducing custom interfaces, but aren’t keeping them up to date with Android, then even if they could, we have to rely on the network providers to release the update after they’ve customised it and started an over the air update (OTA). OTA is probably the most irritating thing ever, people in the same part of the country getting updates weeks before others, but even then, it takes months to release an update, from the manufacturer to the end user.

The network operators need to stay out of this and stick with what they’re good at, leave the handset manufacturers to release the updates, and then the handset manufacturers need a more efficient way to update. This way a developer can stick to one version, and hence one SDK. End users can then stick with one market to get all the apps all the time, if your on 1.5 I’m sure there are apps you are missing that people on 1.6 are enjoying.

Finally just to throw an extra little comment, I reckon the people over at HTC throw some bits of Android 1.6 into their HTC Sense version 1.5 of Android – I’ve only noticed it with wireless support, but could it lead to problem later on down the line?

I’m sure a big company like Google can discuss this and fix some of these issues, at the end of the day, all manufacturers want their handsets to be successful, by releasing an Android device I’m guessing they are edging their bets on Android making them successful, but Android’s success relies as much on developers as it does the platform itself (or does it, is Google apps enough?). If it is developers then manufacturers need to make their life’s easier by staying up to date.

End.

Then just visit the appropriate link in the Android Browser

Yep, that’s right, I said it, iPhone is going to be a winner.

But this isn’t over any hate towards Google, Android, Cyanogen etc. but just more of the fact that Android is at a point of heading down 2 roads. One road leads to Android being fragmented, losing alot of it’s value. The other road leading to a community using different devices but one central set of app’s. I better explain what I’m talking about.

Motorola recently announced that they are using a “derivative” of Android. I knew they weren’t using the stock Android as they have added their own user interface “Blur”. But they’ve recently explained the reason for them having their own SDK (Software development kit). It’s so they can offer developers custom API’s. Now for the non-technical folk, this basically mean I can write an app that will do cool things on Motorola phones, but that same app won’t do the same cool stuff on other non-Motorola phones. In one sense this is great, if Motorola open up the Blur API it might make adding social networking into an app more exciting and easier, BUT on the other hand this means if I’m writing an app and decide to use Motorola API’s, then I won’t get the same experience on another Android device. So I now have to test it on 2 devices to check for differences (Possibly developing for 2 different API’s) or I only develop for Motorola. (Just for the record, I do understand that at the moment all Motorola API’s are fully compatible with the standard Android API, but how long is that going to last?).

Now lets take this whole Cyanogen fiasco, firstly I have to side with Google on them telling Cyanogen to stop distributing the Google apps i.e. Mail, Calendar etc. BUT where I side with Cyanogen is there seem’s to be no reason why he can’t include Android Market. Think about it, if you had an Android device without the Android Market, it loses alot of functionality. Put it on the device you’ll probably download Google Mail and Calendar straight away. So why don’t Google just allow everyone to use Android Market in their ROM’s, then Google can monitor who is downloading their apps.

So the alternative, someone creates their own market place app, which ROM makers like Cyanogen can include with their ROM’s, but even if everyone uploaded their app’s onto this new market, how good would the market itself be? How good would the app’s be, especially the app’s to replace Google Mail etc? And surely this just fragments everything, the Android version running on the phone, the markets on the phone, then chuck in a bit of API differences along with the ROM differences and you run into a bit of a mess.

The beauty of the iPhone is that everything is in one place, sure you have jail broken devices and a market place for those devices but overall people are happy with the stock iPhone OS, using the one app store. Android is heading down the road of multiple app stores, multiple API’s, multiple ROM’s and it’s just going to mess everything up for the common user.

I just hope Google finds a way that they are happy for ROM makers like Cyanogen to continue their work, with out making the ROM’s virtually useless, by simply giving the ROM makers the Android market to include so users can download the Google Apps that, for now, are the best options available.

I played around with a number different layouts trying to work out how I was going to fix this, assuming it was just my layout that was playing up. But I got curious and played around with the 9 patch png used for the new button and quickly realised that was the problem. To explain further, compare the left image (The layout shown in eclipse, how I wanted the view to display, and the right is how it displayed when inflated in the emulator [Ignore the button widths - that was me playing around, just look at the heights]).

The problem was that android was putting in an empty space for text but didn’t know where to put it as I hadn’t set an area the bin image could put content, as soon as I put this in the 9 patch png, I got the required result.

So if anyone gets weird results, play with the 9.png if you are using them as this can mess up your layout as well as the actual layout itself.

I just read an article over at linux insider http://www.linuxinsider.com/rsstory/67982.html and it basically discusses the confusion between Android and Chrome OS – My question is, what confusion?

Android in my eyes has always been a mobile phone platform, now I admit mobile phones, smart phones, smart books, its all blurring a bit so my definition of what should be running Android – a device which can be used for making phone calls, has a small screen making full aplications difficult to use, processor and memory (RAM) is limited. Something along those sort of lines. Or devices that don’t necessarily require a full blown OS, something more light weight.

I don’t think this is suitable for a netbook, so for instance, swiping my screen to view different parts of my desktop. On a phone with touch screen = intuitive, on a netbook, little backward no?

My main gripe about the whole (Android on a netbook) is, what software am I expected to run on an Android netbook? ooo of course the apps I download from the Android Market that are designed on a mobile skinned emulator, tested on a mobile device and also developed to perform tasks you might want a mobile phone to do.

The article also goes on to say how there are companies releasing Android Netbooks, one company I have never heard of, i.e. a small company who thought they could get in there before some of the big players, and also an expected Acer netbook. Now I admit I find it a little odd Acer are still going ahead with this after announcement of Chrome OS. Still I was never surprised by the lack of netbooks running Android,  it wasn’t designed for that. What makes more sense is that Chrome OS is going to be opened to developers in a year or so, which to me says Google have been working on this for quite some time and informed vendors of this.

With Android, Google worked with HTC to release the first phone, and Google again have said they have been working with vendors to create this OS for netbooks and desktops. So if the vendors were asked to participate with a Chrome OS, why bother looking at Android when you can probably guess Chrome OS is going to provide a far superior user experience on a netbook?

These are 2 OS’s for 2 totally different environments of usage. I.e. when I’m out about and only have my mobile phone on me for quick internet, quick tasks on apps etc, and a netbook or laptop would be used for more intensive work, so document editing, photo manipulation, multitasking in essence.

Final remark to finally put the nail in the coffin for Android on a netbook, Android memory management is quite cut throat, if it needs resources, it will kill any background Activities. Now if I was on a netbook and wanted a couple of apps running and the last app restarted, that would kinda get annoying. I mean will like a little music while we work.

Will Google Chrome OS knock Microsoft’s hold on the OS market? Not really, it will shift some linux and tech enthusiasts over to try it out, whether they’ll stick to it I’m not sure. But it will be cool to see if Google makes use of tools already used by other Linux distributions and whether development for programs will be aimed more towards the Chrome OS. Anyway that’s a whole different conversation I’m not getting into for the moment.