There's a discussion that's been swirling since the dawn of computers about the importance of cross-platform compatibility. Before I spout my extreme one-sided rant, I'd like to hear what you have to say about the importance of cross-platform compatibility when it comes to your UAV system purchase.

I'm talking configuration/telemetry software here more so than microcontroller selection or firmware.So you're looking to buy a UAV, how important is it to you that the software run natively on your OS?

To capture the question more specifically. *Most* GCS programs are written in a Windows only developement package like .NET which would require an emulation of Windows to operate on non-Windows operating systems. There most likely will be quirks or things that don't operate correctly, I'm assuming, as Mac and Linux are not the native platform.

So the question would be: Would you rather have the program developed in Java or QT or some other language to allow the program to be used natively on Windows, Mac and Linux even if it means giving up some features or execution speed?

Hi Happy, I'll toss in a couple quick thoughts.

For games that have a 6 month shelf life, you develop for what ever technology is popular today and it doesn't matter what's coming around the corner next year. For applications that plan to stick around a bit longer on the scene, I do think it is wise to consider a cross platform development strategy. You never know when the computer world will shift out from under you. 2 years from now, maybe we'll all be running android based tablets/laptops (even desktops?) all our applications will be on the cloud -- or maybe we'll all be running windows 2015 or pear-pads or maybe we'll all be running something no one has heard of today? Who knows, maybe we'll have an official federal government OS and the choice will be made for us. :-)

So putting all your eggs in the .net basket isn't necessarily a bad thing -- if you can't deliver a good windows solution you'll cut out a lot of your potential customer base. But it does limit you to the windows world and the windows way of thinking and if something new comes along (cough ipad/android) you could be left with all the other dinosaurs fighting for a shrinking water hole.

The challenge is can you do something that looks and runs nice on all your target platforms (and doesn't look like an eye sore or completely out of place compared to all the other apps on your device?)

Another thing to consider: if you don't build in a cross platform strategy from the start, it can be really difficult to reverse engineer this in later. From the start you need to strategically pick portable supporting libraries -- how you handle networking, serial communication, menus, windows, dialog boxes, save and load files (even how do you deal with file names), the list goes on and on.

But at the end of the day, if your users follow the normal trend -- probably close to 80-90% of them will be running windows with the rest divided between mac and linux.

Speaking very generically -- there will be pluses and minuses to any choice you make. The challenge is always to navigate through the various options and pick a set of paths that have a lot of good strengths. But there is always a downside to any choice so you also have to figure out a way to work around that and minimize the amount of negative while maximizing the positives that your choice will bring.

Another thing that could affect your decision is how you plan to license the result. If this a commercial product developed in-house, then whatever you do, it probably should run well on windows and you may not care about anything else (at the end of the day, the cost of a laptop is noise in the total price of an advanced UAV "system" so just include the laptop with the required OS. If it's going to be open-source, then I think that adds a lot of motivation to keep everything cross platform. You'll miss out on a lot of potential talent to help if you lock into a specific OS.

+1 on what clolson said.....

Having developed software for 19 years you have to look at it from a business standpoint and where your biggest return will come from. Not saying you can't look ahead to the other little hand held devices with those little app OS's, but like he said 80-85% of your near future users will be on a windows based OS. So my vote is "Not critical - I run Winblows"

Gene

I agree with whats been said, but I also have a prediction. I truly believe Windows will fall, in the next 6-8 years. Android and Google will take down the giant. Everybody has heard of the Google laptop, complete with its own OS (cloud computing/storage). If I were developing software today, I would take a hard look at Google and thier OS.

You are probably correct there philthyy..... but I could still make a good chunk of change off windows based software in 6-8 years!

Yup! You could, but never stop looking ahead. I think, once a business forgets to look ahead, they are already behind! Just my .2 cents.

