Awesome, what is it? (#52)

News

• What is your biggest challenge with Power Apps graphics? by Tolu Victor
• 6 additional modern controls in Canvas! by Claudio Romano
• Power FX Functions = Low Code Plugns by Matthew Devaney
• Functions in Microsoft Dataverse
• Megan V. Walker post "Finally"
• SharePoint vs. Dataverse by Sean Astrakhan
• Copilot API's for PCF and client API developers by Hemant Gaur
• Private Preview Signup for Agent APIs in Power Apps
• Powerful Devs Conference 2025 - Watch it back!
• Powerful Devs Hack Together
• 10 Lessons to Get Started Building AI Agents
• Building Microsoft AI Agents: Which Tool Should You Use?
• So do you want to become an MVP? by Agnius Bartninkas
• Simplify Microsoft Power Platform deployments by using pipelines

Events

Canadian Power Platform Summit
March 21-22
PPBOOST15 -> 15% off general admission + workshops 

ColorCloud
April 24-25, 2025
Ulrikke's Workshop: "Power Pages: From creation to go-live!"
Session with Andy Wingate: "Business Central + Power Pages = TRUE"

DynamicsCon
May 13th - 16th, 2025
Nick's session on "Crash course in Power Platform Pipelines"
Session with Angeliki Patsiavou "The Ultimate Power Slam: Dynamics 365 vs Power Pages Portals"

DynamicsMinds
May 26-28, 2025
Nick's session on Burnout, Powerliftin

Be sure to subscribe so you don't miss a single episode of Power Platform BOOST!
Thank you for buying us a coffee: buymeacoffee.com

Podcast home page: https://powerplatformboost.com
Email: [e-mail]

Follow us!
Twitter: https://twitter.com/powerplatboost
Instagram: https://www.instagram.com/powerplatformboost/
LinkedIn: https://www.linkedin.com/company/powerplatboost/
Facebook: https://www.facebook.com/profile.php?id=100090444536122
Mastodon:
I can see it.
I have it in a folder.
I even tell you the folder.
And the thing is, if I wrote aPower Automate flow and then it

(00:24):
would give me an error that itcouldn't find it or couldn't
read it, so I know there's anerror.
Co-pilots and these AI tools,they have that confidence level
that it's like, oh, nothingexists.
So, as an end user, I'm like,okay, nothing must exist,
something truly exists.
So this is where I think thisis some of my.
This is, I know, going down adeep rabbit hole here, but this

(00:47):
is my frustration with some ofthese tools where, especially
the AI, the co-pilot, there isthis implicit arrogance of, yes,
this is how it is, and it'slike, no, that's not how it is.
So, from an end customer'sperspective, you want to get
them on board.
You got to start building trustin these tools and if something
, if co-pilot can't do something, it's got to start building
trust in these tools.
And if something, if Copilotcan't do something, it's got to
come back and say hey, I'm sorry, I can't read.

(01:08):
I couldn't find emails there.
Could I can't read the emails,as opposed to saying nope,
there's nothing there.

(01:36):
Welcome everyone to the PowerPlatform Boost podcast, your
weekly source of news andupdates from the world of the
Power Platform and the Microsoftcommunity, with your host, Nick
Doelman, and Ulrikke Akerbæk.

Ulrikke (01:49):
Hi Nick, Are you ready?

Nick (01:51):
Yeah, I guess.
So I just Kat was visiting.

Ulrikke (01:56):
Show Cleo to the.
Did she jump down?

Nick (02:00):
Don't worry, she'll be back.
I'm pretty sure She'll be back.

Ulrikke (02:04):
Hi everyone, Welcome to Power Platform Boost podcast
and cat podcast and everythingAll right, Nick.
So how are you?

Nick (02:11):
I'm good.
Yeah, we had a few technicalissues this morning, so that's
why I'm set up in my alternatestudio here this morning.
And it's also family day herein Canada, so that's why the
rest of my family is sleepingupstairs, so they're not having
to listen to our recording thismorning.

(02:32):
They're sleeping in.
But overall, yeah, I'm good,how are you?

Ulrikke (02:37):
Awesome, I'm good, thank you.
It's a winter holiday here thisweek, so the kids are off of
school, uh.
Luckily, herman does still havehis after-school program,
because he was kind of runningup the walls, uh, from being
home alone this morning, as Iput him there and then we record
, and then afterwards we'regoing to go cross-country skiing

(02:57):
, because it's pretty cold stilland the sun is out, but it's uh
, they have fake snow up at uh,so that's where we're going to
go later.

Nick (03:06):
Oh, because I have probably about three feet of
real snow on my yard right now.
We had two major snowstorms,one earlier this week, one last
night.
You know, I love it, I love thesnowblower, it's very soothing
for me.

Ulrikke (03:22):
Good to use the snowblower Boop boop, boop, boop
.

Nick (03:28):
So we do have it's like minus 15, we have a pile of snow
.
Um, like I said, it's familyday, it is.
Uh, in here in ottawa it's a.
There's a winter festivalcalled winterlude.
Um, we may or may not go andcheck that out, because usually
it's a lot of people, but Ithink I sent you a picture the
other day.
Um, for those of you who don'tknow, in ottawa we have the
world record for the longestskating rink.
There's a canal that goesthrough Ottawa.

(03:49):
Every year that freezes.
They make it into a skatingrink, they flood it, they make
it smooth.
You can get hot chocolate andbeaver tails and you can just
skate for many, many miles orkilometers.
And you see people.
Years ago, when I workeddowntown, you would see people
going to work with skates overtheir shoulders because they
would skate from home just towork.

(04:09):
Um, and it's really funny.
You see people in militaryuniform, because in the capital
city, of course, the nationaldefense is here.
So you'd see guys in fullmilitary gear wearing skates
over their shoulders.
Like, how canadian is thatreally?
You know?

