to our digital services workgroup bi-weekly meeting we're going to go ahead and jump in and i always like to start in the beginning just to level set on what we're actually doing as this group so we do meet every other week and we've been doing that since 2018 and our vision has changed a little bit this year we're primarily focused on digital services last year this this year we're going a little bit broader in looking at digital trans is it just me you're breaking up a little bit john is in and out yeah i think that was uh how about now is that good yeah yeah i think what this announcement keeps coming on to say admit people in the lobby so i'll get pretty quick to our agenda the shared work uh we have on a drive to google drive and we're in the process of moving that to a forum and then the uh the hosting group is we're a group of enterprise architects and our whole group is the basically the ones that host these from the servicenow side uh jeff luby is running that architecture team in the u.s it was like to keep track of our new members that are coming in and that's why i always go through those first couple slides just so you're not you know deer in the headlights jumping in so you know a little bit about what we do in the group so if today is your first time if you can just drop your name and where you're from in the chat we'll be here to recognize you and then welcome you along the way there's also a shared instance i'm not going to demo off the instance today but we do share an instance where we give everybody admin access and if you do want access to that instance there's a bunch of work that we've done on there uh you can just drop me an email that's still pretty informal right now it's another thing we're looking to formalize some more all right so our agenda today is to hand it off to allen and jamie architects from kaiser we're going to talk about object storage as a service and they're looking to get everybody's feedback so they'll be soliciting feedback from you uh to see how others have done this or to solicit ideas i know there are a couple people on from servicenow so feel free to chime in and give them any advice that you've had in this area and then i'm going to introduce a little project that i'm working on for 2022 a work group project and then solicit some feedback as well and then if there's time at the end we'll just go around the horn and see if anybody has anything interesting that they can fill us in on that they're working on with servicenow so alan and jamie i'll hand it off to you you should be able to share now i think it'll just take it from me and you guys can take it away [Music] is it showing yes okay get my other page up here there we go all right um hi everybody i'm alan prochaska um jamie tucker and i work at kaiser um in healthcare rit and what we're bringing to you today is a is what we're ramping up on in terms of object storage our technology team is beginning to employ this and we need to figure out how to represent this in our service management system in in servicenow so i'm going to show a couple diagrams just to provide context so in traditional storage you've got a you've got a storage array and that that contains disks and when somebody wants storage there's an allocation made from it which a storage allocation or the the um the class that gets used to the volume the storage volume a server mounts that volume and then they present a file system out to the client application and the client application would would use that in object storage it's similar except that there is a storage obs there's a service abstraction between the allocation and and the servers so that way the client doesn't have to be aware of okay you know where is the implementation where am i getting my allocation from instead they address this this storage service and in fact ours um you know our technology leads called the storage as a service um you know osas object storage as a service um our the technology that we're using is clarity um and their their organization the level of abstraction is called a ring so the storage servers are all aggregated into a ring and then you address that ring and say give me some storage and back comes a bucket which is really charming um but it seems to have taken hold in terms of the of the terminology and after you get that allocation that bucket comes back to the app location along with the pass key which you use for your for your authorization so that way the application can read in right you know to and from the bucket um pushing information and pulling and pulling information um and um and that way they're insulated from the the deep behind the curtain details of the individual storage servers so so let's elaborate this a little bit more as we're building this out in common service data model so we have this same road that we had here we've got the the storage servers and the ring is an instance of of this overall business application object storage as a service the vendor is is scality or the application the scalability so there are multiple deployments of this we've got three three deploy three production deployments in our production data centers and then one non-prod deployment in the non-prod data center and so clients would would get allocations from those places and out would come buckets those who get associated with an application deployment um which which we represent as that's the mapped application service um in in service now um which is the child of the the business application so as we're and then here's the pass key that goes with that for the authorization in terms of services we think that this is a technical service a storage service and one offering which would be you know object storage and we expect that there will be other offerings like either san or nas or block or file we need to work with the with the technology leads about that but this is what we we have in in mind um in terms of looking for a class for the um for the bucket we were directed to cloud object storage because typically uh object storage