Dear friends,

Welcome! kindly join us in celebrating the completion of phase 1 of the HBL project.

All of us at HBL have much to be proud of.

HBL is possibly the biggest community-created collection of materials and tools for early Cantonese learning. Keeping the project open source and copyright free under a CC 4.0 license has become the gold standard that is now seriously considered by new initiatives in academia and publishing. Our use of written Cantonese is now being followed by more and more authors.

The automation that we built allowed us to write, illustrate, voice record and layout an average of 2 new titles a week. This output is faster than conventional publication methods by many multiples.  In addition, we build the automation to generate interactive H5P videos, TPR Lesson generation and CEFR-based conversational practice.  Remarkably all of this was achieved by one full-time colleague and the founders, working over 2 years with the very generous help of volunteers.

The project, funded by the founders, sales of books and donations, demonstrated that with the right design and clear-headed execution, a lot can be done with very little outlay.

We demonstrated a new way of viable community-based open-source publishing that significantly lowered the entry barrier and included the neediest.

As a result, we now have a diverse selection of stories, lessons, videos and learning materials on our website. Covid-19 restrictions helped us accelerate this process as exceptionally talented volunteers made themselves available for much longer.

Today, to our delight, our faster, more inclusive and more modern method of publishing has been upended.  Large Language model (LLM) based Artificial Intelligence has exploded on the scene and advancing at a dizzying pace. This is a hugely democratizing moment. Education in general and early Language learning, in particular, is being reimagined as you read this. An AI-based personal tutor is a very achievable reality in the near term. For those who want to learn, high-quality education will no longer be the privilege of the entitled few.

All of this at a fraction of the cost and exponentially wider scope than what we achieved! We are overjoyed that the barriers to entry into publishing that we had lowered, AI is completely smashing. Now, anyone with a computer and imagination can generate content in many forms.  As a matter of fact, customised content can be generated on demand. The possibilities are endless. 

Ending on this high note, I am now taking a break and passing the baton to the new wave of content creators. HBL was a terrific experience and privilege for me and I look forward to upcoming initiatives that propel open-source Cantonese learning to higher planes.

Thank you, 

Viveik 

———–

冚唪唥第一階段嚟到尾聲，係時候回顧一下之前三年我哋做咗嘅嘢，同埋之後嘅方向。

三年話長唔長，話短唔短，喺 Viveik，Joanne 同埋團隊所有人嘅努力之下，我哋推出咗實體故事書，整咗超過二百段故事影片，設計咗大量對話學習材料，開發咗粵拼跳跳紮轉換工具，仲聯絡咗海內外對廣東話有興趣嘅朋友，一齊創作咗好多新嘅內容。我覺得作為一個小型嘅群體，呢個成績都算係過得去嘅，可以同自己交代。

我哋不時會反思呢個計劃喺香港、喺粵語群體嘅定位。我哋深知幾個人嘅時間精力，係唔可以代攬晒一切廣東話事務。好好彩，呢個計劃嘅第一年，已經好多人加入我哋嘅行列，喺香港、喺其他地方，用唔同嘅方式推動粵語嘅文字創作，令到學廣東話變成一個「搞得掂」嘅過程。

啱啱開始嗰時，我哋做嘅嘢係各種嘅「範式轉移」。喺推廣粵語嗰陣，如果要新嘅思維，要工具，要人手支援，我哋好樂意坐埋一齊諗計解決，行出嚟向公眾推廣理念。去到某個階段，可以話係已經「做大個餅」之後，我哋係時候換個方式去「揼石仔」，等成員可以用唔同嘅方法去貢獻香港社會。

冚唪唥嘅故事唔會再更新，但係會整理做開源嘅格式發放。

我知道只要有信心，廣東話肯定會繼續傳承落去，亦都肯定會有下一個爆發。唔知係一個月，定係十年，總之廣東話再次需要我哋嘅時候，冚唪唥第二階段就會開始。

Chaak

———-

由2021年頭加入冚唪唥原來已經兩年。由試驗TPR教學、印刷40本圖書、搞書展到推出240個故事、互動影片、對話練習，仲有推廣到學校機構、海外書店團體……原來短短兩年可以為廣東話做到咁多嘢。呢個計劃令我大開眼界，更認識咗唔少朋友。

首先當然要多謝Viveik同阿擇有咁大膽同有意義嘅構思同領導，同時，我都要多謝我自己花咗好多時間照顧呢個計劃。亦希望睇到呢段文字嘅你，無論係捐款買書支持冚唪唥嘅朋友、幫過我哋手嘅義工、定用緊冚唪唥資源嘅老師學生，都好好多謝你自己為廣東話付出嘅心機同時間。呢個計劃就好似一棵樹，靠大家各施其職，曬太陽灌溉施肥，先得以茁莊成長，而結出嚟嘅果又會驅使其他人創作更多粵語資源同活動。

冚唪唥順利噉完成咗第一階段嘅所有目標，推出咗好多前所未有嘅資源畀大家學習廣東話同文化，我哋係時候要唞一唞，同埋諗下下一步可以做咩。其實呢個計劃一直都冇好清晰嘅方向同大綱，而係一路做，一路發掘可行嘅目標同不斷思考計劃嘅定位。適逢最近Large Language model (LLM) AI成為熱門技術，我哋又諗，AI可唔可以幫到我哋創作廣東話嘅故事呢？或者直頭教到廣東話？我哋都唔知，但絕對值得冚唪唥探索一下。

Joanne