Ulrikke (04:24):
yeah, it doesn't get more canadian than that, really.
So that's uh, oh, that's socool.
I hope you, if you do, justtake a little short short video.
Maybe we can add it after thething on the podcast so everyone
else can see.
Yeah, that's so cool.
Yeah, it's really really pretty.
I've seen pictures, so it looksso awesome.
All right, um, but there areother cool things to talk about

(04:45):
today as well.
We have a list, as long as ever, with news and updates from the
community.
So, and also, did you see themassive kind of Rory's poem, and
the thing we did last time wasso much fun.
I really enjoyed that.
And, yeah, talked to Rory alittle bit afterwards and he was
so much fun.
I really enjoyed that.

(05:05):
And, um, yeah, talk to her alittle bit afterwards and he was
so he really liked it.
So, uh, yeah, it's good to dothat for people.
I think absolutely.
Yeah, no poems this time,though.
So, um, yeah, maybe next timewe'll see.
Uh, okay, so let's dive intothe.
The news updates.
First on my list is because,now you've structured it so that
we go Power Apps, power Pages,power AdMate, stuff like that.

(05:28):
First on my list is Power Appsand a post actually on LinkedIn
from a guy called Tulu VictorSanvu where he asks what's your
biggest challenge with PowerApps graphics and kind of.
The thing behind it is that nowwe've had kind of three
generations of Canvas appgraphics from Gen 1, kind of the

(05:53):
out-of-the-box components lookslike something from the 90s.
You've seen the default Canvasapps doesn't really look all
that good.
And then you have Gen 2 withthe modern controls and then
also, sorry, the, the firstcontrols, and then you have kind
of the layouts and the, themodern grid um components that
makes it um kind of adjustableto the screen size and there's a

(06:17):
lot of modern ways to work withthat.
And then, of course, you havegen 3, which is the modern
controls, uh, with ai and othercapabilities.
And he's also kind of nodding toan announcement made from our
friend, claudio uh, romano, whomade an announcement I think
it's january 17th or somethingwhere they announced the ga of

(06:38):
the most commonly used moderncontrols.
So now you'll have things likeum, um, I put it down here just
to be sure I got everything.
So you have text and text input, number input combo, box, state
picker and form, which is nowcompletely modern, um.
But so even though we have allof these modern controls and it

(07:00):
now looks a bit more sexy and upto date than kind of what we
used to have, people stillstruggle with it.
So his kind of question to theaudience was even though you
have all these tools, what is itthat you struggle most with?
And I said for him it's kind ofcolor combination.
It's still kind of a gotcha.
So just got the conversationgoing, I think, and he got a lot

(07:22):
of comments in the LinkedInchat as well from different
people in the community.
I chimed in a little bit aswell.
But, nick, what would be yourkind of biggest challenge with
Power Apps, canvas Apps, interms of the graphics?

Nick (07:39):
You see my design skills right, very, but you know I
think the colors is a good one,because even even um, you see,
you see sort of the mix andmatch of colors, uh, it looks.
It can look almost jarring tolook at a particular app, like I
like something very, you know,very neutral, like not neutral,
but very kind of something thatflows well together.

(08:01):
Of course, really it's the userexperience, like making sure
that you're not jumping all overthe place and you know, and
just making making these appslook professional, like I did
see the post, I did see he had acouple of before and after or
sort of the evolution of things,and one looks like, yeah, a
canvas app that maybe I wouldmake, and then you see something
that, wow, that looks likesomething professional from an

(08:23):
app store.
So you can really see the, youknow, and it's like these little
things do make a hugedifference in terms of user
satisfaction.
Again, I know for me I'm veryfunctional, I want to make sure
the app works, but also I thinkthat once you kind of get these
aesthetics and things out of theway and of course this is
evolving as well I don't do aton of canvas apps purely on

(08:44):
their own.
I do a lot.
I do embedded custom pageswhere you're trying to met in a
custom page situation.
You're trying to kind of matchat least a little bit of the
look and feel of the overallmodel driven app that you're
embedding in, because that couldlook really weird too If all of
a sudden you have this otherwindow popping up that looks
completely different.
The end user is going toquestion wait a minute, am I in

(09:06):
the same app because of thedifferences.
So yeah, it was a great post.
It's something that I'vedefinitely saved for there for
when I do have to dive into someof these design decisions and
make it better, and it'ssomething I am working on and
it's on my bucket list of thingsto get more ramped up on modern

(09:26):
web design and look and feeland that kind of stuff.
I know it's a weak area for me,but you can't know everything
and I rely on very smart peopleand very talented people like
yourself to make things lookgood, but it's something I think
all of us could aspire to getbetter at.

Ulrikke (09:41):
Yeah, thank you.
But um, and for me, I thinkkind of the I think I've cheated
in this area a little bitbecause I have the uh, the
training from school in terms ofcolor, um, theory, that I know
which color goes together andkind of have the theoretical
foundation to understand how theuser navigates with their eyes

(10:01):
and how they kind of look acrossthe screen and what's important
, the stop stop effect, thethings that go together and
doesn't.
But you look at the picturethat he has in that post.
Like I said, the first pictureis kind of gen one where you
just put the elements on thepage and you don't really think
about it.
And the next one is, of course,a lot better in terms of the
graphic.
But also I'm curious to howmuch work actually goes into

(10:23):
making an app that looks likethe cause.
That's second and thirdgeneration graphics, because
before and, as you know so,christine Kononeski, for
instance, she introduces she waskind of the one of the first
one, in terms of my feed atleast who who dived into the
more custom graphic UI in Canvasapps, but she relied heavily on

(10:44):
the HTML component and createdit using HTML, css and
JavaScript.
It's very much my home base interms of what I know coming from
web development.
But I'm curious how much ofthese, the things that we see in
the second screenshot, isactually out of the box
components, because I don't workwith Canvas apps a lot either.
So for me it's still one ofthose things that I need to-box
components because I don't workwith Canvas apps a lot either.