like this is located you know off-prem in the cloud and so that that would make sense in our case it's actually an on-prem installation um and so we're thinking that we might subclass cloud object storage into a class called bucket just to make it very clear to to clients you know what is the ci it's a it's a bucket um but that should be equivalent um if we do have some off-prem allocations that are um that are that are storage and off-prem so this is a whole lot of information so i'm gonna start turning off the fire hose um so in summary what we're i've told you about what we what we have in mind we're ramping up on this um and in particular we're interested in you know what's the class that we should be using for the allocation for the bucket um and that we're planning to represent um the cert the technical service the technical service offering connected into the deployments and wonder if you guys have feedback on that and if any of you have used object storage have object storage running in your environments how did you represent it and did it work so let me pause there i see one hand up anybody from the iton team on and is there a good way to find out who among those 77 people has their hand up or is it just sequential search is this yeah i had my hand up but um i wasn't able to find chat anywhere oh we hear you paul you want to go ahead with europe are you there paul uh yeah i'm here i just wanted to mention and i had my hand up before they started i just can't find chat anywhere on ms teams so um it's okay with me but you'd mention leaving our name and organization or whatever and i don't it's not even in the app so anyway that's all that's what i just wanted to point that out okay i see i see the note from rick from rick williams that it seems you know wrong to use cloud object storage for an on-prem um uh i agree it's it's a little there's some dissonance there um [Music] hello um right do you have a suggestion for an alternative class one second i actually have two things taking myself [Music] while we're waiting jamie um anything to add um i yammered on but i may have missed something so to me real quick that cloud object storage there it's it's in between you know you're going from the you're doing going from the technical domain into the consume domain you got that cloud object storage in between right in the csdm it doesn't account for that right so you have the application service if the application scales you as application service then going over to the right getting into the consume domain right is there it would it would all be it would all be here you know because the it it's all within the managed technical service domain you've got the app service you actually have two app services one is for the client application the other is for the um currently they've they've defined the object storage as a service as a business application so we've got we've got that deployment as well and then the actual bucket is one of these infrastructure cis this is all how the the technology group would assemble something that eventually gets offered over in the in cell consumed so okay so this is this is all in in orange orange got it so basically you have your app yeah you're going to have certain applications that are going to be dependent on that object storage and then you'll have a technical service because there'll be a group that's taking care of that storage but the big question is how is it represented under those those application services that that consume them right yeah yeah i would assume that it would be a a relationship um just like it would be a relationship to a traditional storage storage volume or file system is it exclusively that is the traditional volume or storage system that it's exclusively dependent on or would there be other things too that's one thing that it's dependent on the other thing that a a client application would need would be would be a server and and it could also have a have a database um in the past with in in in the traditional storage the the volume gets mounted onto this onto the server and then they use the file system that the the server presents this decouples that so that they just basically get a get a socket and you can read and from that socket somebody wasn't on mute um which i which i think can work okay the client will the application will have multiple technology requirements so there could also be a database you know there could be other things okay um yeah my only thought on this is that it just seemed and i'm i'm not expert in the space with the object storage and stuff like that it seems that seems like almost if you're receiving object storage as a service it's almost invisible right it's you're going to you should go directly to the service do you need do you need to represent the infrastructure behind that i guess you do yeah it would be um yeah if it's if it's truly a as a service then this would all this would all be off-prem like in the in the case of azure um you know when we're going out to to get azure resources they've they've got a storage object they've got a database object and we just work directly with those and azure has all of this stuff um behind it um but but we don't see that we just we just know yeah yeah there's a there's out there yeah i guess that's where my confusion is that bucket there right that that's that piece by saying it's cloud but you're you're it's but it's really it is really on-prem that bucket is on-prem right it is because we've we've implemented we've actually implemented these rings on prem but you know i asked is there a um what what class in the service now i recommend and and this was the suggestion and i think that it's because most um storage as a service is off-prem and in the cloud and so they've they've named the class this so so if we if we pick this class or the bucket is it's going to bark and quack like a um like a bucket and would bark and quack very much like what a cloud object um storage ci would do it's just unfortunate that it has cloud in the name because that's a