By Simon Bisson, Columnist, InfoWorld | MAR 1, 2017
Microsoft Graph, a collection of APIs that link into many of Microsoft’s core services, is under the hood of several of the newer Office 365 services. It’s a powerful tool that simplifies working with Microsoft’s numerous cloud services, including Azure Active Directory, Exchange Online, SharePoint, OneNote, OneDrive, and the new Planner service.
With Microsoft Graph, there’s no need to build new enterprise applications as Office plugins. Instead, you get direct access to all the data stored in your Office 365 tenant. For example, you could connect an HR request inbox to a HR application, extract content from messages, and fill in forms for approval.
Providing one API for access to all the information stored in your instances of Microsoft’s cloud breaks down many of the silos that in the past required the deployment of large-scale enterprise content management systems. As such, the Graph AI approach also simplifies the building of applications that need to work with Office 365 content by providing access to key objects like mail, files, calendars, and contacts.
Getting started with the Microsoft Graph is easy enough, thanks to SDKs for common platforms and samples for REST connections from both familiar languages (Python, AngularJS, and Swift) and Microsoft’s own platforms. The web-based API Explorer lets you build and test queries either against a demonstration tenant or, once you sign in, your own Office 365 instances.
With Microsoft Graph, you can build your own Office 365 clients
One new feature, delta queries, lets you quickly extract only new or recently changed records or relationships. With Office 365 offering 100GB of mailbox space on some accounts, and 1TB of file space, delta queries are going to save a lot of time and bandwidth.
If you’ve worked with any REST API, you’ll find Microsoft Graph familiar. It offers the usual
post options, along with
delete. You can also use the
patch operator to update specific elements of specific records, so you can update addresses in an contact list or edit a OneNote document.
Underpinning Microsoft Graph is another set of APIs and services, Office Graph. These tools provide the information and the machine-learning-powered insights you can bring into your applications. Office Graph is queried using GQL over the SharePoint APIs, and it lets you extract relationships among objects in Office 365. For example:
- Whether someone has viewed or edited a document
- To identify the people you regularly work with
- To pinpoint popular documents in your team
Delving into Office 365
One example application built using these tools is Microsoft’s Delve. Available as part of an Office 365 subscription, Delve uses your Office Groups and Outlook contacts, to let you see the documents your colleagues are working on. It also gives you a place to highlight and share documents you believe are important to your projects and colleagues.
Working with tools like Delve means you need to commit to using Microsoft’s cloud services. It’s not a local index, so it can only work with files that are stored in OneDrive for Business or in Office 365’s SharePoint libraries. Although Microsoft lets you sync specific folders with OneDrive for Business, you’re relegating PCs to acting as a local cache for files, with master copies in the cloud. Users can control access to shared folders by inviting individual users or groups, which also controls what files other users using Delve can see.
Built on top of Microsoft’s internal machine learning tools, Delve uses Office Graph to determine what’s relevant to you. If you’re part of a team working on a document, Delve surfaces it as a high-priority document. At the same time, it shows other documents you’ve collaborated on with other team members, or documents you may have emailed to them. It also uses organizational structure to help pinpoint relevance. The result is a set of documents that should be relevant to the work you’re currently doing, surfaced automatically.
Delve’s user experience isn’t the traditional Office list of files. Instead, it delivers documents as “content cards” that give you a quick overview of what they contain and who’s been working on it.
Available as an iOS, Android, web, and Windows Store app (with a Windows Mobile version in beta preview), Delve is a good example of how Microsoft is addressing the mobile market, using its cloud services as a hub that delivers similar user experiences to all the devices a business user might have.
Measuring your productivity — and business relationships
Delve’s tools are also at available in other Office applications. The MyAnalytics Outlook plugin, available with an Office 365 E5 subscription and as an add-on to other levels, uses Delve’s tools to see how you work with email and in meetings. By measuring the time you spend in documents, it delivers reports on how you spend your time in work, including how quickly you reply to emails, and who you communicate with the most. A web-based personal dashboard lets you explore your time usage further, to help you be more productive. Although the personalized analytics are private, anonymized information is available for managers to get an overall look at how their department is working.
Understanding Microsoft Graph’s relationship to Office 365 development makes more sense of Microsoft’s LinkedIn acquisition. Where Microsoft Graph delivers links between individuals and their content, LinkedIn’s graph tracks connections between people and their careers. Bringing those two different (but intersecting) graphs together can make it easier to identify expertise inside organizations. It also could give your organization tools to share and manage work more widely — an approach that’s increasingly important with a shift in work patterns to contractors and other temporary staff.
Bringing the Office and LinkedIn graphs together should also add a new dimension to other tools in Microsoft’s cloud, like Dynamics 365. Wrapping more information around a sales prospect can help with all aspects of customer relationship management, from sales to support. If there’s a contact point between a prospect and a colleague, it could be useful to the sales team. Similarly, in support situations, revealing linked documents could help provide a solution more quickly or point to an appropriate expert.
A knowledge network that can map both the relationships among individuals and with their content has long been the goal of knowledge management platforms. To that end, it’s going to be very interesting to see what Microsoft builds on its social and work graphs.