WEBVTT 00:00.000 --> 00:11.480 So, today I am here to talk about interactive and collaborative client-side GIS and Jupyter 00:11.480 --> 00:12.480 Lab. 00:12.480 --> 00:21.920 So, I will discuss the problem statement first and what we are trying to solve and then 00:21.920 --> 00:25.320 walk through the architecture and some concrete workflows. 00:26.200 --> 00:28.920 So, let us first discuss about the current model. 00:28.920 --> 00:36.920 So, many geospital workflows still assume desktop first local single use single user environment. 00:36.920 --> 00:43.000 So, these two tools are most prominent tools in the ecosystem right now and both of these 00:43.000 --> 00:49.320 tools are so powerful and so efficient for use for single user to use. 00:49.320 --> 00:53.360 But with the power they come with a cost as well. 00:54.360 --> 00:56.360 GIS lives in a local machine. 00:56.360 --> 01:01.360 Users need to handle installation dependencies and configuration themselves manually. 01:01.360 --> 01:07.360 And this adds a lot of friction while sharing workflows and reproducing the same. 01:07.360 --> 01:11.360 And while teaching new users. 01:11.360 --> 01:15.360 These tools are driven through the rich desktop interfaces. 01:15.360 --> 01:19.360 These rich desktop interfaces are very powerful and very efficient. 01:19.360 --> 01:24.160 But they are quite overwhelming when you are a beginner and a new user getting started 01:24.160 --> 01:25.400 with these tools. 01:25.400 --> 01:34.360 You can see that there are a couple of tool bars and some nice have to populate it panels. 01:34.360 --> 01:38.600 And collaboration still relies on sharing files manually which creates a lot of friction 01:38.600 --> 01:43.920 because files are still shared by emails, get out drives and they are difficult to iterate 01:43.920 --> 01:47.960 when multiple users are working on them and when you are working on something that requires 01:47.960 --> 01:49.960 multiple people to work on it. 01:49.960 --> 01:53.960 It creates a lot of friction and back-and-forth. 01:53.960 --> 01:58.520 And consider I am working on an environment where my computation environment and my visualization 01:58.520 --> 01:59.960 environment are separate. 01:59.960 --> 02:05.960 So, then I need to do a lot of back-and-forth to between my visualization tool and my computation tool. 02:05.960 --> 02:11.960 Such as when I am using geo pandas in a python environment and I am using this. 02:11.960 --> 02:14.960 I am using QGS to visualize the same. 02:14.960 --> 02:20.960 This separation slows down the workflows and makes reproducibility very hard. 02:20.960 --> 02:23.960 So, here comes Jupiter into the picture. 02:23.960 --> 02:27.960 Jupiter already changed how we work with data. 02:27.960 --> 02:32.960 And as we know, notebooks are very reproducible and computational documents. 02:32.960 --> 02:36.960 All your code, data outputs and narrative all is in one place. 02:36.960 --> 02:41.960 So, you can share your story with a single notebook with a single document. 02:41.960 --> 02:46.960 Notebooks are very easily shareable and reproducible from one user to another. 02:46.960 --> 02:53.960 And they run on your desktop or on your browser as well as per your liking. 02:53.960 --> 02:58.960 It creates a seamless environment where the computation and its story lives together, making 02:58.960 --> 03:05.960 analyzes and analysis easier and to reproduce and analysis easier to reproduce and understand. 03:06.960 --> 03:09.960 Notebooks are not only reproducible. 03:09.960 --> 03:12.960 They are also collaborative thanks to Jupiter collaboration. 03:12.960 --> 03:17.960 Jupiter collaboration enables live multi-user editing and the real-time collaboration 03:17.960 --> 03:23.960 inside your notebook and it files format called .JGIS. 03:23.960 --> 03:29.960 It is a JSON based representation of the entire GIS state that we represent on the map. 03:29.960 --> 03:36.960 It is a fully contained file structure that contains your layers, your symbology, your collaborative 03:36.960 --> 03:38.960 metadata and your user config. 03:38.960 --> 03:44.960 Layers are live, editable objects and symbology updates instantly because it is collaborative 03:44.960 --> 03:52.960 and analysis can run directly inside the notebook. 03:52.960 --> 03:58.960 So, let us understand a bit how Jupiter GIS works and let me give you a brief overview of the 03:58.960 --> 04:01.960 architecture that we have on Jupiter GIS. 04:01.960 --> 04:07.960 So, we merely divide Jupiter GIS into three components, first of which is rendering. 