If they raise the on premise prices that’s it I’m out. I’m struggling now so if the prices increase I’ll just move to another platform.
3 Likes
I would summarise my feeling after watching the roadmap with two words: excitement + uncertainty
I’m excited about the Android client, Linux FM Server on premises, integration with JavaScript, and XML importing for updates.
And I’m uncertain and worried about the future of my business, if we are heading towards a “Cloud only” strategy with a 5 users licence minimum. They insist it’s “Cloud first”, not “Cloud only”, but I wonder if this still will be the case in three or four years time.
4 Likes
It sounds like in your case that it might be a good time then to start writing CF logic in a separate micro-service. Then, if you leave FMP, your code is still totally usable by your next platform. I posted a quick video yesterday showing how, in less then 2 minutes, you can create a working “hello world” (starting point) micro-service.
I keep most of my logic in micro-services (JSON stuff, XML stuff, any looping, etc.) and use it across multiple applications, multiple operating systems, etc… All for free.
The good news is that, using the (SLOW) “new” while loop in FMP 18 as an example, while loops in compiled micro-services (or “for loops” or “do…while” loops) are about 400 times faster than FMP when recently bench-marked. Plus, I didn’t need to upgrade FM to get them. 
Even using FM’s JDBC driver and JDBC logic (JDBC code is mostly generic so it easily migrates to your next DB vendor: MySQL, Oracle, SQL Server, …), I get 3X better performance than queries with FM alone. I’ve posted multiple benchmarks on the “other forum”.
Although I like FMP a lot, I refuse to get boxed in by any proprietary software vendor and I started migrating my logic to micro-services about five years ago. Now, my clients can use the same logic from the browser, terminal, really from any HTTP-enabled application(s).
3 Likes
You need look no further than FMI’s JDBC driver’s marketing-directed crippling so it only works with a “local” FMP database. No other company or organization cripples their JDBC driver.
Guess what? The FMP JDBC driver has no such single machine/server restrictions with FMS.
Hmmm…
It’s actually easy, using a micro-service, to work around this marketing silliness so you can remotely use JDBC with a regular FMP database, but this marketing heavy-handedness – and whoever approved it – is ridiculous.
3 Likes
I’m at the point where I’m looking to go full stack development - yes it will take me longer initially but at least I am in full control. I’ve been pleasantly surprised by how good some of these JS frameworks are and how they speed up the development process - I still struggle with the design side in HTML/CSS but having just discovered CSS Grids even that side is looking up!
5 Likes
Yep, absolutely. I’ve been full stack for 20 years. FMP for personal stuff, mostly. I refuse to let companies with proprietary tech pull ridiculous shenanigans.
2 Likes
Someone shared this funny picture with me recently, figured I would share it here (found via google search):
6 Likes
When I took some time (3-4 years ago) to learn Swift, CloudKit, and Core Data, I didn't really do it as a hedge against FileMaker, I just wanted to learn how to develop for iOS. Although now, the way I architect things, I could spend a weekend refactoring a few Classes and URLSession calls, and use any backend I want (FileMaker, CloudKit, Firebase, etc.).
Once SwiftUI is ready for prime time (mid year), I'll be able to deploy to any Apple device, not just iOS. Hello FileMaker thin client. 
5 Likes
Been busy so missed this till now.
I am excited that Android is possibly in our future. I have been developing for others it would be cool to be be able to develop for myself for once. I am also worried about cloud first and what that means for the future. If it ever is cloud only I will need to move to another platform or close shop.
I don't foresee them ever being cloud-only. It doesn't fit their message, or their positioning. Cloud-first is a technology approach, not necessarily a business approach.
Oops, sorry about that one, perhaps English sense of humour should be added to the community guidelines. LOL.
5 Likes
The great thing is that this Forum SW still leaves me the choice to read your post. Something, the Claris PЯДШDД department wouldn't offer me.
1 Like
Still, I agree with Andy....too many snowflakes here too. There was nothing wrong with his posting. Jeeez. (Will I now be censored too?)
Don’t worry about it guys, the world isn’t going to end (as far as I know).
I still can’t believe what Cecile managed to start here. The forum is an absolute credit to her.
Thanks
Andy
7 Likes
A certain amount of rudeness and incivility became très à la mode, these days...
But: quod licet Iovi, non licet bovi 
1 Like
Are yes, 'bull', that is getting closer
1 Like
We can all agree on the awesome job Cecile did to set up this forum and continues to do. Too bad FMI has their head stuck in the sand. No matter. 
3 Likes
Each customer ultimately decides which is the best solution for his case - as long as the choice is left to the customer. All of these ‘first’ approaches look ill-balanced to me and are patronising.
If you listen to Srinni talk about what Cloud-first means for Claris, he is pretty open and clear about it.
We don't choose the technology that drives the platform now. Cloud-first is about using and working with technologies that are cloud-friendly. That, very often, means they are lighter weight, more easily integrated with web technology, and more standardized. These are all things we have been asking for, for the better part of a decade.
Cloud-first also affect deployment. Setting things up to make deployment easier, more like a cloud deployment would use. That brings us back to a VM deployed on Linux.
Cloud-first is exciting for Claris ( it allows them to develop faster ), and for us ( it allows us to take advantage of standardized tech, and integrate better with external services ).
I hope your views will be confirmed. We'll see.