Development, the store and strategies for enterprises

Question

Hey folks,

I'm an LOB software engineer, specialised in C# and XAML.  I work for a software company that builds custom software.  It's a business-to-business strategy.  Our clients typically are large enterprises with 100+ employees.  In some cases, the enterprises are multi-nationals with thousands of employees.

On my previous job, I worked on a software product that was meant for small businesses/retail.  For example, (general) software for pharmacies, dentists, housedoctors with small practices, etc.

Both of these strategies got me thinking about the new application model of win 8 and how that will work with the store and I'm left in a dark spot with many questions and no answers.  So I thought I'ld introduce these questions here to see what your thoughts are.  A msft representative answering these concerns would be great, but I don't expect that to happen.  On the store blog, a msft employee only got as far as "we don't have anything to say about this at this time".

So, let's consider 2 scenario's... 

1. LOB development for a large business

Now, it's been said allready that enterprises will be able to deploy metro apps to machines within the domain without going through the store.

So, in case of custom development for just 1 client (where the cost is entirely carried by that client), I guess the store is a non-issue.  We'ld just deliver the code (or appx) and they'll take care of the rest.

However, what if we build an application for enterprises as a product?  Will enterprises be able to buy a metro app from the store and then add it to some kind of internal repository from where they'll be able to deploy the applications to several hundreds - or thousands - of machines?  How does that work for licenses?

How do I, as the development company, know to how many machines the application was deployed? 

2. Development of a metro 'companion' for retail

Consider the application I worked on at my previous job, the situation is even less clear.

It was an application for a specific retail sector.  This application was modular.  There was a 'light' version, which was the strict minimum.  Next to that, it included all kinds of modules (like accounting for example).  What was being 'sold' however was a right to USE the application.  There was a one time installation cost (which included data migration if needed) and an annual 'maintainance' fee - which was dependend on the amount of modules included in the installation.

When that maintainance contract was broken, the fee is no longer paid and the customer stopped receiving updates and no more support was delivered.  The retail literally was no longer considered a 'customer' and the application stopped working (it connects to a service to see if the contract allowed it to run - it only works for a month without confirmation of the service). 

This is a very common business model.

Now... what if the development company wants to also add a metro app companion of the desktop application?  This app would also need to be coupled to the maintainance fee.  It would also need to be modular.  And it should also stop working if the fee is no longer paid.

I'm not even sure if it is allowed to be in the store if it is dependend on a desktop application (which needs to be installed by one of the technicians of the development company, as it requires a buttload of settings and database related stuff).

Also: price.  The "light" version of this retail desktop application ran about 3000 bucks.  This high price is off course due to low volume (specific retail sector means small deployment base, yet an entire company needs to survive on those sales!).  The same would go for the metro app.  I'm guessing msft won't like it if a metro app, dependend on a desktop application, would be priced at like 400 bucks in the store, right?

But it gets worse...  The annual maintanance fee...  If a customer also deploys the metro app, his annual fee should go up.  But how do we know which customer installed it if the metro app is distributed through the store?  How do we control this?  And are we allowed to make it stop working if that maintanance contract is broken?

And last but not least: this business model is based on buying a right of USUAGE.  The application is NOT the property of the retail using it.  Only the data in the database is.

It's a very strange situation.  Now, I'm all open for changing those models.  Progress implies change after all.  But that shouldn't have an impact on the revenue of such companies. 

Any thoughts?

Answer 1

Hi Aroush,

I like the way you have presented your issue here. I do hope someone just doesn't put you off with the standard 'this isn't the right forum for the issue', even though that may be the case, I would still want to see some serious opinions on this. Many of the guys here are extremely talented pros and their opinion should open possibilities, even if a direct solution may not emerge at this forum. After all MSFT's gain is linked to the success of businesses like your's.

---

Irfan

Answer 2

Ha, thanks :-)

Yeah, I really don't expect to find answers about this stuff here.  I posted a similar thing on the blog ABOUT the store as well (that's where I got the "we don't have anything to share about this at this time")... and nothing.

I'ld be VERY surprised if msft doesn't give some love to its enterprise customers though.