little bit confusing yeah okay yeah that's my feedback then there's yeah the feed the the naming convention of it is confusing even though it's on-prem yeah okay yeah um do you think that if we subclass that to some to a class name that didn't have cloud in it is that an improvement or is that lipstick on a pig with no offense to pigs or lipstick i see paul has his hand up paul was it kepler yeah sure is absolutely my my thought would be as far as classing is concerned wouldn't it make a little bit more sense if at the top end for storage you had a storage like parent class and underneath that you had on-prem and cloud underneath there and couldn't you put your your buckets under either of them depending upon where they are so you could more easily identify where they're physically located if that's an important aspect to the solution yeah um in the current um in the the the current classing um let me just bring up our our instance of course when you're doing in front of people it decides to slow down just like when you ask your your child a simple question in front of guests you know you know suddenly they you can't respond or they or kids say the darndest things you know that's part of the charm so so there is this concept of storage volume and it just descends it extends right from ci um and then file system is it is a subclass of that and underneath there are these various file systems nas nfs smb there isn't the split in this class hierarchy of on-prem and off-prem and i could see some people arguing well we really don't want to represent that in in class because that's an implementation detail i'm i'm not sure how that that argument works out but the file system is is the closest thing that i can see to a bucket because it's a you know it's it's an area where people can um you know read and write information so that's why i was first thinking of just representing it as a storage volume and maybe subclassing that down i think the um the suggestion for cloud object storage is that the traditional implementation of these is off-prem and so that ci might might match um so what i'm hearing from you is instead of just adopting cloud object storage you know maybe consider you know building off of this hierarchy um whether or not we decided to do the on-prem or off-prem did i get that yeah absolutely yeah that's exactly what i was thinking but then underneath each of those underneath cloud and on-prem you would then end up having to have file systems duplicated as well right yeah now that i see exactly how they're doing that i'm not sure if that's the best way yeah and i think my inclination would be you know rather than to say oh this is a file system is actually to be direct and say it's a bucket because bucket is the is the the concept that's used in um with cloud infrastructure and um and with the more malleable environment so let's just represent that directly and so if i were to go this route i think the question would be do i subclass that from storage volume or from file system you know or from someplace else but good thoughts i appreciate those let's see i mean if you had a magic wand to create your own class would you would you prefer that right are we trying to are you trying to put something that maybe they have not that servicenow hasn't provided an answer for so um so i i do have the the magic wand which is we can we can create a a subclass you know but then you get into the spider-man you know with great power comes great responsibility um you know is it um and if there already is something you know in the model that meets this need generally speaking it's better to go with with something you know from the platform and out of the box uh the cloud object storage if we were if we were using discovery um and if this were an off-prem um we could probably get the the cloud object storage cis automatically built unfortunately we don't have discovery in our implementation but it is useful in terms of um uh understanding the intent of the classes like we can hook it up in a non-pride environment train discovery onto a a target location say go and it builds the ci's and you say oh oh this is how things are supposed to be put together um if we had that you know potentially we could train discovery onto this on-prem storage service and say go and and it would be interesting to see would it build out cloud object storage um cis um but getting back to your to your question what what i what we want is to be able to represent the bucket so that we can tie them into the into the tree of of ci's so we know where it came from it comes from this ring and what it's connected to is to this application so if this goes down then then this application will get will be an impacted ci and an incident can see that change can see that and and we can leverage it that way it's so that it's tied in into the into the the ci tree the ci network i'm not so particular about about where it is but only um thinking about if servicenow has already done work on this we may as well leverage it so long as it's not um a really big mismatch was that from mike yeah okay now just trying to put a little tweak on that conversation to look at through different lenses yeah okay yeah yeah um charles i see your comment do you want to annotate that sure this comes more from my time as an architect before i join service now um so so it doesn't represent official servicenow policy or anything uh disclaimer yeah uh at my employers we swung too far on the side of trying to cram everything into the existing ci model and we ended up overloading a bunch of classes and so you've got a bunch of things that are like mandatory attributes for one group but not for another group that are like sharing the same ci type so what i've been trying to do it and my past employers is swing them a little bit more the other way i've recognized the uniqueness of different objects and then the uniquenesses of the different management processes that go around those objects and treat those as separate ci's rather than overloading them on one ci and then making