Healthcare as an industry has been somewhat slow to embrace big data and analytics technology, but there are some indications that this trend may be changing. For a number of reasons, the future of healthcare is likely be heavily reliant on analytics technology. In this podcast we explore some of the reasons why adoption of analytics has been slow in healthcare but also some of the reasons why most industry watchers expect a coming era of high adoption.
Is there a disconnect – or, an impedance mismatch – in big data development today? The days, applications that run on big data platforms take quite a while to develop, Vin Sharma, director of strategy and product for the big data analytics efforts at Intel’s Data Center Group, says on this edition of the Talking Data Podcast. For big data to happen in a big way, the gap needs to close. Sharma and others are working to reduce the ‘while’ it takes to develop big data apps, and they propose The Trusted Analytics Platform as the means.
There’s been a lot of noise coming from the popular media about how companies like Uber and Airbnb are revolutionizing commerce in the US. One thing that often gets overlooked is the degree to which data and analytics underpin what the companies are able to do. Take a listen to this edition of Talking Data to hear more about how analytics is a key factor in the growth of the sharing economy.
Our latest podcast finds Jack Vaughan on the road, covering the NoSQL Now 2015 event in San Jose, Calif., and a Boston Chapter meeting of TDWI. At the latter confab he caught up with Mark Madsen, a noted figure among data industry followers. Madsen said the new architectures for analytics disrupt the basic database as we know it – splitting it into components. For background, these levels are: abstraction layer, data engine, Storage management and file system. Much of the rethinking in today’s Hadoop and Spark world revolves around this. Listen in!
Data preparation is a technology somewhere between a sleepy backwater and an active no man’s land. As Hadoop and other odd new takes on data processing have come on line, as the data that the enterprise is dealing with has become more various and voluminous, data preparation methods have begun to change. Check out Ed Burns and Jack Vaughn for this edition of the Talking Data Podcast, which digs into Hadoop-on-SQL, data curation in the hands of Michael Stonebraker, and the effect of machine learning on the data preparation mileu.
Today we go back into the vaults to enjoy a Talking Data podcast of yesteryear. In the summer of 2013, our editors visited Information Builders’ Summit, GraphConnect, and RedHat Summit, and lived to tell the tale. Under discussion were multitenant data bases, HTML 5, and – can you believe it? – Hadoop!
A fellow who used Front Page to build 5 fair-sized Web sites back in the mid-1990s can attest to Microsoft’s ability to bring better technology to the wider public. Front Page is often joked about now, but it was basically a $5000 (or more) 4GL tool that Microsoft bought (from Vermeer), in the process reducing the price to $99. We mention this because, in the Spring, Microsoft CEO Satya Nadella talked about Microsoft’s enduring mission to make technology available to the masses. It is an assertion that has some grounding, but it is hard to speculate whether Microsoft can find that kind of magic again. This podcast is a recap of Ignite 2015 and Build 2015 SQL Server and Azure announcements that look to move the traditional mission forward. Can Microsoft steal a march on Amazon Web Services? Today, developers’ ability to whip out a credit card and build an app quickly on the cloud may be the closet analogy to the effect Microsoft had when its software crept into the organization without first going through IT, then known as the Glass House. But in the cloud, Amazon has the lead. – Jack Vaughan
The long term prospects of MongoDB and other NoSQL databases will be judged by feature improvements for enterprises. We saw such updates when we visited this year’s MongoDB World convocation. A case study from ADP is discussed in the podcast.
At last week’s Spark Summit, both Databricks, whose founders gave the original Spark framework, and IBM introduced cloud-hosted versions of Spark. These announcements created quite a stir among data scientists who are hungry for the processing power of Spark, but who would rather not put in the time and effort of managing the clusters.
Take a listen to this edition of Talking Data to hear more about why users are so excited about the prospect of hosted Spark implementations and how it could change computing.
Content strategist Diana Hwang joins the Talking Podcast crew to talk about data journalism. Hwang and editor Jack Vaughan took part in a data journalism workshop sponsored by the New England Science Writers group, and they outline their experiences with Python, CartoDB, GeoJSON … and this strange new software they’ve found called ‘Excel’!