(11:05):
So for me it's still one ofthose things that I need to dive
more into and I can't wait toactually have a Canvas app
project on my hands so that Ican dive more into it.
But also it's funny because inhis post he says so repost.
Can we stop with first-gen appsfor good?
And you make an interestingpoint because canvas apps and
the design around it right, it'sthe same one for custom pages,

(11:28):
so the components they crossover right we should yeah yeah,
so then you can't really removeit because you're dependent on
the old kind of interface tomake it look like a model driven
app, for instance, you stillhave to have some of that so
that it can align.
Because I completely agree withyou, if you suddenly draw or
you put a modern UI canvas appor custom page into a

(11:51):
model-driven app, it will lookvery odd.
So you need to make sure thatthe UI they have in model-driven
apps they can kind of matchthat in canvas apps as well,
because it's an important thing.
So it is not to say that we canjust kind of get rid of the
first end all together, uh, butsure, moving in that direction,
I think, yeah, absolutely forsure, yeah, so, uh, next on the

(12:16):
list is and I hope that you knowmore about this than I do,
because this is where I'mgetting, I'm walking on thin ice
PowerFx functions used to becalled local plugins, which
didn't even get it make it outof preview before they changed
the name.
So if you see functions orPowerFx functions, it is what

(12:38):
used to be local plugins and I'mtold that kind of the plugin
words scared people off becausethat's so pro-codey that they
removed it purposely to call itfunctions, and now you can also
reuse them across canvas appsand and and power automate.
So how much have you kind ofworked with this nick?

Nick (12:59):
a little bit at the start, when they first came out,
definitely played around with itand I did find that there was
some limitations.
Um, it became very obvious thelimitations compared to what we
would call traditional pluginsthat we would write in c-sharp
code and dot net and of coursethat those are those are trigger
words for people that are verylow code.
Um, but I could see where theevolution is going.

(13:22):
It was pretty exciting becauseI thought, wow, if we can get to
a point where I can just gointo the regular maker portal
and actually build myfunctionality using Power Fx.
Not that I don't like writingplugins, but in my particular
role I haven't written.
Honestly and we've talked aboutthis before I haven't written a
plugin in like four or fiveyears.
I haven't written a plugin inlike four or five years.
So because I've been involvedin projects where I've had other

(13:45):
access to other developers, Iunderstand how it works.
I could write a plugin if I wasforced to, but it is something
I don't do every day, so it's acase of you know that.
Okay, I need to relearn orremember stuff to get into it.
Power effects functions isgreat because if you're already

(14:07):
using power effects in differentareas and I'm beginning to find
that, yeah, you know we talkedabout canvas apps previously,
but power effects is nowbeginning to pop its head a
little bit into power pages.
It's uh, showing up in thecommanding when we're doing the
command bar stuff, uh, and ofcourse, in, you know, in the
custom pages as well, there'spower effects.
Again, we can extend this.
So, yeah, I did a video onusing the low-code plugins in

(14:31):
Power Pages and how all thatworks, probably last year or the
year before, if I remember.
But now seeing what's comingwith the Power Facts functions
and also seeing Matthew's poston being able to call these
directly from a flow, that'salso pretty exciting, because
now we're getting into the worldwhere we were before with
classic workflows.
With classic workflows, you'reable to create what they call

(14:54):
workflow plugins or somethinglike that.
It was a special plugin thatyou would call from a classic
workflow, where it'd be this,this whole world, and that
opened up a lot of opportunities.
This is now now being broughtinto Power Automate as well, so
you can write a Power Fxfunction that's reusable and you
can actually call it from aPower Automate flow, which is
really cool.
This, to me, is we're just onthe tip of the iceberg with some

(15:17):
of this stuff.
We even have our friend, likeSarah Lagerquist has done
presentations I think she didone in Tallinn on the new power
effects functions as well.
So we're really getting.
You know, we're bringing a lotof that power that would before
was very much, uh, limited tothe pro developers, bringing it
into the hands of the low codemakers, which is great, but in

(15:38):
leaving and again we still needdevelopers and letting
developers again focus on thosemore advanced scenarios.
So again it's just sort of ashifting of work, because
developers are really good.
Developers are hard to find.
I do know very good developers,but they're always super busy,
they're always tied up in otherprojects.
So the more we can kind ofbring some of that down to the

(16:00):
regular low-code maker folksPowerFx functions is a great way
to do that.
So I'm excited to see theevolution of this.

Ulrikke (16:07):
Yeah, me too and also the fact that you can add it to
solutions.
I think that's also one of themajor things, right, so you can,
and I tried it out.
I just went into one of thesolutions I had.
I took add new and thefunctions.
They're in the automation thing.
It showed up and I can selectsources, I can select inputs and
outputs and I can choose whichtables in Dataverse, for

(16:27):
instance, that they're going toconnect to, and then I can write
my Power Fx function.
And that was really neat.
Even I could do it right.
So what I was missing?
Actually, for once and hold onto your socks For once I missed
the Copilot thing, because I'mnot good at writing Power Fx.
For once I missed the co-pilotthing because I'm not good at
writing power fx and I'm used toworking with power automate
where the prompt thing asks mewhat is it you really want to do

(16:47):
?
And I say I want to calculatethese two fields together and it
says here you go, and it givesme the prompt, it gives me the
power fx function.
It was not, it wasn't in there.
So I was like oh no, now whatdo I do?
I have to write this power fxon my own from scratch.
Where's my Copilot?
And then there's a first timein forever that I had actually

(17:07):
missed the Copilot.

Nick (17:09):
So there you go, there you go.

Ulrikke (17:12):
Even for me there was hope.

Nick (17:14):
You had to write your own code like a chump.
I know right when do.

Ulrikke (17:18):
I start, I don't even know where to start.
Oh, it's so much fun, but nowyeah, of course.
This is the preview, though.
So kind of, use it with care,don't use it in production and
do what we say, not what we do,et cetera, et cetera.
My dad always used to say thatwhen I was young, when I was a
kid, and I said, well, you do it.

Nick (17:41):
He said do like I say, not as I do.
You did it.