everything more complex later so yeah that's more customization in your cmdb but it's contained customization and it's not the type of stuff where you're trying to parse out like is this the the customized version or is this the out of the box and it's not intertwined and you're doing less filtering and all the rest of that so what about the decision about okay if i don't want to overload this class do i subclass it or do i make a sibling class did you face situations a decision like that yeah and i think it really comes on down to like how do you say so like this one where it's saying cloud object storage i think your inclination of no it's a bucket it's something unique that's up here maybe of cloud object storage is the right way to go at the end of the day i think it's more important that you decide and have clear rationale than what the actual answer is like you're consistent in how you try and do things right rather than trying to be perfect with each thing more important for it to be defensible in court um yes than what the actual document said exactly okay that's helpful here's a crazy thought that could be way off base yeah right because it's not because it's not almost like with that and maybe you still need that class but could that be i mean you have a storage service is it is it are you offering it to the application in different ways are you offering it through this cloud object it could be as simple as as an an offering i know that's not disguised because an offering is a ci right well offer that's no the no the offering is how it's through the offering that they will get the ci the our our our service it's a technical service of storage and then the various offerings from that are the various types of of storage services object block file nas san and so each of those would would have its requestable items you know if i want to create modify or or remove a um you know something from from that storage or from that from that offering but the bucket is what they're actually going to use and so they address a request and say you know create bucket or allocate storage they're going to address that through the requestable item associated with this offering um but the offering does that make sense yeah yeah yep so what i'd argue there is you've got two unique infrastructure cis a bucket and a pass key you've got relationships of that bucket and it's pass key and you've got relationships of that bucket and the pass key to your client application and then to your technical service down below so so all of those things are delivered through a single offer in your catalog right right and in the fullness of time you know we would build all of that out um and in the crude mvp those those offers may go through you know a an email template that arrives to somebody and says please please make me one of these and it's a manual check that that gets the allocation you know approved and then done eventually they would just be going to the requestable item generating the offer which would get consumed fulfilled and back would come the bucket which is already connected related to all of those things i'm not seeing more questions um the conversation is really really helpful if you run across other situations or information about object storage or storage as a service yeah i'd be glad to receive the the input um you know my email is available through through john or maybe in the the meeting invite um john should we wrap this up and do you want to go on to your next um topic yeah sounds good good topic i learned more than i was able to contribute unfortunately but thank you for that sure thank you everybody for the for the thoughts all right just a few more minutes that i want to go over some stuff the over the last two years i did a project each john you're dropping out again yeah the uh those announcements i can't stop i am tired of teams man every time i get somebody in the lobby it cuts me off but i'm doing a project i'm going to kick it off and um and basically what we have is a lot of people have been asking me for this where they run a central i.t department they have i t service desk that service desk provides incident management and technology management for some assets right because they're central i t they don't support all the assets or all the applications a lot of other cert departments have their own service desks and they offer services just like the ones that we offer in our central service desk so as part of a consolidation effort right instead of having 25 different i've gone into some organizations that have 15 different service desk solutions right so when you talk about application deduplication as an ea that's that's definitely a big target so in some cases the cios has said you know can we consolidate these systems can we centralize the offerings that the service desk has not necessarily the service desk function itself but the offerings so basically saying hey you can come in and use my incident management offering on my instance so that we're sharing more than we are you know standing up other systems and administering other systems a lot of these i work in government so a lot of times you know the sheriff and department of health will all have their own systems and they're all doing it incident management and i.t request management so there's a lot of ways to do this and um a lot of people have come to me and talking about like acl hell you know we have to use a bunch of acls to do this and you don't really have to do that there's a lot of simpler ways where you simply say you treat level one or tier one service desk is another assignment group and you're able to get to tier one one organization we did it as a workflow when the incidents get logged and it says if there's no assignment group in the incident or no service then we'll go and look at the department of the requester and then assign it to their tier one so so there's other ways to do this besides acls everybody assumes acls right away uh which is i think a bad assumption so so that's the problem definition and um what i'm thinking about is a deliverable i don't know if any of you guys have gone through the exercise of of an instant architecture we actually go through and there's some um hello fun day oh somebody's not okay yeah you could just mute do a mute all john from your instance okay we'll try it there all right so the um so similar to the way that we recommend instance architecture whether you do multiple instances you do domain sep instances you do apps up there's app separation and then there's deliberate customization so there's like four models for deploying an instance or having an instant strategy in your organization and we've been down that path a few times and i want to do something similar here where we have a presentation where we talk about the pros and cons of different ways of like implementation models so i don't know if we come up with three or five different models where we can say you know here's one incident that you could come in through a portal or come in through a phone and we could properly route it so they give different options there down to the acl option where we we go ahead and really forcefully separate everything a questionnaire would help quantify it and it's very similar again to those instance recommendations is we get have a questionnaire and it would lead them to a decision on which method is best for them to use when they're supporting multiple tier ones or multiple i guess service desk stacks and then a document at the end and this is one thing with the instance architecture that we do that's really useful is we memorialize the decision so that you know two years down the road they say why the heck did they do it this way those those decision points are actually listed okay so the next steps here is to pull you guys so if you're comfortable coming off mute and if anybody is uh either working on this today or you have some opinions on this or you're being asked for this would be really ideal um or if you've been down this this path before anything that you can provide me along those lines would be great so that i don't have to start from scratch a lot of times you end up doing stuff you're like wow somebody spent a really long time on that before so i saw some stuff on the communities through some stuff on the guru sites about ways to do it there's even a couple apps up there there's an app called simple separation that's really popular for doing multiple groups like that okay after that it's a matter of pulling the material together and um all that material will help determine the scope and then also if anybody wants to be wants to contribute to this if they want to be part of like a little working group on it where we meet monthly and get together let me know and we can uh we can definitely do that and it wouldn't be a big time commitment but if the more people i have the bigger we can go with the scope or the more you know i think the better that body of work will end up being in the end so the body of work i'm looking to create is that that set of things that i talked about earlier like a presentation of model options for doing multiple multiple tier one service desks but also to um maybe even have it in now learning but have it as a group of things that somebody from at a minimum somebody from our working group here can pick up and use and help them make their decision in their organization but ideally we'd even have this as a now learning course so that one i talked to you about earlier where you have the deployment options for instances that's available on now learning today so we basically want to redo that for this scenario because this is a scenario i get a lot where you know we installed servicenow and centrality and we're using it for 10 of our service desk requests because the other 90 are spread across all these departments that all do their own thing so if anybody wants to come off mute if they have anything to share or just drop comments in there if you're interested in contributing or have something you can use contribute as far as material i'd really appreciate it and i'll share the information going forward that we start to assemble hey john uh what would this scenario include any type of instance e-bonding would that be something else that would be something else okay it could be yeah it could include that i guess because like e-bonding could be that we well you're not you're not really sharing it on the the single instance what you're doing is you're saying okay there's a service on this instance of service on that instance and we're kind of letting them use each other's service right what i'm talking about here is actually offering like incident management as a service to tier one and request management as a service to tier one got it okay hey on it's alan um very familiar problem um and i'd be glad to um to join a working group to to provide feedback um the it's a version of the i think it's a version of of the shadow i.t problem where you introduce the centralized tool and and you know like you said it gets used for 10 because everybody has has their own systems and um the the trap here is that's fundamentally a process problem that and and the stakeholder management problem because all stakeholders um will will assert oh no we're we're different we're different we're special we've got really complicated needs and we couldn't possibly use a centralized solution and and an experience shows that that may be the case for for 10 to 20 but generally speaking 80 percent of their of their work could follow a standard solution but there's a lot of resistance to that and the trap is to try to is to try to find some tool magic to address that process problem and what you really need to do is to back it up and solve the process first and go to the stakeholders and say okay so so tell me about you know how you do incident request and what are those what are the flavors of incident what are the flavors of request you know what kind of services do you offer um you know after getting all that information you're going to see some clustering