DSC Weekly Digest 28 December 2021: An Auld Lang Syne (and Cosyne Too)

The last week of the year has traditionally been about reflections and planning, taking stock of the old (or auld), and preparing for the changes of the new. I’ve added my own thoughts about what 2022 will bring in the macro-scheme, but wanted to use this opportunity to write about what has happened, and will happen, with Data Science Central (DSC)  and TechTarget in general.

This is my second New Year’s helming DSC. A lot has changed in that year. Data Science Central is in many respects something of an island at TechTarget, with a different operating model, a broader focus than most, and with more of a community focus than the TechTarget search sites. It is also a site that was intended from the outset to be more technically focused, especially within the realm of data science, machine learning, and artificial intelligence.

When DSC was first started by Tim Matheson and Vincent Granville, Data Science as a discipline was just getting started and could be described as the use of programming like languages in order to do statistical analysis. Certainly doing statistical analysis via computers was far from new. Languages such as SAS and SRSS were in use back in the 1960s and 70s, and even the ancient language of Fortran contained many statistical packages. 

What differed in 2012, when DSC debuted, was the rise of R, a language that had been created as an open-source package that drew upon SAS and SRSS as antecedents. Not long thereafter, programmers added a number of numerics packages to the Python language, including Numpy, Pandas, SciKit, and others, and soon the two camps were launching cows, sheep, and chickens at one another using virtual trebuchets trying to stake a claim as to which language was best for data science.

Who won? Arguably, Python seems to have outlasted R, but as with many “religious” wars in programming, what seems to have happened instead is that those parts of data science that could be encoded as pipelines were, well, encoded as pipelines. Data science moved into the cloud or into databases (which moved into the cloud). There are still people arguing about Python vs. R, of course, but the rest of the world appears to have ascended into the electronic firmament.

Much of the context of DSC changed over the year from the best distribution to use for dealing with biased data to the best techniques for monetizing data products, the convergence of semantics and machine learning, issues faced with metaverse development, and the rise of data meshes. Put another way, the technology that DSC covers is changing rapidly and dramatically, and DSC has needed to change in response.

Our team at Data Science Central has changed as well. In January of 2021, DSC received a new Executive Editor in Ben J. Cole, even as we lost the last of our interns and the transition team that had been in place since DSC was acquired by tech target in 2000. Ben has been responsible for managing a number of sites including DSC, but he has also pitched in to do social media work and made sure that specific contributor invoices got paid. By mid-2021, we had also hired DSC’s new Assistant Editor, Scott Thompson, who took over much of the social media function so that I could concentrate on the editorial side of things. Starting in 2022, Scott will also become our first regular news editor.

This is worth discussing in a bit more depth. In the tech publishing field, most of the content that we (DSC) publish is considered editorial in nature – essays, how-tos, analyses, surveys and so forth. Much of it may be news to you, gentle reader, but the term itself usually means content about events – companies, trade shows, product announcements, and releases, and so forth. This content is something that DSC has mostly lacked, and it was one of my objectives when I took over the reins of the site to cultivate a new news presence.

With Scott taking on the role as news editor, we are planning on rolling out the opportunity for regular contributors to also become paid news contributors. The details for this are still being worked out, but we believe it provides both an outlet for new tech journalists to spread their wings and for companies to better get the word out about what they are doing. We’ll have more details about DSC News later next month.

A second innovation will be the shift to the BrightTalk platform, which TechTarget also acquired this year, as our preferred video and podcast platform of choice. We believe that BrightTalk sets the standard for technical media content, and moreover will provide a vehicle to expand coverage beyond written articles to general media production, something that your editors plan on taking advantage of into the new year.

Another big change, which will roll out on January 10th, 2021, is the switch to a new publishing platform. Outside of the weekly newsletter and managing editor functions, this has been the focus of my efforts through the last year. Even before I came on board, the decision to switch to a new platform had mostly been made, moving from the commercial platform that it had been on for the last decade over to a self-hosted WordPress system. This was driven in part by platform sunsetting issues and in part by the decision to make DSC, like most of TechTarget, something under better control of the company.

This also has provided Data Science Central to make some much-needed overhauls. The biggest is the switch to the new Gutenberg publishing system, which provides a powerful mechanism for adding new functionality and makes it easier to build a consistent look and feel. This also incorporates several new modules, including support for inline formatted code based upon language, the ability to write LateX notation making it far easier to incorporate mathematical formulas and expressions in articles, support for scalable vector graphics and more modern CSS, and support for inline graphs, charts, and tables, presented in ways that worked not just for web browsers, but for mobile devices as well..

DSC has long believed in providing access to new, free, and innovative papers and useful cheat sheets for developers and data scientists, but as we were plotting out our transition strategy, I had deliberately chosen to move away from the ad hoc presentation of such content until such time as we made it easier to maintain and search for it. The new platform is built with automated taxonomy semantics in mind, making it easier to both add new functionality and to make adding content into this framework more manageable, and it will also offer up additional opportunities for targeted advertising for partners and vendors. With that in place, we are now working with publishers to provide free access to up-and-coming book chapters, courseware, datasets, and perhaps even software.

The final change is a shift in the way that we handle contributing articles. DSC has, for a long time, been an open community for writers, with appropriate editorial control for content and quality. After a lot of internal discussions (and for a number of reasons), we are moving away from that model somewhat to a controlled community model. Contributing writers will need to have separate accounts to write new content, which will be given by open invitation. We will be contacting regular writers over the next couple of weeks to transition them in automatically, but if you are interested in writing for Data Science Central, please contact us at [email protected] and we will work towards getting you on board.

We’re excited to show off the new changes to the site, especially as it makes it easier to find relevant articles, provide superior content, and better follow the work of the superb writers who make up our community.

So, with a bit of bubbly (or fizzy apple juice, my preferred alternative) I wish you a Happy and data-filled 2022, and invite you back for the NEW Data Science Central.

Kurt Cagle
Community Editor,
Data Science Central

To subscribe to the DSC Newsletter, go to Data Science Central and become a member today. It’s free!