Decoding the new economy

Getting academics onto the cloud

Discount and free software programs are good for educational programs but they do risk industry wide vendor lock in.

Offering free products to students and academics has long been a tactic used by software companies to build their market presence. The current fight for dominance in the cloud is seeing the same tactics being used.

Last week I had the opportunity to talk to Amazon Web Services’ Glenn Gore about his company’s academic support program.

Part of that conversation ended up in a story for The Australian about how researchers are now using cloud computing services and it’s worthwhile looking at how AWS are using this program to cement their products’ market positions.

“We work with the majority of universities across Australia,” Gore said. “It’s part of an international focus around how we support the education sector in general.”

In some respects AWS’s behaviour isn’t new, for years Microsoft, Autodesk and Adobe have had programs offering free or deeply discounted products for academic or student use. The success of those schemes in becoming defacto industry standards is no small reason why these companies have dominated many sectors.

Microsoft themselves have the similar Bizspark program for tech startups and it’s easy to see how that initiative is helping push Azure’s adoption into a field that has been dominated by AWS.

One of the drawbacks though with cloud computing services is the risk of ‘sticker shock’ where customers end up with big bills. One of the universities I spoke to in researching the story recounted how 0ne of their faculties was presented with a huge AWS invoice because their engineers didn’t provision the services correctly.

This is where AWS’s team steps in with advice for researchers, “in the case of Koala Genome Project use the on-demand model, the standing pricing model for the cloud,” recounts Gore in pointing out the nature of their work could use spot-pricing to take advantage of cheaper prices in off-peak times. “As a result of making that one change they were able to do eighty percent more research.”

Getting more research time is always attractive for researchers and Dr Rebecca Johnson who leads the Australian Museum’s part of the koala consortium was particularly effusive about the support from AWS staff,

“What we have been able to access via this partnership with AWS is compute time and compute capacity that we just would not have had access too,” Dr Johnson said in a media release. “It would have cost us thousands and thousands of dollars to create and we just would not build such a computer system these days. You would not create your own computer infrastructure as we would only use a fraction of it anyway. So, it is great for us to piggy back off these already built systems.”

Being a relatively small institution, the Australian Museum is a good example of how cloud computing can work for those without the resources of big universities or corporations in the same way small businesses and startups can access resources formerly only available to enterprises.

Amazon’s programs though show the Microsoft model of getting students and startups onto their systems early pays dividends. It’s good for academic institutions but one wonders whether it’s also another form of vendor lock in.

Cloud computing’s elusive gold

Microsoft and Alphabet’s Google show the fragility of even the most profitable online business models

Alphabet, aka Google, and Microsoft yesterday announced their quarterly results and despite both making healthy profits the numbers show the online world is a tough place to make money.

Microsoft’s stockholders took a five percent hit to their wallets after the company announced weaker than expected results for the last quarter.

Notable in the results were the stunning sales growth of its cloud services with Azure boasting a 120% year on year on year increase.

Yet Microsoft’s Intelligent Cloud division which includes Azure saw its profits fall nearly 13%, showing the company’s products may be making inroads against Amazon Web Services but making profits in that market is very tough indeed.

Similarly Alphabet’s results still show the company is sill totally dependent upon the advertising river of gold for its profits.

Particularly concerning for Alphabet is its ‘other bets’ division doubled its sales but saw losses increase by 20%. Overall Google’s advertising revenues made up 89% of Alphabet’s total revenues this quarter compared to 90% last year.

While both companies have very healthy profits – about five billion dollars this quarter for each – Alphabet’s continued dependence on Google advertising and Microsoft’s declining profitability should be a worrying sign for shareholders in both companies.

Both companies show that despite the apparent riches of the technology sector, making profits is getting tougher. Shareholders of both companies should be watching carefully for any disruption to either business.

What happens when machines start to learn

Deep reinforcement learning promises to change the way robots are taught to do tasks

Computer programming is one of the jobs of the future. Right?

Maybe not, as Japanese industrial robot maker Fanuc demonstrates with their latest robot that learns on the job.

The MIT Technology Review describes how the robot analyses a task and fine tunes its own operations to do the task properly.

Fanuc’s robot uses a technique known as deep reinforcement learning to train itself, over time, how to learn a new task. It tries picking up objects while capturing video footage of the process. Each time it succeeds or fails, it remembers how the object looked, knowledge that is used to refine a deep learning model, or a large neural network, that controls its action.

While machines running on deep reinforcement learning won’t completely make programmers totally redundant, it shows basic operations even in those fields are going to be increasingly automated. Just knowing a programming language is not necessarily a passport to future prosperity.

Another aspect flagged in the MIT article is how robots can learn in parallel, so groups can work together to understand and optimise tasks.

