It’s Always Surreal in Philadelphia, Where Art Meets AI in a Sweeping Space


Climbing up flights of stairs in a bank building full of rooms draped in surrealist art, tunnels with lurking beasts called “skin horses” and exhibits of keepsakes imaginary and real, I find myself looking at an art mural across a domed ceiling that I can explore with instruments next to me. Speaking into a microphone, I see my words scroll across the edges. My hands, thrust into a small chamber, are projected across the ceiling, highlighting parts of the mural. Suddenly, AI-generated descriptions emerge where I’d put my hands.

This is the Ministry of Awe, a new installation experience in Philadelphia that I was lucky enough to visit ahead of its opening, and it’s a welcome East Coast dose of strangeness. Created by Meg Saligman and over 100 other artists, it’s a six-story space that makes me think of Meow Wolf or long-time LA oddity the Museum of Jurassic Technology — or even London’s very real Sir John Soane’s Museum.

A creepy sculpture of a long-nosed beast with humanlike eyes and hair.

This “skin horse” lurks in the basement, if you look hard enough.

Scott Stein/CNET

The former bank building’s now an immersive art gallery full of hands-on experiences to unravel and a storyline too: messages in drawers, phones that can be dialed or answered, bathrooms that record your “deposits” with audio messages. Everything at the Ministry is an exploration of the meaning of banks and their associated power. But what drew me here just as much was the idea of how tech would fold into a space like this.

Watch this: I Saw the Future of Tech Art in Philly

Much like Meow Wolf’s explorations of layers of tech into artist installations, something I talked about at SXSW recently, Ministry of Awe is playing with tiny doses of AI — nothing that generates or replaces the work of artists but rather in a way that highlights and possibly enhances. The Ministry of Awe’s signature fifth-floor artwork, The Heavens, is a giant mural work by Saligman that’s projected across the segments of the ceiling. Angled seats let visitors hang around and gaze up, but several “instruments” in the room let you play with the space, too, created by the tech company Spatial Pixel.

A room with glowing ceiling murals, white chairs and instruments, including a microphone in a glass container.

A full look at the projector-filled room where the Heavens mural exists, along with interaction instruments. This is just one room of many in the Ministry.

Scott Stein/CNET

Spatial Pixel is focused on “spatial computing for spaces, not faces,” and was founded by Violet Whitney, former director of product and associate director of design at Google Sidewalk Labs, and William Martin, an architect and designer. Both also teach a course in spatial AI at Columbia University. 

Exploring AI through art

The Heavens’ interaction tools and how they’re designed to feel integrated and somewhat invisible are part of Whitney and Martin’s explorations of where AI could work in subtler space-aware ways. This fascinates me because AI, smart glasses in particular, are already trying to solve for this with very mixed success. What I’ve found is that art and entertainment can often be better places to explore ideas of AI in contained ways, with rules deliberately made to respect the work and art.

Two people, who are the founders of Spatial Pixel, stand in a room filled with glowing murals.

Spatial Pixel’s team in the room they helped design.

Scott Stein/CNET

Whitney and Martin met Saligman in the same Philadelphia neighborhood, which is how they ended up collaborating on the Ministry of Awe’s exhibits. The Heavens experience is run using Spatial Pixel’s open-source platform, called Procession, that blends multiple AI models into a system that works in physical spaces. Whitney and Martin already have an interactive lab space for it at Columbia, but the Ministry of Awe is a public test-bed, working off art that they want to keep sacred. 

“A lot of what we’ve been doing is finding ways of changing the mural, or the way that you see the mural through light. A core way we’ve been trying to allow visitors to interact with it is to pick up on the things that they’re saying in the space,” said Whitney. “We want to take the things they’re saying and change the mural based on their words and what they point at.”

A vaulted ceiling in a lobby with art and windows for Ministry of Awe in Philadelphia.

The Ministry of Awe’s multilevel former bank building has many rooms inside, many of them interactive, and they were designed differently by different artists.

Scott Stein/CNET

Right now, a lot of the mural interactions are simple and ephemeral: My words disappear, my highlights fade. But the Ministry of Awe’s toying with the theme of banking in personal data, too. And the software being used to run the installation is programmable, so Spatial Pixel aims to keep evolving what happens over time.

“Our goal is eventually to record what the people are contributing, with the right consent. But then maybe those ideas become like this bank. It is a bank, after all, to store these ideas, and then Meg can use them and review them and use it to evolve the painting and physical space. And so it becomes this sort of perpetual dialogue with the muralist,” said Martin.

It’s part of the thinking that Spatial Pixel wants artists to play with, as opposed to tech companies. 

A mural with an angel and words highlighting parts of the painting.

Words overlay with art, depending on how you interact. The work changes slightly over time.

