Effortless GIS, CAD & BIM data exchange with Speckle
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
| Title |
| |
| Title of Series | ||
| Number of Parts | 156 | |
| Author | ||
| Contributors | ||
| License | CC Attribution 3.0 Unported: You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor. | |
| Identifiers | 10.5446/68460 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language |
Content Metadata
| Subject Area | ||
| Genre | ||
| Abstract |
| |
| Keywords |
00:00
Computer-aided designSoftwareComputational physicsScale (map)Real numberVirtual realityDigital signalComputer fileSoftwareComputer fileCombinational logicStandard deviationComputer architectureProcess (computing)FrictionFile formatOpen sourceConstructor (object-oriented programming)Different (Kate Ryan album)Software engineeringTheory of relativityFrustrationWage labour1 (number)Connected spaceCodeDiallyl disulfideProjective planeArithmetic progressionSource codeComputer animation
02:19
SoftwareCollaborationismPay televisionOpen setAutomationBuildingScale (map)Multiplication signConnected spaceProduct (business)DatabaseAdditionProjective planeWeb 2.0CollaborationismInterface (computing)SoftwareGame controllerMobile appTask (computing)Computer fileFile formatCloud computingDifferent (Kate Ryan album)ExergieWeb applicationServer (computing)AutomationComputer animation
04:06
Scripting languageMobile appSoftwareModel theoryComputer architecturePhysical systemEndliche ModelltheorieFile formatPoint cloudComputer animation
04:51
UsabilityModel theoryMobile appSoftwareGeometryInformationBuildingData exchangeAttribute grammarMobile appComputer fileCategory of beingLevel (video gaming)Data structureEndliche ModelltheorieIterationFeedbackComplete metric spaceProjective planeMassDifferent (Kate Ryan album)Cartesian coordinate systemMathematicsRouter (computing)Materialization (paranormal)Uniform resource locatorModel theoryOpen setDiallyl disulfidePolygonPhysical systemWebsiteMathematical analysisSatelliteComputer animation
07:50
Data modelModel theoryAutomationDigital filterRevision controlWeb applicationModel theoryLink (knot theory)Multiplication signDifferenz <Mathematik>File viewerDigital rights managementSoftwareCategory of beingElement (mathematics)Projective planeEndliche ModelltheorieDialectDiscrete groupCollaborationismMoment (mathematics)Different (Kate Ryan album)Overlay-NetzMereologyAnalytic continuationWeb 2.0Revision controlObject (grammar)Group actionSheaf (mathematics)Latent heatCoordinate systemCASE <Informatik>Exception handlingComputer animation
10:12
Continuous functionDisintegrationTelecommunicationAutomationData modelDigital signalScale (map)AutomationComputer architectureMassMathematical analysisFunctional (mathematics)Model theorySimulationComputational fluid dynamicsInstallation artBuildingMobile appRepetitionTask (computing)Beta functionFormal languageLevel (video gaming)Regulator geneSoftwareMathematicsEndliche ModelltheorieOpen setAnalytic continuationContinuous integrationConstructor (object-oriented programming)Computer animation
11:38
Point cloudGoogolSource codeControl flowMobile appObject (grammar)Social classOnline helpTask (computing)Group actionSpacetimeReduction of orderModel theoryComa BerenicesLink (knot theory)Game controllerMobile appPhysical systemDemo (music)Computing platformSoftwareSource codeInternet forumFunctional (mathematics)Compilation albumGoodness of fitTraffic reportingSoftware bugServer (computing)Endliche ModelltheorieLatent heatQuery languageInternet service providerWebsiteCoefficient of determinationMereologyElectronic program guideComputer architectureElement (mathematics)Uniform resource locatorSheaf (mathematics)Cartesian coordinate systemSoftware testingRepresentational state transferBuildingComputer animation
14:18
SoftwareAliasingDigitizingStrategy gameCollaborationismInterface (computing)File viewerScale (map)Task (computing)Focus (optics)Compilation albumJames Waddell Alexander IIDisk read-and-write headKolmogorov complexityImplementationData modelVideo trackingMathematicsReal numberSoftwareTask (computing)Parameter (computer programming)File viewerEndliche ModelltheorieTrailModel theoryFeasibility studyMathematicsCASE <Informatik>Source codeTraffic reportingDifferent (Kate Ryan album)Multiplication signData structureComplex (psychology)Web 2.0Projective planeIterationWorkloadComputer simulationComputer animation
16:25
Open sourceSource codeCodeCASE <Informatik>Observational studyLink (knot theory)Type theoryComputer animation
17:06
System programmingOscillationProgrammable read-only memoryMultiplication signDemo (music)Presentation of a groupObservational studyProjective planeReal numberOpen sourceCASE <Informatik>Error messageTerm (mathematics)CodeComputer animationLecture/ConferenceMeeting/Interview
18:42
Computer-assisted translationComputer animation
Transcript: English(auto-generated)
00:00
So from the title of the talk you can guess I'm going to not talk only about your spatial data, but about the combination of architectural and engineering data as well. And if there is one sentence that I can explain what SPACEL is, because it's a lot of different things, it would be the data hub for architecture, engineering, and construction.
