Discussion forum - Post and discuss topics here.

Mobile needs young developers, how to lower the burden to develop

Picture of Ron Verweij
Mobile needs young developers, how to lower the burden to develop
by Ron Verweij - Tuesday, 11 December 2007, 4:11 PM

Just a thought.

The Internet really was taken to the next level late 90-s when young kids with no concept or constriant started to do things they wanted to do and changed the Internet. To me many developments around mobile still are translations of things that worked on the Internet or evolutions of even older means of communication. The real innovation comes from regions in Asia that embraced Mobile 5-8 years ago and are a lightyear ahead.

How can the mobile world lower the burdens to develop on mobile phones and create mobile phone magic? Your thoughts please.

Cheers, Ron

Picture of Mike Rowehl
Re: Mobile needs young developers, how to lower the burden to develop
by Mike Rowehl - Tuesday, 11 December 2007, 4:58 PM
I completely agree. There are two things that made the early web really take off. The first was that anyone who wanted to learn about "this web thing" could take a look at how the websites they enjoyed were constructed. The "show source" option is always there, making it easy for potential authors to learn from existing content. The second is that constructing new pages can be done with very simple and minimal tools, usually using just the computer you use for browsing without having to install much more.

Both of those things are when you get when you provide applications in scripting languages like Python for example:


users can cross over and become developers pretty easily by just hacking the features and functions they need into the applications without having to constantly reinvent the wheel. Unfortunately the security models used by mobile operating systems are forcing stuff like this out of the market. Using Python even on Nokia devices is becoming harder and harder.

Perhaps once some of the other handsets make it out to market we'll see this turn back in the right direction. Linux based systems either like Android or OpenMoko should be very easy to provide scripting environment that hook into any level of functionality on the phone. Of course the whole open source argument is implicit in this, it's necessary but not sufficient. The applications need to be open source, and they need to much easier to rework than a J2ME application for example.
Picture of Alex Kerr
Re: Mobile needs young developers, how to lower the burden to develop
by Alex Kerr - Tuesday, 11 December 2007, 5:47 PM
Good call on Python. I think the other big thing that will achieve this is widgets, and cross-platform compatibility for widgets (achievable?). Something like Apple's Dashcode on OS X, where widgets can be made by *anyone* is what's needed in the mobile space. Then we'll see an explosion of creativity.

I tell you what won't see an explosion of creativity from the masses - requiring people to write software in C++ or .NET. That's for sure.

This reminds me of the home computer days 20 years ago, when BASIC ruled the world. One of the main reasons you've got people in the industry with passion who founded lots of great software companies big and small is not due to production-line IT courses which we have nowadays, but due to the power and flexibility of home computers like the Amiga, the Atari ST, the C64, the Spectrum and so on, and most of all the ease of access to these systems, via BASIC. IMHO the rise of the beige box PC and Windows killed a lot of this creativity and potential (and no, Visual BASIC does not cut it). It's great we have the diversity back in mobile, now we just need the ease of creativity. "User Generated Content" does not have to be limited to stupid videos, crap pictures, and annoying sounds to share with your mates ;-). It could also include cool software, if we put the right tools in place.
Picture of Peter Cranstone
Re: Mobile needs young developers, how to lower the burden to develop
by Peter Cranstone - Tuesday, 11 December 2007, 8:56 PM
The key is the "tools".

If Google wants to really succeed in mobile then it has to give the programmers more options than the competitors.

So far the winner is Microsoft.

Right now we're programming for both Windows Mobile devices and Blackberry/Symbian in Java, and I will tell you it's simply no contest.


Picture of Alex Kerr
Re: Mobile needs young developers, how to lower the burden to develop
by Alex Kerr - Thursday, 13 December 2007, 12:12 AM
Ah, but Peter, you're an example of a highly experienced/knowledgeable/talented developer, with some money. I'm not just saying that for flattery or to make a point, you/the people in your company, really are.

And therein lies the rub, as they say. The subject of this post is the opposite of that - new, inexperienced, not much money etc. Microsoft are simply not the new developer's friend, even if they do sometimes give tools away for free. Windows Mobile is, like Windows, not a very nice platform to develop for, IMHO. It's badly designed and engineered. That's why it's a dog of a phone operating system. (And why Windows has always been a dog of a PC operating system). I don't understand why any sane young inexperienced developer would go through the hassle of development on Windows, esp. when it's such a minority platform on a "big picture" level, to be blunt, compared to hitting the big time potentially with a mobile app that more easily targets SMS, mobile web, or even Symbian. J2ME's a bit of a big "Hmmmm" because of the fragmentation issues, but it does in fact attract the smaller developers even so.
Something like Android is coasting on mostly hype and promises right now, so IMHO is equally unappealing a platform for the context we're talking about, as WM is. But it does have promise and vendor support, so we'll see.

