Software application development
This blog is intended to provoke discussion and exchange between like minded software application developers, engineers, architects, project managers - and keen hobbyists too.
Tuesday 9 February 2010, 8:40 AM
Palm’s webOS: if you can build a website, you’ve got an app
The company says that apps on webOS are built with HTML, CSS, and JavaScript, so, in the most basic terms, if you've built a website, you can build a webOS app. Palm also offers a developers a 70/30 split of revenue from any apps sold from its online catalog, which I suppose compares quite nicely with some of the other programmes out there.
'Does it work then? Is it over-hyped marketing spin? I spoke with Palm’s Dion Almaer and Ben Galbraith to try and get some insight, so here’s the gospel according to Palm’s developer outreach glitterati:
The web will be the route for ALL application usage (or enough to make it a viable channel for this proposition to hold water) so widening the doorway for online apps and web services to migrate to devices is a good thing.
NB: Almaer & Galbraith actually said, “We’re glad that Palm did not just produce another proprietary OS or development framework. We align to the openness of the Internet and open web standards in general.”
Palm will continue to monitor the quality and nature of the content being presented to its app catalog. But it will not stand in the way of innovation even if you are trying to reinvent the wheel. I asked if they would advise against developing something as basic as a calculator app (the devices ship with one anyway of course) and Palms says ‘bring it on’, you might just develop something new and compelling.
Palm is starting to move towards hosting its own developer events, although the company appears to still be keeping close quarters with operators for this kind of thing. Palm Developer World Las Vegas 2010 is not quite ready yet (that's a joke OK?) and the best route for developer tuition and learning continues to be the website linked at the start of this blog.
There’s also a plug-in development kit waiting in the wings (soon to launch apparently) with a set of tools and APIs to make it easy for developers to integrate C and C++ code into their webOS applications – good for “immersive” 3D game functionality so they say.
“We are excited to offer several on-ramps for developers to reach Palm customers and monetise their applications. If you are a web developer you can now employ and direct your skills for rich mobile apps. If you have C/C++ code laying around, or just prefer working in that environment, the PDK has something to offer you – and Flash developers also have a route on our platform. The worlds of ‘mobile’ and the web have converged,” said Galbraith.
Palm has also just unleashed a "Hot Apps" program that infuses a million dollars into the app ecosystem.
What I think all this points to is a new proposition for previous ‘desktop web use only’ focused web designers. A website is an application then and further – a website is a mobile application. Personally I think we need to draw a few caveats and provisos here, we’re not quite there are we? While Palm also mentioned provisioning for offline app functionality, until we reach the next tier of broadband penetration, device proliferation and user adoption, some of this may be clever positioning for a technology ‘just about to happen’. How long do we have to wait? Ah, now if I knew that!
Monday 8 February 2010, 8:28 AM
The economics of computing
The paper was attached to Sybase’s ASE product. Originally created for UNIX platforms in 1987, Adaptive Server Enterprise (ASE) has been around for a few years now. So I was primarily interested to find out what factors could make a technology leap forward when it has already been in existence (in one form or other) for more than two decades.
After all, you can’t fit a catalytic converter and a nitro fuel turbo charge onto a Hillman Imp can you? So how do you super boost an (albeit successful) product that is so long in the tooth?
In-memory database management system (or, IMDB) usage emerged as a means of boosting performance and scalability while containing storage costs in extremely high-speed data systems. But something has changed and this technology has become more mainstream. The reason for this change?
The economics of computing have changed.
Hardware has changed – we now have cheaper, faster multicore processors.
Memory has changed – cheaper 64-bit memory makes large in-memory deployment affordable.
Software has changed - internal DBMS data management software exploits both of the above to form a combined offering that is even greater than the sum of the parts.
Analysts, industry commentators and even vendors themselves are keen to talk up these trends and argue that these developments mean that we are better positioned to be ready for virtualised desktops and cloud-based environments.
IDC points out that data centres today need to reduce their physical footprint and operational cost while keeping the door open for growth and still delivering better performance. So how do they do it?
“The new economics of computing, which derive from large memory models, 64-bit addressability, fast processors and cheap memory, make it possible to design core database technology that is far faster and more scalable than was possible when the only option was to base data management on spinning disks,” says IDC.
Clustering and virtualisation have been already employed to help achieve scalability and data consolidation – and now, in-memory database technology joins the party too.
I have merely touched on the fringe of this subject, but I think it points to a few telling considerations.
1) We do not talk about database technology enough in the context of software application development as we tend to focus on the fringes, the periphery and the allure of GUI or mobile handset based stories.
2) In-memory database technology may be quite a beautiful coming together of both hardware and software development that gives us power advantages, economic advantages, ecological advantages and operational advantages. Yet it is unlikely to make many IT headlines this week.
3) Significant IT shifts generally only occur when a combination of technologies, usage patterns and market conditions all align consecutively. Andy Grove would call this a
strategic inflexion point I think. This could well be one. I could be wrong, but it kept me from reading up on Lark Rise to Candleford that’s for sure!
Thursday 4 February 2010, 10:35 AM
Who wants to be a Euro developer millionaire?
'To put things in context, Vodafone is championing its new developer competition around the latest services in the Vodafone 360 ‘offering’. Simply put, this is a suite of Internet services for phone and PC/Mac heavily themed around social media and lifestyle apps. All of a user’s contacts, status updates and messaging services are brought together in one place alongside music, photos and mapping services with the option to share any of these elements with friends.
