O$10: what is the rationale to skip O$9 ?

Bastiaan Olij Bastiaan.Olij at instinctsystems.com.au
Mon Nov 5 17:21:05 EST 2018

Hey Brian,

I'm getting "old" too, though only in my 40ies :P I wasn't having a dig at age  but just commenting there are an awful lot of Omnis developers I know who are indeed retiring and I do think it is a worry Omnis isn't attracting enough new developers. A concern voiced clearly by Omnis Software that it is their top priority to get fresh developers interested in Omnis and one of their main motivations for the choices they have been making.
Though I do have to also counter that with the great news that over 27% of the participants at EurOmnis were first time attendees.

But I fully echo your sentiments here. It is part of why we're looking at augmenting our offering with other tools and we're investing in technology outside of Omnis as well. Now that we are moving from a client-server to a 3-tier solution it is likely that the middle tier will not be Omnis though I still think Omnis remains a strong contender for the front end of our application but instead of talking to a PostgreSQL database directly, it will be talking to our middle tier solution.

There are however two things I want to highlight which really struck me as interesting realisations. Omnis basically stood still for 11 years, that isn't a fully fair statement as things have been added but for all the reasons you highlighted I think it is a statement we can still make and I think Omnis Software has acknowledged this and is addressing this. But you don't catch up on 11 years over night.

In light of that statement the first thing that strikes me is that I still can't think of any better platform to develop a fat client in. Yes it looks clunky, yes it seriously needs to be modernised, but I still have not found any other product that lets me develop a fat client application as easily and quickly as Omnis can and that is saying something as especially Omnis Studio requires you to write far more boilerplate code to get started than its competition. I just can't make something that looks like it belongs in 2018.
I share Dougs remark that all I want is for the Omnis runtime to run on iOS. I don't need a dev version for iOS, I don't care if it is stripped of the UI parts that do not work on that platform (menus, toolbars, single window?), but I would take an Omnis runtime for iOS over both Xcode or any HTML based solution any day and as long as I know I'm on that platform I can have my library make the necessary concessions. 

The web is here to stay but as you say, there are far stronger players than Omnis in this market and they are larger, have far more money behind them, and are "free" for developers to use. I just don't see Omnis winning here.

However I believe that the app market is going to be equally as strong and Omnis is incredibly well posed to be a big player in this market space. A native Omnis runtime for both iOS and Android that allows us to easily deploy our applications in the AppStore or Play store together with modernised controls, that is where Omnis imho can take on the competition and grab good market share. Combine that with being able to publish the same app through the existing runtimes for Linux/Windows and Mac OS X desktops and you have something nobody else has. So far all the offerings for native deployment to mobile/tablet devices stay within that eco system with maybe the exception of Xcode though that ties you into the Apple eco system (well, and there is Godot, but that's a game engine).

The other is that Omnis' Achilles heel is also its strongest point.  Omnis is a runtime language which in itself isn't all that special though it could potentially benefit from a build in JIT compiler. But Omnis did something with their runtime I have yet to see anyone else do, and they have done it for far more then 11 years, I've worked with Omnis for over 20 now and it was there when I first picked up Omnis 7, it probably was there long before that but my memory doesn't go back any further.

The Omnis runtime language in combination with its debugger in combination with the runtime and development environment not being separate is a strong point that I have not seen anyone do as effectively as Omnis has done for 20+ years. Yes web development does come closer because you can change server side logic while keeping the browser client alive but it's still a miles away and you still can't debug and change Javascript in any meaningfully comparable way. I was blatantly reminded of how bloody good this part of Omnis was when I had to do some XComp work yesterday. The endless edit, compile, restart, get to the place where the problem is, step through code, quit, make a small change and having to go through all of that again to see if your change worked, I love C++, it is my favourite language by a mile, but this is where I loose hours of time.

The ability to fix problems while you debug and see the effect of your changes right there and then without ever leaving the environment you are in nor having to restart anything, that is what makes it possible to double if not quadruple your productivity compared to other languages.

20 years ago the Omnis debugger was 20 years ahead of its time, and it still is today. 

Kindest Regards,

Bastiaan Olij
Head of development - Instinct Systems: The JobBag People
Ground Floor, 48 Chandos Street
St Leonards NSW 2065

Phone +61 2 8115 8000
Mobile +61 4 321 44833
bastiaan.olij at instinctsystems.com.au

 From:   Bryan Brodie <brb at appimatic.com> 
 To:   "omnisdev-en at lists.omnis-dev.com" <omnisdev-en at lists.omnis-dev.com> 
 Sent:   11/6/2018 5:18 AM 
 Subject:   Re: O$10: what is the rationale to skip O$9 ? 

