Parse is a Y Combinator backed startup that provides APIs so that Android and iOS mobile developers "don’t have to develop, deploy, or maintain their own server-side stack". They have just raised $1.1 million.

Parse allows you to save key-value pairs to the server and then later query them in relational database-esque ways. Parse also provides app push notifications and in the case of Android provides an alternative to C2DM that is particularly interesting in that it works on Android 1.5 and later (rather than 2.2 for C2DM).

If you are interested in giving it a go, TechCrunch have negotiated a limited number of beta signups.

Pricing hasn’t yet been announced but "will be free for small apps, and will scale up in a reasonable fashion for heavier usage apps and add-ons". The solution should scale as it’s based on Amazon EC2.

From a mobile developer viewpoint, it provides a quick and easy way to provide for the server side. It also provides backup of your data and an easy API for programming from the phone. One problem I can see is that you won’t have easy access to your data. While you can import via CSV, you can’t easily ’see’ your data for reporting, admin or your own backup purposes. You will need to create an app do these things and access programmatically rather than via database viewing/reporting or CRUD tools easily created if you control your own server side.

Another downside is that you will be making your app dependant on the continuing existence of Parse. Hence, for now, it’s probably ok for hobby apps, trivial apps or apps with a known limited lifetime. Hopefully, in time, it will become self-funding or some large brand or company will end up using the service and will end up buying out or sponsoring it to guarantee it’s longer term existence. 

There’s a very interesting and useful new free report on M-commerce conducted by Lightspeed Research for Figaro Digital. 

Here are the charts I think will be of most interest to developers…

 

Predictably, I suppose, the top things being bought are Apps. However, music, clothes and books score much higher than I would have anticipated. This suggests that it’s increasingly becoming viable to sell physical goods on mobile.

Following on from this, it seems that many people believe mobile is well suited for making purchases… 

 

Finally, if you are wondering what payment method should be used for physical goods…

To be honest, I find some of these results surprising. I hadn’t realised the selling of physical goods was so strong on mobile. This means that there some new immediate opportunities in this area. There are many more insights to be found in the free full report. 

 

Canalys released their worldwide smartphone statistics for Q2 2011 yesterday. An amazing 107.7 million units shipped in the second quarter of 2011. Only a few years ago I was saying the smartphone market was small compared to the market for all phones. Today, smartphones are ubiquitous and mobile development is no longer experimental but mainstream.

Here’s a graph I have created from the numbers…

Canalys omitted the Symbian percentage from their press release but we know it’s between 12% and 19% as they said Symbian is now in third place. I have been kind and given Symbian 18%.

Canalys said "demand for its Symbian-based smart phones has dissipated very rapidly" and "Fewer than 1.5 million Microsoft-based smart phones shipped during the quarter, equating to a mere 1% share of the global market, down 52% against shipments a year ago". These figures don’t bode well for Nokia’s forthcoming Windows Phone devices.

Android is now shipping over twice as many phones as iOS and had the strongest growth in Q2 2011. However, I think the end game will be more interesting. 

While the native apps vs web debate is fairly boring and over-discussed for mobile developers themselves, it’s nevertheless an important issue for companies considering mobile development. It nearly always comes up in initial conversations I have with companies and I have to explain the pros, cons and midway hybrid solutions.

Today, I came across a presentation on scribd by Worklight on "Native Web or Hybrid Mobile App Development?". It provides a great visual description of the various types of development and the tradeoffs. 

From experience, one observation I have is that the type of development tends to be determined more by what the company wants to do vs what’s available via the web browser over and above  other considerations such as ease/time of development, how to distribute and the end user experience.

I am currently getting a very large number of calls and emails from offshore companies in India. Either I am on some kind of list or offshore companies are concentrating more on mobile.

The thing is, I do all development myself. I sometimes recommend companies (offshore and non-offshore) to people but that’s the extent of it. Tellingly, of the ones I have continued the conversation with, none have been able to point me to contactable references for mobile work they have performed. They are all words and no proof. If you are thinking of offshore developers, see my previous thoughts on choosing a mobile developer. If you are an offshore developer, please don’t contact me unless you really have provable capability.

I am often asked by potential clients for my views on offshoring. I have nothing against offshoring and I often do feasibility studies for projects that end up being implemented offshore. I don’t mind as I am paid for the consultancy.

Development costs can be up to 1/5 of those of US or European developers. It’s a case of balancing risk against cost. It’s generally acknowledged that offshore projects need more management. You typically need to have a good grasp of what you want and what’s possible as you typically won’t get much advice. You are expected to be the technical expert on what you are trying to create. There can be cultural problems such as developers saying ‘yes’ to everything to please you rather than saying a difficult ‘no’ that would be helpful in the longer term. It can also be difficult to extract trustworthy time estimates. 

There are also problems, particularly in India, with high employee attrition rates. It’s not uncommon for the average employment period to be less than the length of your project. As an illustration of the scope of the problem, salaries in the first 6 months of 2011 of India’s IT services increased by 12% due to need to try to stem attrition rates of 20%. There’s also a trend for IT outsourcing companies to have a presence in clients’ countries in order to win work. All this is actually driving up costs for Indian contractors and making outsourcing a bit less competitive. Hence, you might try Russia, Eastern Europe or even Vietnam if you are looking for the very cheapest mobile development.

One tip I do have, especially for large projects, is that you don’t have to have an offshore developer do all the work at once. Also, you needn’t just choose one developer. Pick a very small, difficult part of your project and ask several developers to implement that. Use this to narrow down the developers and give the super-set of your project to the ‘best’ developer… or the experience might even make you re-consider whether offshore is right for you.

If you are a mobile entrepreneur you need as many facts and figures as you can find to evolve and validate your ideas. One source I came across recently is a study by thinkinsights with Google. It’s in the form of a slideshare presentation that provides information on how smartphones are used in daily life, how consumers multi-task with their smartphones, the types of information consumers search for on mobile, the role smartphones play in decision making for products and services and how consumers respond to mobile ads.