IP theft of First-ever Bangla search engine 'Pipilika': Unmasking Dr. Ruhul Amin Shajib - Part 2

With heavy hearts, we, the original inventors Chisty and Burhan, present the truth behind the IP robbery of Pipilika, the first search engine in Bangla. The fradulent lecturer Ruhul Amin Shajib, now associated with Fordham University in America and still engaged in several govt and private projects in Bangladesh, got involved in this Pipilika initiative and gradually took it over entirely, step by step. I briefly discussed some of this in the previous post.

Subsequently, 'fradulent lecturer' Ruhul Amin Shajib and his cronies dared to undermine our contributions, only acknowledging our work in creating logos and naming our brainchild 'Pipilika'. But can they deny the countless sleepless nights and sacrifices we made?

Naming of Pipilika

In the 3rd year of varsity back in 2008, Burhan and Chisty embarked on an ambitious journey, starting to work on their very own search engine. The project began as a personal endeavor, crafting a project named Pipilika (Ant in Bangla) with Oracle Database under the guidance of Tanveer Sir, also known as Khondokar Intenam Unayes.

It was during this time that the idea of building a search engine took shape, a unique small scale version different from Google, yet something closely connected. And thus, the name 'Pipilika' was born. Burhan, with his creativity, came up with the perfect analogy for the name. Just as food cannot escape the hands of ants, the 'Pipilika' they built would tirelessly gather information from every corner of the web. It had a beautiful Bengali touch to it, adding that cultural essence they held dear.

Later, during the 4th year, Sir M Zafar Iqbal questioned the choice of the name, expressing his concerns about its convenience. Burhan and Chisty offered their heartfelt explanation behind the name, hoping to find an alternative if needed. However, to their surprise, even after 2-3 years, during the official inauguration, the name remained unchanged. At the launch event in 2013, Sir Zafar Iqbal himself admitted that they couldn't come up with a better name than the one Burhan and Chisty had given - 'Pipilika' (Ant). It was a validation of their creativity and the meaningful thought behind it.

Burhan also had a talent for graphics design, a passion he pursued as a hobby. The Pipilika's logo was his masterpiece, skillfully crafting it by transforming an English 'p'. He meticulously cut the tail of the 'p' and shaped it into eyes, adding two antennae on top, perfectly capturing the essence of an ant. The original logo was a striking combination of red and black, resembling the colors of a real ant. After Shajib's takeover, despite owing a vast funded team, the sad truth emerged that they failed to bring any real innovation to the table, resorting to mere superficial changes in the name and logo.

Ruhul Amin appeared a year after Pipilika's inception.

In 2008, as an academic project under Tanveer sir's guidance, we started working on a search engine with a database. As we progressed into our 4th year, we decided to delve deeper and chose 'search engine' as our research topic. Initially, we developed a corporate search engine, later shifting to general search engines as part of our thesis.

Meanwhile, in 2009, Ruhul Amin Shajib (now known as 'Assistant Professor' at Fordham University) joined the CSE department. Initially, we referred to him as Shajib Bhai, considering him an elder brother during our first year. Soon, he expressed interest in working with students for research, revealing his expertise in search engines at work. And we already had developed a small search engine. That's how Ruhul Amin became our 4th year thesis supervisor.

Now, the real journey began—what exactly is a search engine, what it entails, and what did we accomplish?

Details of Pipilika

Search engines serve the fundamental purpose of retrieving necessary information from the vast expanse of the internet. In the case of Bengali in 2000s, a unique requirement for Bengali morphology arose, and conventional search engines at that time struggled to handle joint alphabates of Bengali properly, resulting in numerous Bengali texts going undiscovered on the internet.

At that time, our initial design for the search engine focused on addressing these challenges and incorporated the following key points:

• Creating knowledge base
• Web crawler
• Web parser
• Text analyzer, with a special Bangla text analyzer for Bangla search
• Categorization algorithm
• Data indexing
• Ranking
• Text highlighting
• Web searcher

Later, we added some other elements to the search engine. Our unwavering dedication and relentless efforts bore fruit, leading to the remarkable success of our Pipilika Search Engine. It proudly secured the 1st Prize in Sylhet Divisional Fair in 2010 and triumphed at the National Collegiate Software Contest in 2011. These accolades are a testament to the hard work we poured into our creation.

The significance of our work is undeniable. Otherwise, why would GPIT (Grameenphone IT) show keen interest in collaborating with SUST on the Bengali search engine around October 2011? We must ask, did they base their interest solely on Ruhul Amin's CGPA? But the truth stands unwavering - the credit for Pipilika's development belongs to us, who tirelessly worked on it for two years as part of our thesis. Ruhul Amin disregarded our hard work in the name of an arbitrary imaginary group of students.

It might seem tedious with all these technical details, but there's a significant reason behind it. Ruhul Amin attempted to discredit our contributions using various tricks, claiming that we didn't create a search engine but a mere prototype. He went as far as asserting that Pipilika was solely his creation. However, during the initial launch, he himself acknowledged it as the result of the Pipilika thesis project, though he still attempted to mask our involvement. When the truth became widely known, he resorted to labeling it as a separate project, questioning our qualifications, skills, and even resorting to abuse.