Hi $all, 
What I've wanted since the iPhone shipped in 2007 was simply an Omnis runtime that 'just worked' on mobile.   
The same way it 'just worked' back then on Windows and Mac.  
Omnis 5 and 7 were excellent for rapidly creating cross-platform multiuser applications.  
Then came the web - with as big an impact on client-server, as client-server had on the original mainframe/minicomputer paradigm. 
Omnis did not keep up. Too much neglect by Raining Data and Tiger Logic.  
Client-server, Omnis's core, began a steady decline from the top of the IT heap. 
It’s a shame that the Omnis owners in 2007 didn't have the vision to begin creating a mobile runtime within their multiplatform development environment.  
They would have _dominated_ the web development industry. 
Instead, here we are 11 years later and the current-release Omnis end-user user interface is as crufty as it was in 2007.  
11. years. later. 
Why would any 'young' (since we're being ageist and I am 59 years old) developer opt for Omnis over, say Xamarin or Visual Studio, Xcode or a free open source web application development environment?  
While there are rationales for / against these alternatives, there is one big advantage: Stack Exchange. A smart, motivated novice can pretty much teach themselves those other environments.  
Not Omnis. It's proprietary. 
I taught myself LAMP in 2013 using Google, Stack Exchange and other resources.  
LAMP isn't cutting edge, but it 'just works' and is compatible with virtually everything out there.  
- Google API? Yes  
- Payment processing? You betcha! 
- Responsive design? Absolutely. 
- Cloud adaptable? Indeed 
- Free as in beer? Of course. 
I use a lot of free open source tools. When something changes and my tools break, I am able to go online and benefit from the experience of hundreds of thousands of other developers using the same tools.  
For free. 
Here we are in 2018 and Omnis is no longer a rapid application development environment.  
Everything that worked well in the 80s and 90s has been subsumed. This is the short-attention-span era.  
Someone here stated "the old timers are retiring". Well, I know of several that have moved on from Omnis.  
I am helping some migrate. I know Omnis and I know native web application protocols. It’s a nice niche.  
There's a ton of fantastic Omnis apps that are not going to ever work in Omnis v10 or v11 or even v19.  
I really hope Omnis survives, because I spent five years (off and on) developing my software design pattern toolkit to utilize it. 
But -  I am considering how my self-developed toolkit could be ported to a native web application environment.  
If I am going to learn an entirely different way of doing things, I need to do it in a viable and cost effective way. 
That's my two cents worth. 
PS - I really like O$ V8! 
   > From: Andrew Stolarz <stolarz at gmail.com> 
   > Subject: Re: O$10: what is the rationale to skip O$9 ? 
   >   One of the main things im looking for in the next release of Omnis is a 
   >  complete GUI overhaul so that our apps look modern etc., that and bug fixes 
   >   of existing features. 
   >   Even if it delays new features I'm ok with that in favour of above. 
   >   Andrew 
    > From:  Bruno Del Sol <bruno.delsol at bydesign.fr>     
    >  > Hello Bas, 
    >  > Thanks for this update on the Wednesday session and your tests. 
    >  > I also like the new editor although it will not make me code any faster, 
    >  > as far as i can tell. 
    >  > New but mostly updated GUI controls should really be part of Studio 10. 
    >  > The main issue is with the default styling of 
    >  > the basic controls that is obsolete. We need Omnis to add support to 
    >  > today's mainstream design themes, a new $styletheme 
    >  > property (for forms and windows) could be added with 3 values  : 
    >  > - Default Omnis Studio (for compatibility) 
    >  > - New Ios theme ( 
    >  > https://developer.apple.com/design/human-interface-guidelines/ios/overview/themes/ 
    >  > ) 
    >  > - New Material theme (https://material.io) 
    >  > When switching to one of the new theme all controls should change their 
    >  > default styling to match as closely as possible 
    >  > the expected appearance according to the theme guidelines. 
