【AI x learning】 AI Is a Co‑Creator, Not an Answer Vending Machine: Crafting a Good Opening Line

“Your very first sentence decides whether you and AI become soulmates—or just plastic acquaintances.”

Published on: 2026/01/24

---

Introduction

Is this you?

Ting‑an is a freelancer. Recently, projects have been scarce, and the pressure of rent and living expenses has left him feeling crushed by life.

He heard that chatting with ChatGPT can sometimes feel even better than venting to a real person, so he opens ChatGPT and types: “Work has been stressing me out lately. Can we talk about it?”

ChatGPT quickly replies: “Of course, I’m here with you. Would you like to start with the situation itself, or how you’re feeling about it?”

In truth, Ting‑an was hoping for concrete suggestions about his direction as a freelancer. But ChatGPT’s response feels more like emotional support, which doesn’t quite match what he needs.

“Aren’t people saying ChatGPT is great at solving problems? This doesn’t feel that impressive…” he thinks to himself, slightly disappointed.

The scene shifts to Zhe, a marketing planner struggling to write copy for a vacuum cleaner campaign.

He opens Gemini, hoping to brainstorm with AI and see whether human creativity and AI can spark something new together.

Zhe has taken AI training courses, so he roughly knows how to talk to AI and set roles. He types: “Please act as a senior marketing copywriter and write a humorous marketing copy for me, within 300 words.”

He assumes that once AI gives him the copy, he can call it a day and go home. But instead, AI produces a vague, atmospheric piece and ends with: “You know what? Even your cat would want to buy this vacuum cleaner!”

Zhe feels the copy doesn’t highlight the product’s unique features at all. The target audience also seems oddly limited to cat owners. It sounds humorous, but something feels off.

The camera moves again to Hsiao‑yun, a graduate student working hard on her thesis.

Before writing the thesis, she needs to submit a proposal to the department office. She has a rough idea in mind but hasn’t organized it into a structure she can actually hand in, which makes her quite anxious.

So she opens Copilot and types: “I need an outline for a thesis in educational technology. Please help me organize five main sections, each with a short explanation.”

Copilot generates a lot of content about the history of educational technology, tool applications, teacher training, and more. It looks like a lot of information.

But as she reads, a thought surfaces: “None of this really looks like something I’d actually submit.”

These situations feel frustrating and discouraging. Yet not everyone seems to have this problem.

A good beginning is half the journey

There’s a group of people who, before asking AI anything, first clarify what they want to gain from the conversation. They think about what context and background AI needs to know, and what their final Goal is.

If you happen to see their screen while they’re using AI, it looks a bit strange. They keep typing, but AI isn’t replying yet. They don’t seem anxious at all—in fact, they almost enjoy this process of laying the groundwork.

It looks time‑consuming, but in the end they often get AI outputs that fit their needs with minimal back‑and‑forth or revision.

For them, AI feels like a soul‑level collaborator and a thoughtful thinking partner. AI seems to really “get” them and can provide precise, tailored content and information.

This is the power of a well‑crafted Opening Line.

The first sentence decides the quality of the conversation

Not assigning a task, but inviting AI into your thinking

The Opening Line is the very first question you type into the chat box when you start an AI conversation.

People who can turn AI into a truly supportive assistant aren’t necessarily those who know the most about prompt design or templates. They’re the ones who know how to open their mouth—and what to say in which situation.

Very often, when we use AI, we treat the Opening Line like a task assignment: “Help me write an article,” “Act as this or that role,” “Give me three suggestions.”

This kind of language structure is fundamentally command‑based. It’s one‑way, and it treats AI as a cold tool rather than a co‑creator in the AI conversation.

But truly high‑quality dialogue doesn’t start from tasks. It starts from context. It’s not one‑directional or purely commanding, but a two‑way interaction where back‑and‑forth exchange creates better outcomes.

That’s why we emphasize the design of the Opening Line.

The Opening Line is not just “what I want.” It’s “where I am right now, how I hope you’ll understand me, and what we’re going to accomplish together.” It’s a way of opening the Conversation Framework, not just dropping a task into a slot.