04:07.960 --> 04:14.960 So, for rendering we use open layers that enables us to dynamically render vector and 04:14.960 --> 04:18.960 raster data and also to dynamically style style it. 04:18.960 --> 04:23.960 And for the second, secondly we have analysis component for the analysis path we use 04:24.960 --> 04:26.960 gdile compiled to web assembly. 04:26.960 --> 04:32.960 It also enables us to do a lot of things inside the browser itself and some of the examples 04:32.960 --> 04:38.960 are that we can do processing right inside the browser such as buffer, dissolve, rasterization 04:38.960 --> 04:39.960 etc. 04:39.960 --> 04:43.960 And third and the most important component is obviously collaboration. 04:43.960 --> 04:50.960 We use YJC addities to enable live multi-user editing without conflicts and live 04:50.960 --> 04:55.960 synchronization between multiple user sitting different parts of the world. 04:55.960 --> 05:01.960 And in all of this that dot gdice file works as a single source of truth for the entire 05:01.960 --> 05:04.960 GIS state that you are working in. 05:04.960 --> 05:13.960 So, now that we know how Jupiter GIS works lets dive a little bit into the demo and see 05:13.960 --> 05:19.960 how it works and what other things that we can do with Jupiter GIS. 05:19.960 --> 05:34.960 So, when you would open Jupiter GIS in your just give me a second. 05:34.960 --> 05:43.960 So, when you would open Jupiter GIS you would see a window like this and here is your 05:43.960 --> 05:49.960 Saturn to create a GIS project by clicking on which your GIS project will create and you 05:49.960 --> 05:56.960 would see a nice UI with a single toolbar that is not too overwhelming with some like you 05:56.960 --> 06:03.960 can see that these panels are not those populated like not that populated as I shown previously. 06:03.960 --> 06:09.960 So, in this toolbar we have some buttons I will later explain in the demo what we can do 06:09.960 --> 06:10.960 with these buttons. 06:10.960 --> 06:15.960 But, for now we can look into the layer browser with the help of layer browser you have 06:15.960 --> 06:18.960 many example base maps that you can add to your map. 06:18.960 --> 06:27.960 I will just add open sheet map and see how it looks. 06:27.960 --> 06:34.960 Next we have so, Jupiter GIS comes with vector and like vector and raster data support. 06:35.960 --> 06:41.960 You can directly load your vector and raster data either it is either it is saved locally 06:41.960 --> 06:50.960 or it is served on cloud. 06:50.960 --> 06:55.960 Next so, I want to show sphere. 06:55.960 --> 07:01.960 So, in this visualization so, this is live demo working inside the presentation only. 07:01.960 --> 07:08.960 It is I frame I frame so, these blue dots are vector data representing some vector data 07:08.960 --> 07:15.960 and this green layer you see is the raster data that I can toggle it on and off I can just 07:15.960 --> 07:27.960 see the vector data if I want I can zoom in and same I can do with the raster data as well. 07:28.960 --> 07:32.960 So, next we have so, we also support because we use open layers. 07:32.960 --> 07:36.960 We also support advanced symbology with vector data. 07:36.960 --> 07:38.960 So, I would like to show you how it works. 07:38.960 --> 07:44.960 So, I would select this layer and right click on it and you see this option to add it 07:44.960 --> 07:50.960 symbology I would select this and I would select a symbology style based upon my like 07:51.960 --> 07:54.960 I would select graduated. 07:54.960 --> 08:01.960 The value for which I want to classify my symbology would be magnitude of earthquake 08:01.960 --> 08:06.960 and I would select a nice color them based upon my liking may be select cool 08:06.960 --> 08:11.960 and I would classify it would automatically classify your data based upon the ranges of 08:11.960 --> 08:16.960 magnitude like for instance this is magnitude and I would apply this symbology and you 08:16.960 --> 08:25.960 can see my layer has been updated with the nice symbology. 