Hello everyone,
Nearly twenty years ago when I was still working for someone else, I got
chewed out by my FileMaker sales rep for cannibalizing client seat license
sales by moving to the web. This was back in the CDML days, when you could
still web-enable any FMP client license and run as many web connections to
it as you wanted. That ended, I think, with FileMaker 6, and it honestly
had to end. (Or was it 4?) I have no problem with FMI figuring out how to
make money off web traffic, which I though they did fairly well with FMSA,
as Bev mentioned. If you need the web components, pay for them on the
server side. If you don't need it, it's not a cost item.
Shortly after my clash with my FMI rep (so much for the customer is always
right, eh?), I went out on my own as a web developer, and the first decade
of my consulting tenure was largely devoted to quietly replacing FileMaker
installations with PHP/SQL web apps. It's really hard to make the business
case for a dedicated server machine and an FMS license when my clients all
have MySQL or even SQL Server and can stand up a virtual machine with linux
and Apache in the blink of an eye. (And forget about client seats.) It's
far easier to get a project approved if there's no hardware or software
purchase necessary, and that fits well with an IT department's existing
supported infrastructure.
The other thing I learned was that FileMaker isn't really a very good data
source for web applications. I've migrated a lot of data out of FMP and
into SQL systems, and the performance increase is tremendous. (Don't get me
started on mixed and/or queries, and wicked-fast queries for summaries and
counts.)
Like Derrick, I have abandoned FileMaker as a development platform. I'm
down to one FieMaker-based client, and they run their entire business with
a professionally-designed FileMaker inventory and accounting system by one
of the preeminent FMP development companies. My client is not likely to
abandon FileMaker, but as far back as 2012 I pitched them on mirroring some
pieces of their FMP system to a cloud-based SQL setup. They didn't bite, so
I did a complete redesign of their web-based FMP system that launched in
2014. That system is still in heavy daily use, and it's not going anywhere
anytime soon. We've added a lot to it in the last four years, and I'm
actually working on more enhancements this week.
In 2011, I started looking for an MVC PHP framework to use as my
development environment. I settled on Cake.php, because it had built-in
support for MySQL and MS SQL Server (the two data sources I encounter the
most). Plus, a 3rd party had even written a database connector for
FileMaker that uses FX.php! It was out of date, so I had to adapt it for
newer versions of FMS, but it works, and works well. So I have one code
base and one framework underneath all my web apps, including the one that
uses FMS as a datasource.
I have been bracing myself for FMI to drop XML support for several versions
now. I figured I'd use that as leverage to do the SQL mirroring thing I'd
proposed to my client years ago. But... if I can write a new data connector
from Cake.php to the Data API, and licensing the Data API doesn't end up
costing my client an arm and a leg, then that's what I'll pursue.
That's a heck of a long way to getting to answering Joel's question! I
think for me it'll be door #2: Keep writing PHP but switch to using the new
Data API. Like Steve, I've been writing PHP for a long time, and I have no
reason to switch now.
My 2¢... or maybe more. Sorry, I did run a bit long there.
Dale
On Thu, Feb 15, 2018 at 3:00 PM, Fogle, Derrick <fogled@xxxxxxxxxxxx> wrote:
My strategy: Use PHP with the new Data API for legacy / interim needs, but
abandon the platform entirely for any truly new development.
For the record, I’ve been in the “no new Filemaker app development” mode,
only doing modest system mods on what I’ve already built, for nearly a
decade. Best ROI I’ve ever gotten from a platform and my labor investment
in it, by a long shot.
But, I’m already in a pinch between FM13-to-16, and High Sierra
compatibility now. I haven’t even thought about the fx.php compatibility
but that’s probably an issue too, I suspect? Not to sound apocalyptic or
anything, but the “no good thing lasts forever” meme looms large on the
horizon. Once Filemaker figures out how to monetize every transaction, it
will lose any competitive advantage against other platforms, because let’s
face it, that’s one of the few advantages of it now.
Heavy sigh…
Oh, and Hi everyone!
Derrick