Announcements（公告）

Foundry Map: Geospatial analysis in Foundry

Date published: 2022-11-22

Map out the digital twin of your world

With Foundry Map, you can now bring together and analyze data in a singular, geospatial context. Once data is imported into the Ontology, objects can be brought into Map for analysis alongside raster imagery and geospatial layers. Users of varying technical ability can now build, manage, and search through self-directed geospatial applications that can be easily shared and used across Foundry.

Use Foundry Map to:

• Explore connections between geospatial objects and traverse physical networks.
• Visualize contextual geospatial data from a variety of sources, including high-scale vector data and satellite imagery.
• Review temporal data such as events or paths of object movements over time.
• Search geospatially for point and polygon data, using bounding box and polygon intersection queries.
• Interact with maps by drawing shapes and add annotations or perform geospatial actions such as zone or area creation.

Follow your assets with track analysis

Foundry Map provides both live and historical location data, enabling users to visualize the tracks and movement of specified objects over time. This capability can be applied to any mobile asset, such as aircraft, ships and packages.

Track analysis can also be combined with other temporal data, such as sensors, IoT data streams, and events. Start by selecting any historical, current, and/or future time range, then identify a specific point in time to analyze the locations of objects. This enables you to better understand the general state of the world at a precise moment.

Add color to your time series data

With robust controls on color, Foundry Map makes it easier to visualize patterns, make predictions, and answer key questions. Users can change the color and style of objects on the map according to data values and bring together time series, time-dependent properties, and associated events for targeted situational analysis. For instance, this map shows a combination of avalanche regions, weather stations, and SNOTEL stations that display snow depth, location, and color-coded severity over time. Obtain high-level intuition in one screen, then interrogate the specific points of interest easily with the chart or jump into Quiver for dedicated time series analysis.

When should I use Foundry Map?

We recommend using Foundry Map for geospatial workflows, particularly in analyses that rely on both geospatial and temporal information. Foundry Map can be used in stand-alone analyses, as a template to generate new maps, or embedded into a Workshop application as a powerful geospatial component.

The introduction of Foundry Map does not affect the usage of existing maps in other Foundry applications. Existing map components, such as Slate, Contour, and Quiver, will continue to function as before.

Visit the Foundry Map documentation to learn how to set up your first map.

---

Slate: Redesigned Dependency Graph

Date published: 2022-11-10

Today, we are proud to introduce a complete overhaul of the dependency graph! The dependency graph supports users to investigate and understand the connections between different components in a Slate application and manage the complexity of mature applications.

In this update, additional application components like Actions, Events, and Global Variables are now represented in the graph. A node visual redesign and new layout engine improve legibility, especially in complex views. And it's easier than ever to jump into the graph view with quick links within widget, function, and query configuration panels.

Learn more about the new dependency graph in the docs.

Sunsetting Foundry Reports

Date published: 2022-11-01

Note: This notice announces the intent to deprecate the Foundry Reports functionality in favor of other reporting, dashboarding, and collaboration tools within Foundry.

Beginning early May 2023, new Foundry Report creation will be disabled for most customers[1] and existing ("legacy") Foundry Reports will become read-only. Legacy Foundry Reports will not need to be converted or transitioned to new formats unless they need to be edited.

All customers now have the following options available to replace the dashboarding and reporting functionality of Reports:

• Contour dashboards for content that is driven by a Contour analysis. Contour dashboards provide improved interactivity over Reports, with features like chart-to-chart filtering and an easy drag-and-drop interface to build the dashboard while iterating on the analysis. Contour dashboards also support a fullscreen presentation view and PDF exports. Get started with Contour dashboards.
• Quiver dashboards for content that is driven by a Quiver analysis. Quiver dashboards provide functionality unavailable with Foundry Reports, since Reports does not support embedding most Quiver visualization types. Get started with Quiver dashboards.
• Notepad for cross-product reporting. With Notepad, you can create both live-updating and static ("point-in-time") documents, annotate documents, share documents in-platform, and export documents as PDFs. Get started with Notepad.