Scott Stein/CNET

“What if you could actually talk to a painting? What if you could actually interact with a work of art and then explore it in new ways? We realized,” said Martin, “that accessing these tangible computing techniques, like being able to recognize gesture, move objects around — there’s certainly a lot of academic groups that are discussing this, but it’s still really inaccessible to the actual designers that want to make experiences in that way.”

The idea echoes experimental AI art I saw in Austin at SXSW just days after my Ministry of Awe visit — questions about agency and ownership, where boundaries between AI and personal work get drawn. And as I toured the Ministry space with Meta’s smart glasses on my face, it made me think about how smart glasses — and most AI tools — right now have almost no consideration for this delicate line. 

But they’ll need to. And maybe art spaces are the places to begin to think it out, with no glasses or personal wearable tech needed at all.





Source link

Leave a Reply

Subscribe to Our Newsletter

Get our latest articles delivered straight to your inbox. No spam, we promise.

Recent Reviews


Power BI Datasets – Table of Content

What is Power BI?

Power BI is a set of software services, apps, and connectors that work together to turn disparate data sources into coherent, visually immersive, and interactive insights. Your data could be in the form of an Excel spreadsheet or a hybrid data warehouse that is both on-premises and cloud-based. Power BI makes it simple to connect to your data sources, visualize and uncover what matters, and share your findings with whomever you choose.

 Become a Power BI Certified professional by learning this HKR Power BI Training !

What are Datasets in Power BI?

A dataset is a data collection that you can connect to or import. Power BI allows you to connect to and import all kinds of datasets, allowing you to put everything together in one place. Dataflows can also be used for sourcing the data for Datasets. Workspaces are associated with datasets, and a single dataset can be used in multiple workspaces.
We have selected “My workspace” and then the “Datasets + dataflows” tab in the example below

Power BI workspace

Let us now look into the different types of Datasets in Power BI.

Types of Datasets

Datasets in Power BI are ready to report and visualize the source of data. There are five different types of datasets, each of which can be constructed in one of the following ways:

  • An existing data model will be connected that is not hosted in a Power BI capability.
  • Power BI Desktop file needs to be uploaded which includes a model.
  • Uploading a CSV (comma-separated values) file, or uploading an Excel workbook (Includes one or more Excel tables and/or a workbook data model).
  • Creating a push dataset using the Power BI service.
  • Creating streaming or dataset with hybrid streaming using the Power BI service.

Let us now explore different types of Datasets.

1) External-hosted models

Azure Analysis Services and SQL Server Analysis Services are the two types of externally hosted models. Installing the on-premises data gateway, whether on-premises or VM-hosted infrastructure-as-a-service (IaaS), is required to connect to a SQL Server Analysis Services model. A gateway isn’t required for Azure Analysis Services.

When there are existing model investments, such as those that form part of an enterprise data warehouse(EDW), connecting to Analysis Services makes sense. By utilizing the identity of the Power BI report user, Power BI can establish a live connection to Analysis Services, enforcing data permissions. Both tabular models and multidimensional (cubes) are supported by SQL Server Analysis Services. A live connection dataset sends queries to externally hosted models, as demonstrated in the accompanying 

External-hosted models

2) Power BI Desktop-developed models

A model can be created using Power BI Desktop, a client application for Power BI development. The model is essentially a tabular Analysis Services model. Models can be created by importing data from dataflows and blending it with data from external sources. While the characteristics of how modeling can be accomplished are outside the subject of this article, it’s crucial to note that Power BI Desktop supports three different types, or modes, of models. We are going to discuss the datasets in the coming sections.

Row-Level Security (RLS) can be used in externally hosted models and Power BI desktop models to restrict the amount of data that can be obtained for a certain user. Users in the Salespeople security group, for instance, can only see report data for the sales region(s) to which they’ve been assigned. Roles in RLS can be either static or dynamic. Static roles apply the same filters to all users allocated to the position, whereas dynamic roles filter by the report user.

3) Excel workbook models

The creation of a model is automatic when datasets are created from Excel workbooks or CSV files. To construct model tables, Excel tables, and CSV data are imported, and an Excel workbook data model is translated to produce a Power BI model. In every scenario, data from a file is imported into a model.

4) Push Dataset

A Power BI dataset that can only be created and populated using the Power BI API is known as a push dataset. However, the lack of a good user interface for creating a push dataset restricted its adoption to scenarios where a single table was inhabited with real-time data streaming.

5) Hybrid Streaming Dataset

Real-time streaming in Power BI allows you to stream data and update dashboards in real-time. Real-time data and visuals can be displayed and updated in any Power BI visual or dashboard. Factory sensors, social media sources, service usage metrics, and a variety of other time-sensitive data collectors or transmitters can all be used to collect and transmit streaming data.