Ulrikke (17:43):
I made a vow that day to never say that to my kids,
but maybe, maybe, and actuallysomething else, okay.
So, talking about things thathave no idea what it means, this
is one Okay.
So this is for all you Dynamics365 Customer Insights people.
I saw a post this week fromMegan Walker on LinkedIn.

(18:04):
It was a post and it just said,finally, and there's a
screenshot of something incustomer insights and the whole
comment section is a people inSwedish and Danish and Norwegian
going it's here.
Does that mean what I think itmeans?
And people go, yes, it'sfinally here.
I have no idea what it is, butI thought if this many important

(18:25):
people thinks this is awesome,I'm going to highlight it.
So this is what the screenshotis.
You have customer insights, amodel driven app, open.
You're in the customer insightsyou have on the far bottom you
have assets libraries, emailtemplate, task templates.
You got form templates.

(18:46):
You open that and then you canselect active form templates on
the tops.
You kind of have a viewselector thing at the far top
and you can see galleries andcustom templates and favorites
and stuff.
Isn't that fantastic?
Yeah, I mean nick.
Finally, finally, it's here itis.

Nick (19:09):
It's funny.
I like megan who's like hooray,and then I f?
K, you know, I f?
Y k, y k, like, if you know,you know, like you know, you
know, I'm not sure if there'sthat.
There's that scene in thelatest spider-man movie where
the three Spider-Mans from thedifferent multiverse are
together and the Tom HollandSpider-Man says he says oh, I
was once part of the Avengers.

(19:30):
And the Tobey MaguireSpider-Man was like wow, that's
so cool.
What is that?
What is that?
That's how I feel.

Ulrikke (19:41):
Yeah, yeah, that that's how I feel.
Yeah, yeah, no, so so let'sjust, you know, just be a part
of it, like whoo finallycelebrate the wins.

Nick (19:50):
Yeah, whoa awesome awesome .

Ulrikke (19:54):
It has a lot of comments.
I love it.
Go marketing people.
Sorry, customer insights people.

Nick (20:01):
And this goes like in all seriousness, this goes to show
the breadth and the depth of thePower Platform and again, we've
said this before, you cannot bean expert in the entire Power
Platform.
There are so many areas.
Market like, customer insightsand journeys is one of them and,
as you know, because I've beenranting about it, I've been

(20:23):
fighting with some of thosesolutions this week in our own
project of things that aregetting deprecated and re-added
and trying to introduce allsorts of dependency issues.
So I'm seeing a bit of the backend of all of this going ah.
But again, it shows that thisis a very important piece of the
Power Platform.
It's used by a lot of people,it's very powerful, what it does

(20:45):
, I mean, I understand what itdoes, but just the whole getting
there kind of thing.
So, anyways, this if, if Megansays this is awesome news, it's
awesome news.

Ulrikke (20:54):
It's awesome news, exactly, yeah, and and the the
thing that you're now bashingyour head against that.
You inherited that from me andyou've gotten further in a week
than I did in three months, sokudos to you.

Nick (21:06):
That was me having a meltdown on Microsoft support
yeah, yeah, yeah.

Ulrikke (21:10):
And I watched that from afar thinking, oh, I'm so happy
that that's you and not me,because I am Norwegian and you
are Canadian and they are luckyit's your call.
And then I counted slowlybackwards from 10.
So let's move on to anotherkind of existential question,

(21:31):
and I feel Sean Asterkin is fullof these lately.
So this is another oneSharePoint.
And Thomas Sunser, please takea breath, we're just kidding.
Sharepoint versus Dataversewhat do you think?

Nick (21:51):
So I mean, I am very much the like friends, don't let
friends use SharePoint as a puredata source.
Now I get why.
Um, and here's, here's the tome the sort of dividing line and
sean and and we love you, sean,uh, but you are a bit of a shit
disturber, you know that, right?

(22:11):
Um, I think it does it onpurpose, to be honest.
Oh, totally, oh, yeah, yeah,but that's, that's how he rolls
man.

Ulrikke (22:18):
So yeah, it's good marketing.

Nick (22:19):
You get mentioned on podcasts yes, there's no such
thing as bad breasts I hug outwith shot in chicago at ignite.
Uh, this is he's.
He's a super, super smart guy,very like.
He just fits in with our crewso well.
Um, he helped out with the labs.
He was one of the proctors, um,very knowledgeable, uh.
But yeah, he does kind of throwlike there's certain topics

(22:41):
that it.
So here's a here's next lessonon how to get a lot of things on
LinkedIn Just post somethingvery controversial, just put
managed versus unmanagedsolutions.
What do you think?
And then just get everybody topile up and do whatever he talks
about.
You know, and my thing withwith Dataverse, if you're
building anything enterprisewhere it's in a company, even in

(23:02):
a, you know, even a smallmedium business, if you're
working in a team where you'reworking on an application that
needs data storage and needssecurity, dataverse is the
obvious choice.
As soon as someone startssaying, well, price and
licensing, like, come on, it'sless than a cup of cost of a cup
of coffee per employee permonth, per day, per month, kind
of thing.
So again, it's like we saidlast episode it depends, but for

(23:28):
the most case, if it is anenterprise or business solution,
dataverse is the way to go.
If you're building a power appmore for yourself.
And again, this is where, ifit's more of a personal
productivity type app thatyou're building, then, yeah,
doing it against a sharepointlist is probably very acceptable
and that's and I know that'sblasphemy from based on things

(23:48):
I've said before.
Um, but really you know theright tool for the right job,
even if it's an excel onlinelist, if it's something very
simple.
But as soon as you startexpanding to more than a few
people, where there is someissues, where you want to make
sure your data is protected dataversus the way to go.
Now sean does have a lot oflike pros and cons listed here.
You can read his post on thatum, contribute to your own um,

(24:11):
your own ideas.
He even has a little chart here, the pros and cons and the
whole little you know gartnerstyle chart there.
So, yeah, um, very, it's alsoone of these things that, like,
really it really comes down to.
It depends on what you'rebuilding.
That's my attitude toward it.
If I build, if I'm buildinganything in power apps for me,
like for customers, or even manyprojects.

