Open Source RPA: TagUI — Part IV

In the previous blog series Part III, we discussed two very popular Open-source RPA tools, Automagica and Robocorp. Let’s talk about yet another popular open-source, cross-platform open-source RPA Tool. We come to the concluding part of our blog series on Open-Source RPA.

TagUI

TagUI is a command-line RPA tool that is easy to install, write code, and use. In Tag UI automated workflows are called flows which are basically text files with instructions. These flows are made of steps. E.g. “click button ‘submit’” is a step. Flows can be described in easy to understand language that contains TagUI commands used with Identifiers to create steps.

Identifiers are used to Target elements for automation. Identifiers are DOM, CSS, and XPATH elements which usually contain an id, a name, and class attributes. XPATH is a more explicit and powerful way of targeting web elements. This support is only for Google Chrome. For other browsers, image recognition is used. Positional vectors (point with X-Y coordinates, Region with 2D area co-ordinates) and image files can be used for image recognition.

TagUI will instantly convert the automation flow above into 100+ lines of JavaScript code and execute it. Under the hood, it uses Chrome DevTools Protocol, Sikuli, CasperJS, PhantomJS, and SlimerJS.

Source: TagUI AI Singapore

Some of the Key Features:

1. Web Automation — Chrome extension for recording web actions or an invisible (headless) browser
2. Visual Automation of desktop and websites using Sikuli.
3. Natural Language — like Syntax. Makes it easy to develop, deploy and maintain.
4. Supports 20 Human Languages — Write scripts in >20 human languages. Flow file can be translated to any of the other supported languages.
5. R & Python Integration — R and Python integration for Big Data, AI, or Machine Learning.
6. Cross-Platform — Available in Windows, macOS, and Linux.
7. The automation flow can be triggered by scheduling, command line, API URL, email. A flow can run another flow by calling the flow’s .tag file.
8. API or command calls can be made with a single line to integrate with other services or apps.
9. Continuous integration with CI/CD tools is possible using the CasperJS framework and TagUI’s Chrome integration.
10. Provides ‘live-mode’ to enable interactive mode when you can see the results of the code step dynamically. This feature is very useful for development and debugging.
11. Has a feature for low-level control of the keyboard and mouse.
12. Allows the use of conditionals (if-then), loops, and helper functions. Some helper functions are inbuilt. Some can be custom created using JavaScript.
13. Object repositories are CSV files that store variables used inflows. They help to separate flows from the data. Within the flow, TagUI can use the objects and they will be replaced directly by the definitions before it is run.

Limitation

1. Lacks a drag-and-drop interface. A low-code/ No-code has become the norm for the RPA toolsets.
2. Relies on the Sikuli Visual Recognition engine is not selector-based.
3. Lack of support for IoT devices.
4. Lacks orchestration and reporting facilities, Analytics and Visualizations

Why you can’t Adopt an Open Source RPA?

1. The commercial viability of the project is always a question mark. Owners and maintainers may ditch the project anytime jeopardizing all efforts and progress been made.
2. Slow response to queries and troubleshooting needs. This can impact productivity.
3. Upgrades and Updates can be messy. Release management and stable builds are areas of concern.
4. Enterprise-level security features and Orchestration facilities are generally lacking when compared to other commercial players.
5. Open-Source projects cannot meet user-needs in a time-bound way as it happens in commercial software development.

To conclude

Commercial players can raise capital. This helps create the reach within the developer community and marketing prowess which is hard to compete with.

Open-Source isn’t always about being better, but more so about being distinct in an empowering way.

There is great diversity in the contributors who follow the project, and this can potentially fuel innovations. But those features can easily get hawked and used by a commercial player. There isn’t much of a level-playing field between Open-Source proponents and corporations in the creation of the IT Automation stack.

But this may well change, with ease of funding, technical reputation, better operationalization of projects, business awareness, and business maturity. The most critical need is the evolve the product at a speed the market demands, and open-source solutions may tend to overlook this business character of competitiveness.

Technical success doesn’t necessarily lead to business adoption and it’s the latter part that needs to be focused more on.

With that been said, Open-Source RPA or for that matter any open-source software serves a very utopian need; the democratization of technology to make it available to individuals and organizations big and small; to place more power in them to re-define their activities to compete and to upscale.

What’s Next?

Ask us anything. We are with you every step of the way.

Keep watching this blog space as we‘ll be posting more on open-source RPA…

With so much noise around the ROI of RPA, here is a webinar recording that looks closely at actual costs involved in RPA implementations, a realistic savings timeline, and the levers that impact the ROI.