Hybrid Streaming Dataset

Want to know more about Power BI,visit here Power BI Tutorial !

Power BI Training

  • Master Your Craft
  • Lifetime LMS & Faculty Access
  • 24/7 online expert support
  • Real-world & Project Based Learning

How to Create a Power BI Dataset?

Before discussing the steps of creation. It is necessary to know that there are three basic ways to retrieve data in Power BI Desktop that you will use to create your visualizations:

1) Live:

Here you will be connecting to a server that carries all the data. Although no data is sent, the model’s metadata is imported into Power BI Desktop. A query is transmitted to the server when you build visualizations, and it is then executed. The outcomes are then visualized and returned to Desktop. With SQL Server Analysis Services (SSAS) models, whether multidimensional or Tabular, live connections are commonly employed. Power BI Desktop behaves like any other thin client in this scenario, like Excel or Reporting Services (SSRS). It is not possible to make major modifications to the model, but you can add new measurements that will be available in that  .pbix file.

2) DirectQuery:

You can make more modifications to the model here than you can with a Live connection. The data is kept on the server, and queries are run on the server, just like in Live. The Power BI Desktop model, for instance, allows for the creation of relationships.

3) Import:

Power Query queries are used to import the data into a Power BI Desktop file (.pbix). The data is compressed highly so it’s feasible to load records in millions into a file on your system. A model, comparable to an SSAS Tabular model, is built behind the scenes. This is the most versatile mode, as it allows you to blend data from any source. However, all data must be loaded into your model, which can take a long time to refresh.

Now, let’s move to create the dataset. Below are the steps which make you comprehend the creation of the Power BI Dataset.

1) A dataset is connected to the .pbix file where it was created one by one. When you first launch PBI Desktop, click “Get Data” to create a new dataset.

Get Data

Alternatively, you can choose a source from the dropdown menu as shown below:

dropdown menu

2) Let’s assume we imported a few tables from the WideWorldImporters SQL Server sample database (The .pbix file can be downloaded here). The tables and their relationships are visible in the Model view:

.pbix file downloaded

3) You can view the actual data of one table at a time in the “Data view”.

Data view

4) You can create, view, and interact with visualizations built on top of the data and model in the “Report view”. 

Report view

 The dataset is made up of the data as well as the model view. Now, let’s move to the different modes of Dataset available in Power BI.  

[ Related Article : msbi ]

HKR Trainings Logo

Subscribe to our YouTube channel to get new updates..!

Dataset modes in the Power BI

These modes of Dataset in Power BI ascertain whether or not data is imported into the model or retained in the data source. The following are the three Dataset modes in Power BI:

  1. Import
  2. DirectQuery
  3. Composite
1) Import

The most popular mode for developing datasets is the import mode. Because of in-memory querying, this mode provides incredibly quick performance. Modelers can also benefit from design flexibility and support for certain Power BI service capabilities (Quick Insights, Q&A, etc.). It’s the default mode when developing a new Power BI Desktop solution because of these advantages.

It’s crucial to realize that all imported data is saved on disk. When the data is refreshed or queried, it should be fully loaded into the memory of Power BI. Import models can yield very rapid query results once they are in memory. It’s also crucial to note that there’s no such thing as a partially loaded Import model in memory. An Import model can also integrate data from any number of supported data source types. The following image illustrates it. 

Import model

2) DirectQuery

Import mode can be replaced by DirectQuery mode. Data is not imported into models created in DirectQuery mode. Instead, they are made up entirely of metadata that defines the model’s structure. If the model is queried, data is retrieved by using the native queries from the underlying data source.

DirectQuery Model

3) Composite

The composite mode can blend DirectQuery and Import modes, or integrate multiple data sources for DirectQuery. The storage mode for every model table can be configured for models created in Composite mode. Calculated tables (defined with DAX) can also be used in this mode.

Composite Model

Import and DirectQuery modes are used in composite models to give you the best of both modes. They can blend the high query performance of in-memory models with the capacity to access near real-time data from data sources when set properly.

Top 50 frequently asked Power BI Interview Questions

Power BI Training

Weekday / Weekend Batches

 Conclusion:
We have successfully learned that Power BI lets you connect various datasets for importing and bringing them all together in one place. In this blog, we explored the topics of Datasets in Power BI in a systematic flow by understanding Power BI, then Datasets in Power BI, different types of Datasets and models used for reporting and visualizing data, creating a Dataset for connecting files, and various modes of Datasets in Power BI.

Related Article:

  1. MSBI vs Power BI
  2. Looker vs Power BI
  3. KPI in Power BI
  4. DAX In Power BI
  5. Power BI Architecture
  6. Power BI Components
  7. Power BI Dashboard
  8. Power BI Data Modeling
  9. Power BI Documentation



Source link