A good Opening Line is like rolling out a carpet so AI can walk into your context. It’s not “here’s what I want you to do,” but “here’s how we can begin together.” This design mindset comes from understanding language—and from your expectations for this shared journey.

At the same time, the way you ask questions shapes the depth of the conversation. If you only say “help me organize five aspects,” AI will give you something structurally correct but not necessarily aligned with your thinking.

But if you say, “I’m exploring directions for a thesis in educational technology and want to clarify which aspects are worth diving into. Please help me organize five angles with brief explanations,” AI will step into your thought process instead of just completing a surface‑level task.

Shift your mindset and treat the Opening Line as an invitation letter to a conversation. How you write this invitation determines whether AI shows up, how it shows up, and in what way it walks alongside you.

For you as the user: higher willingness to engage

A good Opening Line doesn’t just help AI understand you better. From your perspective as the question‑asker, it also makes you more willing to keep talking.

It acts like a psychological “on switch”—a sense that “I’m intentionally designing this conversation,” which boosts your feeling of control.

According to Self‑Determination Theory (SDT)_[1], when Autonomy, Competence, and Relatedness are supported, people’s motivation shifts into a higher‑quality form. They become more willing to engage in the behavior.

When you can lead the conversation, clearly explain the background, and feel AI’s responses becoming more aligned, you naturally want to invest more in the interaction. The AI conversation stops being a series of cold commands and becomes a meaningful exchange.

In Expectancy‑Value Theory (EVT)_[2], motivation comes from Expectancy and Value. When you believe “this conversation will be useful for me” and that it “might help solve my problems or worries,” you’re more inclined to say a bit more in your Opening Line.

That extra context doesn’t just help AI align with your Goal. It also helps you trust that this dialogue is worth having.

Research by Tversky and Kahneman (1981)_[3] shows that even when content is logically equivalent, framing it in positive or negative terms can significantly influence people’s decision preferences. The way a sentence is phrased shifts what people focus on.

This reminds us that in AI conversations, the design of the Opening Line—and the overall rhythm and tone of the dialogue—can influence the decisions we eventually make about our problems.

For AI: helping it truly understand you

AI may look like it can solve almost anything, but it’s definitely not living inside your head.

You might secretly hope it will behave like a partner who understands everything about you without you having to say much.

But even between partners, accurately guessing someone’s preferences, style, or size when buying a gift is a low‑probability event.

It’s the same with AI. If you say nothing, it can’t magically know what you’re thinking. It can’t pre‑assume your preferred dialogue rhythm or decision criteria. At best, it can only infer from the language cues you give it.

If we’ve already formed a mental picture of what the “right answer” looks like, we’re very likely to feel disappointed when the conversation doesn’t match that expectation.

In a 2020 study on GPT‑3, Brown et al._[4] pointed out that GPT‑3’s performance is closely tied to the user’s text prompts. In other words, the model’s understanding is driven by your language and the cues you provide.

This “understanding” doesn’t come from some innate wisdom in the model. It comes from how you guide it with your words.

The same study also showed that when given enough examples or detailed user requirements, the model can display behavior that looks like understanding and can handle abstract reasoning or tasks.

In short, the design of contextual prompting directly affects the quality of AI’s replies. When you choose to say more, AI can respond in a way that’s much closer to your needs and expectations.

More recently, Zhou et al. (2023)_[5] found that the quality of large language model responses is strongly related to how the initial prompt is phrased. These strategies don’t require extra training—just better language design to improve alignment.

Vague Opening Lines make it easy for the model to drift away from your intent, while precise contextual design improves consistency and relevance.

Traditional prompts that lack context‑driven guidance tend to cause over‑generalization or “context drift,” making the replies feel less tailored.

All of this shows that the Opening Line is not just a task instruction. It’s your way of saying, “Here’s how I want you to understand me.”

If you’re willing to share more about your background, purpose, and even preferred style, AI will interpret you in the way you hope. That, in turn, shapes what role it takes, what tone it uses, and what decision logic it applies in its replies.

If we’ve already pre‑decided what the answer should roughly look like, why not proactively tell AI?

