SDL, part of RWS Holdings plc, announced a strategic partnership with Fuji Xerox Co., Ltd. to offer SDL Contenta Publishing Suite to manufacturers and aerospace and defense organizations in Japan, with a future plan to expand into other Asia-Pacific countries/regions for technical content creation, management and publishing. Technical documents for these industries are highly complex, and often include hundreds of thousands of instructions and graphics to support the operations, maintenance and inspection of complex assets. The combination of SDL and Fuji Xerox will help these organizations to consolidate, standardize, and adopt practices, based on standards including the ASD S1000D, to achieve efficiencies across their content supply chain. The SDL Contenta Publishing Suite is an integrated, publishing solution for technical content, with functionality optimized for each step of the publishing process. Based on the S1000D standard, it helps organizations manage millions of pages of complex technical documents and deliver interactive electronic technical publications (IETP) enabling maintenance professionals to meet mission objectives, reduce mean time to repair (MTTR), and keep assets deployed.
Translations.com announced its certification of Contentserv’s integration for GlobalLink Connect. The certified solution gives users a new way to leverage GlobalLink Connect’s translation workflow management while creating and processing translation requests within an interface familiar to Contentserv users. Contentserv’s integration with GlobalLink Connect provides an all-in-one solution to initiate, automate, control, track, and complete all facets of the translation process. Contentserv’s product experience platform combines with GlobalLink Connect’s extended localization workflow capabilities to create a seamless plug-and-play content management solution with minimal effort. Contentserv combines Product Information Management (PIM), Master Data Management (MDM), and Marketing Experience Management (MXM) to give brands and retailers the ability to offer high-quality and engaging end-to-end product experiences. By leveraging GlobalLink AI, Contentserv customers can reduce costs and project timelines while maintaining quality control over translations. The integration allows users to:
Save time and money when translating content
Streamline the translation process for all product content across all sales channels
Schedule and request on-demand translation via the Contentserv UI
Gain transparency of translation spend, turnaround time, and other KPIs
Optimize internal or external vendor management
Utilize flexible workflows using machine translation, human translation, or both
Achieve ROI via reduced IT involvement and project management overhead
Cortical.io announced a new release of its Cortical.io software. Utilizing a natural language understanding (NLU) approach based on semantic folding theory, the software analyzes the content of large quantities of documents. It automatically searches, extracts, classifies and compares key information from agreements, contracts, and other unstructured documents like policies and financial reports. The Cortical.io Contract Intelligence solution understands the meaning of whole sentences and concepts, instead of just keywords. The new version capabilities include: high-fidelity rendering of documents, improved extraction capabilities and advanced search (i.e. the ability to perform range queries that allow you to search for numerical and date ranges).
Cortical.io Contract Intelligence helps reduce the time and costs for any organization that needs to review and extract information from a large number of unstructured documents. It also helps reduce human errors inherent to a boring repetitive task and make better use of expensive subject matter experts. This helps markets such as insurance, that review and extract sensitive information from policies and loss run reports on a large scale. Pricing is based on annual volume of documents.
Acquia announced Acquia Digital Commerce, a solution to enable marketers to unify data, content, commerce, and digital merchandising into a single data layer to deliver a seamless omnichannel experience across the customer lifecycle. Using Acquia Digital Commerce with Acquia Open Digital Experience Platform (DXP), marketers can drive real-time, personalized, shoppable experiences at every customer touchpoint. Together with partners commercetools and Lucidworks, Acquia has built a composable commerce solution, giving marketers flexibility to create digital experiences across the customer journey. Acquia Digital Commerce delivers the agility to build digital experiences across every channel, and the flexibility to support multi-tenant architectures and composable, multi-site experiences. With a microservices-based architecture, Acquia Digital Commerce helps ensure repeatability and reuse to drive standardization and compliance, and enables continuous refinement and testing to optimize results.
Teams can maximize their commerce investments by integrating this headless, cloud-native shopping platform with Acquia Open DXP. commercetools provides an omnichannel shopping platform and the Lucidworks AI-powered product discovery solution delivers personally relevant products and content to customers. Today also marks the launch of the Acquia DX Alliance, the company’s open technology partner community, to drive collaboration through an ecosystem of leading technology vendors providing choice and jointly delivering interoperable solutions that extend Acquia Open DXP.
eccenca released version 20.12 of its knowledge graph software eccenca Corporate Memory. The latest releases consolidates their mission to make semantic data management technologies enterprise-ready and usable for business users. By focussing on user experience and performance eccenca Corporate Memory provides a wide array of access and integration points, a ready-to-use query catalogue, tools for data automation as well as the means for a detailed and transparent data governance. In past releases eccenca introduced enterprise-ready interactive graph visualization, automation tools for data migration and normalization, REST-API connections to query data from 3rd party applications, and mechanisms for data protection and access management. eccenca Corporate Memory 20.12 benefits:
Simplified building process: The DataIntegration workbench unifies all relevant views.
