Hello everyone,
I have got an email from rcconsulting that could be of interrest to you folks:
Tomorrow (August 6th) at 9AM PDT / (12 noon Eastern) we will be broadcasting live from FileMaker Developer Conference 2019 in Orlando, Florida!
We are told to expect some big announcements. So join us for the latest news.
Watch Live...at Twitch
I will watch, will you ?
The ābroadcastingā is just going to be Richard talking about what was announced. If itās not under NDA. It wonāt be live broadcasting of the actual presentation.
Sounds like the announcements will be big, though.
Iām a subscriber to Richard Carltonās tutorials/video courses - I received the mail as well - and will join
hopefully, itās not just a new episode of farmer Johnās ābetter lighting on the workbenchā - thing 
I donāt believe so. Iāve been told by more than one source, itās not just hype.
Oh yes, the NDA . . .
What could be the biggest announcement:
. Looking at the number of bugs in FMPA and FMS 18, that doesnāt prompt me to try yet.What are your predictions ?
The only four facts I know.
Doesnāt necessarily mean anything, but itās interesting.
Apple acquires Stamplay, but the CEO now works for FileMaker 
.
But wait, Stamplay is Italian . . . letās hope itās not another Fi Again Tony.
Draco switching to SQL : YEAH ! 
Not going to happen. Same as an Android version of Go isnt going to happen.
FileMaker IS Apple. They are controlled by and owned by Apple.
I donāt get the reference. But pay attention to what Stamplay is/was.
Which is a major strategic mistake. Tragedic.
But unlike the Android version (which is also a terrible, but less, strategic mistake), thereās no stupid politics involved here.
There are so many things you would lose if they moved to SQL. Iāve had that discussion with the FM engineers and PMs.
Xojo is a good example of what I mean. Devs there that have moved to Xojo, spend a lot of time trying to get it to work like FM. Once watched a video of a dec demoing how easy it is to hook up the data source. The video was 90 mins long. Then the devs went to work on getting that live broadcast and update mechanism of FM. Nothing against Xojo. It has many positives, but the feature set is vastly different. The experience is vastly different.
FIAT means fix it again Tony . . . the cars.
Thereās nothing worthwhile weāll loose going to sql, or postgress.
Thereās only 4 functions to rewrite in sql :
The filemaker āadvantagesā : quick type changes (sql is so fats comapred to draco that changing type will be quicker), multikey, even unstored calc can be emulated easily in SQL, and not to mention that SQL is open source, so they could tweak it if necessary.
Moreover, will gain a ton more : true scalability, true data separation, IT people respectā¦
Iāve detailed this in the comunity idea thread, I wonāt repeat it here.
I donāt trust FMI PMs at all : they were able to create a 40% Slower FMS and write in there doc, how by the way, if youāve imports just stop the server !
Theyāre stuck in there faulty mindset.
Last year you came back from devcn overjoyed about the speed gain weāll get, but we got FMS 18 which realize the exploit of being slower while utilizing more cores.
Soā¦
Frankly, if Stamlay CEO would have been promoted FMI CEO, heāll probably do this on the spot, but alas we got a salesman instead
That oversimplification of what FileMaker does is disrespectful to the development team at FileMaker, and to all developers. Please, show me a demo of your upcoming productā¦
Which of the PMs have you met? The problem with quoting 40% slower, is that it only applies to single user tests, and it a difference of milliseconds. Iāve done several tests, as have many others, that have shown that under load FMS 18 is significantly faster. So trying to say something bad was done, is a faulty presentation. In tech, everything is about compromise. To get speed in one area, you sacrifice something somewhere else. Given what theāve done with a v1 rewrite, Iām impressed.
Possibly next week, Iāll have some numbers to show actual affect of multiple users on a low-end mini as a Server. Basic idea, individually the machines finish 1,000 very complex finds in about 11 mins. Running 3 machines doing the same searches, takes 36 mins. When you scale up to 4 cores, you get back to much better numbers. And 18 handles multiple users way better, and it never hits that bottleneck I am seeing with this low-end mini.
Now, letās get back to the discussion at handā¦exciting new coming out of DevCon.