some affinity clustering of things that multiple people do and that could be the first segment of okay here's some common here are some common services um and and then the challenge is to offer those in a way that is attractive and competitive so that it's in there it's to their advantage to use the central solution cheaper and simpler rather than maintaining their own and that's when you get in into the tooling but the the the first step has to be understanding you know how they see the world um i'll go ahead and say it quacked as it may be you know you know customers see things the way they do but if you're going to get them onto a into a central solution they can only move from where they are so you really need to understand where they are in order to paint a path for them absolutely and so you basically become the vendor right as that central i.t organization and i think a lot of cios now this is becoming uh you're hearing like gbs global business services and in the government sector that i work in mostly is cio as broker where you're actually trying to become a broker or these services that people can share and um so that's the trend right now so a lot of people are asking about it and i think doing incident and requests is good with this group because people are familiar but when we get to the bottom of it it's not going to be much different than let's say onboarding right if you're doing certain hr functions or different service desks for facilities so i think it's going to be the same thing as like how can we one provide that service and then two how can we provide it as a service to others right right the uh so in terms of of the advantages um to to pitch one is is cost and and simplicity the central solutions should be cheaper and and simpler for them to use other advantages is if there is cross organization coordination i'm required if an incident may start in one place and either involve another organization in the solution or may have to move if those can be in a in a single process that's more efficient and then finally when it if things have have a common definition it provides greater opportunities for um for um shared resource management and for coverage um you know between the between the functions and we had esri on a few weeks ago the ea team from esri and they talked a lot about moving from demand-based planning to capability-based planning right and this is more moves you into that more mature capability based planning where you're providing a capability that everybody uses instead of everybody providing that capability for themselves so i think it's a it's a real maturity step right so that that would have to be recognized in there that there's going to be some growing pains in doing it but there is a possibility i think a lot of people have gotten here too um gotten here already right and that um that'll be an interesting part of the study i think too to see who's actually been through this already and figured it out on their own sort of like that one i'm working with a city that that actually has a workflow that will come in and do the assignment where people have different they do it at the catalog level so that depending on who you're logged in as you'll have permissions to certain catalog items that'll populate an incident differently right and assign it to the right group so there's all these different approaches you could potentially use acls being the one that everybody cringes at like i said before but it'd be interesting to see what's really working and then what's been a pain in the neck to everybody okay cool thanks for thanks for the input on that and uh i will reach out definitely to uh anybody what we're also doing here over the next few meetings we're gonna create i don't know if anybody's a member of the servicenow communities but it's it's been a little bit of a because there's about 800 of us on that email invite now so it gets hard our email caps out at 500. so what we're doing is we're going to move this to a community to a forum sort of like our developer forum and we have a couple other business forums too and servicenow and this will be a good forum topic that we'll be able to you know post that body of work on and then actually have a collaboration um on that community with with different questions and also you know draft documents and such we can put on there instead of having it all on the google drive like we have it today so i think this could be that first body of work to to kick that off i have a couple people on there that are helping me with this uh that are all co-owners of the digital service work forum uh so that'll that'll be a huge help for me and i'm looking forward to really maturing this so that we can get some of this this shared with everybody all right that's all i had for today is anybody i want to open it up because we haven't done it like a round table in a while but i just want to open it up to general comments anybody getting stuck anybody trying dpm any other things we've talked about over the last few meetings that you need to follow up on or you know do you want to come on and run something through with a group like they did today from kaiser open forum this is we'll give it a few minutes yeah this is paul gebois um real quickly mike van bergen forward me forwarded me this invite for today's session how do i make sure that i'm invited to all the future sessions because i'm finding this to be fascinating and i'd like to be involved so oh excellent so once you're on the uh my email list i just have a spreadsheet right now that i'm putting everybody in and once the meeting gets forwarded to you i put you on that list until you tell me i'll take you off okay what we'll do is i've been drafted i've been drafted okay you've been drafted perfect you just have to tell me to stop if you want it to stop yeah no no keep them coming and i do have one other another quick question too it would be for me just based on where i am right now at musc uh it would be great for me to get a better understanding of how servicenow uses