00:21
And even that one sentence is not generic enough, so I had to add a star over there to show that it's also open for any kind of related data in 1D, 2D, or 3D. By the way, my name is Katarina, I'm a software engineer at SPACEL working specifically with the GIS data.
00:41
And to briefly introduce the history of SPACEL, it was born from the PhD research of Dimitri in 2015, and it was born out of the frustration of how inefficient the architectural industry is technology-wise. It was an open source project, it's grown organically over years.
01:01
I myself have discovered it somewhere between 2015 and 2016, when I was trying to connect software between Unity and Grasshopper, and that was literally the only piece of code I could find that was open source. And a couple of years later it got adopted by larger AC companies, and in the middle of the pandemic the two founders had a great idea to start a company out of it,
01:22
and here we are today. We are almost 30 people right now, and if again there is one sentence to describe what is our mission, it's to make the AC better, and I will describe in which ways we are trying to make it better. I will not go as deep to the history as the previous speaker.
01:42
I'll go just 50 years back and show that in fact also the architectural industry has moved from the manual labor and paper-based files and documents to the digital ones. The underlying frictions of the fragmented and disconnected formats is still there,
02:02
and although in the geospatial data there might be more data standards, in AC there are also data standards, but AC is a design process, so most of the data is work in progress, and it's harder to adjust to the standards while you are in the design process basically.
02:21
The AC tech is ages behind, not in the way of what capabilities it's providing us, but in the way of what workflow it opens to the users. It's still very dependent on the fragmented data, so we have different file formats, we have engineering software that have their own formats,
02:41
we have architectural software, we have GIS software, and rarely your teammate or a person from another company, they might not even be able to open your file. Then we have the vendor login, you have the ArcGIS project file, and you won't be able to open it until you purchase a license, and all of this together comes down to the lack of productivity
03:02
and wasting time on coordinating things instead of actually doing useful work. So the three ways we're trying to make AC better is to connect people, projects, and software into the meaningful and seamless workflows. We are also bringing the collaboration interface
03:21
through the web app, which pulls the data directly from the database where you uploaded your data, and it's a central place where you can explore the data, manage, and collaborate with others. And Automate is a recent addition where basically you can run a small repetitive task
03:41
in an automated way on your data. And the underlying thing under all of it is the control that you are in full control of your data. You cannot necessarily rely on the public specular server, you can host your own, you can set up all the permits and access policies that you need, you're flexible to deploy to any cloud provider in any region
04:03
and deploy any custom tools and apps on the top of it. The principle of how Speckle works on the back is taking the data from the host software, whether it's architecture engineering, either the Microsoft Excel or any other app,
04:20
converting it into the data, which is written in a very simple, again, open format, reminding the JSON format, and uploading it to the cloud, where once you publish it, you can access it in any other host software. You can pull it back there, you can pull it to your Python script and run some publishing dashboards,
04:41
or you can access it through the VAP app. And you also can access the versioning system of all the models, basically following the GitHub principle, but for 3D models. So the principles of connecting the data through different workflows, I have tried to pull some of the more
05:01
geospatial related examples. These are the tutorials we have on our website, speckle.systems, and you can find kind of most common or most interesting workflows over there, but it's not limited to these workflows. You can implement your own. And I will show you a couple examples of how to connect from software A to software B.