Clark and Brennan’s 1991 study_[6] also reminds us that effective dialogue relies on building “common ground.” Beyond tone, purpose, and background, speakers use strategies like paraphrasing, asking questions, and backchannel cues (“mm‑hmm,” “oh”) to confirm mutual understanding.

This kind of conversation is a continuous negotiation process, not a one‑time understanding.

AI isn’t human, but its language model still depends on background information and contextual scaffolding. Only then can its replies better match what you actually want.

Deep conversations aren’t just for humans—they apply to AI too

We hope AI can understand us and give the responses we’re looking for because we’re used to the feeling of being understood in human relationships.

But human understanding isn’t innate either. With strangers we’ve never met, we rely on emotional tone, responses, shared background, accumulated memories, and the courage to ask deeper questions to gradually understand each other.

Kardas and Schroeder (2022)_[7] found that people often talk too little in conversations and underestimate how much joy and closeness deep conversations can bring.

This may come from social anxiety around “deep questions” and worries like, “Will this be too abrupt?” or “Will they not want to talk about this?”

When using AI, question‑askers may have similar concerns: “What if I type all this and AI still replies randomly?” or “Isn’t this inefficient?”

In reality, more background and clearer purpose help AI better approximate the shape of the answer in your mind.

When you’re willing to add just one more sentence like “Here’s my current state…,” it becomes much easier for the other side to enter your world. AI is no different.

Not wrong—just not yet taken care of

Maybe you’ve searched online for “good prompts” to set roles or tone. Maybe you simply open an AI window and start issuing commands, or just type whatever comes to mind.

None of these approaches are wrong. They’re all part of exploring how to build a connection with AI—a way of saying, “I’m trying to help you understand me.”

Sometimes our sentences look complete on the surface, but we haven’t clearly communicated the underlying purpose.

For example, “Please help me write an article” is not inherently a bad sentence. But without background, audience, tone, or purpose, AI may produce something that doesn’t really match your needs.

At that point, you might think, “Why is this so shallow?” But it’s not really AI’s fault, nor is it yours. The conversational space simply hasn’t been prepared yet.

Other times, we use role settings like “Please act as a senior marketing planner.” In our minds, we’re thinking, “Now you should know how to talk, right?” But without context, purpose, and follow‑up dialogue, it can turn into mere imitation rather than understanding. AI’s reply might sound right on the surface, but you still feel, “It’s kind of right, but not quite.”

None of this is a mistake. It’s a reminder about the design mindset behind the Opening Line. It’s not just a technical trick or a template with variables—it’s a way of saying, “I want you to really understand me.”

It asks us to say a bit more and think one layer deeper so AI can truly enter our world instead of circling around the surface of the task.

Our goal is not to dismiss these attempts, but to take care of the expectations behind them.

We believe that when Opening Lines are designed to better support psychological engagement, clearer context, and emotional cues, AI doesn’t just respond more accurately. It also makes it easier for people to feel, “I’m genuinely being understood.”

This isn’t about being “smarter.” It’s about starting more gently.

Crafting a good Opening Line is the starting point of high‑quality dialogue

Throughout this series, we’ve moved from users’ psychological needs, to how AI understands language, to the deep structure of human conversation. All of it points to the same thing: the Opening Line is not a technical trick, but a way of designing dialogue.

It’s not “how to make AI do things,” but “how to let AI enter your thinking.” It’s not “what I want,” but “how we begin together.”

When you’re willing to share more background and purpose in your first sentence, you’re not only helping AI understand you better. You’re also making it easier for yourself to keep asking. You’re not just operating a tool—you’re inviting a co‑creator to work with you.

So here’s the question: what exactly counts as a good Opening Line? How should we design that first sentence so AI can respond accurately, deeply, and in a way that truly fits?

Next, we’ll walk through the design logic of the Opening Line Generator, break down the elements of a good Opening Line, and actually practice crafting one that’s uniquely yours.

This isn’t just an exercise. It’s a real beginning. Ready? Let’s step into your very first sentence together.

The magician of opening lines: The Opening Line Generator

Before designing your Opening Line

In earlier articles, we talked about conversational Tone, helping you understand how much control you take in AI dialogue and how you guide the flow.

Then we explored Motivation types and Goal Anchors, so you can see what kind of Motivation you bring into each question and what final result you’re aiming for.