(24:32):
I'm using Dataverse becauseI've been working with it for 20
years.
I know how it works.
I know how the security set up.
I know the table relationships.
I know all of this stuff.
Sharepoint to me.
I use it like document storage,those types of things,
intranets I know it has.
I know SharePoint.
When you say SharePoint, itactually has a whole bunch of
different things it can do.
There's things that it used tobe able to do that it can't do

(24:53):
anymore.
So it in all in all the end ofthe day, right tool for the
right job.

Ulrikke (24:59):
Yeah, exactly, and I used to be a SharePoint
consultant in my previous lifeand even I know that and I've
seen some crazy shit.
To be honest, I mean peopletrying to creating relational
databases using SharePoint.
I mean SharePoint lists aren'ta database.
People Come on, even though youhave rows and columns, it's not
a relational table.
So I mean, but you can do it ifyou know how to work, create

(25:23):
workarounds.
Even last week I saw a customerwho they are very cost conscious
and he's created so many coolapps and solutions using
sharepoint as a database.
It works until he reaches a2000 role limit and then he has
to do some creative something,something, and I'm in awe of
this guy, what he's actuallybeen able to do.

(25:43):
But I'm also worried.
It's like you said, I I heneeds more bandwidth because now
he's he's the only one whoknows how any of this works.
Sure, and then last weekend,you know, last summer, when he
was on holiday in greece, theyhit the 2000 limit and he had to
, you know, kind of grab hiscomputer and get off from the
beach because he was the onlyone who could do anything about
it.
So I mean there are things that, um, that you should use

(26:07):
SharePoint for, but as adatabase, uh, no, it's a no-go
Documents, sure.
So, yeah, right thing for theright, uh, right tool for the
right job.
I think that's kind of whatwe're preaching, but it's fun to
see this kind of side-by-sidecomparisons what is the pros,
what is the cons, and how tokind of know, and from a high
level.
If you have no idea what any ofthis is, of course this will

(26:30):
help.
So, good job, sean, and keepthe PR stuff coming.
We'll continue to shout yourname in this podcast.
Also speaking, I think this isgoing to be the episode called
lurker talks about things shehas no idea about, but things is
really cool because corepalettes and ai guys, um, I

(26:52):
tuned into powerful devs lastweek, um, and I saw, for
instance, I saw this and, firstof all, uh, uh, scott and kendra
very good job at kind ofbringing everything together,
and I saw scott's video frombehind the scenes.
That was a lot of fun to seehow all this comes together.
Robin did a huge job at kind ofmanaging the back office.

(27:14):
It looks so much fun, um, andalso, if you're kind of, and
manganus did something on, uh,did a session or something, and
we had Leon coming on announcingnew capabilities for PCF
co-pilot APIs, for instance.
I don't know, the first thingabout it.
So I kind of dove in to seewhat this is.
Okay, PCF co-pilot APIsannounced new co-pilot APIs for

(27:40):
PCF and client API developers.
So, from what I can understand,this is going to enable pro
developers to create inlineAI-driven experiences powered by
Copa, the studio agent, in aPCF component, Right.
And then so there's anannouncement.

(28:03):
There's a guy called, yeah,Hermant Gur.
He posted on LinkedIn, kind ofgrabbed the announcement from
Leon from Powerful DevsConference, pulled it into
LinkedIn and kind of talked alittle bit about it and then
also highlighted the URL for howto sign up for the private

(28:24):
preview of this.
So you would get.
We have the links in our shownotes.
If this is for you, you cansign up for the private preview
now.
But even with his descriptionand Leo's description, I don't
know PCFs.
I'm not that familiar with CopaStudio yet to understand what
this can unlock.
And also, looking at thecomment section of the post that

(28:45):
Hermant posted, nor can anyoneelse.
So it's one of those thingswhere everyone goes what Whoa,
Awesome, Fantastic, Whoop, whoop, whoop, what can we use it for?
What can we use it for?
And there's crickets and I go.

(29:06):
No wonder, I don't understandit.
So I'm like and I love theseannouncements, right, and it
kind of baffles me that there'snot a lot of use cases out.
And actually someone also wrotethat as a comment on Herman's
post Like great, what are yougoing to use it for?
And he's like I don't know.

(29:32):
So this is the challenge for meto all of you listening If you
can tell me what you can usethis for and no, Scott, you
cannot answer then let me knowand I'll talk about it on the
next episode.
How's that?
Because I'm really curious andthis is a way for me to learn.
What is the big deal?
What does this enable?
What does this unlock that wecouldn't do before?
Help me understand, please.

(29:53):
So I'm looking forward togetting smushed by a million
gazillion messages from peoplegoing.
This is so obvious.
This is the use case, this iswhat you want to use it for,
this is what I couldn't dobefore that this enables me to
do.
Et cetera, et cetera, et cetera.
Right go, people go yeah go.

Nick (30:13):
I mean I have some ideas, but uh, I haven't shoot.

Ulrikke (30:17):
I didn't know you worked with this so much, so you
kind of had this.
So no, no, I don't.

Nick (30:21):
I just I'm.
I just say, in terms of ideas,let's say, like all I can think
of.
Let's say, if you're doing apcf control, let's say, sir,
like a like, think of a pcfcontrol that we've used before
rich, rich text editor, forexample.
That's a PCF control that youapply to to like a text there.
Potentially could you actuallyhave that tied into AI, to a

(30:43):
co-pilot, to actually, you knowthe help summarize this text or
help me write this text out alittle bit better.
Or you know, again, I'm justI'm spitballing here because I
don't write.
Like I said we talked about, wehave to learn how to write PCFs
.
This is sort of the first thingthat kind of comes to mind,
because even now, built intoyour browser, I have this little

(31:05):
co-pilot.
You know you want me to cleanup your text or summarize the
text that you just wrote.
I could see that being builtinto something like that, being
built into a PCF control or youknow so, either some sort of
calculations or, you know, beable to paste in something,
extract certain values out of ablock of text.
You might paste into a control,those types of things.

