Licensing - Project Oakdale

[baileyscott]

I've been experimenting with Project Oakdale - PowerApps created in Teams. I'd read so much about Project Oakdale in various articles plastered all over the internet and I was excited to read rumours that users would not need the per app or per user license. 

My investigations discovered that actually there are no changes to the licensing of PowerApps created in Teams. Premium Connectors still require the per app or per user license. I was hoping that Microsoft would have at least allowed a Standard Connector to CDS or to Azure SQL.

I'm frustrated. Everytime I wow my customers with PowerApps and then explain the licensing the project dies a death. The licensing cost is prohibitive for my customers. 

Can I suggest that a positive way for customers to move from PowerApps Portal to PowerApps created in Teams is to include CDS or Azure SQL as a Standard Connector. We need a relational database Standard Connector. Microsoft make money from CDS capacity and from Azure SQL so surely making it a Standard Connector would increase usage and adoption?

Without these changes to the licensing I'm afraid I see no point in Project Oakdale. 

Please vote for a Standard Relational Database Connector in Project Oakdale so that we can increase adoption and usage.

Thanks!

[GarryPope]

I whole heartily agree. Great suggestion. If CDS was a standard connector, we'd pump all our data into CDS, which would increase the storage size, which would require the purchasing of more storage. It's a win for everyone.

Project Oakdale will make a difference, as we'll no longer have to bend SharePoint to breaking point. But having it in Teams means the database is siloed, meaning the real benefit of having all company data in a single CDS environment is impossible.  

[PaulD1]

Without a sensible database option Power Apps is too limited to build anything truly useful - if your data is important, you shouldn't be storing it in SharePoint lists, if it is important and you have more than a handful of users, you are better off paying for development up front in more flexible (license-free) tools rather than paying the ongoing licenses to connect to real databases.

I've up-voted the idea even though I think CDS is a a much poorer option than Azure SQL. CDS it is not properly relational, not supporting joins across multiple entities within Canvas PowerApps or WorkFlows and so encouraging redundant relationships and duplicated/inconsistent data. It also incorporates non-relational 'additions' such as multi-value fields, option sets and polymorphic relationships which make it quicker to design build databases but store up all sorts of issues for the future.

[JC75NATS]

Agree - although this about the full licensing costs, i work for a average size company, and PowerApps has been well adopted, but is now being curtailed because of license costs for what is often a relatively simple app, but needs to connect to CDS or Azure SQl  - its often cheaper and probably better to outsource this as an external app development not using PowerApps, but as that often then as its own challenges to get off the ground results in it not occurring either. 

So at the moment we have a simple solution that at its heart should enable this community and quick development but is totally cost prohibitive, because company isn't prepared to pay on top of the office e3 licence a further 40 dollars for every user, some of who won't use many apps.. roughly a million dollars a year so people can have some simple apps is clearly not going to happen especially in todays climate.

M$ simply need to be smarter with licensing and realise that a free connector to Azure Sql, CDS and instead charge for storage will result in a  proportionate charge that companies can actually justify.

[Craig_Humphrey]

Isn't one of the points of Project Oakdale, that there's a "lite" version of CDS available at no cost? (you know, Dataflex vs Dataflex Pro), or have I missed something?

[PaulD1]

@Craig_Humphrey 

A few issues with Project Oakdale as I understand it:

1. It is only in Preview, not ready for production use.
2. There are features missing in the preview (according to forum posts I have seen) so not sure if these will be supported (think these are primarily mobile features such as GPS, accelerometer, SaveData/LoadData).
3. It can only be used from inside Teams.
4. I haven't seen any commitment (may have missed it) that this will not require additional licensing in the future. 

[Craig_Humphrey]

@PaulD1 

Those aren't issues, those are the boundaries of it's current state.

1. Being Preview is not an issue, it's just where the product is up to.

2. It's not unusual for Previews to be missing functionality, if they delivered it all at the start of a preview, there would probably be more bugs...

3. The WHOLE POINT is that it can only be used inside of Teams.

4. This is hardly an issue, MS almost never commit long term to a licensing structure.  It's not uncommon to start out with a free product and charge later - MS are not alone in that.  And while that's not a great thing, it is, what it is.  If you decide to implement some mission critical application using the latest from a vendor, then you're the one taking the risk that the landscape may change...

I'm not saying that Project Oakdale doesn't have issues, I'm sure it does, but during it's Preview, that's the time to feed those issues back to MS and get some votes behind them to MS actually take note...

<rant>The original post is a gripe that Project Oakdale didn't change the licensing, but from what I'd heard and read, it never promised to.  So this isn't really about Project Oakdale, it's about MS' approach to Power Platform licensing in general, which is a huge issue and there are dozens of posts in here about it.  Vote for them, add your comments to them, don't fragment and diminish the issue by spreading it over so many different posts... </rant>

[PaulD1]

@Craig_Humphrey 

Agree this is more a discussion about licensing than Oakdale per se, but I think reflects the disappointment with the way licensing has evolved with Power Apps. I think there was some hope in the community that perhaps Project Oakdale might address some of those concerns (that you can't connect to a reliable/robust data source without paying extra licensing).

For the audience that are interested in Oakdale because they want to build Apps on a better data source than SharePoint (but are not in a position to pay the extra licensing) rather than because they have any interest in Teams, I think my points are valid.

Given the history of licensing for SQL Connections I am very wary to spend time/effort on Oakdale as a solution to the data source issue without a commitment that the pricing won't go the same way as SQL in 12 to 24 months. I also think it is valid to point out the current limitations related to it being in preview.