Powerful, easy-to-use reporting: Integrated connectors allow the creation of dashboards and data visualizations directly in Microsoft Power BI and Redash.
Business-user friendly data exploration: The catalogue of declarative data queries allows business users to access and explore data without coding.
Workflow automation: The updated cmemc command line tool simplifies the execution of workflows like dataset creation, update and deletion as well as updating vocabularies.
Ready for internationalization: Localization of the user interface and metadata with i18n language integration.
Enhanced data transparency and understanding: Statement annotation allows definition and documentation of additional metadata for a shared understanding of enterprise data across departments.
Payload CMS launched a content management system and application framework built with NodeJS, React and MongoDB. Payload is a “headless” CMS that allows you to edit and publish content in one place, but use your content from any number of devices—including websites, mobile apps, smart TVs, wearables, etc. They don’t place restrictions on how developers build their apps and focus solely on managing and delivering content. Payload was built to deliver the JavaScript community a “silver-bullet” content management solution. Although it’s not the first JavaScript CMS, many JS developers still use WordPress in a headless role for their CMS—which is built with PHP and initially only meant to be a blogging platform. Payload features:
GraphQL, REST, and NodeJS APIs
Easily customizable ReactJS Admin
Fully self-hosted
Extensible user Authentication & Access Control
Field-based content Localization
Powerful field types including a layout builder
Payload is free for local and development purposes, and production licenses are priced competitively at $22 per month if paid annually.
This certainly caught me off guard. Graph databases have been leading the popularity contests for the last five or six years, but in the last twenty four months Time Series databases have leapt ahead, as this DB-Engines chart dramatically demonstrates. Peter Wayner looks at why.
Business models based on being both a publisher and a platform have always been fraught. In some ways Medium has managed this better than most. They just acquired “social ebook platform” Glose, but it’s not clear how this fits into their platform/publisher model. One clue may be Ev Williams’ earlier statement that Medium’s…
top-line metric is “TTR,” which stands for total time reading. It’s an imperfect measure of time people spend on story pages. We think this is a better estimate of whether people are actually getting value out of Medium.
But in a short post about the acquisition Williams says they “are not planning to bundle books into Medium Membership, though there could be book-related benefits. TBD.”
Ethical issues in privacy, advertising and machine learning
Informed and interesting interview with Oxford philosopher Dr. Carissa Véliz. Don’t worry, this is not a long dry treatise, but an engaging and accessible discussion that does not require a technical or philosophical background.
Both search engine developers and users treat facets as useful for refining broad search queries. But there’s a tendency to conflate broad queries with ambiguous queries. There’s an important distinction between the two.
Fortunately, we have the ever-reliable Daniel Tunkelang to explain.
The Gilbane Advisor is curated by Frank Gilbane for content technology, computing, and digital experience professionals. The focus is on strategic technologies. We publish more or less twice a month except for August and December. We also publish curated content technology news weekly We do not sell or share personal data.
Google introduced “ToTTo: A Controlled Table-To-Text Generation Dataset”, an open domain table-to-text generation dataset created using a novel annotation process (via sentence revision) along with a controlled text generation task that can be used to assess model hallucination. ToTTo (shorthand for “Table-To-Text”) consists of 121,000 training examples, along with 7,500 examples each for development and test. Due to the accuracy of annotations, this dataset is suitable as a challenging benchmark for research in high precision text generation. The dataset and code are open-sourced on our GitHub repo.
In the last few years, research in natural language generation, used for tasks like text summarization, has made tremendous progress. Yet, despite achieving high levels of fluency, neural systems can still be prone to hallucination (i.e.generating text that is understandable, but not faithful to the source), which can prohibit these systems from being used in many applications that require high degrees of accuracy.
While the process of assessing the faithfulness of generated text to the source content can be challenging, it is often easier when the source content is structured (e.g., in tabular format). Moreover, structured data can also test a model’s ability for reasoning and numerical inference. However, existing large scale structured datasets are often noisy (i.e., the reference sentence cannot be fully inferred from the tabular data), making them unreliable for the measurement of hallucination in model development.