You have a reason to target WM, because of the nature of your products - WM is an obvious and good target platform, given your potential customers. I'm sure some young devs do target WM (esp in the States). But most young techy types, I think, will either own Symbian smartphones (specifically Symbian) and thus something like Python for tinkering will appeal, or if they prefer targetting non-smartphones for whatever reason, you're looking at SMS, mobile web, or the slightly more daunting prospect of J2ME.
Either way, development will need to be low cost or free and easy just to do something non-serious. These are not people with a mission, a business plan and a wad of cash (and if they are, I don't think they fall into the context of this discussion!)

Picture of Ron Verweij
Re: Mobile needs young developers, how to lower the burden to develop
by Ron Verweij - Thursday, 13 December 2007, 11:03 AM

Hi Alex,

Exectly my point. I want teens that just begin doing something cool they would like to do with a phone and create magic. Yes this will be shocking for the established developers, this will be shocking for large companies who trow away big budgets on services nobody use but good for the developments of mobile technology and services because it taps into the real mobile user mindset and create a new reality.

For this I see we need: good tools, free availability of tools, standardised phones, open standards, etc. To me its irrelevant what will be the best tools, OS, platform, etc. The glass ceiling is there now and needs to break.

Thanks for all the response.

Regards, Ron

Picture of Peter Cranstone
Re: Mobile needs young developers, how to lower the burden to develop
by Peter Cranstone - Thursday, 13 December 2007, 3:49 PM

Thanks for the compliment. I'm lucky to have an incredible team around me. But it's more than that - it's in part how we approach the task at hand.

In 17 years I've worked with 3 brilliant engineers. The first one was my uncle, David James, who was incredibly gifted mathematician. I spent 4 years with him, until he died in 1996. It was then I met Kevin who I've worked with ever since. Along the way I worked with one other person - Dr. Bill Worley, the former Chief Scientist of HP.

David and I worked on very complex data compression algorithms - Kevin and I worked on HTTP communications and Bill and I worked on a secure OS. Of the three I've worked with Kevin the longest. Of the three both Bill and David James where scientifically brilliant. Two incredible minds - and I took huge risks with both of them. One died - one built a secure operating system (see this independent review).

Both taught me to stretch my mind to places I would never normally go. Which brings me to Kevin - Kevin is in a class of his own - and he differs from the other two in a unique way - practicality. The two big problems we've worked on together - making the internet go faster (mod_gzip) and making the Internet contextually aware (mod_mobile) have both been his ideas. They are both rooted in solving real problems that affect millions of people.

We never set out to do something "Cool"... we set out to solve real problems for real people. We also don't want to ask people to change their behaviors.

Bill and David were different - David was so far out there it did defy reality. But I got to travel to places (math wise) that were just incredible. Bill solved the problem of building an un-hackable operating system. It's what the world needs - whether they will adopt it or not who knows. It requires a behavior change. But it was worth doing.

Kevin builds/solves problems that the world needs - and the two of us make sure the design is done in such a way that doesn't require behavior changes. Hence the practical aspect of what we do.

With regard to what we develop for or with - we don't care - that's NOT important - what is important is solving a really meaningful problem for someone. We're platform and developer tool agnostic. We focus on the horizontal not the vertical. If we do our job right then it will scale to any operating system or developer tool.

Example - Mod_Gzip - developed in C. Works on every hardware platform, every operating system and every version of all the web servers out there. (might need a few tweaks in a port). We did our job right by focusing on the problem - not by inventing something cool.

We've done the same with Mod_Mobile - it's designed to work on everything - the other client component is also designed that way. And while neither Kevin or I particularly like Java - that's NOT a consideration - the customer wants it to "simply work" and if that means that we need to write a Java client for Blackberry and RIM that works the same was as WM then that's what we do - I mean why wouldn't you?

At the end of the day too many programmers focus on what's cool - instead of focusing on solving a meaningful problem for an identifiable group of customers.

If you focus on the later then the tools will fall as they may. In fact you won't care what you use because it's all about the customer.


PS. To build all 4 of my startups to the point of showing real working prototypes has never taken more than 18 months and consumed well less than $100k of funding. The hard part is the idea and a prototype - that doesn't consume a lot of money. The next hard part is scaling the idea - that does require money, but less than everyone thinks if you have a brilliant product marketing person on board. Which as luck would have it I do. I would put Liz in a class alongside Kevin - she's the one who turns the idea into something that customer can understand. Which as they say is - priceless.

Picture of Jason Delport
Re: Mobile needs young developers, how to lower the burden to develop
by Jason Delport - Tuesday, 11 December 2007, 6:23 PM
Guido van Rossum works for Google so I reckon there's a good chance we will see Python as a first class citizen on Android at some time.