(31:28):
Again, I don't know, maybe I'mcompletely off base here, but
that's just sort of the ideasthat are beginning to pop up in
my head.
On building something like this, building that within a model
driven app, um, you know, tryingto do some of the stuff that
I've been trying to experimentwith as well, um, and I'm still
at a stage where I'mexperimenting with stuff.
I'm building some stuff and I'mlike, is this really the right

(31:49):
tool there's?
There should be a better way.
This, this to me seems like apure power automate thing or
that kind of thing.
So that's just kind of goingoff the co-pilot studio side.
But, yeah, I think it.
I think this is a gooddiscussion to have.
Yes, there's a lot of greattech, but what is?
Give me what you can do withthis.
Look at, like, customer serviceand what they've built in there

(32:11):
in terms of being able tolocate and find information,
whatever.
Yeah, that's great, I get thattotally.
But what are some of the other?
What are some of the day-to-daystuff that we can use this
stuff for?
And I think that's where weneed to be right now to get
people really on board andexcited about this.
The days of demoing, the yes,you can do.

(32:31):
Here's what Copilot can do on aconference stage.
That, yes, you can do.
Here's what Copilot can do on aconference stage.
That's great.
Give me some real world, andbeyond the customer service side
and beyond the sales side.
Yes, those are great scenarios,but what else can we do with
that?
We've talked about this too inour own projects in terms of
volunteer matching and that kindof stuff.
I think that exists.
But there's where you take thatscenario and, yes, we can begin

(32:52):
even to build that into the appa little bit, potentially with
some of these controls, I thinkyeah, but I'm also waiting for
the customers to be more in linewith this and be more mature.

Ulrikke (33:03):
I can't tell you how many times the last year I've
I've demoed the simplest usecase, the thing that ai builder
could do for years and yearsalready.
And the customer goes awesome,yeah, that looks so cool.
Let's put that on ice for nowand just let's get the MVP up
and running first, and thenwe'll readdress the AI later.

(33:25):
And I go please don't, becausethis AI builder thing is going
to save you and your end usersso much time.
If we start with this, that toimplement it later means we have
to do the same work twice.
Why wait?
And they go ah, well, you knowit's, it's us.
We're not really that mature andwe're not sure if our users are

(33:47):
ready for ai yet.
And it's like well, actuallythe user don't know they're
using AI, they simply upload animage.
The work we do is on the back,and they go no, we're not ready.
It's like well, where do you gofrom there?
I mean, it frustrates the crapout of me, because I want to, I
want to, I want to start.
We'll start working with thisright.

(34:08):
And it's only so far I can go onmy own.
I have implemented a few thingson my own and it's fun and all
that, but it's that's not wherethe the learning is for me.
Going like you said, goingthrough the learning material
fine, uh, setting up some smallthings for you on your own,
doing it on a hackathon fun.
But actually when, when thekind of the rubber hits the road
in a real world project whereyou hit those little

(34:30):
nitty-gritty, that's where Iactually learned the most and
it's kind of those little nittygritty, that's where I actually
learned the most and it's kindof it's you can't get the boy,
my biggest challenge with all ofthis, and I think we might've
talked about this before.

Nick (34:45):
So, for instance, let's say I so, for for those of you
and I do want to talk some aboutthe MVP program.
But we're on distribution lists.
We know discussions onparticular product groups.
Every day there's adistribution list.
Emails come in, people askingquestions, product managers
clarifying things, uh that kindof thing.

(35:05):
Very interesting, a lot ofgreat information.
But these as you, you knowyourself, these distribution
lists can result in dozens, ifnot hundreds, of emails per day.
I don't have time to readthrough these emails.
Oh, wait a minute.
I can use Copilot to summarizethat for me.
I asked Microsoft 365 Copilotto summarize these emails.

(35:28):
It keeps failing.
It comes back and tells me withthe utmost confidence that hey,
there has been no emails sentto this distribution list in the
last 24 hours.
There's a hundred emailssitting there.
I can see it.
I have it in a folder.
I even tell you the folder.
And the thing is, if I wrote aPower Automate flow and then it

(35:49):
would give me an error that itcouldn't find it or couldn't
read it, so I know there's anerror.
Co-pilots and these AI tools,they have that confidence level
that it's like, oh, nothingexists.
So, as an end user, I'm like,okay, nothing must exist,
something truly exists.
So this is where I think thisis some of my.
This is, I know, going down adeep rabbit hole here, but this

(36:10):
is my frustration with some ofthese tools where, especially
the AI, the co-pilot there isthis implicit arrogance of, yes,
this is how it is, and it'slike, no, that's not how it is.
So, from an end customer'sperspective, you want to get
them on board.
You've got to start buildingtrust in these tools and if
something, if co-pilot can't dosomething, it's got to come back

(36:30):
and say, hey, I'm sorry, Ican't read I.
It's got to come back and say,hey, I'm sorry, I can't read, I,
I couldn't find emails therecould I can't read the emails as
opposed to saying nothing?
there's nothing there.
Yeah, yeah, I know, bries mebananas and I even pinged.
I know, daniel, I sent you thescreenshots and like how the f
do I get this working and evenanyways.

(36:50):
So that's just sort of I knowI'm kind of I love these tools.
I love, like I've been, I havesometimes I had this you know,
this comma separated file, athroat hit copilot or chat GPT
and say format this into a nicetable.
It will do all that.
Um, I put together a whole bigtraining plan.
I want to cover this, coverthis, put it in a training plan,

(37:12):
break it out.
The number of weeks.
Do this, do that?
Like this kind of stuff.
It's great because you knowthis is that that part of AI
absolutely love.
It's actually saved me tons oftime.
You know, putting togetherplans, putting in agendas,
reiterating on that Again.
For those of you who don'tsubscribe to Donna and
Jeremiah's third rodeo weeklynewsletter, it is so good.