If you haven’t read the related articles in this series yet, I recommend revisiting these three pieces first:

【AI x learning】 From Passive Use to Thinking Loops: Why You Must Learn to Ask Before Using AI

【AI x learning】 Asking Is More Than Input — It Is the External Shape of Your Inner Direction: Finding Your Starting Line

【AI x learning】 Conversations Aren’t Driftwood — They Move With Coordinates: Learn to Set Goals So AI Truly Knows Where You Want to Go

These articles help you gently examine your own state and clarify your Motivation and Goal before you start an AI conversation.

The fields in the Opening Line Generator are actually an extension of the previous Goal Anchor Generator. Those fields are exactly the elements needed to build a good Opening Line for prompt design.

If you haven’t clarified your Goal or Motivation yet, you might find it hard to fill in.

So I suggest going back to those earlier articles, doing the checks, and then returning here to fill things out.

Take it slow. Settle in and complete it with patience. You’ll find that the process itself is a form of thinking and organizing.

Of course, if you already know yourself well, you can jump straight into the Opening Line Generator below.

What makes a good Opening Line?

So, what kind of Opening Line counts as “good”?

Recent research on how to talk to AI and design prompts for large language models has become quite detailed. In a 2023 study, Lin_[9] proposed “Ten Simple Rules for Crafting Effective Prompts for Large Language Models.”

In 2024, Hewing and Leinhos_[10] introduced the “Prompt Canvas,” which integrates commonly mentioned elements in the literature, including role setting, context, task instructions, format requirements, style and tone, and constraints.

In 2025, Do et al._[11] went further by synthesizing over 150 papers and proposing 21 prompt attributes grouped into six dimensions: context, clarity, style, task orientation, interactivity, and constraints.

Based on these studies and our previous three articles, we distilled the key elements a good Opening Line needs to balance:

Context

Context is the background information and situation of the conversation. Without context, AI can only guess, and it’s easy for it to miss the point.

In our Opening Line Generator, context is supplemented through Motivation‑specific fields. For example, Problem‑oriented users specify task outcomes and deadlines, while Thought‑organizing users describe confusion points and constraints. These help AI quickly enter your thought process.

Clarity

Clarity is about being concrete and specific, avoiding vague or overly broad wording. The clearer your Opening Line, the more precisely AI can target your needs.

In the generator, this is implemented through the “Goal Anchor Sentence,” which asks you to state your Goal in one sentence, such as “Finish the outline for 10 slides before 17:00 today.”

Even if the Goal Anchor is short, the earlier background fields make the overall context more complete.

Role & Style

Role and tone decide “who” AI is in the conversation and “how” it speaks. This directly shapes the atmosphere and how closely the replies match your expectations.

In the Opening Line Generator, this is handled through the “AI Role” and “Reply Tone” fields. You can describe them freely, such as “technical consultant with a precise, direct tone” or “creative collaborator with a gentle but structured tone.”

Still, remember that role and tone may shift as the conversation unfolds, as AI continues to infer from your language cues.

Task Orientation

Task orientation means making it clear what concrete task you want to accomplish. Without it, AI’s replies can easily drift into small talk or stay unfocused.

In our Opening Line Generator, task orientation is expressed through the Motivation modules and their dedicated fields. For example, Creative‑support users specify the creative form, purpose, style references, and so on, which all make the task more concrete.

Interactivity

Interactivity is about whether the prompt invites a dialogue rather than a one‑way command. Prior research also notes that iterative questioning and confirmation gradually build understanding, highlighting the importance of interactivity.

The Opening Line Generator may not fully capture interactivity on its own, since it only shapes the first sentence. But once you start asking follow‑up questions, you’ll notice that a well‑designed Opening Line gives AI more room to surprise you and co‑create insights.

Constraints

Constraints help AI narrow the scope, such as “no more than 500 words” or “don’t talk about technical details, only mindset.”

In our Opening Line Generator, constraints are embedded in Motivation‑specific fields and conditions, such as “preconditions” for Thought‑organizing or “what not to cover” for Foggy‑exploration. These help AI stay within the boundaries of your needs.