Prior to disabling new Report creation, Palantir will provide the following functionality:

• Ability to export Contour dashboards as PDFs (expected late 2022).
• Ability to send Notepad documents via email on a schedule (expected early 2023).
• Ability to convert legacy Reports to Contour dashboards or Notepad documents (expected early 2023).

Contact Palantir Support with any questions.

[1] We will still support editing existing Reports and creating new Reports for customers who need to create dashboards with content from Code Workbooks. Read more about this use case here.

Additional highlights

Administration | Resource Management

Analysis tab added to the app | The analysis tab in the Resource Management app gives users the opportunity to explore their usage data in more depth. Here users can explore their usage by usage account, project, ontology, and source. Usage can be grouped together and explored in aggregate or viewed at the level of an individual Foundry dataset or resource.

Analytics | Contour

Add choropleth custom bucket color style for Contour map board | Choropleth layers in the new map board can now be colored using user-defined custom buckets. Users can define custom ranges for buckets and the colors that represent those buckets. The map will use this information to decide which color should be shown in each region.

Analytics | Notepad

Allow PDF export from Notepad Templates without persisting Notepad | Previously, Notepad documents that were generated from templates for PDF download via Workshop's Notepad Template button always needed to be saved in Compass. The new "Save Notepad from PDF export" option (default for new widgets) provides the option to not persist the resulting document.

Add print page size options | Introduces a new print option for Notepad documents to configure the page size. Options available are: letter (current default), legal, tabloid, A3, A4, and A5. See more in the Notepad documentation.

Analytics | Quiver

New Create Array transform | Added the ability to define an array by referencing an existing array or by manually specifying values (static, referencing a column, or referencing a parameter). This can be used as an individual card or apply in a transform table.

An improved Events plot backed by transform tables | The updated events plot can now plot event data from transform tables in addition to object sets. With transform tables you can modify (such as date shift) or enrich the underlying event data before plotting it. The chart plots events over a timeline, highlighting the number of events in each interval. Hovering over an interval shows a list of events in that interval, and hovering over each event will show the details of that event. Colors and event details are also configurable.

New dependencies view in card editor | Added a new Dependencies view in the card editor showing the data lineage (input and output cards) of the selected card in order to simplify dependency tracking without leaving the canvas. To streamline dependency navigation, clicking on an input or output card will change the focus to that card.

A new input type (time series) to transform tables | Time series plots can now be an input to a transform table. This converts the time series data into tabular format where it can be manipulated, edited, or enriched.

Added array input type support for calling functions on objects (FoO) and actions from Quiver | It is now possible to pass array primitives in Quiver into FoO functions and Actions that have array inputs defined.

Date/time Boolean transforms released | Added the ability to compare timestamps using Boolean date/time transforms with the following transforms: before, on or before, after, on or after, on, not on, and between.

New output type (Tabular time series) for transform tables | Added the ability to directly convert tabular format time series data (timestamp and value) to a Quiver time series via transform tables, to then be able to use all the time series visualizations and transformations available in Quiver.

App Building | Slate

Add dependency graph popout | The dependency viewer can now be opened in a new full-screen window like other editor panels.

Removing code sandbox permission check | The code sandbox widget is available to all users and no longer requires privileged permissions.

Shift-click allow selecting widgets | You can now add widgets from the Slate widgets sidebar to a multiple selection group by holding Shift or Cmd/Ctrl. Previously only selections made on the canvas could be added to a multiple selection group.

Data Integration | Code Repositories

Enable input prefiltering in Authoring Preview for more column types | Authoring preview now supports additional column types for input sample prefiltering. For more details on how to use previews for pipelining, refer to this guidance. Supported column types now include String, Boolean, Number types (Integer, Double, Decimal, Float, Long, Short), Date, and Timestamp.

Data Integration | Pipeline Builder

Autosave support | Pipeline Builder will now autosave changes related to a specific pipeline and branch. Autosaved changes are private to the user making those changes, and they are not committed to the version history of the pipeline. These changes will be restored automatically on page refresh and persist through log out/log in cycles. Once a user saves a new version to a branch, their autosaved changes are cleared.