But we are determined to share the genuine story of how we created the first Bangla search engine - an endeavor that earned awards and warm reception. It was indeed a search engine, and anyone can attest to that. The piles of papercuts, email threads, and thesis documents bear witness - except for Ruhul Amin, who seems intent on confusing everyone with his counterclaims. We must understand his motive - by dividing the original inventors and positioning himself as the leader, he secures the power while relegating the rest of us to mere employees.

Our journey since 2008 has been marked by persistent efforts to create something extraordinary, even in the face of limited resources. Though we had access to a Google research paper on search engine architecture, it wasn't sufficient to build the entire system. Additionally, unique challenges specific to Bangla demanded our relentless focus as we tackled each fundamental issue one by one.

Knowledge Base

The first challenge we faced in building a search engine was constructing a knowledge base. At that time, to the best of our knowledge, there was no search engine available for the Bengali language. Moreover, the required information wasn't accessible for search engines to process effectively. Once we laid the foundation for the entire architecture, understanding how to move forward became a time-consuming endeavor. Burhan and I grew weary of the constant struggles with poor 2G internet and dormitory load-shedding, prompting us to relocate to the university lab with mattresses and pillows. For nearly a year, I toiled away in the lab, working diligently on the search engine. The journey presented numerous significant challenges that demanded innovative solutions.

Data Collection

In the beginning, our data collection process revolved around gathering information from different yellow pages and online business directories. Talha Bhai (2004) developed a crawler called Domain Specific Crawler, while some of our fellow brothers from the 2004 batch worked on a spell checker for their thesis/projects. Back then, these were distinct theses, seemingly unrelated to search engines. The journey was just starting, and little did we know how each thread would eventually weave together to form the intricate tapestry of our search engine venture.

Indexing

After numerous dramatic efforts, I finally managed to complete an indexer. It was a significant milestone as it allowed us to save the crawled data to Lucene after parsing it with our generic parser and running our custom categorization algorithm. Achieving this feat demanded relentless days and nights of effort, spanning over 6-7 months. Anyone who was a student at that time can vouch for the intensity of our endeavors.

However, a daunting challenge surfaced - there were no Bengali analyzers available for Lucene at that time, causing everything to break. Despite my best attempts, I encountered failure after failure. Seeking assistance, I turned to my thesis supervisor, Shajib Vai. Together, we attempted Pair Programming, investing several days in the process, yet the problem remained unsolved.

Then, one day during Ramadan, after Sehri, I had a breakthrough. I successfully resolved the issue, making it easier to display Bengali text without any broken characters. It was a moment of triumph, and the satisfaction of overcoming the hurdle was immense.

But I didn't stop there. I wanted others to benefit from our work and ensure proper representation of the Bengali language in search engines. So, I contributed my solution by adding a post to the Lucene Jira tracker (link in comment), hoping it would make a difference in the broader context of language processing.

Searcher and Distributed Computing

After completing the indexer, we proceeded to build a searcher, and initiated the work on distributed indexing. Together, we successfully configured the distributed system using the alpha version of Hadoop. Before diving deeper into this aspect, Shajib Vai reassigned Burhan to a project involving the generic parser. Burhan dedicated significant time to fine-tune a highly efficient system for certain websites, leaving an impressive mark on our journey. Zafar Sir himself acknowledged his exceptional work, praising it as "Very Impressive" during the thesis viva.

Throughout our pursuit, we faced challenges with slow PCs in the university lab, often encountering frustrating blue screens. Trials and errors seemed unending, and smaller tasks weighed us down. There were moments of disappointment, and help was scarce, with no resources like today's ChatGPT or online forums to rely on. However, we clung to our dream of creating a search engine for Bengali, pushing forward despite the setbacks. The day when the entire system finally came together and worked in harmony felt like an Eid festival. Our confidence soared, reinforced by each success.

The day of the final thesis presentation felt like a grand celebration. It showcased the fruit of our labor, the culmination of days of perseverance and struggle. Throngs of people gathered to witness our demo, and Pipilika participated in the Sylhet Divisional IT Fair in 2010, securing the prestigious 1st prize, with the medal presented by the minister himself. Computer Jagat featured Pipilika prominently on its cover page, acknowledging our contributions with pride. Later, in 2011, Grameenphone IT approached to fund Pipilika, transforming it into a multi-crore project that garnered widespread recognition. Even after starting our jobs, we continued to support the project for a significant period, witnessing its growth and hiring new talents for further development.

From Pride to Heartache

However, the story takes a bitter turn from here onwards. After a few years, the sight of massive billboards across Dhaka featuring our familiar logo left us astounded. Every pixel seemed so familiar, reminiscent of the days when I meticulously crafted the logo, cutting off the head of a 'P' to create the eye and adding antennae on the halved 'p.' The feeling of pride was overshadowed by a sense of being overlooked and underrated as we saw our teacher, Ruhul Amin, representing Pipilika on a talk show, never mentioning our names even once. The immense success and praise of Pipilika left us with mixed emotions - immense pride followed by the sting of being excluded.

Never in SUST's history has a project achieved such remarkable success. We could have never imagined that our brainchild would attain such prominence and fame at the national level, with Pipilika's name being echoed throughout social media and mass media. Amidst the joy, there is an underlying pain - the pain of seeing our teacher on TV as the face of Pipilika, omitting our contributions entirely. The billboards that once filled us with pride now make us feel small and disregarded. We can't help but wonder, will 'teacher' Ruhul Amin ever provide an explanation for this heartache?