FileMaker is not SQL, so what ??? Those IT people that disregard FileMaker just listen to SQL database vendors, the never put any effort in trying to understand what FileMaker is.
That being said FileMaker and FMI are not perfect. Just read messages on the official Community and look at bugs that are reported about FMPA and FMS 18. I am not a FileMaker opponent, but FMI seems to have dropped V 18. Is it that 1 year between new releases is not enough ? Today I saw that FMI acknowledges a new bug - unfortunately can't recall which one. I couldn't understand how come this went unnoticed during testing.
I also know Xojo. It's a very good tool for OOP development, implementing functionalities to help boost developers productivity, but you can't develop an app as fast as with FileMaker though.
Unnoticed. Unreported. Both the wrong words. Fixing a bug is about figuring out what is causing it, changing the code to address it, without breaking other stuff. This is the same stuff we deal with when doing our own development. However, FMI operates at a different scale than we do. There is so much more variety they have to account for. Add on that they are writing code that has to read other code...so much can break so easy. Go build a node.js project with Express. So many moving parts...so fragile. Run through that for a bit, and it will give you more respect ( in general, not speaking specifically about you ) for what FileMaker delivers to us.
Many bugs had been reported during testing. At that point, it is about fixing bugs that will cripple files running. Most bugs don't, and we can work around them. Other issues are caused by external technology ( see the issue with Gmail and the Send Mail step as an example ). Hopefully FMI gets a fix for them. But honestly, most of the bugs reported, I've never seen or run into.
Testing is another beast.
I tested V18 much more than any other version since the .fp7 run - and it was phantastic. As Go18 came out, I was surprised because it solved the biggest issue here.
The main reason for that testing was, that I worked a looong day long with V18 without recognizing that - the icon was identical (to 17) on my Mac (I usually have own icons, but due to the fact that I had to upgrade my osā for 18, I set up a vanilla, new Mac - and did not have the icons on that machineā¦). After recognizing that I was working with 18 and that was so smooth, I decided to keep on working with 18 - so I tested all āmy perspectivesāā¦
But: I tested āmy environmentā, just my own habits and needsā¦
Some of our customers are ISO certified (900x, 27001, etc.) and they must test (one reason that makes it impossible to establish every version of filemaker; one year is way too short, impossibleā¦). The last time, we introduced the new version of fm, user groups were defined to check as much as possible before going live. Some of the groups were late, users couldnāt test as fast as planned.
So, the CEO asked those groups to complete testing until a specific date (it was a short time, just to be āon planā).
All groups finished their testing, sending protocols to dev (Iām the leader of the ādevā there). We had to invoke the new version over a weekend.
Not even printing was workingā¦ (but was testedā¦)
During macOS classic, there was a software called āOne Clickā by Westcode Software, phenomenal, one could do unbelievable things, interface, scripting, almost whateverā¦
I was a beta-tester at that time and one westcode-guy was leading all testings, holding us testers straight on a line. We knew every single point and there was an open (for everyone visible) list of issues with their priority.
Filemaker (Iām not a beta-tester for fm anymore) was similar - but not as āstraightā - but beta-testing for Westcode took much more timeā¦
At some point, a software company has to publish a āv.x01ā, knowing that there must be updates. This is normal. What is not āthat normalā is the fact that fba testers could not send in any reports (I received a mail telling me that there is no more testing before release)
-> this was changed in the V18 run - we could send in our reports!
What remains the big problem for me is the fact that fmi still has no release notes (for go almost zero, a bit more for desktop version - but far away from āfullā). Makes it hard to test, to install new versions/patches (I mean, the Japanese-thing doesnāt bother here in any wayā¦)
In short: Today, full-testing became impossible. It has to be a compromise. (nothing newā¦)
Xojo: I got a professional license (all platform) here. Working on interfaces is much better with filemaker. Overall, filemaker is still the fastest environment that Iām aware of
According to information available on the Internet, Stamplay is a low-code workflow automation tool, similar to Automator but more modern, versatile and cloud-enabled.
When a company wants to make an important announcement to its customer base, it is done through a public statement, preferably a written one. I personally prefer this over the hang-on-the-lips-of-those-who-are-in-the-knowing consumer-world marketing style.