While Fanuc and the MIT article are discussing small groups of similar computers working together it’s not hard to see this working on a global scale. What happens when your home vacuum cleaner starts talking to a US Air Force autonomous drone remains to be seen.

Amazon Web Services and the new rules of business

Amazon Web Service CTO Werner Vogels lists the lessons from a decade of AWS operations. These could be the new rules of business.

The one company that has driven both the adoption of cloud computing and the current tech startup mania is Amazon Web Services.

Later this week AWS celebrates its tenth birthday and Werner Vogels, the company’s Chief Technical Officer, has listed the ten most important things he’s learned over the last decade.

The article is a useful roadmap for almost any business, not just a tech organisation, particularly in the importance of building systems that can evolve and understanding that things will inevitably break.

Importantly Vogels flags that encryption and security have to be built into technology, today they are key parts of a product and no longer features to be added later.

Most contentious though is Vogels’ view that “APIs are forever”, that breaking a data connection causes so much trouble for customers that it’s best to leave them alone.

Few companies are going to take that advice, particularly in a world where changing business needs mean APIs have to evolve.

There’s also the real risk for businesses that their vendors will depreciate or abandon APIs leaving key operational functions stranded, this could cause major problems for organisations in a world that’s increasingly automated.

Vogel’s commitment to maintaining APIs may well prove to be a competitive advantage for Amazon Web Services in their competition with Microsoft Azure, Google and an army of smaller vendors.

Werner Vogel’s lessons are worth a read by all c-level executives as well as startup founders looking to build a long term venture, in many ways they could define the new rules of business.

Keeping the IoT simple and safe

Making the IoT simple and safe is the most important tasks facing Internet of Things vendors

Ten years ago a joke going around was “what if Microsoft built cars?” The answer summed up the frustrations users had with personal computers and the differences in engineering standards between traditional industries and that of the IT sector.

As we enter the Internet of Things era, that tension between consumer devices and good engineering continues as shown by a software bug that rendered Nest thermostats useless.

That poor software would drain the battery without warning the user, illustrates how poorly designed many of these devices are.

Ironically Nest’s owners, Google, held a conference earlier this week where the company’s leaders flagged the importance of standards, security and privacy.

In a call to action for the IoT industry, Google’s lead advocate Vint Cerf, also known as one of the “fathers of the Internet,” warned that compatibility, security, and privacy could be obstacles to the IoT’s success.

Reliability is also important, particularly when talking about safety and security – Nest also make carbon monoxide detectors – where a device crashing or failing can have terrible consequences.

At present most of the Internet of Things is about the gimmick of connecting devices to the cloud and controlling them from your mobile phone. Consumers are not going to embrace IoT products if they add cost, complexity and risk to their lives.

Keeping it simple and safe are probably the most important things designers of IoT devices can do.

Creating a digital spaghetti divide

A business digital divide could be fatal for those caught on the wrong side of it.

Are we seeing a new digital divide develop between big and small businesses, particularly in areas like retail and hospitality?

This thought occurred to me during a radio spot earlier today where we were talking about Apple Pay’s Australian launch. Many small businesses don’t have the capital or expertise to implement many of these new technologies.

A number of factors contribute to this including the legacy systems installed in small businesses, the proprietors having a poor understanding of technology and, most importantly, the lack of either capital for reinvestment or cashflow to fund the monthly charges that are standard for cloud computing services.

The expensive cloud

One unstated factor with cloud computing services is how the cost of services add up. For example a Premium 10 Xero customer with Receiptbank attached is looking at a $100 a month in charges. It’s not hard to see how adding cloud based Point of Sale, rostering and customer service software could see a small business incurring $400 a month in fees, throw in Salesforce and you could be looking at a very expensive exercise.

No doubt for those companies that can afford these services this is money well spent but for many margin or low turnover businesses, the charges could be a deal breaker.

Spaghetti Junction

Another aspect to the cloud services is the myriad of different platforms that need to be stitched together in most businesses, one cloud service founder calls it “digital spaghetti.”

Managing this bowl of complexity isn’t easy and raises a number of business risks as different services apply varying policies and practices to the data they collect and store. A breach or service failure at one could cause a ripple effect through all business operations.

For many small business owners, particularly older proprietors, managing this complexity is intimidating if not downright scary.

It may well be there’s a number of opportunities for a canny service provider to offer an out of the box small business solution, but for many older small operators with limited capital and restricted cashflow affording such a product might also be difficult.

The risk though for those businesses is they will find themselves falling further behind as markets, consumer demands and the workforce’s expectations evolve. A business digital divide could be fatal for those caught on the wrong side of it.

Category: cloud computing

What happens when machines start to learn

Similar posts:

Keeping the IoT simple and safe

Similar posts: