Microsoft 365 Community Graph, 11 Mar 2020

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Microsoft Graph developer community call-March 2020

Description

Hosted by Jeremy Thake, this month's agenda included:
-TLS 1.0/1.1 Deprecation for change notification subscriptions (1.02)
-GetResourceSubscriptionPath (4:45)
-Proposal for Versioning & API Evolution in Microsoft Graph (11:30/43:28)
-Partner Demo – Klynke.com (24:32)

Resource links:
News on TLS Deprecation:
https://developer.microsoft.com/en-us/office/blogs/microsoft-graph-subscriptions-deprecating-tls-1-0-and-1-1/?wt.mc_id=YT_CCrecording
GetResourceSubscriptionPath:
https://github.com/microsoftgraph/msgraph-sdk-dotnet-contrib
View the presentation slides: https://www.slideshare.net/OfficeDev/microsoft-graph-developer-community-callmarch-2020
Microsoft Graph documentation:
https://docs.microsoft.com/en-us/graph?wt.mc_id=YT_CCrecording

A

Thank you for attending the Microsoft graph community call for March. We have three speakers this month. We were really striving to try and get external speakers in each of these course to talk about what they did with the graph and what they've been focusing on with a graph and some of the challenges and pains they've had with a graph they can share. So you can learn from that as well and obviously give us inside of the product seeing the feedback.

A

So today we have partner, as your mom overseas, that's going to talk, but we're gonna first have Vincent Bray talking and then pipe hither from the identity side of the graph services team, both fairly notes of the p.m. world. You might be familiar with Vincent from being inside of the kind of the SPF X SharePoint e well for a long long time as well, and so it's nice to have you on the inside Vincent I. Will let Vincent jump straight in to talk about his stuff, because I am 3/4 of time this week.

A

With this call hi.

B

Everyone thank you for having me I'm gonna, be talking about two things today, the first one really quick. We are deprecating TLS, 1.0 and 1.1 for change. Notifications subscriptions aka what hooks across Microsoft graph. What's changing well starting May, 15th change. Notifications won't be delivered anymore, other Challis, 1.0 or 1.0. Why are we doing that? Well, a couple of months ago now we there was a few exploits that came out about TLS 1.0 1.1, allowing attackers to compromise the privacy material data over TLS.

B

So the industry has been since then gradually moving over from chance, 1.0 and 1.1 to TLS 1.2. How do you know if it affects you? Well, first, the question you must ask yourselves are easy: are you using web hooks with Microsoft craft today and is your notification URL not supporting. Tell us one point you if this is the case when you need to take action and we'll see what actions you can take, if you say no to both of those questions, you don't need to do anything at this point.

B

So, instead of just pulling the plug May 15th and saying hey, we're not delivering change notifications anymore, other choice, one point: zero and one point one: we are introducing a gradual rollout of this change, starting February 17th. It was a couple of weeks ago, new subscriptions creations, so post must either upgrade to TLS 1.2 if you're not reporting it already or include an additional property. But I'll explain a little bit later on. The next change is marked. 16Th post request must now great to see us 1.2.

B

There is no other choice and patch request must either so renewing the subscriptions existing subscriptions receiver upgrade to TLS 1.2 or include the additional property I'll be talking about just a little bit starting April 15th batch request must also upgrade to TOS 1.2 and lastly, May 15 over TLS, 1.0 and 1.1 will not be delivered anymore. Your notification URL must support LS 1.2 to keep receiving notifications when I was talking about is new property. We are introducing a new property called latest supported, TLS version.

B

This property is optional and allows you to quote unquote buy more time when earlier I was showing the calendar to say you must either implement TS 1.2 or provide that that new property, if you cannot support TLS 1.1, just yeah HS 1.2 just yet, you can set up this property to V 1.1 for some time, so you can keep creating and renewing subscriptions until we require you to upgrade to tell us 1.2, so this property allows you to buy more time on the subscriptions.

B

So if you want to have more information about our plans and a detailed information about, why are we duplicated? Tell us one point: zero and one point one: what can you do? What are the different alternatives and things you need to do? You can go to aka.ms/offweb books, TLS the link displayed on the screen and with that, if there are no other questions, I'll move to the second part.

B

Let's jump right away to the demonstration, so here I'm going to be showing code about creating subscriptions. Usually, when you create subscriptions into the Microsoft graph and I'm going to make it bigger, you create a new subscription object here, you say: hey I want to be notified about updating or created or deleted objects, and so on and so forth, and then you provide a resource path. All of that is works very well. However, there's reason why strings in ideas are often color-coded to red. This is because strings are usually error-prone.

B

You can even miss typos, not validated at compile time and that can lead to mistakes and code crashing in production right. So you want to avoid as much as possible to have hard-coded strings. On top of, that is not very obvious to know exactly what you need to put here. Should you include your version or not? What is the exact path you should put here? You can include trailing slash, it's not very obvious. Of course, all of that is detailed in the documentation, but that kind of pulls you out of your context.

B

You have to go, see the documentation and read up to understand what exactly has that resource path is accepting here so working with the Microsoft graph community contribution library and I'll switch over to my browser for a second I included, a new extension method that I'm going to show you in a few seconds. These Microsoft graph, SDK dotnet contribu, is a library that allows the community to augment Microsoft graph SDK without net decay, with new things that could be useful. That could save you time and so on and so forth.

B

Every effort is led by polish chef Lynn and if you have our community contributors and of course we do accept to pull request, and if you have ideas you can also discuss that in the issues and so on and so forth. Now what benefits do we get? Well? The first thing I'm going to do is here in dependencies I'm, going to add the Nugget package, which already did before we call to save as time, and you can see that here in installed. I have this new library called granted community.

B

This is exactly what's here: I installed for this demonstration of the 1.20 1.0 preview 1. It's really important because right now the method I'm going to be showing you is still in preview, but it will be released as part of the 1.22 later during the next few weeks, once I've done that the next thing I'm going to show you here is what can I do with this.

B

So, instead of hard-coding that path, I can use the Microsoft graph SDK fluent API, plus that new method to get the resource subscription path for me and so I don't have to hard code. That path so see here, I'm using my regular graph service, client, dot users typing in the user, and you can imagine that coming from variable coming from your code and other things like that, this is just for demo purposes here, dot events, and then this is the new method we're talking about get resource subscription path.

B

So that is going to give you the exact same string I had earlier without having without requiring you to look in the docs or maybe exposing you to any kind of typing error, or anything like that right. So now, if we debug that code loading loading loading.

B

We have breakpoints, we have everything, so we can see that I already I already started and rocked before. So we save time I'm going to create my subscription here.

B

Hopefully, in my breakpoint will break there, we go so now. If we look at the subscription object here by doing shift f9 in Visual Studio, we can see that that subscription object has resource path here that is automatically set. We did not have to hard code that resource path. It has been provided by this new extension method here, which takes advantage of a fluent API, all the Microsoft graph SDK.

B

So now I'm just going to go with five and remove the breakpoint, so that is going to create my new subscription and just to show you it's actually working and I'm, not cheating here I'm going to create a subscribe. So in case you didn't see it I subscribe to the events on this users calendar, so any new or updated events will trigger a notification to be delivered. So here I am going to add a new event.

B

And if we do that, hopefully there fairly soon, we should get a notification which we just received and if we look at the notification we should get. The information of this new event has been created. Here is the idea of event, and now I could write some code to Google get that event and do whatever I need to do with this processing to process that unit it's to f5. So again, this handy little method from the graph community contrabass decay will help you take advantage of the fluent api to get the resource path.

B

When you create new subscriptions on the Microsoft graph and with that, if there are any other questions, I'll take questions and then I will hand it back to Jeremy.

A

Thanks Vincent.

B

Thank you Paul for writing. The URL of the repository in the chat.

A

Yeah and support just saying if there any gaps at SDK feel free to create an issue Darryl and the dotnet team, actually look at the contract thing and there's going to be kind of a little bit of a journey as we kind of do. The thank you for it makes sense to pull into the core library will do it.

A

Otherwise it will stay in the contract library, but if there are things that you're doing that you're adding to kind of decorate the current SDK to be great to understand how you're doing that and to contribute back to that repo. So thanks for contributing Vincent and obviously thanks to Paul for driving that effort in that repo as well. So next up, we have our Peter to talk about all the deprecation plans and proposals. We have so I'm going to hand it straight to you up if oh great.

C

Thanks reading and thanks insect yeah, well I'm just going to share my screen greetings, everyone hope you're. Also staying safe in the global epidemic worries that's happening right now, but my name is Arpita. I am a senior PM on the graph core infrastructure services team on the other side of the organization. Today, I'm going to present on a proposal for API will usually in Wersching strategy within Microsoft grass. Does this topic I'm mostly hoping to get a lot of input, sound and discussion going forward?

C

This has been something that's that we've been showcasing with multiple forums, internal and external for the past four and a half to two months, receiving feedback and evolving as we go forward. But the key thing here is to understand: what is how do we wall, graphs and one of the challenges has always been.

C

As you all know, Microsoft graph has been there for almost 45 years and the core policy states that we will introduce a new major version in graph and and we'll have a close to two years of support, post interaction of a new version which would enable our developers to have enough time to to move into the next major version.

C

All of this great great introduction and but we've also been hearing both from the products or the workflows under made graph so-called work loads, which are necessarily the products underneath graph as to what is the next step that we do to make changes that enable our customers to to get the right value for each of these products, for example, SharePoint exchange and at the same time multiple developers have expressed their questions or concern. Sir hey, how do we be prepared for what's gonna change in graphs like given that serpent everything else?

C

So this has been an ongoing conversation and and as you can see, the goal here is to to make sure that we balanced their application developer experience. That's your experience with the agility of changing the product. We want to make sure that we provide you the best exposure and experience to the products underneath crash, but at the same time ensure that you're getting the best developer experience.

C

So the goal of this particular API evolution strategy is how do we fix the things that are coming currently broken in Grasse and evolve, rafts and add new features, but at the same time, like I, said, provide enough best developer experience to ensure that you have good head, sir good signaling and, at the same time, you're not caught by surprise if your applications are not working, but basically we want to make sure that we walk current in hand with you and make sure your timelines are aligned with your release. Timelines are aligned with.

C

However, we can support you. So that's the goal of this entire work and proposal so.

C

Feel free to stop me at any point. In this conversation with respect to the API strategy or versioning strategy, some of the major options we considered back in December. We like a group of core graph architects, met together and talked about versioning strategy that we should go forth with graph of the major options considered.

C

One is the unified major version, which is, which is something that is very familiar to the industry, going from v1 to v2 VN to VN, plus one one that is very familiar to industry, like I'm I speak to a cons in the context stuffs graph, which is as y'all understand graph is a unified API for all of products. Underneath Microsoft ecosystem are underneath graph, meaning there's n different products, 50 to 60 different products that are all coming together to create this ecosystem of grass and enabling or showcasing the relationship across these products.

C

And hence it's a true graph of the user, centered user or divine self centered. It's usually hard to coordinate among and when I say API developers, the product teams which are feeding into graphs of the identity or a SharePoint or exchange or Intune devices. It's really hard to do all of these API developers are the workload product teams to come together.

C

Allah underlined their timelines and, and you know, have a hard deadline to make jump from one version to another version and at same time it also impacts all developers, meaning whether or not you are using a specific API.

C

Okay, let's move on to the next versioning strategy, consider which is per service or per product versioning, where we'll allow one of like any of these workloads that are underneath graph two to have their own independent questioning, for example, SharePoint could use a dependent versioning from exchange internet versioning, and all of them will be served through the same end point as graft at microsoft.com.

C

Yes, it provides a good independence to each of these businesses, making them move at the pace that their business is evolving, taking into consideration that the feedback that they're receiving about the product etc. However, one of the key key major selling points are major major promises. Our graph is that hey. This is a singular ecosystem that you're seeing and it establishes the relationships between these products, and it becomes really really hard to track and understand.

C

It starts showing the service silo switches which defeats the purpose of graph I sure if you asked who is doing it currently, as you is doing currently, and primarily because of the architecture that Asscher is built on right now, meaning all the services underneath assure have their own independent, like that they're not interdependent on each other, like compute, is not interdependent on storage, in terms of finding relationships similar to how graphs history and in that sense graph, is really unique and and nothing like what majorly a well in the industry right now and the last option that we we did some research into and looked into intently was this version, independent API evolution, and we will dig more deep into what this is about, because this is the option that that we have put forth to be considered and talking about as a proposal to go forward will and this.

C

This particular strategy in terms of pros it provides lead impact to the application developer, meaning only those APL guys that are impacted are only those applications that are using the epeans that are impacted will be required to make any changes or make sure that they are not affected, but at the same time it's also very flexible for product evolution and then providing like I, said very agile product value for for the applications being developed.

C

However, of course it is hard to understand- and that's where we are looking at all the various strategies within the product to make the most easiest to be possible. But the primary thing here to know is this is not really an enterprise way, especially not Microsoft, something that Microsoft has done. However, good API companies out there that are like very well acknowledged within the industry like snack and and Facebook, have followed this method, and this is completely this is currently coming.

C

You know becoming a becoming a thing that application developers do understand, so those are some of the options we consider and the respective cadence. It's quite obvious. As it's presented here, we looked at various cadence in terms of three to five years: quarterly, on-demand, hybrid and and landing on the second one we just Gordon, let's dig into more of what that means.

C

So the version independent EAP evolution strategy is the one that we are currently getting a lot of feedback on and and people leaning towards, which means that for the aps, within graphs, we will publish the deprecation of the api elements in a quarterly fashion. That means each of these API elements, whichever has to be deprecated, will take. They will show up as deprecated in the schema.

C

That means they're going to be more at some point and still will be fully supported for the time period of the deprecation for that inquiries and responses, and the two parallel tracks here to notice one the product will showcase the deprecation and then communications will show. This is this: is the major pair electron that I'm talking about where it will make sure that all the required communication go out to the application developers and there will be a minimum duplication period of two years.

C

The parallel track of announcing definitely I mean a communications as one of the key things that we are looking at, and a minimum support feed it off of two years will still be taken care of with the gradual retirement with while, while we monitor the telemetry again, the key thing here is to note that only those applications affected will need to react. All of this primarily again.

C

The key goals here is to make sure that the product evolution happens at a quick pace, independent of aligning each of the different workloads at the same time and making sure that we fix the things that are currently broken in graphin and in the communications primary aspect to be can get off another in a good way. I'm skipping this light because we talked about why we are going with this proposed strategy and there's a question about: will there ever be a new version released for graph from v1 to v2?

C

And, of course, the answer is yes: even with this strategy, we will look at releasing a new end point for graph, but then it doesn't have to be as it's going to be much less frequent and there will be strong reasons of major protocol changes.

C

That'll affect all the payloads underneath the grass could lead to a new version, but we're not forcing that to happen anytime soon in the future, because because there'll be a path for each of these api's to introduce the changes in a you know in a in a developer, friendly fashion within the same endpoint.

C

The next two slides are the ones that are going to be major focus here. Where for the application developer, here are the four cycles that that are key for an application developer. Api lifecycle is, as we look at these four stages. An API can be like. It is in a completely supported, State, it's functional.

C

All the abs are encouraged to use these api's when, if whenever something is decided to be like deprecated or changed, will mark the schema accordingly and the developer experiences will then start showcasing, discouraging the use of that API element and then we'll walk through what that looks like in the developer experience and then, as we go into the next phase after two to two and a half to 2.5 years of this deprecated period, minimum support policy we'll start looking into a gradual retirement with telemetry into consideration and whitelisting if required, and then one ensure that deprecated element will remove.

C

That and API element will no longer be soluble, so the app developer experience here is the major slide which we talked about. While we break the breaking changes. These are all the key elements of developer experience that we'll be looking at to to make sure this goes as smooth as possible. One place is to look at the change largely within the changelog URL. There will be a section for breaking changes which will be laughs of all the changes that are getting introduced as part of this ap evolution and when I say ap evolution.

C

Addition of new things will of course be part of the changelog, but if at all, there is something that's going to be removed or changed or renamed that will also be called out, but that will be called with specific tags to ensure that we get the right attention from the developers. So there's a section for breaking changes with the right tags of whether it's deprecated or retired, and, of course, one one thing that we are looking into keel is providing ability for developers to register for notifications on the deprecation changes.

C

While changelog is one experience, the other key key important experiences is sticky updates. So within this differ graph very compelling. The applications will still show compiling warning compilation, warnings for deprecated attributes, and that will happen again over a period of two years, minimum period of two years. So to speak, because as you all, the support philosopher graph is two years, but at the same time the products underneath graph could have a different support policy promise to our customers.

C

For example, SharePoint has much longer support period, so it will be based on an independent product support policy. However, it like a minimum of tools, will be ensured for the product that is within graph, so SDK updates they'll be SDKs, I mean whenever you recompile your application with the reduce ticket.

C

Once there will be compilation, warnings for the deprecated attributes and the graph explorer there will be changes within the graph explorer where, if at all, you're using any deprecated attribute, it will start showcasing warnings as well as redirect you to the order Terminator responses for duplication headers. We guide you to look for these replication headers. You will get notifications on what responses are throwing.

A

Okay, sorry so I'm gonna transfer hop over now to our third presenter, we'll come back to our theater and so give you guys time to think about the feedback a bit more. So our third presenter is actually a partner. That's talking about how they've built using the Microsoft graph for solution, clink comm! So do you want to introduce yourself and just cook through what you've built and some of the learnings you've had there? Yes,.

D

Hello, my name is Oscar Mayer and I will introduce to you the service that we have built and let me just share the screen here. We have created the service on top of office 365 that is meant to serve as a sole interface to the office 365. We are there, the graph we created the very common business case, like case management, project management and timesheet reporting that is built on on the top of the graph. So what we do is that we try to glue together all the main components of the office 365.

D

So this is the way we think of it. The service is using the graph to glue together, SharePoint planner outlook, conversation groups and try to capture everything together in the way that we can make sure that we have a complete solution, so we can see all the cases or the projects and etc over the solution space.

D

So what I want to give you a demonstration of is how we actually are managing the cases that are entering into the company, where we turn the group conversations into cases automatically using the market graph service of web hooks and then signal r2 to actually present it to the users if I go into the application.

D

This is the main screen that you see. What we are looking at is really a SharePoint list that is hosting or working. We are using that as a database. You could see with all the metadata.

D

We have a case that has entered the company and if we look at the case, we can see that there was an email coming into the system requesting if there was a position available and with the CV attached, and we have then created some conversation like thankful and asked for the cover letter and- and we have different communication between a team.

D

So we find out that this application was quite good. We have a tasks associated with the case and we can here jump directly into the planner task. If we want to work on that one, as you can see, it could start directly into the tasks and planner and we can see in notes that it's marked for Klink. So we have a direct communication between the planner and and the system.

D

We have also the documents and we can save the documents directly in the repository it's really saved in the SharePoint repository, and you can see the cover letter and CV that we have received from the actually image and, of course we can go into the SharePoint. If you want to be there and you can see, we have the same data same documents in the system, and so it we can go back and forth.

D

We can edit things in either application and the clink will always capture everything together and make sure that whatever is belonging to the case will actually show up. So I will ask my college David to sent an email into the system and see how it pops up in here in the case list.

D

So it takes only a few seconds. We are now cutting the case. You can see it, the screen is updated and we got a red icon here. We can see this is the newest case coming in, and so what really happens is that he sent an email to the male of the group. The group was actually indicating that had received new email and we captured it and put it automatically into the SharePoint list, where we can then continue the team, the operation team here to work on the cases.

D

This is a totally bundled with the teams, so we can create create as many teams as you like, and we keep the you could say, the threat of the communications always together and all documents, tasks and cetera coming from different system and office 365. So if I go back to the slides now we can just look at how this is done and we have now on the left side.

D

This is the catchment management system where we actually start to in it the the web book and we put up our end point: the listen endpoint and as soon as that is pinked by the graph. We start our services and go directly into our application. We have actually moved all the services into a specific dll that you can distribute as a new cat, so that is the yellow things here. So we can hit the web books that we call our service to edit the web box and really what we do.

D

There is as we go into the ICU. As you all know, the client, autograph and subscription request similarly saw earlier in the presentation. What we have added to this is that we need to put subscription and authority into cos. We use distributed cast to our the capture the subscription for for its user, and when we are hit on the endpoint, we grab it again from the casts they take out there.

D

We have registered this group conversation resource and from the subscription we grab the group ID and the conversation ID, where we then look up all the threats for this conversation, and then we can get to the posts and the attachments and everything that is belonging to email, where we then call our group site and add to the SharePoint list they sent.

D

That becomes then part as we saw earlier it pops up on the screen of the users, but at the same time, when we have actually taking this case and registrated to the system, we request the signal our to notify the client. So we have here the signal part that is the other core technology we use, so we have in it set the team. Its group has team, has its own group and we need to register it.

D

That with signal are so we can communicate this event or this web hook to all the clients where we actually call the teams we send request signal are to notify the client, and we are shifting here between CC, our pant and and JavaScript code depends on if we are on the signal, our hub or or in the JavaScript client.

D

So in the end we write to the screen, so that covers the how we really are a capturing, the email automatically using web hooks and put in putting all together, try and- and this way we can lose the the over solution of the SharePoint, planner and and outlook and in a holistic manners. So we have done the next part that I want to show you.

D

That is how we use the system, the graph again to registrate time on directly from Outlook events or directly from the the case or projects within the system to the SharePoint list and and how we can make sure that we can capture time and regulate the time. So this is then how we create directly from Outlook calendar and Planet tasks using Microsoft graph. So if we go back to the system, we have here the worker service. We can see here that this is the timesheet part of the solution we were just in the case.

D

This is time as we have been regulating, underneath we can see all the Outlook calendar events and you can see the marked one are already registrated with the time seed, the number of hours etc, and we have then two here that have not been registrated and if we go to, let's say other browser here to this to see the this calendar you can see. This is Outlook office calendar. You can see that everything that clink has been regulating is it marked with.

D

It was the K and you can see here, the simonkal is not marked in Outlook. So if we go back to the system, we could register ate this car call here, and you can see that we capture the time and minutes directly from the calendar. We can classify this and and put it to another team or or link it to a specific work like the planet, Oscar and the chatter. But if we just do it like this straight from Outlook, then we just registrated and now it is marked here.

D

You can see, call for Simone and now seen here it's in the character and that we go back to the outlook you can see. We have now marked the seamen calendar immediately. If we go back to the system- and let's say we want to remove it again- we just edit this registration, and now we delete it from here. You can see it's gone and if we go to Outlook, it's also gone. It's not marked anymore.

D

So by this way we can utilize the graph entirely to direct a time on events directly from from Outlook, or we can also release it. Of course, within the system we can register eight tasks that are coming from elsewhere, so we can add to the day registration and then we can just put the hours and minutes and and on what team. In this case, we have only one team and we can classify the work in a different way, a building code and etc.

D

So that's how we have done it, and if we look at the code slides for this, then we have it like on the screen. Now in the timesheet, we read from the Outlook calendar from planner from protect the shipment list and the case list, and, as you can see here, we are also with everything in services in our look at packets, but underneath we call the graph client the current view.

D

As you know, this is a common thing in the SDK, the planner tasks and the main touch not me have been looking at now, and this is how we registrate everything in the protein and a-team case.

D

So we write to the Outlook calendar, then this mark, so we can make sure that the user knows that he has already registered it or not and in the shipment list we again use SharePoint as a database, so we are not using any extra services like sequel, server or anything else, and that was the goal to actually use office 365 solely as it stands just add value with a graph by I think common business service just total there, the solution, so this actually covers our presentation of this solution. So if you have any questions or.

A

That was great I love the fact that it highlights a lot of the different connection, points that Microsoft 365 provides and allows the user not only to be inside of your product, but also to continue to use, outlook and planner and so forth in the way they're using it right now and summer season. Our marketing team was asking what what the things that you run into. There were unexpected as you went through, and did this like what their challenges did.

A

You have it's great to see you wrapping things in NuGet packages and you know Drinker wrapping those, but there were. There are other things you did that you learn over time, communicating with god. Yes,.

D

The first thing that we had to overcome was that the fact that the bat box is really meant for for an interface update, but as we are working in the solution towards teams, we have all the teams to registrate the bat book, and then we have. We did not decide to use Azure functions or or or.

B

D

Or anything else outside of the solution, so we need we have to create complex identification of so we are not duplicating the data that this was really already served. So that was one of the challenges that we had. So we captured the web book and then we had to create some mechanism to actually make sure that it was unique in the system, because otherwise everyone in the group were updating the same, because we are not only reading the data. We are writing the data also, but it need to be a eunuch.

D

That was one of the things that we had to overcome. There are still things that we find strength in the graph that was specifically when we as we are working on. You could say, conversation groups working on email part.

D

We have to forward messages between the groups and between conversation, and it seems to be that that is still an open question for us actually to the graft team, really that if we forward a message into a group conflict conversation, it loses the the body of the forwarded messages. So that is, it keep all the attachment and everything. But the actual body itself is not there, so that is all still an authentic problem with the sober solution, but otherwise it has been interesting to work with the graph.

D

It has been a great experience, but then we had there were a few things about latency and etc. We had to wait for the systems and in many cases, then we have to look. You could say and make sure that we waited enough time to be to get everything available.

D

Specifically in the beginning, when we are configuring, the solution, because we are self configuring, the solution for its t-nut. So it's enough to lock into the system and consent as an admin. Then we set up all the groups and planner and their cetera, and then we had some problem with the latency and we had to wait for for some time, but we have all become this by by actually doing some looping and aunt, etc. Yeah tata solder,.

A

And so I I have wondered that with the signal are in terms of the webhooks, because they aren't defined with an SLA inside the system and depending on the workloads there's. Certainly our time is worth what the web books don't come through kind of instantaneously, and so that's interesting. If your users give you that feedback with regards to you know, then doing something and not seeing the web book pop with the signal a straightaway. The live stream are also mentioned. Do use graph batch calls and how you mitigate it with like kind of throttling.

A

Is your system hitting any throttling issues, yeah.

D

We have not had any throttling issues. What we have done is that we captured the token immediate.

D

We are using I'm salt for the token service and we will then capture the the token as soon as we can and I make sure that we don't request too many is too often to the graph and but when we need to wait that request until we get an error back, we wait and we Kristin could request again. So we are really managing all the calls to different services by ourselves. So we are not using passes or we are not trying to tell well.

A

Look I really appreciate your time today. Thank you. I want to jump to a people to talk a bit more about the deprecation and so again just thank you so much for sharing that and we'll make sure those people asking for the slides in the deck, so I would definitely make sure these are included in the community course light deck that we put out with a recording later in the week.

A

So thank you again for contributing and if other partners would like to share what they've been doing with the graph and I love the fact that you can share it via a working product. This super useful to understand the context and not just jump straight into here in the core. So thank you for demoing and showing the api's in that way. Polka thinks so I pit there I'm, hoping that we manage to iron out the audio issues. Yes, can you know yeah? That's why Clara?

A

Yes, so I'm gonna hand it straight back to you, so thank you, Ramona pre-show, that okay thank thank.

C

You again, our apologies, everyone I'm just talking to share my screen back. Hopefully you can see there and are able to hear me well, because we have only a few more minutes left on this call. I was just thinking what's the best way to to capture this I, hopefully shared the essence of what we are trying to put together as a proposal for for versioning and APA.

C

Well, you should within graph and the two key things, but for this entire group to take away is what are the best developer experiences that Microsoft can invest in such that you are like I, said: you're getting the right best developer experience to ensure that your business is continuing, but at the same time you are able to grasp, are captured, changes being made to enhance the applications being developed.

C

Some of the changes I talked about, which is, which is what we presented: two different forums before, like the SDK experiences, header responses, graph, Explorer and then the documents, basically the changelog api reference documents and blog posts, community calls, etc. These are some of the things that we shared and yeah I also received a very good feedback from community both on the azure side, as well as the EM 365 side on ideas floated around for hey Cheers.

C

Why don't you consider this option that option and I captured those options and putting it in here for one like Jeremy mentioned, I'll share this deck out, so you'll have some time to go through this absorb and see if you have any feedback to share, but then, since we have only two more minutes, instead of going deep into this, please type in your questions in the chat. If you have any feedback inputs or would like to have a discussion further, so I'm happy to set up something else.

C

But but this is one thing: is synthesis has been going on for almost two months in terms of proposal and I've received great feedback on this. We are looking forward to moving very fast, and this and we'll also be discussing this during the the remote MVP summit. That's going to happen in mid-march, so there's also another opportunity for discussion and feedback a pass here. Please feel free to type in the chat or, if required, we could have another discussion around this.

A

Yeah I think what we might do is we'll share these slides with people and then in the April community call. We can get you back with clay, a bunch of questions if we can use the the Microsoft graph hashtag we with the hashtag deprecation and then that way, we'll be able to collect the questions together and then we can have you presenting back our petha in in the April call. This is really important to us as a team.

A

My path has been driving this and there's been obviously a lot of internal discussion, and you can imagine the opinions that we all have, which this has done a great job of collating and trying to get one agreement across all the different API review count and various different divisions of the company that want to have an opinion. The strategy here we do, but we also want to hear from you as well, and so what we'll do is we'll share these slides with this recording, and then we will.

A

Basically, you know have encourage you guys to comment in the blog post, because we are able to add comments now in the blog posts and then we'll get up at the back on in the April call to discuss this further and obviously answer any of your questions in the blog comments as well. Thanks, Jeremy excellent, what we're bang-on an hour. So thank you so much again for Vincent up ether and as grimmer for presenting today and if people are interested in contributing and presenting in the community called.

A

Please let me know- and we obviously, as you can see, we're interested in what you're doing an open source perspective, but also love to see what partners are doing, because I think that really does show the value of the graph in terms of the breadth of the API is across the Microsoft 365 surface and the power of what what you can do with that. So we'll see you next month in the April call and it's April 7th at 8 a.m. and there are plenty of other calls that are available where you can see.

A

Other community calls on our mics or 365 developer. Youtube channel, and so it's a Kao, MSM 365 dev YouTube you can subscribe to that channel. Every single community core is recorded and published there as I, say house is on April 7th and you can see that we don't just have Microsoft graph and SharePoint calls which I know you're very familiar with on these, that with the people I can see in the lists, but there are power-ups community calls and office. Add-Ins community calls and the Microsoft teams community call and the identity and adaptive car.

A

So there's a lot of content. We're sharing with you all on a monthly basis. So you know if you've got things you're. Building that map to these things, we'd love to hear from you to make sure that we can have the community presenting back and not just having the pg present to you. So a big thank you for everyone for attending, and we should be at this recording out by the end of the week with the slide decks, and we encourage you to communicate there on the blog post.

A

Thanks again folks and see you in April.