Vodafone says that the 360 service now has more than 7,000 apps available since its launch 3 months ago and it expects to have shipped more than two million devices incorporating 50 different handsets that are capable of accessing the Vodafone 360 Apps Shop.
So what was that about a competition again? Oh yes sorry – along with offering chilled bottles of San Miguel and plates of octopus tapas at Barcelona’s Mobile World Congress on Monday 15th February 2010, Vodafone will launch of its €1m ‘Vodafone App Star’ prize fund to reward developers of the best app in Europe.
Gosh! Best app in Europe eh? They had better be careful what they pick. What if they are all rubbish? Will they have to give the prize out anyway? There’s very little new innovation in this space these days, so isn’t one million Euro knicker a bit much? They don’t seem to think so…
“The sheer volume of applications that have been submitted and approved to ship in our Vodafone 360 Apps Shop is testament to our continued commitment to open web standards,” said Lee Epting, director of content services at Vodafone Group.
You can check out the details here if you wish. I don’t think they have really detailed the bones of this competition enough in terms of what the app will be judged on and how it will be analysed, but perhaps that will be more evident at the show.
As for me, I hope to get out there, but accommodation will be running short now – so if you hear about an English journalist getting thrown in a Spanish jail for pitching his tent outside the Sagrada Familia then you don’t know me OK?
Tuesday 2 February 2010, 2:50 PM
Beating Steve Martin & Alec Baldwin with Business Intelligence
'In a bid to create some PR out of nothing, BI consultancy firm Concentra has ambitiously predicted the Oscar winners in the run up to the nominations tonight. Personally, I’ve seen so many ‘developer competition’ and ‘new platform augmentation’ stories already this year that I’m open to a bit of cheesy publicity. So here’s the scam…
Concentra says that it has used the Guardian’s DataStore collection of public data to feed its own Tableau BI platform and try to predict the mostly likely candidates for this year’s ceremony based on past awards won by films and individuals at the European Film Awards, Golden Globes and Screen Critics teamed with previous Oscar winners.
According to BI logic, Christoph Waltz, Mo’Nique and Jeff Bridges are among the top actors expected to receive the highly prestigious awards this year.
Now, any of you who are keen film buffs will know that there are some wild curve balls out there this year such as Tom Ford’s debut A Single Man starring Colin Firth. So did Concentra supply its BI engine with enough intelligence to track for genuinely remarkable talent that shines unexpectedly?
Would they for example have predicted Slumdog Millionaire?
I fear the answer is no.
You’ll see more mentions of Cameron’s Avatar and Tarentino’s Inglorious Basterds on this visual dashboard than you will of the latest Polish art house or gripping film noir love triangle.
But does this teach us a lesson for BI analysis in the workplace? Yes we know that the team tasked with data collation, management and preparation for the BI system will look for all the most important variables. But will they look for the unimportant ones too and help program for random unexpected events? This was a fun story, but it leads one to think that maybe we are being too superficial with our data.
Ask Steve Martin what he thinks and he’d just tell you that he is one wild and crazy guy – right?
Monday 1 February 2010, 3:18 PM
My multiscreen mantra
'Of course, there are solid code development reasons for wanting a multi-screen set up. For example, would you find it useful to have one screen tracking a software change and configuration management (SCM or SCCM) tool and another showing your as-it-happens development environment?
Author of the Coding Horror blog Jeff Atwood wrote some time ago now that, ”Multiple monitors [are] most useful when an application has palettes or when two or three windows need to be open, such as for programming/debugging.”
One might logically suppose that multiple screens would be fairly mandatory in extreme programming environments, so just how many do you need? Developers working on multifaceted GUI related development projects may even want to break up and individually view parts of the total user interface if it is large, complex and/or intensely rich in nature.
Of course some games developers will need multiple-screens from the outset as they may be programming for game play options that allow additional screens to support camera views and/or partial ‘surround-view’ with the use of three or even four screens built to be housed in a row.
'Going back to my reference to change management and the use of multiple screens, SCM vendor PureCM points out that if you’re writing code or fixing bugs in, say, Visual Studio – then you’ll be wanting to view your SCM tool separately to reduce the possibility of introducing errors caused by switching from one to the other.
PureCM’s technical director Mike Shepherd is on the record with the following statement, “The roots of using two screens go back a few years. Looking specifically at the Visual Studio environment, Microsoft provided an SCM interface to give access to source control. While this provided a common interface to many external programs and tools, its use was limited in scope. The SCM interface route doesn’t sit easily with more modern development concepts like task-based development and refactoring, making the two-screen approach an ergonomic compromise with the potential to introduce issues.”
OK so this might sound like the most obvious subject in the world to talk about as so many of us are using more than one screen now, but hey – there’s a technical justification to be made here and sound reasoning to back it up isn’t there?
The argument that it ought to be easier to work with your main development tool and have your SCM tool seamlessly integrated with it must hold water surely? This way you can perform tasks such as managing changesets, merging changes and getting line-by-line histories to find bugs more directly within Visual Studio itself.
Right that's it for today, now to plan the integration of my xBox 360 Live somewhere into my blossoming hedgerow of electronic windows.


