Creating the Wiki and Homepage
This page provides tips and steps to create your wiki–of course, the specifics depend on the software you choose.
What to Do
1️⃣Get to know your software. Gather existing documentation, user manuals, release notes, and any relevant information about the wiki software, and then identify your chosen core functionalities and unique features. Group them logically based on usage scenarios or workflows. You must understand how your co-workers will view, create, and edit pages inside the wiki.
2️⃣Capture screenshots, diagrams, or short videos to demonstrate the functionalities that make the wiki useful for your workplace. The functions might be drag/drop tools, shortcuts, collaborations, and special resources. You will need to discuss these functions in your project completion report to justify why this wiki software was the best choice for your project. You can also use the visuals on your homepage to show your co-workers what the software can do.
3️⃣Draft the homepage content using a word processor (Word or Docs) so you can easily edit and proofread. Your draft should begin with a welcome and purpose statement so Karen, Richard, and all your co-workers know they are on the company wiki landing page. Then, you should write a new paragraph that describes how this wiki generally works (using your notes and relevant images from step 2). The final section needs to include best practices for writing technical descriptions using your notes from the Documentation Structure lecture and bonus materials.
4️⃣Proofread and finalize your copy using an Editor (Word) or a Proofreader (Docs). Once you are satisfied with the content, copy your text onto the homepage of your wiki.
5️⃣Build the rest of the wiki pages using appropriate navigation and information architecture. That means you must add pages for your object/equipment and technical process. The software you select will affect how you build the wiki, and you should use your notes from Step 2 to select your software.
Software documentation explains how the software works and how to use it. It includes information about installation and a description of the software’s functionality. Balancing technical accuracy with user-friendly language is crucial. Presenting information in simple, easy-to-understand terms ensures that even non-technical users can effortlessly navigate the software. You should focus on the features and how to use them (the opposite of the last project on a tutorial) because you want to describe what the software can do rather than specifics of how to use it.
AI Prompting!
The AI interns are very good at this task. You should write these descriptions independently so you know how to do it. If you work for a company making proprietary stuff, they will likely become BIG MAD at you for inputting details to the AI intern instead of describing it yourself (because then your intern’s developers will have all those details and such).
So… don’t use the intern this time. Below is an example you can analyze and model, though.
Gemini Prompt: write a technical description for yourself.
1. Overview:
Gemini is a large language model (LLM) developed by Google AI and trained on a massive dataset of text and code. I can generate text, translate languages, write different kinds of creative content, and answer your questions informally.
2. Key Components:
- Transformer Architecture: My core architecture is based on a neural network called a Transformer, allowing me to process and understand language at a deep level.
- Massive Dataset: I am trained on a massive dataset of text and code, including books, articles, code repositories, and web pages. This allows me to access and process information from various domains.
- Attention Mechanism: I use an attention mechanism to focus on specific parts of the input text, allowing me to understand the context and relationships between words.
- Parameterization: My large number of parameters (over 137B) allows me to learn complex patterns and relationships within the data, leading to more nuanced and informative responses.
3. Functionality:
- Text Generation: Based on your instructions and preferences, I can generate different creative text formats, such as poems, code, scripts, musical pieces, emails, letters, etc.
- Translation: I can translate languages accurately and fluently, supporting over 100 languages.
- Question Answering: I can answer your questions in an informative way, even if they are open-ended, challenging, or strange.
- Information Retrieval: I can access and process information from the real world through Google Search and keep my response consistent with search results.
- Learning and Adaptation: I can learn and adapt to new information and tasks, improving my performance.
4. Technical Specifications:
- Model Architecture: Transformer-based, with 137B parameters.
- Dataset: Text and code size varies depending on the specific model version.
- Training Process: Supervised learning with various techniques like backpropagation.
- Hardware: Large-scale TPU clusters for efficient training and inference.
5. Limitations:
- Limited understanding of the real world: I can access and process information, but I do not have my own experiences or emotions.
- Potential for bias: My responses might reflect biases present in the training data.
- Continuous learning: I require ongoing training and updates to stay relevant and accurate.
Overall, Gemini is a powerful language model with many capabilities. I constantly learn and evolve, aiming to be a valuable tool for understanding and interacting with the world.
