The Future of Programming
​SPLIT THE APP
Make Creators, Not Users

TriSplit is built upon a new open source programming architecture that splits application ("apps") into three independently developed and validated external objects that can be located anywhere on the web and combined by users to create new and automated solutions without writing a line of code.  Calculations and data processing (Code Engines -CE) are split off from user interfaces (User Interface Frameworks - UIF) and written in any programing language. Open source XML data file schemas are defined that will connect all horizontal, vertical, and longitudinal markets from the bottom-up. This organization provides automated solutions that cannot be created any other way. TriSplit has gone through three working revisions over the last six years on the desktop and is being prepared for release on the web.

TriSplit defines an eXternal Object-Oriented Programming (XOOP) "soup" environment where data definitions and methods are split into two separate objects and controlled by industry and subject matter expert (SME) stakeholders. Industries and data experts define common Universal XML (UXML) data schemas and SMEs create plug-in calculation tools. TriSplit provides open source I/O code to allow everyone to work together on compatible tools from small calculations to systems like PLM and Digital Twin. Experts become creators of new solutions, not simple users who have to rely on independent programs that have limited customization and poor data integration. After 50+ years of software development, many needed programs are lost or not created and many markets have missed great opportunities.

TriSplit Coming Soon! Plug-and-play repositories of unified data and executable Code Engines running on common UIF "viewers."

Traditional "App" Programming Architecture

In a traditional app programming structure, programmers tie a custom user interface to a fixed set of calculations and data processing. They also define their own proprietary data file formats which require neutral file translators to use other apps. These translators might not exist, may not work correctly, and they slow down the work process. One PLM paper refers to this as "data integration hell." More critically, apps with user interfaces do not allow their data processing code to be separately sequenced and automated for new tasks. That's a fundamental flaw of the interactive app programming architecture.

Traditional apps are also more costly to create because they require cooperation between both programming experts and subject experts. For a huge number of markets and application areas, this greatly limits the development of programs. When technology changed, as when DOS changed to Windows, many calculations were lost. The calculations may still be valuable, but the cost of converting them to a new technological app format was not justified. Now, with a change to web-centric app programming, good calculations are again being lost.​ TriSplit breaks apps into more basic and independent programming objects.

New TriSplit Programming Architecture

TriSplit breaks the app into three separately developed and tested objects that can be located anywhere on the web and combined in many new and creative ways by users and SMEs. The first objects are external subroutines called "Code Engines" (CE). They are batch programs that read XML text files of input, do some calculations or data processing, and then write out the results into the same or different XML file. Code Engines are like Unix "filters" on steroids, but they read and write XML files containing  any program data, not just text string data.

​The next objects are the open source User Interface Framework (UIF) front-end programs (interactive or batch) that can automatically launch one or more CEs without any custom containerization. Many different classes of UIF, like game engines, textbook frameworks, and 2D/3D design programs and optimizers, can be defined to allow mixing and sequencing of CEs from anyone. This goes far beyond the traditional scripting and command files of Unix filters. The future of programming will be the launching of complex CEs from common UIFs.

​The third object specifies XML text files (UXML) that can contain any program variable or data structure. TriSplit provides an Open Data Definition (ODD) community where data experts collaborate to define unifying XML schemas that can be applied to any program data or structure along with open source code for file I/O. UXML is to general programming as HTML is to web programming - one XML data file for any program data. JSON compatibility will be provided.

One UIF offered by TriSplit is a spreadsheet that handles general calculation and analysis needs. Code engines are launched in the background and it looks like the calculations are done in the spreadsheet. Users can combine CEs in many new and creative ways without writing a line of code. Code engines can be developed in any programming language, tested separately, protected, have long lives independent of technological changes, and located anywhere on the web.

​Processing speed, fast storage, and web flexibility now make this possible. There is no longer any reason not to separate code even if you are writing a traditional custom UI/app. ​

TriSplit Manifesto

Make Creators, Not Users
Democratize the Web
Empower Others
Divide and Collaborate Independently
Conquer Complexity and Cost
Unify Data Horizontally, Vertically, and Longitudinally
Encapsulate Knowledge, Skills, and Learning Curves
Plug-and-Play in the Same Computer Sandbox
Eliminate Zero-Sum Thinking
​Open New Markets
Win-Win-Win
Create Computer Lib/Dream Machines (Ted Nelson)

Move People's Cheese

FOLLOW TriSplit

Leave Your Email and Get the Latest News!