Highlights of the Opening Line Generator

From these studies, we can see that a good Opening Line must balance context, clarity, role and tone, task orientation, interactivity, and constraints.

Our Opening Line Generator breaks these elements into easy‑to‑fill fields so that, almost without noticing, you end up with a high‑quality Opening Line for your AI conversation.

It doesn’t ask you to write a perfect sentence in one go. Instead, it guides you step by step to fill in what matters.

The final generated text box also leaves room for edits, so you can adjust it to your needs. This design turns the Opening Line from a source of pressure into a supported process.

Once you finish filling it out and click “Generate Opening Line,” the system automatically combines everything into a complete paragraph and displays it in the text box below.

You can freely edit, adjust, and then copy it with one click.

Its strength lies in being dynamic rather than a rigid template. It adapts to your Motivation and Goal, turns your psychological needs into language structure, and makes it easier for AI to understand you. At the same time, it preserves flexibility for you to fine‑tune.

You’ll also notice that the generator automatically adds a final sentence: “Please don’t output yet; I’ll provide the dialogue rhythm later.” That’s because the next article will focus on dialogue rhythm with AI.

If you want the full experience, you can wait until you’ve read that article and then use the generator. But if you’re ready now, you can simply delete that sentence and start right away.

Now you can slow down, put on some soft music, and start filling in your Opening Line Generator.

AI Opening Line Generator

Fine‑tuning your Opening Line

The Opening Line you get from the generator is only the first step. The real conversation begins when you adjust it with your own hands.

We’ve emphasized that the Opening Line is the most important opening sentence in your AI conversation. There’s no single perfect answer—only “better fits” for your style and needs.

The elements that research highlights—context, clarity, role and tone, task orientation, interactivity, and constraints—are already built into the generator.

But the final adjustments still need to be made by you. Through these small tweaks, you make the Opening Line more aligned with your needs and your current state.

If you feel the context in the generated Opening Line is still too vague or the constraints are too loose, you can add more background details. For example, specify the audience or scenario: “This is a talk for a high school audience.” That helps AI focus its reply.

If you’re worried that AI’s reply style won’t match your expectations, you can refine the role or tone more precisely. For instance, add “Please use an encouraging but non‑preachy tone” or “Please be gentle and non‑judgmental.”

Tone adjustments can shape both the atmosphere of your AI conversation and the style of any content AI generates, such as blog posts or speeches.

If you find that the generated Opening Line doesn’t fully match your task, you can explicitly highlight what matters most. For example, add “The most important outcome is…” so AI can focus on your top priority.

As long as the key elements of a good Opening Line are present, the content itself is flexible. Only you truly know what your AI conversation should look like.

That’s why this final fine‑tuning can’t be outsourced. It has to be done by you. Once you’ve adjusted it, this becomes your unique, most suitable Opening Line for this particular AI conversation.

Example of filling out the Opening Line Generator

Tzu‑ning is a graduate student who needs to submit a report but is stuck on the structure and unsure about the writing style.

From the Motivation check, he discovers he has Dual‑Primary Motivation: Problem‑oriented and Thought‑organizing. He decides to prioritize the Thought‑organizing Motivation.

His Goal Anchor sentence is: “The first draft of my final report needs to be completed by this Friday, including three main sections and a conclusion.”

For AI’s reply tone, he writes: “Clear but not stiff, helping me clarify my thinking.”

For the AI role, he writes: “Thought organizer, helping me clarify structure and style choices.”

In the Problem‑oriented fields, Tzu‑ning fills in:

Task goal and outcome: Finish a 1,500‑word final report on “The intersection of technology and education.”

Deadline: By 11 p.m. this Friday.

Success criteria: The professor wants to see a clear argument structure and critical thinking.

Extended exploration: Common report structures and how to avoid overly loose arguments.

Additional curiosity: What are common academic writing styles? How can I choose one that suits me?

In the Thought‑organizing fields, he writes:

Confusion point: I’m not sure whether to use a comparative or argumentative structure, and whether to approach it from the education side or the technology side.

Decision goal: I want to choose a clear main thread and decide on the overall style of argumentation.

Preconditions: It must not drift away from the course topic and must not be overly technical.