service now to support their enterprise architect practice um are you using uh tools like lean ix you know with servicenow and or other tools along with servicenow in order to enable and support your enterprise architecture capabilities yeah can you drop me a note in the chat because we have we use inspi which is a ea tool that's runs on the platform we had them in not too long ago to go through their products and we have a white paper from our ea team that talks about like a now on now paper they call it where they publish their their approach so they use a mixture of apm and in spy which is one of our partner products okay great i'll yeah i'll send you over a message okay john this is uh michael argonne um i shot you an email and i don't know if anybody else has run into this we're uh starting out on the dpm path and with service builder and the issue we're having and i don't know if anybody else had this if it's a roles issue or a configuration on our side but we found that with service builder we can go all the way to publish on the first time we uh create the service but if we come back in on a draft it's not allowing us to approve it it's saying we have duplicate services has anybody else run into that i have yeah i have myself run into that and um i don't know the process to get into that situation uh but sometimes i do and then you have to go back and delete the draft and then start again it like creates a copy of the object when you check it out and then when you check it back in it it rationalizes the copy from in the regular object so i would log a ticket for now and um i have a ticket on that as well so the service builder yeah it's i can't figure out how i'm getting out of sync but it only happens sometimes well there's a couple of things that almost appears with uh the current taxonomy taxonomy node service definition structures where you you have to do like you said where you have to actually go back and delete everything and then reload uh we've run across that also when we were building our our service portfolios and nodes so it was quite challenging so we're uh we're learning it's yeah i hadn't never had anything where i went back in though when that happened usually i get it in the beginning like if i stop in the middle and then i go in and i edit the main object after i started with the draft i do it like from another screen i know i can reproduce it that way but i'm you know i guess the solution to that is to exclusively use the service builder but i still got all habits right going into those old forums well we're trying to yeah we're even trying to do it exclusively using the service builder but it seems like you said there's a draft copy and a published copy and then when you go in it says no can't store the draft copy over the published because it's you know you've already existed and it's like oh okay yeah yeah let's we'll do tickets for now i'm sure i'm sure caitlyn has some visibility to that i don't think she's on today though okay great thanks anybody else yes service builders interesting we had a conversation the other day with service builder on you know do we let especially with the business services now do our people that weren't defining services before now able to come in like the service owners can able to come in and define those properties and it was a really interesting conversation because the servicenow team gets stuck with doing all that today but sort of like what you're doing with the citizen developer right the uh the low code no code on on app engine kind of the same thing here with the service builder you can let the the service owners take it the first mile and then the last mile would might get picked up by by an architect team or somebody on the servicenow team that knows the cmdb really well well that's what we're we're trying to do for lack of a term enterprise service management because right now everything comes through our it service management organization and we maintain service now and you know that's one of the powers i think a service builder is for lack of you know another lack of term it's kind of idiot proof you could fairly step by step through and at the end uh you end up with the defined service that you want cool glad to hear you're going through it everybody else using service builder pardon i haven't seen it i wonder you can hear me hey mitch yeah so i'm playing around with the service builder in in your in the devon since right here your your personal instance and uh you know i'd love to see more more functionality to include things that we typically define in a service the different attributes do you foresee something like that coming yeah there's uh like what's in scope and what's out of scope and those types of being able to add things yeah yeah because i love the idea of kind of a wysiwyg for our service owners to take control and ownership of of managing their own um you know their own offerings and services don't know that's what i have to bring up to caitlyn uh she's been reaching out to some of the people on the group for feedback so if you i'm happy to drop my name in there kind of functionality we would be interested in leveraging great yeah great i'll ever reach out because uh i know lou had a conversation with her a few weeks ago i don't know if they went into service builder they just did dpm but um she's definitely in there because these are both v1 products right so we're definitely reaching out to get as much feedback as we can thanks john you're welcome all right i think that's all we had for today anybody else has anything else or thinks of anything else afterwards just drop it in that chat i always go through there and in the newsletter i answer as much as i can that was that we didn't close the loop on so um oh i had one other really nice presentation to share today that we came out with but i'll share that next time i'll save that for next time all right any last comments going once going twice all right thank you everybody thank you kaiser for presenting and sharing that today great session thank you