Diagram

Advanced handling of scale and rotation component parameters | Updated the controls for editing scale and rotation parameters for diagram components to allow for more fine-grained control. In addition, these parameters can now be bound to object or series data, bringing them inline with other component parameters where this functionality is already available.

Ontology | Ontology Management

Enable the properties panel V2 for everyone | Redesigned properties panel, splitting different property configurations up into different tabs for easier navigation and more space for each configuration category.

Redesigned Action Form Sidebar | The parameter sidebar has been completely redesigned and now allows quick access to other parameters without requiring you to return to the form overview to do so.

Ontology | Vertex

New home page | We've overhauled the Vertex home page, which opens when you select Vertex from the Foundry Workspace sidebar. It now allows you to explore all available Vertex resources, through quick access options and search. The home page gives you an overview of the number of graphs, graph templates, and Search Around files, as well as quick access to your favorites and recently used files. The search view allows you to search for graphs by name, and for templates and Search Arounds you can also search by object types they reference.

Security | Checkpoints

Add frequency setting for Login checkpoints | Login checkpoints can now be configured to display with a specified frequency. When creating a new login checkpoint, the checkpoint can be set to display for a user only after some specified amount of time has passed since the user last saw the same checkpoint.

---

中文翻译

---

公告

Foundry Map：Foundry 中的地理空间分析

发布日期：2022-11-22

绘制您世界的数字孪生

借助 Foundry Map，您现在可以在单一的地理空间上下文中汇集和分析数据。数据导入 Ontology 后，对象可以连同栅格影像和地理空间图层一起引入 Map 进行分析。不同技术水平的用户现在都可以构建、管理和搜索自主式地理空间应用，这些应用可以在 Foundry 中轻松共享和使用。

使用 Foundry Map 可以：

• 探索地理空间对象之间的关联，并遍历物理网络。
• 可视化来自多种来源的上下文地理空间数据，包括大规模矢量数据和卫星影像。
• 查看随时间变化的事件或对象移动路径等时序数据。
• 使用边界框和多边形相交查询，对点和多边形数据进行地理空间搜索。
• 通过绘制形状与地图交互，添加注释，或执行区域创建等地理空间操作。

通过轨迹分析追踪您的资产

Foundry Map 提供实时和历史位置数据，使用户能够可视化指定对象随时间变化的轨迹和移动。此功能可应用于任何移动资产，例如飞机、船舶和包裹。

轨迹分析还可以与其他时序数据（如传感器、IoT 数据流和事件）结合使用。首先选择任意历史、当前和/或未来时间范围，然后确定一个特定时间点来分析对象的位置。这使您能够更准确地了解特定时刻的整体状况。

为您的时序数据增添色彩

借助强大的颜色控制功能，Foundry Map 使可视化模式、做出预测和回答关键问题变得更加容易。用户可以根据数据值更改地图上对象的颜色和样式，并将时间序列、时间相关属性和关联事件结合起来进行有针对性的情境分析。例如，此地图显示了雪崩区域、气象站和 SNOTEL 站的组合，这些站点随时间显示积雪深度、位置和按颜色编码的严重程度。在一个屏幕上获得高层次直觉，然后通过图表轻松查询特定兴趣点，或跳转到 Quiver 进行专门的时序分析。

何时应使用 Foundry Map？

我们建议将 Foundry Map 用于地理空间工作流，特别是那些同时依赖地理空间和时间信息的分析。Foundry Map 可用于独立分析、作为生成新地图的模板，或作为强大的地理空间组件嵌入到 Workshop 应用中。

Foundry Map 的引入不会影响其他 Foundry 应用中现有地图的使用。现有的地图组件，如 Slate、Contour 和 Quiver，将继续像以前一样运行。