Google's "vision" is to push all the apps into the browser with all the data storage in the cloud. (Or for mobile platforms have a dedicated app that parallels the web app and shares data storage in the cloud.) Apple seems to be signing up for a similar approach (at least as far as data storage in the cloud goes.) If google's vision takes over in the long term, then the actual hardware and OS you are running is pretty irrelevant. Probably still a lot of MS Windows out there, but who cares. Google's vision really started to take shape once they jumped into the mobile market -- interesting times but google wields tremendous power ... if the king passes the throne down to an evil son, then look out!

I would agree that Android and to a lesser extent iPad is an upcoming force to be reckoned with. I don't see Windows going away, but clearly their market share is at stake. I don't see Linux going anywhere....ever.

Never said "don't look ahead", just being completely pragmatic from a business standpoint. Windows is not going away for a while and Happy has a firm foundation with his GCS in that OS. Having been an entrepreneur for the past 20 years I learned you don't walk away from ANY revenue stream just to be "the first" with an Android GCS. There is already a shell of an app out there called ZPDashboard that provides a heads up display, wouldn't take much to add controls in on it.

Gene

For what it's worth, Android is built on top of Linux -- so I do see linux going lots of places in the future. Linux is also used heavily on the backend of things -- servers, internet infrastructure, cloud infrastructure. I also see a strong future for Linux "under the hood" in lots of embedded devices as low end uP's become more and more capable. (Just the way it looks from my vantage point -- I'm not claiming any special knowledge or even that I'm right, well ok, even though my opinions may change frequently, the fact that I am right does not.) :-)

Here's what it boils down to for me:

If you are making a pure commercial product, then make it for windows, and *maybe* out of the goodness of your heart, you could bless the mac world with a lesser mac version, and if you are especially nice, bless the linux world with an even lesser linux version.

If you are making an open-source product, then definitely go as cross platform as possible.

Either way, definitely keep your eye on android and be worried about whatever google is up to -- they are building momentum and could definitely put a big dent in microsoft and apple.

Dude.... I think you summed it up perfectly!

Yup! I think Closon summed it up perfectly, also!

On a side note, Centexflyer said "Never said don't look ahead", but that he is being completely pragmatic from a business standpoint. I find this an interesting example of the conflict (too strong of a word, but I don't know what else to use) between innovators/creators and business people. There is a fascinating book called Crossing the Chasm which outlines this exactly.

I bring this up because I found it to be an incredibly informative and useful book for understanding business in the dynamic tech world.

dichotomy?

Last night, I tried running the Android SDK. There were a bunch of updates available, so I went ahead and tried to download. It kept coming up with an error message that the main SDK app was in use and couldn't be overwritten.....well duh! How do you think I'm downloading these updates? So I found a work-around and had to copy the folder elsewhere and run a .bat file. It downloaded about 2.5GB of stuff. In the morning, it didn't download to the right place so I had to copy the temp folder back over top of the original location.

Then I tried to launch Eclipse. It kept complaining I was on version 12 and I needed 14....well that's what I was doing with the SDK update I thought. Turns out, nope, I have to go to Help, Update Software and enter a cryptic URL for it to download. Ok, finally updated.

Now, when I launch the SDK a DOS command window launches and stays open while the SDK program is running. Turns out that program is a .jar file and the .bat file launches the Java app...but doesn't seem to close gracefully.

It just seems to me that Mac/Linux/Java developers seem to think differently than Windows users and developers. The update process, the unprofessional launch of the SDK app, the clunky update process of Eclipse. It's all bizarre. You MUST RTFM or you simply cannot do it. It's not like trying to learn the nuance of the programming language. When I made the switch from VB6 to .NET I was Googling things constantly looking for how to do the things I had been doing one way for years and years....and now the syntax was slightly different. C# vs VB are just syntax differences for the most part. Plenty of Googling to switch between those two. But the step from VB.NET to try and write a Java app that runs on Android in Eclipse has been a nightmare so far. It suspect it will take me a year or so and several hundred thousand Google searches to become comfortable with the backwards nature of programming in Eclipse. VB6 was FAR more advanced than Eclispe so far.

Everything about running Java on Windows is just plain clunky. I hope that's not the case for running it on Mac or Linux or the guy/group who made Java should be ashamed of themselves.

GCS?

What type of application are you thinking of?