(37:34):
Every week they come withdifferent scenarios for stuff to
try and to learn.
I've learned so much from theirnewsletter.
We should probably put a linkinto that again.
Those types of things are great.
But now we're into thescenarios where we're trying to
build these and trying toincorporate into our own tools
and it's, it's, it's a struggle,the real that part.
So it's struggling for us toget things working from an end

(37:56):
customer.
They're not seeing the value,because the value is it's, it's,
it's it's, I won't say broken.
In certain it's working, butother areas it's not where we're
being marketed, that it isfantastic.
Sorry, that was a.

Ulrikke (38:10):
I know I'm all over the place on this one, but no, no,
it's absolutely fine, and I,because I think a lot of people
will uh find this useful becauseit kind of speaks to what
they're going through, uh, andwe, we are honest about these
things.
We don't give you just asugar-coated marketing message,
because we live and breathe thisevery single day and actually
just let's just kind of pick upthat a little bit, because we

(38:33):
are talking about powerful devsand we're talking about
hackathons and how to make AIwork, and there is going to be a
powerful devs hack for a coupleof weeks, so it's actually
going to it has started thisweekend, this past weekend, I
think you can still sign up.

(39:00):
Yeah, and there's the deadlineas well, of march, isn't it?
Yeah, so we're going to put alink to it.
And also on the powerful devsconference, robin talks about
the hack, so I'll just put alink to that in the show notes
where you can sign up and what'sexpected and what you need to
do.
There needs to be a submittedvideo, kind of going through the
solution that you made, and Iread somewhere, nick, that you
will be a judge.

Nick (39:22):
Yeah, and you're a judge as well.

Ulrikke (39:26):
Yeah, I was asked to be and I was a bit late responding
, so we'll see.
Maybe I'm a judge.
I didn't really hear back.
So I'm just pinging Danielielas we speak and he's like I'm
gonna check and then we'll see.
Maybe I'm a judge, maybe I'mnot.
Um, okay, yeah, that's going tobe a lot of fun, even though I
hope I get in and I hope I canbe a judge to see all the crazy

(39:46):
submissions coming in.
Um, because also for us it's ahuge.
It's a nice way to kind of seewhat people are actually working
with, what's the use cases,exactly the way we talk about.

Nick (39:56):
And real world use cases.
It just shut us up.
If you can come up with realworld use cases that are awesome
, I'm more than happy to like.
We're more than happy to showthose off and tell the world.

Ulrikke (40:06):
Oh, definitely so, if we're both judges and or, of
course, even if you're just ajudge.
And then the next episode,after we're finished, let's
highlight the top things that wesaw, because that's, that is
exactly kind of the solution tothe things that we've been
ranting about this episode, Ithink, isn't it?
I think?
That's pretty cool absolutelyokay, so we're kind of running

(40:29):
way over.
Uh, you have a couple of thingshere okay, yeah, yeah, there
was something.

Nick (40:38):
Just I won't dive too much into it because, again, if
you're building, you want tolearn, if you're more of a pro,
from a pro dev aspect of it,using ai foundry, there is, uh,
building ai, ai agents forbeginners.
I think donna posted thatsomewhere and it's actually a
github repo that has a bunch oflessons to go through.
And that's more going beyondthe Copilot Studio side.
This is going into the Pro Code.

(40:59):
That's your jam.
Check that out.
I think that looks prettyinteresting.
Speaking of building AI agents,our good friend, lisa Crosby
posted a video this week.
I love her videos.
She is my calming voice on allof this because I'll watch her
stuff and it makes sense to meafter watching some of her
videos.
She, she is my calming voiceand all of this because I'll
watch her stuff and she can makeit makes sense to me after
watching some of her videos.
So she's done one of buildingai agents.

(41:21):
Which tool should you use?
Um, including the pro dev stuff, which I know lisa's not.
That's not exactly her comfortzone, but she does a good
overview of all of that, so youshould check out that.
I mean, you should check outyour her youtube video.
She, you should check out yourher YouTube video.
She has almost like a 70 oralmost 80,000 subscribers, so
you're probably alreadysubscribing to your, uh, if
that's your thing, um, and thenthe other thing, um, yeah, maybe

(41:45):
I'm going to.
I'll.
I'll sort of save this.
There was a post by Agnes, uhBarnicus, um, about so, do you
want to become an MVP?
And that it's funny becausethat generated a lot of chatter
and I still go to my LinkedInand that was posted a couple of
weeks ago and there's stillcomments being generated and all
of that.
Maybe this is something youknow, because that could

(42:08):
actually go into another rabbithole of things.
But to really summarize,there's been a lot of, there's
been some posts about how tobecome an MVP, what the roadmap
should be, trying to treat itmore like a certification or
something that you can achieve.
That's the wrong way to thinkabout it and that's really the

(42:28):
focus of his posts.
To me, getting the MVP was morethe result of work and what I
posted there and I've said thisbefore to become an MVP is it do
you want to become an MVP or doyou want to do the stuff that
an MVP does?
I want to do the stuff that anMVP does.
Even when I was at Microsoft, Iwas doing it because I wasn't
an MVP.
Then, even if I was no longeran MVP, I would still want to do

(42:50):
the stuff like the podcast, thepresentations at events, doing
the blogging and things likethat.
Mvp yes, there are benefits.
You do get some stuff.
There's, you know, a lot ofcool things that comes with that
program but it isn't so muchsomething you can work.
It isn't something you shouldbe working towards.
It should just be somethingthat comes to you.
So check out that.

(43:11):
We'll put the link into thatpost and maybe we'll pick up
more in that conversation later,because I think this is a
bigger, a bigger thing.