访问 Foundry Map 文档了解如何设置您的第一个地图。

---

Slate：重新设计的依赖关系图

发布日期：2022-11-10

今天，我们自豪地推出依赖关系图的全面革新！依赖关系图支持用户调查和了解 Slate 应用中不同组件之间的连接，并管理成熟应用的复杂性。

在此更新中，其他应用组件（如 Actions、Events 和 Global Variables）现在也显示在图中。节点视觉重新设计和新的布局引擎提高了可读性，尤其是在复杂视图中。并且，通过微件、函数和查询配置面板中的快速链接，进入图形视图比以往任何时候都更容易。

在文档中了解有关新依赖关系图的更多信息。

Foundry Reports 即将停用

发布日期：2022-11-01

注意： 本公告宣布计划弃用 Foundry Reports 功能，转而使用 Foundry 内的其他报告、仪表板和协作工具。

从 2023 年 5 月初开始，大多数客户将无法创建新的 Foundry Report[1]，现有的（"旧版"）Foundry Reports 将变为只读。除非需要编辑，否则旧版 Foundry Reports 无需转换或迁移到新格式。

所有客户现在都有以下选项来替代 Reports 的仪表板和报告功能：

• Contour 仪表板：适用于由 Contour 分析驱动的内容。Contour 仪表板比 Reports 提供了更好的交互性，具有图表间过滤和简单的拖放界面等功能，可以在迭代分析的同时构建仪表板。Contour 仪表板还支持全屏演示视图和 PDF 导出。开始使用 Contour 仪表板。
• Quiver 仪表板：适用于由 Quiver 分析驱动的内容。Quiver 仪表板提供了 Foundry Reports 无法实现的功能，因为 Reports 不支持嵌入大多数 Quiver 可视化类型。开始使用 Quiver 仪表板。
• Notepad：用于跨产品报告。使用 Notepad，您可以创建实时更新和静态（"时间点"）文档、注释文档、在平台内共享文档以及将文档导出为 PDF。开始使用 Notepad。

在禁用新 Report 创建之前，Palantir 将提供以下功能：

• 能够将 Contour 仪表板导出为 PDF（预计 2022 年底）。
• 能够按计划通过电子邮件发送 Notepad 文档（预计 2023 年初）。
• 能够将旧版 Reports 转换为 Contour 仪表板或 Notepad 文档（预计 2023 年初）。

如有任何疑问，请联系 Palantir 支持。

[1] 对于需要使用 Code Workbooks 中的内容创建仪表板的客户，我们仍将支持编辑现有 Reports 和创建新 Reports。在此处阅读有关此用例的更多信息](https://palantir.com/docs/foundry/code-workbook/present-visualizations/#adding-visualizations-to-reports)。

其他亮点

管理 | 资源管理

应用中新增分析选项卡 | 资源管理应用中的分析选项卡让用户能够更深入地探索其使用数据。在这里，用户可以按使用账户、项目、Ontology 和来源探索其使用情况。使用情况可以分组并以聚合方式探索，也可以在单个 Foundry 数据集或资源级别查看。

分析 | Contour

为 Contour 地图板块添加等值区域图自定义桶颜色样式 | 现在可以使用用户定义的自定义桶为新地图板块中的等值区域图层着色。用户可以定义桶的自定义范围以及代表这些桶的颜色。地图将使用此信息来决定在每个区域显示哪种颜色。

分析 | Notepad

允许从 Notepad 模板导出 PDF 而无需持久化 Notepad | 以前，通过 Workshop 的 Notepad 模板按钮生成用于 PDF 下载的 Notepad 模板文档始终需要保存在 Compass 中。新的"从 PDF 导出保存 Notepad"选项（新微件的默认设置）提供了不持久化结果文档的选项。

添加打印页面大小选项 | 为 Notepad 文档引入了新的打印选项，用于配置页面大小。可用选项包括：letter（当前默认）、legal、tabloid、A3、A4 和 A5。更多信息请参见 Notepad 文档。

分析 | Quiver

新的创建数组转换 | 增加了通过引用现有数组或手动指定值（静态、引用列或引用参数）来定义数组的能力。这可以用作单个卡片或应用于转换表。

由转换表支持的改进事件图 | 更新后的事件图现在除了可以绘制对象集的事件数据外，还可以绘制来自转换表的事件数据。使用转换表，您可以在绘制之前修改（例如日期偏移）或丰富底层事件数据。该图表在时间线上绘制事件，突出显示每个区间内的事件数量。悬停在一个区间上会显示该区间内的事件列表，悬停在每个事件上会显示该事件的详细信息。颜色和事件详细信息也是可配置的。