08:25.960 --> 08:30.960 Exactly similar to vector data you can also style raster data as well. 08:30.960 --> 08:35.960 You can do single bands symbology and multi band rendering or two raster data. 08:35.960 --> 08:42.960 So, similarly I would select my layer and I would do the same to add it symbology it 08:42.960 --> 08:43.960 would open up our dialog box. 08:43.960 --> 08:49.960 This time the options are a little different because obviously it is raster data raster 08:49.960 --> 08:50.960 data. 08:50.960 --> 08:57.960 So, I would let it be single band symbology and I would select a color ramp maybe you 08:57.960 --> 09:04.960 use bring this time I would apply it it takes a little bit time because it is cloud 09:04.960 --> 09:16.960 optimized you to but yeah you can see that the symbology has been updated. 09:16.960 --> 09:23.960 So, now I want to show you like that we can do dynamic layer management and you like in 09:23.960 --> 09:30.960 jupiter GIS just similar to QGIS you can reorder layers like by dragging them up and down 09:30.960 --> 09:40.960 I just did and as I shown you you can toggle a layer on and off like this you can 09:40.960 --> 09:48.960 if I just want to see one layer I can do that I can even remove a layer like this I can even 09:48.960 --> 09:58.960 rename a layer and if I and there are several options that I can do I can zoom to a layer 09:58.960 --> 10:07.960 and yeah so this is like this is like we are layer browser is also like the layer panel 10:07.960 --> 10:12.960 is also inspired by QGIS similar to QGIS we can do a lot of things with this. 10:12.960 --> 10:19.960 So, we also have identify panel you can select and click on features to which you want 10:19.960 --> 10:26.960 to see properties and attributes of so I would show you how it works so I would select 10:26.960 --> 10:33.960 the layer again and I would enable the identify panel and then I would select some attributes 10:33.960 --> 10:41.960 over here it would show me which attributes I have selected it would give me a list of all 10:41.960 --> 10:47.960 attributes that are selected I have this nice search button which I can use to highlight the 10:47.960 --> 10:59.960 attribute which I want to which I want to focus on I can do the same for others it also goes 10:59.960 --> 11:06.960 with a nice zoom animation and I can see I can do it with all the type of vector data 11:06.960 --> 11:10.960 I can do it with a polygon I can do it with a line and I can do it with a point as well 11:10.960 --> 11:16.960 so I would select some polygons over here so it gives me all the list of all the selected 11:16.960 --> 11:22.960 polygons and we also have a search button over here so I can use it to highlight my 11:22.960 --> 11:30.960 attribute that I am looking at I can study and see all the things that this feature 11:30.960 --> 11:38.960 have next I have next we have time slider so this is some fancy stuff that that is also 11:38.960 --> 11:45.960 that Jupiter GIS is also enabling so we can visualize data with respect to time so I 11:45.960 --> 11:52.960 would take this for example I would enable time slider and then we have this nice menu 11:52.960 --> 11:58.960 over here which has this slider I can move this slider to see how my data changes over time 11:58.960 --> 12:09.960 so I would I can maybe also change the time interval that we are rendering the data 12:09.960 --> 12:14.960 for instead of week I would select day then I can see how it changes over time 12:14.960 --> 12:21.960 and if you and if you do not want to move this slider we have this nice play and pause 12:21.960 --> 12:30.960 button you can use this as well to automatically see how your data transforms over time 12:30.960 --> 12:40.960 I would may pause it and next so people who are already familiar with Q GIS and who are already using Q GIS 12:40.960 --> 12:48.960 and do not want to like directly jump into Jupiter GIS we also have Q GIS compatibility so you can 12:48.960 --> 12:56.960 export dot GIS file to Q GIS format and you can even import Q GIS format to Jupiter GIS directly 12:56.960 --> 13:03.960 although we do not claim that we support everything that Q GIS have but we are still in process to 13:03.960 --> 13:12.960 support all the all the things that Q GIS provides but yeah it works nicely with it works nicely with 13:12.960 --> 13:23.960 exporting styles and all the properties that your map have so next we have annotation and 13:23.960 --> 13:28.960 collaboration like I have already talked a lot about collaboration and already emphasized a lot 13:28.960 --> 13:36.960 now you can see how it works so you consider that there are two different users working on this project file 13:36.960 --> 13:42.960 the one on the left and one on the right because there there are these are two different browsers so 13:42.960 --> 13:48.960 if I change the map on the left you can already see that it automatically changes in the right as well 13:48.960 --> 13:58.960 and I can also like talk on the map I can annotate I can make comment I can reply to those comments and I can see what's going on 13:58.960 --> 14:04.960 this makes like obviously this makes collaborative work flows much smoother than file based 14:04.960 --> 14:15.960 shading so we also have Python API for Jupiter GIS that lets you interact with your GIS document 14:15.960 --> 14:21.960 programmatically you can use it to create layers and perform operations right inside your notebook 14:21.960 --> 14:29.960 Python API works same as same as a collaborative working on your file so it is also like real time collaborative 14:29.960 --> 14:36.960 and everything that you do with Python API as well they this show up on the map to you and to the other user that 14:36.960 --> 14:42.960 is working with you on the same file. 14:42.960 --> 14:48.960 We also have client site processing so I in between of the presentation I talked about that we use 14:49.960 --> 14:55.960 to perform operations inside the browser so everything that I will show you here would be 14:55.960 --> 15:03.960 happening inside the browser I am not using any server so I will select a layer and we have this processing option 15:03.960 --> 15:11.960 in which we have some option that what we can do with our data so this is the earthquake data that you are seeing 15:11.960 --> 15:18.960 so maybe I will perform buffer to see what is the risk risk area of these earthquakes 15:18.960 --> 15:30.960 I will give it 10 units that lets consider that everything in the radius of 10 is in the risk of these earthquakes 15:30.960 --> 15:45.960 so I will just give it another name at quicks buffered and I would yeah so you can see that 15:45.960 --> 15:53.960 our new layer has been created which is a polygon layer it uses earthquakes as a base to calculate 15:53.960 --> 15:59.960 what is the risk area of all these earthquakes and you can do a lot more with the processing options 15:59.960 --> 16:09.960 processing menu that we have and we are still extending that menu with time. 16:09.960 --> 16:16.960 So we also have stack browser so Jupyter GIS stack browser lets you connect to your stack catalogs 16:16.960 --> 16:23.960 and load satellite imagery right inside your notebook or geospatial data directly into 16:23.960 --> 16:30.960 Jupyter GIS you can use it to load any satellite imagery that is available in your stack 16:30.960 --> 16:44.960 API just like this. 16:44.960 --> 16:50.960 And we also have a detailed integration so a detailed it is a nice tool a detailed 16:50.960 --> 16:57.960 let's you serve and explore like big vector and asset data sets dynamically using Python plus 16:57.960 --> 17:05.960 T Tyler this lets you explore large data sets interactively although this requires a server 17:05.960 --> 17:13.960 that's why we cannot yet do it in Jupyter light because it requires a kernel component but yes 17:13.960 --> 17:30.960 if you want to explore large data sets with Jupyter GIS, T Tyler is the answer for you 17:30.960 --> 17:35.960 So there are some in progress things that we are working on one of these is story maps 17:35.960 --> 17:41.960 so with the help of Jupyter GIS story maps you can tell I story with your map 17:41.960 --> 17:44.960 you can like you can present your map like just like a presentation. 17:44.960 --> 17:51.960 So story maps are approach to turn GIS into narratives and text combines with interactive 17:51.960 --> 17:57.960 stories live live nice text for teaching communication or decision making. 17:57.960 --> 18:03.960 This is like this looks good it is still in progress but yeah it is close to being 18:03.960 --> 18:11.960 completion so in next release maybe you would be able to use it. 18:11.960 --> 18:19.960 Next we are also exploring integrating the power of LLMs and agents within Jupyter GIS 18:19.960 --> 18:27.960 we are exploring natural language interaction with GIS so that if you are like if you are a very 18:27.960 --> 18:32.960 beginner if you do not want to know like if you do not want to learn GIS concepts if you just 18:32.960 --> 18:36.960 want to do something with GIS you can just interact with Jupyter GIS in natural language 18:36.960 --> 18:42.960 and ask questions manipulate data or analyze maps by simply typing instructions in your 18:42.960 --> 18:50.960 natural language and yeah obviously no prior GIS knowledge required. 18:50.960 --> 18:56.960 So yeah that was mostly about the demo yeah maybe you would want to see. 18:56.960 --> 19:02.960 In the left I am interacting with an LLM I am telling it to add a new layer although 19:02.960 --> 19:07.960 I because I am using an old Gemini model so that is why I need to confirm it every 19:07.960 --> 19:13.960 time so it adds layer automatically and then I would tell it to create a buffer of the same 19:13.960 --> 19:19.960 like the same I showed you with processing manually this time it would be happening automatically 19:19.960 --> 19:28.960 in the LLM. 19:28.960 --> 19:39.960 Yeah I need to confirm it every time because it is acting. 19:39.960 --> 19:48.960 So again I again give it 10 units and yeah this is what that results look like. 19:48.960 --> 19:56.960 It is exactly same to what I showed you manually but you have this time with an LLM. 19:56.960 --> 20:03.960 So this was all about the demo but all the live demos that you saw today were entirely 20:03.960 --> 20:07.960 made within the browser and with the help of Jupyter light. 20:07.960 --> 20:14.960 So it can be it works within fully Jupyter light setup and it can be embedded in your 20:14.960 --> 20:15.960 frames. 20:15.960 --> 20:21.960 No installation required python runs in the client as well because we use use python kernel 20:21.960 --> 20:26.960 and G dial is compiled to web assembly so still we do not require as a component there. 20:26.960 --> 20:31.960 And even the full Jupyter environment works inside your browser this makes it like very easy 20:31.960 --> 20:35.960 for anyone to try or share Jupyter GIS instantly. 20:35.960 --> 20:44.960 So next if you want to like try out Jupyter GIS right now this is the easiest way that I can 20:44.960 --> 20:45.960 think of. 20:45.960 --> 20:51.960 So for anyone who wants to try it notebook link make it image it. 20:51.960 --> 20:57.960 You can open Jupyter GIS repository in your browser and just prefix the link with notebook 20:57.960 --> 21:00.960 dot link and you have the full environment ready. 21:00.960 --> 21:04.960 So no set up no installs just start exploring GIS in your browser. 21:04.960 --> 21:10.960 So I would just prefix the URL and I would accept the terms and it would open Jupyter GIS 21:10.960 --> 21:12.960 inside my browser. 21:12.960 --> 21:24.960 And I can do everything that I show you. 21:24.960 --> 21:31.960 So now that you have seen that what Jupyter GIS can do next we can discuss what you can do 21:31.960 --> 21:37.960 with Jupyter GIS there there is a lot that you can do with Jupyter GIS obviously but 21:37.960 --> 21:42.960 there are some main ways which people can use Jupyter GIS for obviously for environment 21:42.960 --> 21:49.960 and environmental analysis you can combine climate land use and biodiversity mapping data 21:49.960 --> 21:54.960 to do environmental analysis and secondly we have earth observation we already have 21:54.960 --> 22:00.960 stack browsers you can load satellite clusters and you can do times time forecasting like 22:00.960 --> 22:03.960 you can visualize time forecasting data as well. 22:03.960 --> 22:07.960 So yeah it helps you to do earth observation as well. 22:07.960 --> 22:15.960 And thirdly let's you do urban planning as well because you can load infrastructure data 22:16.960 --> 22:21.960 and what not and obviously for teaching and workshop because the installation and 22:21.960 --> 22:24.960 like the reproducibility of this tool is very easy. 22:24.960 --> 22:28.960 It makes teaching and workshops really easy then compared to these traditional tools 22:28.960 --> 22:33.960 which require big installations and hefty setups. 22:33.960 --> 22:39.960 And next we have reproducibility search obviously I have emphasized on it a lot that it is 22:40.960 --> 22:46.960 really easy to reproduce and again collaborative mapping you can collaborate 22:46.960 --> 22:55.960 with a lot of people multiple people at the same time and you can communicate what 22:55.960 --> 23:01.960 you are seeing right in real time. 23:01.960 --> 23:05.960 So what next for Jupyter GIS? 23:06.960 --> 23:10.960 So looking at there is a lot that we want to do with Jupyter GIS one of which is 23:10.960 --> 23:12.960 suggestion support. 23:12.960 --> 23:18.960 So suggestion suggestion support lets you do a split view comparison when you collaborate 23:18.960 --> 23:20.960 in real time. 23:20.960 --> 23:26.960 It lets you create a precise suggestion to the other user that you are working with 23:26.960 --> 23:32.960 that makes the chance of confusion very minute very little. 