But I must admit that I'm a little worried.  A lot of what we have seen so far about the store and metro app distribution seems to mirror the way google and apple are doing it.  But those platforms are literally geared towards consumers who spent their computing time on facebook, twitter and angry birds.

Microsoft, to me, is about enterprise FIRST and consumers second.  I feel like the emphasis of such new application models should be geared on businesses - or at least, enterprise should be a first class citizen in this story.  But it doesn't really look like that is the case.  At least not based on the information found on official msft blogs.

I understand that they don't want to share any information of which they aren't yet 99.999999% sure that they will be able to deliver it, but this kind of puts people like us, business-to-business LOB developers, in a rough spot.

I'm actually counting on enterprise to adopt win8 tablets 'en mass', believe it or not.  I constantly hear our customers yapping about them wanting to have tablets for meetings and while on the road and stuff while complaining that ipads and droids simply aren't viable solutions for the reason given above: they are consumer oriented.  They also require different skill sets (whereas metro is C# and xaml - the two most popular enterprise development technologies) and are hard to integrate into their networks (whereas win 8 is 'just windows').

But not having the information described in my opening post makes it extremely hard for us to start planning all this.

We have a couple of products (currently running on silverlight) that would be absolutely, completely and utterly awesome on a tablet.  From the day we saw the keynote of //Build, we all agreed that these SL applications are simply screaming for a metro version. 

But without knowing how this works for licenses, maintainance fees, control over the apps on our end, custom releases for specific customers that want/need it, max prices, if dependencies on local desktop apps (or databases) are allowed, etc etc etc....  We simply cannot justify starting development of these apps. We cannot justify mobilising 5 developers, a project manager, an analyst and a designer during 6 months to jump into such a black hole.

So yeah, I do hope that all of this will be addressed very soon. 

Answer 3

Please just give me a day or so and I will look up some right links/addresses for you.

---

Irfan

Answer 4

Hi Aroush,

to get better informed on your issue I think the link appropriate for you is: Windows Store for developers, while your question may be put up at the forum: Windows Metro style apps. Please try these links and post back the result. You may also wait for a better informed reply on this forum.

---

Irfan

Answer 5

Hey,

Thanks for the links.  I actually allready posted on the store blog.  They said they couldn't share any info at this time. :-/

Thanks for the trouble though.

Answer 6

Aroush,

they seldom reply back at the blogs since it's only the author who has to answer your question. You are more likely to get a reply from the second link I gave you above (Windows Metro style apps), choose the forum head appropriate to you at that page and ask your Qs.

---

Irfan

Answer 7

They're certainly interesting questions and hopefully as information on the enterprise deployment scenarios pan out it will become more obvious. More widely however, it seems the kind of business model you are using lends itself to cloud-hosted SaaS provision, which is surely something the store will already be catering for.

Answer 8

More widely however, it seems the kind of business model you are using lends itself to cloud-hosted SaaS provision, which is surely something the store will already be catering for.

Yeah, I know.  But there are issues with this.

On my previous job, I actually explored that path.  We were thinking about a new version of the software and I was a proponent of moving everything online and turning it into a web-application.

It was shot down by a vast majority of the customers.  They don't want their stuff to be out there in "the cloud".  Seriously, they were really outraged.

They want local management of it all.  They want their database to be on a machine under their own control.  Which is something I really can understand.

Maybe it's fear, maybe it's a lack of understanding the technical implications and/or advantages of cloud computing...  But the vast majority of these businesses are really not at all interested in putting all their data in an online database.  I would even bet that quite a few of them would even prefer to return to physical files.

Also, for quite a few sectors, I think SaaS simply isn't an option.

For example, the application of the Pharma sector also controls a robot arm that fetches medication products from the stock.  The application for the Optician sector controls machinery like refractors, lensmeters, phoroptors, etc.  While it is perhaps possible to do this through SaaS solutions, it surely will be quite a pain in the a*s to make it work :-)

And a lot of this stuff also varries from customer to customer.

Also, even if SaaS would be ok for the customers, we are talking about really, really, really big applications here that have big and complex infrastructure backing it.  It would all have to be re-written.

But you are absolutely correct off course.  The model kinda scream SaaS.  From an engineering point of view, that would most certainly be the way to go.