05:22
We have an architect or designer working with Rhino software and creating some building mass model, sending it to Speckle, and immediately the engineer working in Revit can attach some structural properties, materials, and other attributes to the data. Once the designer has any changes, it can update their model,
05:40
and the structural engineer will immediately receive the updates and have the new model with already attached structural properties. Another example is when you don't need to pull the data to the other host application, but instead you would like to get some live insights from your data. And whenever a stage of completion your project is, you can always access that data from the server
06:04
and either run your own dashboard or, again, use Speckle Connector for something like Power BI and visualize the live insights from there. And basically everything I have just mentioned and a couple more examples of how these data exchanges can be more meaningful
06:21
is, for example, work between the engineer and the rendering person who works in Blender. Then you can do it in a much quicker way. Or the designer and the structural engineer running through different iterations and getting instant feedback on where analyses have failed and what needs to be changed
06:41
without, again, sending files and asking, can you please check? You can also extract the GIS data from QGIS in a very quick way, having just a simple, basically not doing anything, just opening the QGIS and pulling the base layers with a digital elevation model and Google Satellite
07:01
or any other raster and extrude it into 3D. You can also put your OpenStreetMap buildings or any other models, which are simple polygons, then choose the attributes that you want to use for the height of the buildings and extrude them in 3D. Basically what a QGIS 3D viewer already does, but this is actual geometry that can be usable by anyone else.
07:24
And you can use that data to pull it into another design and engineering apps that have, for example, any information about a geographical location, and Revit Software usually has a setting where you specify your geographic location,
07:41
and then you can either write your script or adjust your GIS data to pull the data in a specific location and support your model in the other software. The ways of collaboration. So we have this nice web app with management and permissions and handling all the bureaucratic things,
08:01
but also the most interesting part of it is the web viewer. And the web viewer does not only allow you to explore the data yourself, you can also send a link to your partner and the person will open the link. You will see each other's presence in that model basically
08:20
through the small tags. You can click on the other person tag and you will be following the exact actions and the camera movements that the other person is doing. You can make the sections, you can filter the data, or you can pinpoint specifically what you want. You can leave the comments on the model, you can send attachments to it, and so on.
08:42
You can also combine different models where you have one project and both architect and master planner and engineer are working on it and they're sending their data as different models. You can still use the viewer to combine those different models, overlay them into the same coordinate system
09:01
and hide or show whatever elements you need at that moment. In this example, I added some network analysis that you can run again from another software and upload it to the same project. Except for just exploration of the data, the way the Speckle data is structured
09:21
allows us to easily identify what properties each of the elements have, also see whether it's a discrete values or continuous values, and relative to that, allow you to filter data by those custom properties. In this case, the custom property was temperature
09:41
and I have just filtered the Northern American data to see what regions are more chill in that specific time. And you can also compare the versions because once you are sending the same model over and over again from the software,
10:02
Speckle is tracking what objects have remained the same, has been deleted, changed or added, and this is an easy way to make the visual diff of your data. And the automate, which is again the latest edition, is currently in the open beta stage, so you can try it out.
10:20
It's a CI-CD, continuous integration, continuous delivery principle applied to architecture. So the way we deploy the software, once we push the tag on GitHub and build the installer and deploy it somewhere, the same principle can be applied to architecture. Once you upload your model, you can run the trigger to run the specific analysis
10:40
on the top of it, build you a specific report and notify you in the specific app. And the example of it is to basically automate small or annoying repetitive tasks and run it, for example, on every model change once you need to make sure that whatever you have edited in structural software
11:02
is still compliant with all the regulations. There is plenty of opportunities here. You can run your custom functions in your language of preference. You can see it on the left example with a function that's running computational fluid dynamic simulation on the building mass in models of the specific district.
11:25
And on the right side, you see the automation function that's running the compliance check on the construction model and replacing you the work of, I don't know, hours of work of the engineers. And the control part of it, again, as I mentioned,
11:42
the Speckle servers are provider agnostic and location agnostic. You can host it by yourself. You can host it with us. And you define all the criteria for that. Again, because you are in full control of your data, you know what you have there and what you need out of it.
