Make ServiceStudio a cross programming language IDE

By Hans Bruins on 29 Sep 2010
Today I was thinking about a personal software project of my. Since it isn't a webapplication I can't use the OS service studio. The next thing I thought was: why not?

So, there this entry originated: what if the service studio was a cross programming language IDE and the visual programming a type of abstract programming language?!

To be fair: SS is already doing that a little bit by supporting C# and Java but that is narrowed down to a specific area:webapplications (or asp/jsp).

I am thinking here about a much broader aproach so that for instance also Windows applications can be created. Even embedded application support is not impossible since I saw last week a visual editor for that purpose.

The way i see this is that OS delivers the IDE, someone creates mappins and an external party delivers the compiler (for your favorite programming language).

I can think of more additions or angles but I guess that the base of the idea is clear now :)

Added 2010-10-06:

A little background about this entry: 
For me, this idea is not nothing like new. Before I started to work with Outsystems (now 5 years ago) I was working for a company that specialized in working with modeling languages. Most of the time we worked with UML and I found the limitations of UML very frustrating and nothing like "universal" at all. 

In that time I developed a big interest for visual languages. I also participated in a government project where we worked together with an University to develop next generation modeling languages. Besides that  I did a lot of investigation on a personal level into visual languages and wrote/draw/sketched a lot of possible approaches.

Then I met OutSystems and saw imidiatly how it solved most of the problems I encounterd with modeling languages before. So, I threw away my sketchbooks and started working with OutSystems on a daily base. That's now 5 years ago and I am still happy with that choice.

I think OS did a marvel job by creating the service studio. But I also think the concept can be taken (much) further. OS has made a visual environment for webapplications. Now it's time for going further then just webapplications.

It also creates a new market opportunity: sell the service studio as an separate entity. 

I added an image to display a global architectural overview of my idea.

J.1 Oct 2010
What would be the advantage of that specifically?

Web-apps are the most universal cross-browser, cross-platform you will ever get today...
Targeting for normal client-apps will make the tooling much much more complex. which OS is it running on (Linux-distro, XP, Vista, 7, 98,95).
You have to worry about the proper requirements on every machine it will be running on  (instead of the only requirement of a proper browser)

Furthermore, there needs to be a different fee-model for the clients I think.

Communication of the client-server needs to be setup. will it be tcp/ip or something else?
If it needs to be secure, we have to provide it (or OS) instead of the relatively simple way of SSL's and handling the security on ISS-level.

How do you handle new releases?
How do you handle with old versions running on client-pc's ?

About different programming languages....
We have integration studio for it. No one holds you back to make javas, python, c/c++ libraries and wrap them in c#.
however java and c# are managed code, which is handy in many many ways :)


I will stop now, but my opinion is, there is no advantage of your idea.





Hans Bruins1 Oct 2010
Joost,

I am sorry that you failed to grasp the concept of my idea. Maybe my explanation was not up to scratch.

The idea is not meant or targeted for web applications. 
There is more then just webapplications around : )

Of course there will be challenges to implemented this, but I think it can be done.

The big and ideal picture: 1 abstract development environment for every language. 

Like Ms. Visual Studio but this time a *real* Visual Studio!
J.1 Oct 2010
I didn't fail to grasp mind you.

OS is a tool to maken web-apps fast and on the agile method,
without actually having to KNOW a programming language.


what is the point of such an abstract development tool?
IDE's are already an abstract-layer and you want to create another abstract-layer on top of it?

sorry to be blunt, but you want to re-invent the esperanto language for programming languages and there is simply no use/market for it, because everyone wants certain stuff done and will get done in their own specific tooling/language.

And let us think, if OS does decide to go the road you have envisioned. what will the market-growth of their toolset possibly be?
How many competitors will they have to defeat?
If you google a bit, there are already dozens of toolset which do almost the same as OS does now. (good thing that OS does it much better most of the time ;) )
If they make an universal IDE, they have to compete with the much large competitors like Microsoft, Eclipse and probably much many more.

Hans Bruins1 Oct 2010
Let's not fight about it,

I think it's a cool idea, you think it's stupid.

That's fine by me.
J.1 Oct 2010
Who is fighting?

Just wondering if you can defend your idea or not.

In a way it's a good idea, but don't think SS would be the tool for it.

Hans Bruins6 Oct 2010
Joost, you probably don't mean it like that but you are coming across as very agressive.

 I don't think the idea needs defending; you get it or you don't. Furthermore I think SS is *ideal* for this purpose.

I will add some extra explaining to my idea later.

J.6 Oct 2010
First of all, I'm sorry to come over as aggressive. That is not my intension.

Second of all, defending might be the wrong word, convincing others might be a better word for it.

And again, I don't think your idea is stupid. I said that in other words already.
There is a difference of not supporting an idea and thinking it's a stupid idea.
For example, I like the fact it's an idea outside the web-box, so kudos for that.


Roy Grob7 Oct 2010
Everyone,
 
I can imagine this could be a very "principal discussion". OS is of course a platform for web-development. 
 
We bought a license for the OS platform 4 years ago as I was immediately convinced by the benefits. We immediately started developing online and mobile web applications.
 
Until a year ago I was totally convinced that online and mobile web applications would serve all our business needs for the years to come.
 
However! I (and we all) must acknowledge that (for instance) mobile apps (like for Android and Iphone) have far more possibilities then mobile websites, for instance, they give you the possibility to interact with hardware (camera, phone, GPS etc..) and that's something that a web app will never do.
 
Not to speak of the marketing and image value of these apps for a company...For me, these apps are a must have for the future of our business, 
 
So from a business point of view, not being able to develop (compile) these kind of apps with OS is very unfortunate.
 
That forces me to either invest in this knowledge, or to hire an external supplier to do it for me. Both ways, I'll end up reprogramming business logic  that is available for 95% in our espaces!
 
So, developing more then web alone with OS, sounds  pretty fantastic too me...
Hans Bruins28 Jul 2011
I would like to thank OS for complete suprising us to support the build of Hybrid mobile apps! We are very happy and excited about this enhancement of the OS platform.

During the build of a Proof Of Concept  with OS we saw a glimpse of what is possible if you combine Web en Native apps and it looked very well.

It is a good step forward and in line with the idea here mentioned. Of course I still think much more is possible for the SS of OS. If you aim high you'll archieve greatness.