卡片编辑器中的新依赖关系视图 | 在卡片编辑器中添加了一个新的依赖关系视图，显示所选卡片的数据沿袭（输入和输出卡片），以便在不离开画布的情况下简化依赖关系跟踪。为了简化依赖关系导航，单击输入或输出卡片会将焦点切换到该卡片。

转换表的新输入类型（时间序列） | 时间序列图现在可以作为转换表的输入。这将时间序列数据转换为表格格式，以便进行操作、编辑或丰富。

添加了对从 Quiver 调用对象函数和 Actions 的数组输入类型支持 | 现在可以将 Quiver 中的数组基元传递到定义了数组输入的 FoO 函数和 Actions 中。

日期/时间布尔转换已发布 | 增加了使用布尔日期/时间转换比较时间戳的能力，可用的转换包括：before、on or before、after、on or after、on、not on 和 between。

转换表的新输出类型（表格时间序列） | 增加了通过转换表直接将表格格式的时间序列数据（时间戳和值）转换为 Quiver 时间序列的能力，以便能够使用 Quiver 中所有可用的时间序列可视化和转换。

应用构建 | Slate

添加依赖关系图弹出窗口 | 依赖关系查看器现在可以像其他编辑器面板一样在新的全屏窗口中打开。

移除代码沙箱权限检查 | 代码沙箱微件现在对所有用户可用，不再需要特权权限。

Shift+点击允许选择微件 | 您现在可以通过按住 Shift 或 Cmd/Ctrl 键，从 Slate 微件侧边栏将微件添加到多选组。以前，只有画布上的选择才能添加到多选组。

数据集成 | 代码仓库

在创作预览中为更多列类型启用输入预过滤 | 创作预览现在支持更多列类型用于输入样本预过滤。有关如何使用预览进行流水线处理的更多详细信息，请参阅此指南。现在支持的列类型包括 String、Boolean、Number 类型（Integer、Double、Decimal、Float、Long、Short）、Date 和 Timestamp。

数据集成 | Pipeline Builder

支持自动保存 | Pipeline Builder 现在将自动保存与特定流水线和分支相关的更改。自动保存的更改对进行更改的用户是私有的，不会提交到流水线的版本历史中。这些更改将在页面刷新时自动恢复，并在注销/登录周期中持续存在。一旦用户将新版本保存到分支，他们的自动保存更改将被清除。

Diagram

缩放和旋转组件参数的高级处理 | 更新了用于编辑 diagram 组件缩放和旋转参数的控件，以允许更精细的控制。此外，这些参数现在可以绑定到对象或系列数据，使其与已具备此功能的其他组件参数保持一致。

Ontology | Ontology 管理

为所有人启用属性面板 V2 | 重新设计的属性面板，将不同的属性配置拆分到不同的选项卡中，以便于导航并为每个配置类别提供更多空间。

重新设计的 Action 表单侧边栏 | 参数侧边栏已完全重新设计，现在允许快速访问其他参数，而无需返回表单概览。

Ontology | Vertex

新主页 | 我们彻底改造了 Vertex 主页，当您从 Foundry Workspace 侧边栏选择 Vertex 时会打开该页面。现在，它允许您通过快速访问选项和搜索来探索所有可用的 Vertex 资源。主页为您提供了图形、图形模板和 Search Around 文件数量的概览，以及快速访问您的收藏夹和最近使用的文件。搜索视图允许您按名称搜索图形，对于模板和 Search Around，您还可以按它们引用的对象类型进行搜索。

安全 | 检查点

为登录检查点添加频率设置 | 现在可以配置登录检查点以指定的频率显示。创建新的登录检查点时，可以将其设置为仅在用户上次看到同一检查点后经过指定时间后才向用户显示。