    >  > imho, embracing these new modern look and feel will satisfy both new and 
    >  > existing user base and kick Omnis sales up as 
    >  > soon as they'll put online a few nice app examples to demo it. 
    >  > Regards 
    >  > bruno 
    >  > By Design 
    >  > http://www.bydesign.fr 
    >  > Bruno Del Sol 
    >  > bruno.delsol at bydesign.fr 
    >  > tel (33) 01 48 78 47 37 
    >  > 46, rue de La Tour d'Auvergne 
    >  > 75009 Paris (France) 
    >  > Le 05/11/2018 ? 02:58, Bastiaan Olij a ?crit : 
    >  > > Hi Bruno, 
    >  > > 
    >  > > 
    >  > > All three points you mention here were addressed at EurOmnis during the 
    >  > Wednesday session with Omnis Software. 
    >  > > 
    >  > > 
    >  > > The method editor indeed is something that many Omnis old-timers won't 
    >  > care about, but in there lies the problem for Omnis Software. The 
    >  > old-timers are retiring, it is the new generation that they need to entice 
    >  > in order for this wonderful product to survive, and the new generation 
    >  > doesn't care for the old method editor. I actually spend the time using the 
    >  > new method editor during my EurOmnis sessions to get some more experience 
    >  > with it. While there were a few weak points, most of which I reported to 
    >  > tech support, some of which were already fixed and I've seen in action, it 
    >  > was actually a joy to work with and pretty solid for a first release. 
    >  > > 
    >  > > 
    >  > > The beta 1 was brought out some time ago already and indeed mainly is 
    >  > around the new editor that they wanted to put in front of developers to 
    >  > find issues with and they have been very clear in communicating that. Omnis 
    >  > Software showed off the remote debugger and several improvements to the GUI 
    >  > which will be in the beta 2 which is hopefully coming sooner rather than 
    >  > later (and if Omnis Software is reading along, I much rather have beta 2 
    >  > now with what is already available and then a beta 3 later on in the year, 
    >  > than wait for beta 2 any longer :) ). You can call it an alpha but that to 
    >  > me is semantics. 
    >  > > 
    >  > > 
    >  > > More modern GUI controls may or may not make it into the next beta or 
    >  > even into Studio 10. Omnis Software did show off a few new GUI controls 
    >  > that will likely be in those releases but some others may require bigger 
    >  > changes. Bob Whiting showed off some new controls with proper alpha 
    >  > transparency for instance that they are working on that I am really looking 
    >  > forward to using. 
    >  > > Now that they've made the move to Cocoa things have become easier on the 
    >  > Mac (I presume) but on the Windows side this requires changing the drawing 
    >  > architecture from the old Window GUI to something like the Direct Draw API 
    >  > to make use of the hardware acceleration (what Cocoa currently uses OpenGL 
    >  > for, now moving to Metal). That is a pretty structural change but one I 
    >  > hope they make. 
    >  > > 
    >  > > 
    >  > > While I wish Studio 10 would come out before the end of the year, I 
    >  > agree with you I hope they take the time to hit this one out of the 
    >  > ballpark. But I do believe they are sailing a strong course, if what we saw 
    >  > at EurOmnis comes to fruition I think Studio 10 will be a strong offering. 
    >  > > 
    >  > > Kindest Regards, 
    >  > > 
    >  > > 
    >  > > Bastiaan Olij 
    >  > > Head of development - Instinct Systems: The JobBag People 
    >  > > Ground Floor, 48 Chandos Street 
    >  > > St Leonards NSW 2065 
    >  > > Australia 
    >  > > 
    >  > > 
    >  > > Phone +61 2 8115 8000 
    >  > > Mobile +61 4 321 44833 
    >  > > bastiaan.olij at instinctsystems.com.au 
    >  > > http://www.jobbag.com 
    >  > > 
    >  > > 
    >  > > 
    >  > >   From:   Bydesign <bruno.delsol at bydesign.fr> 
    >  > >   To:   OmnisDev List - English <omnisdev-en at lists.omnis-dev.com> 
    >  > >   Sent:   11/2/2018 5:17 AM 
    >  > >   Subject:   Re: O$10: what is the rationale to skip O$9 ? 
    >  > > 
    >  > > I would guess they put themselves in a tricky position : 
    >  > > 
    >  > > - They?ve put a lot of time and effort into a new method editor that 
    >  > probably almost none of the existing customers really care about 
    >  > > - beta 1 is missing some of the key features that were advertised, 
    >  > mainly remote debug and better js grids, which makes it more an alpha 
    >  > version than a beta, there is probably a lot more work to do before shipping 
    >  > > - meanwhile the costumers (us) are as ever complaining about what  they 
    >  > really care : we beg for more modern, responsive, consistent, fancy, 
    >  > groovy, top of the notch, animated gui controls, because that?s the only 
    >  > thing our clients ask for IRL ! 
    >  > > 
    >  > > So, if they need more time to do it, that?s ok with me, and i don?t mind 
    >  > if the efforts they put in the right direction causes some version 
    >  > numbering jumps. 
    >  > > 
    >  > > Just 2 cents 
    >  > > Bruno 
    >  > > 
    >  > > 
    >  > > Envoy? de mon iPhone 
    >  > > 
    >  > >> Le 1 nov. 2018 ? 18:54, Alan Schmidt <alan at goconnected.com> a ?crit : 
    >  > >> 
    >  > >> As far as I'm concerned, they can call it whatever they want. 
    >  > >> 
    >  > >> The fact that they've skipped over 8.2 to go to 10 is what is 
    >  > problematic for me, as we were counting on that version... 
    >  > >> 
    >  > >> -Alan 
    >  > >> 
    >  > >> 
    >  > >> ?On 11/1/18, 10:40 AM, "omnisdev-en on behalf of Grzegorz (Greg) 
    >  > Pasternak" <omnisdev-en-bounces at lists.omnis-dev.com on behalf of 
    >  > gpasternak at cogeco.ca> wrote: 
    >  > >> 
    >  > >>     Well, when decision was made to move from O%6 to O%8 it was based 
    >  > on the idea of not having confusion with O7 (aka classic). 
    >  > >>     As far as I know, there was no O9 classic, so why 10 is all of the 
    >  > sudden better then 9? 
    >  > >> 
    >  > >>     Btw, I don't think there is such a thing as "lucky" number.  People 
    >  > do come up with all sorts of ideologically driven explanations but I am 
    >  > interested why rational people chose to follow the irrational "lucky 
    >  > number" philosophy. 
    >  > >> 
    >  > >>     I am just curious, that is all. 
    >  > >> 
    >  > >>     Greg 
    >  > >> 
    >  > >> 
    >  > >> 
    >  > >> 
    >  > >>>     On Nov 1, 2018, at 1:22 PM, Andrew Stolarz <stolarz at gmail.com> 
    >  > wrote: 
    >  > >>> 
    >  > >>> In the email from omnis it stated: 
    >  > >>> 
    >  > >>> "We believe the significant development of version 10 marks a new era 
    >  > for 
    >  > >>> Omnis Software and the renumbering reflects this step change in our 
    >  > >>> product, marketing and ambition." 
    >  > >>> 
    >  > >>> 
    >  > >>> If you look at Apple, they are on version 10 since 2001 & they 
    >  > continue to 
    >  > >>> release product running of 10 (X). Im sure apples marketing dept is 
    >  > >>> sticking to lucky number 10 for a reason (marketing wise). 
    >  > >>> 
    >  > >>> 
    >  > >>> Andrew 
    >  > >>> 
    >  > >>> 
    >  > >>> 
    >  > >>> 
    >  > >>> On Thu, Nov 1, 2018 at 1:18 PM Grzegorz (Greg) Pasternak < 
    >  > >>> gpasternak at cogeco.ca> wrote: 
    >  > >>> 
    >  > >>>> Does anyone know the rationale behind the decision to release O$10 
    >  > instead 
    >  > >>>> of O$9? 
    >  > >>>> What is more interesting, why is the announcement made today about 
    >  > >>>> something that is supposedly planned to be released some time in the 
    >  > >>>> future? 
    >  > >>>> 
    >  > >>>> I have expected O$9 to be released during Euromnis 2018 back in 
    >  > October 
    >  > >>>> but that did not happen, no explanation was given either. 
    >  > >>>> Is this done for some marketing reasons again? 
    >  > >>>> 
    >  > >>>> Greg 
    >  > >>>> _____________________________________________________________ 
    >  > >>>> Manage your list subscriptions at http://lists.omnis-dev.com 
    >  > >>>> Start a new message -> mailto:omnisdev-en at lists.omnis-dev.com 
    >  > >>>> 
Manage your list subscriptions at http://lists.omnis-dev.com
Start a new message -> mailto:omnisdev-en at lists.omnis-dev.com 

More information about the omnisdev-en mailing list