Ulrikke (43:18):
I don't know if you wanted to add anything to that.
Yeah, just the benefits.
I mean, come on a few.
What credit?
A few azure credits, a few umlicensing things?
If you accumulated the timethat I spend on the content,
that makes me an mvp over theyear.
If, if I did that, if I spentthose hours doing my job over
time, I would earn, I could buythose benefits a million times

(43:42):
over.
It's not about the benefits,you guys.
If you're not doing the thing,that means you're doing enough
to be an MVP.
Then you shouldn't be an MVP.
It's not worth it.
The majority of us, we get tobe MVPs because we do this
anyways, because we love doingit, not because we're called

(44:07):
MVPs for it.
And that's how it should be.
I know that there's some chatterabout how the whole program is
watered down and how it's waytoo easy to get in now and
what's happening with the wholeprogram.
I don't know.
I don't have any insights intothat, but I know that some
segments, some part of the world, some demographic it's easier
than others.
If you're a white man in the US, probably going to be harder,

(44:27):
because I think that bucket isalready kind of full.
So I mean sure things like thatapplies because marks don't
need to be diverse um.
So there is some in kind ofimbalance there.
I think that's where some of thethe chatter and the comments
are coming from.
Um, but, as I said, from aneconomic point of view, or a
kind of a um um, what you getfrom, from what you give point

(44:51):
of view, if that's the wayyou're looking at it, it's not
worth it really.
Um, make sure you do it for theright reasons and that you're
willing to keep it up.
Anyone can become an mvp andrenew, but if you don't do the
work over time, you're not goingto be even considered.
So yeah, that's my two cents onit cool.

Nick (45:11):
Um, now, as we should wrap up, speaking of events, um,
there's, of course, as much.
As I said I wasn't going to doas many this year.
Uh, here we go.
Um.
So march 21st, 22nd canadianpower platform summit happening
in vancouver.
Still seats available, eventheir seats available in the
workshops.
Use PP Boost 15 to get 15% off.

(45:35):
Still amazing deals.
I think the co-pilot studio one, speaking of which my gut
feeling is that will sell out,so you might want to consider
getting on top of that.
But there's other great ones aswell, on Power Facts and ALM,
so great opportunity to do somelearning.
If you're in Vancouver orwanted to do a trip to Vancouver

(45:55):
, I'll turn the next one over toyou, because that's the one
you're going to in April.

Ulrikke (46:01):
That's ColorCloud.
It's April 24th to 25th andthen, of course, we have a
workshop day before, so that's apre-day workshop.
On the Thursday I'm doing thePower Pages from Creation to Go
Live whole day workshop on myown.
I'm so excited.
We have such a great curriculumgoing and some visitors from
some good friends of mine tohelp us get through how to make

(46:24):
a Power Pages fit for Go Live.
And then I'm doing a sessionwith Andy Wingate on Business
Central and Power Pages togetherand that's yeah, that's going
to be awesome.
I'm so excited.

Nick (46:36):
Yep and then.
So yeah, I'll be missing out onthat.
I'll have the major FOMO thatweekend, for sure, I know.
And then in mid-May,dynamicscon happening in Chicago
, I have a session, crash coursein PowerPages or Power, not
power pages pipelines.
So that was something we'vebeen dealing with a lot with.

(47:00):
I know there's a new learningpath from Daniel that we didn't
I think it was in our list, wedidn't mention it.
So lots of resources.
But you know, check out mysession.
I can tell you some of the goodand the bad and the ugly about
pipelines.
And then I'm teaming again withAngeliki on this.
We're doing the same redoingthe session that we did in

(47:22):
Prague about, uh, the ultimatepower slam, dynamics 365 versus
power pages, kind of a veryunique way of kind of where to
use one versus the other, wherethey work together and that kind
of thing.
So that would be prettyexciting.
Um, on that one Um, and thenthe next one, uh, happening a
few weeks later, dynamics, minds, um, and I have a session on
burnout, powerlifting and mentalhealth.
So again, I've talked about alittle bit about this one.

(47:42):
This is one I'm a little bitnervous about, but uh, should be
good.
And then I'm teaming with, uh,matt's necker and George
Dubinsky on a session on uh, wecalled it the portals cage match
, uh, power pages versus theworld.
So this is going to be.
I think that to me is worth theprice of admission the three of
us in that one, if I was to goum.
And then we found out this week, um yeah, european power

(48:09):
platform conference.
Uh, we have a session together.
Um, we, I have my anothersession as well on power pages,
and it looks like you and victorare doing a workshop, but I
think I'm helping out with thatone as well.
So, yeah, what's going on?

Ulrikke (48:26):
yeah, one of those things are do you realize that
you have a workshop withvictor's like?
No, do I?
It's like, and I'm going tocheck and go, nick, do you
realize that you're on it too?
You go what, go what, I'm on it.
It's like oh, okay, someoneneeds to start getting on top of
their emails, I think.
But yeah, we're doing a workshopon Power Pages with Victor,
expect to see those top gunsuits because we need to wear

(48:47):
them enough to kind of, yeah,keep it going so definitely, so
definitely looking forward to it.

Nick (48:58):
It's going to be fun.

Ulrikke (48:58):
You and me get to do the top tips and tricks for
Power Pages together againthat's always fun, so that's
kind of my highlight, eventhough, yeah, I love doing the
Top Gun thing with you andVictor.
The 10 tips and tricks forPower Pages is the one I'm
looking forward to the most coolokay next episode March 5th.
March 5th.
It scares me to see that.

(49:18):
Okay, it goes so fast I'm blownaway.
Okay, this is a long one, youguys, a long one where I talk a
lot about things I don't knowanything about.
So please help me out andcomment on the thing where you
saw this thing, and call me outand tell me all the use cases,
please, for the things I don'tknow.

(49:42):
Thank you so much for listeningand we'll catch you next time.
Bye-bye, thanks for listeningand if you liked this episode,
please make sure to share itwith your friends and colleagues
in the community.
Make sure to leave a rating andreview your favorite streaming
service and makes it easier forothers to find us.

(50:02):
Follow us on the social mediaplatforms and make sure you
don't miss an episode.
Thanks for listening to thePower Platform Boost podcast
with your hosts, ulrika Akerbeckand Nick Dolman, and see you
next time for your timely boostof Power Platform news.