12:01
You can build your custom apps and the AI application, I had to mention it. And basically here you have the dashboard, which is kind of a repetitive requirement in architecture, the carbon footprint dashboard where every material, every element has specific carbon footprint
12:21
and in the end you need to optimize it. And once you are changing your model, you can have constant insight of how this carbon footprint is changing. You can also use our SDK and C sharp, Python and JavaScript to pull the data, modify the data, analyze the data,
12:41
build new connectors with new software and, again, do whatever you need with the data. You can use the REST API and GraphQL to query the data from the server, again, depending on your permissions. And this is a new compilation, so our community team has shared this with me and they call it the Wall of Love.
13:02
It's basically the good things people say about Speckle, mainly in our community forum and the practical reason behind this is that Speckle has grown so big or so wide in functionality that it's doing so many things and sometimes we can't even tell what is the exact benefit for the users
13:20
that they can take out of it. And the thing that keeps coming up is that people are using Speckle as their model data hub, so to store, modify and use. The Speckle.community is a place to not only leave compliments, but usually it's a place to report bugs, ask questions.
13:41
People are very active there. Some users are helping other users. People are requesting new features, requesting new connectors. People are also developing their own connectors and come to the forum to say, hey, we have built the new connector. You can try it out. These are other resources. The first one, app.speckle.system, is an actual platform where you can register,
14:00
get immediate demo with some test model and explore how it works. Speckle.guide is a documentation when nobody reads, but they should. Speckle.system tutorials, that's the main website and one of the sections is the tutorials and the last one is community, which I have already mentioned.
14:20
And I will just show briefly how real people and real companies are using it. So one was the hackathon project that Speckle organized earlier this year and we had a team who collaborated across four disciplines, five softwares. They were from four different companies in five different countries and they had used geotechnical data, ground investigation data,
14:42
underground structures to find the optimal path for the underground tunnel and run a feasibility study on it. We had another company who summarized their experience with Speckle as it removes the complexity of advanced workloads and help them focus on the task
15:01
rather than on handling the tools themselves. And what they did is they hacked the web viewer instead of allowing it to just view the data. They gave the access to the engineers to change the parameters of the model that designers have uploaded and run the sketch of what it will be
15:24
instead of giving the engineers the source file with beam data and have a risk of them changing something irreversibly. Another example was the geotechnical engineering project in New Zealand with a lot of seismic activity where there is plenty of iterations of how designers are working with engineers
15:42
and making simulations and their conclusion was that Speckle allowed them to work each person or each professional in their own software that they are most comfortable and not focus on the differences in those softwares. And the last one is my favorite, it's saving time.
16:02
And in this case, the company had to run the live monitoring on the construction site and they had people updating the data in different ways from the Excel sheet through the 3D models to the other reports and they had built a custom dashboard to help them to monitor it and they also took advantage of the 3D change tracking
16:22
with the 3D model in the web viewer. And again, this is open source. You can always introduce yourself in the community. You can go to the GitHub, you can see the codes, you can add your pull requests and what I will ask you to do, please everyone, to join the community, this link will lead you
16:42
directly to the community post saying please introduce yourself and you can just write one sentence of who you are and what you do and what do you do with geospatial data or other types of data. And in this way, we can stay in touch and if you have interesting case study for us, you can share.
17:01
If we have interesting case study for you, we also can reach out to you. So yeah, thank you very much. Thank you, Katerina. Any questions?
17:29
Yeah, thanks for the presentation. Just the examples and the documentation are really good. Compared to most open source projects, getting people to do examples and documentation
17:41
is a real issue. So I'm just wondering how people found time to work and make it look so good for the demos. How people who have reported their case studies had time to make it look good?
18:04
The documentation regarding the code is mainly us because we are responsible for people who read the code and if I find any errors, the tutorials are also written mainly by our team. The case studies are the companies that we have reached out to share their experience
18:22
and they found the time to do that. Other questions? Yeah, I think that's all. Thank you, Katerina.