23:32.960 --> 23:37.960 So it is something that already works in the sister project of Jupyter GIS which is 23:37.960 --> 23:43.960 Jupyter CAD it is the same like it uses the same infrastructure the same backend 23:43.960 --> 23:49.960 but for instead of GIS for CAD for 3D modeling and all. 23:49.960 --> 23:53.960 So we want to also make it work in Jupyter GIS as well. 23:53.960 --> 23:57.960 We would also like to do more cloud optimization support. 23:57.960 --> 24:04.960 So we would also want to extend our five support to Geoparkage Euro so that we can load 24:04.960 --> 24:12.960 big data very seamlessly in Jupyter GIS and we are also working on vector layer editing 24:12.960 --> 24:17.960 and creation so that you can create new vector layers in real time and they are also 24:17.960 --> 24:22.960 collaborative and you can edit the existing vector layers. 24:23.960 --> 24:28.960 And next we have advanced table viewer just like the attribute table in QRIS you would also 24:28.960 --> 24:33.960 like to have some we are working on this as well. 24:33.960 --> 24:38.960 We would also like to have some advanced table viewer for Jupyter GIS with which you 24:38.960 --> 24:45.960 can do analysis you can export data and all. 24:45.960 --> 24:51.960 So here I would like to acknowledge the support that Jupyter GIS has gained over time. 24:51.960 --> 24:58.960 So yes funded the initial development of Jupyter GIS under permanent open call with 24:58.960 --> 25:06.960 for EU innovation and stack and XRA extension had been funded by Venice. 25:06.960 --> 25:12.960 It Jupyter GIS has been built by Constack in collaboration with similar research lab with 25:12.960 --> 25:20.960 major contributors from UC Berkeley DSE and over 30 open source contributors from all around the world. 25:20.960 --> 25:29.960 So if you want to get involved we welcome contributions of all kinds you can contribute code, 25:29.960 --> 25:35.960 bug fixes or develop plugins you can help with documentation tutorials or example 25:35.960 --> 25:38.960 notebooks and your feedback is very crucial for us. 25:38.960 --> 25:46.960 It lets us prioritize things and this is how the community keeps growing and if you want 25:46.960 --> 25:53.960 to join Jupyter GIS community check out geojupyter.org we have bi-weekly we organize bi-weekly 25:53.960 --> 26:01.960 hackathons and we also have a geolip channel which you can join and interact with us. 26:01.960 --> 26:03.960 And thank you so much. 26:03.960 --> 26:11.960 These are QR codes that you can scan for the live deployment of Jupyter CAD and Jupyter GIS. 26:11.960 --> 26:21.960 And the questions. 26:21.960 --> 26:36.960 No problem if there's no questions. 26:36.960 --> 26:43.960 Thanks, that looks very fantastic. 26:43.960 --> 26:50.960 Maybe you can comment like why did you choose open layers as the map? 26:50.960 --> 26:57.960 So it has been like it has been quite directos that we have discussed a lot about this. 26:57.960 --> 27:05.960 We in past we have been using map library although we wanted to enable advanced styling 27:05.960 --> 27:08.960 for vector layers and oscillators. 27:08.960 --> 27:14.960 That's when we realized that we have hit a ceiling over there. 27:14.960 --> 27:20.960 But we are also exploring that we are not just settling to open layers. 27:20.960 --> 27:27.960 We are still exploring to have custom like you can plug in any viewport and customizable viewport 27:27.960 --> 27:31.960 you can use any other viewport we are exploring that feature as well. 27:31.960 --> 27:37.960 And maybe in future we would also like to support 3D rendering for our maps. 27:37.960 --> 27:48.960 So we would also like to extend the view like to explore that direction as well. 27:48.960 --> 27:53.960 I saw in your one of your timers that with a real time element. 27:53.960 --> 27:56.960 Does that require some sort of backhand of some point? 27:56.960 --> 27:57.960 Yes. 27:57.960 --> 28:03.960 So Jupyter collaboration yet does not work in Jupyter light because it requires us 28:03.960 --> 28:08.960 I have a component it requires some sockets and stuff. 28:08.960 --> 28:15.960 But there has been work going on in order to support collaboration inside Jupyter light 28:15.960 --> 28:18.960 in our team. 28:18.960 --> 28:25.960 Obviously we cannot do that yet but maybe in the near future we would be able to collaborate 28:25.960 --> 28:28.960 without a server component as well. 28:28.960 --> 28:29.960 Okay, great. 28:29.960 --> 28:30.960 Thank you.