Extended exploration: How to build a central argument and how style choices affect readers.

Additional curiosity: Will different structures affect how readers understand the content? What style does my professor prefer?

He also realizes he wants to add more, such as: “My professor prefers critical analysis, but I’m more comfortable with a narrative style. Can you help me find a middle ground?” “I’m a bit anxious, so I hope your replies can help me feel calmer and not be too complex.” “Please reply in bullet points, with no more than 100 words per point.”

He decides to put these extra notes at the end of the Opening Line and ends up with something like this:

I’d like you to act as a thought organizer, helping me clarify structure and style choices, and use a clear but not stiff tone that helps me sort out my thinking. Please help me complete the first draft of my final report, which needs to be finished by this Friday and include three main sections and a conclusion. For these tasks, please start from the Thought‑organizing Motivation. If needed, my task goal and outcome are to complete a 1,500‑word final report on “The intersection of technology and education.” The deadline is 11 p.m. this Friday. The success criteria are that my professor wants to see a clear argument structure and critical thinking. I’d also like to explore common report structures and how to avoid overly loose arguments. I’m also curious about common academic writing styles and how to choose one that suits me. Right now, I’m stuck because I’m not sure whether to use a comparative or argumentative structure, or whether to approach it from the education side or the technology side. My decision goal is to choose a clear main thread and decide on the overall style. The precondition is that I can’t drift away from the course topic or become too technical. I’d also like to explore how to build a central argument and how style choices affect readers. I also want to know whether different structures affect reader understanding and what style my professor prefers. My professor prefers critical analysis, but I’m more comfortable with a narrative style—please help me find a middle ground. I’m a bit anxious, so I hope your replies can help me feel calmer and not be too complex. Please reply in bullet points, with no more than 100 words per point. Please don’t output yet; I’ll provide the dialogue rhythm later.

The final result doesn’t have to be one giant block of text like this. I merged it into a single paragraph here because it’s long. You can try entering it yourself and pasting it into an AI conversation.

Summary

You no longer get stuck at “I don’t know how to start.” Instead, you can naturally build context. The first sentence you say becomes the switch that activates your thinking.

From that first sentence on, you let AI understand you better.

The more clearly you express who you are, the more AI can take you where you want to go. But remember, the Opening Line is not a perfect sentence pattern—it’s a thinking activator. Only you know what suits you best.

Everyone’s Opening Line is a unique way to begin a conversation. It’s not just the start of language, but an invitation into your mind.

And beyond that first sentence, your dialogue rhythm also shapes how you think.

In the next chapter, we’ll help you find the questioning and dialogue rhythm that feels most natural and authentic to you.

References

[1] Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. *American Psychologist*, 55(1), 68–78. https://doi.org/10.1037/0003-066X.55.1.68

[2] Wigfield, A., Tonks, S., & Klauda, S. L. (2009). Expectancy-value theory. In K. R. Wentzel & A. Wigfield (Eds.), *Handbook of motivation at school* (pp. 55–75). Routledge. https://doi.org/10.4324/9780203879498

[3] Tversky, A., & Kahneman, D. (1981). The framing of decisions and the psychology of choice. *Science*, 211(4481), 453–458. https://doi.org/10.1126/science.7455683

[4] Brown, T. B., et al. (2020). Language models are few-shot learners. *arXiv preprint* arXiv:2005.14165. https://arxiv.org/abs/2005.14165

[5] Zhou, W., Zhang, S., Poon, H., & Chen, M. (2023). Context-faithful prompting for large language models. *Findings of EMNLP 2023*, 14544–14556. https://doi.org/10.18653/v1/2023.findings-emnlp.968

[6] Clark, H. H., & Brennan, S. E. (1991). Grounding in communication. In *Perspectives on socially shared cognition* (pp. 127–149).

[7] Kardas, M., & Schroeder, J. (2022). Keep talking: (Mis)understanding the hedonic trajectory of conversation. *Journal of Personality and Social Psychology*, 123(4), 717–740. https://doi.org/10.1037/pspi0000379

[8] Graesser, A. C., & Person, N. K. (1994). Question asking during tutoring. *American Educational Research Journal*, 31(1), 104–137. https://doi.org/10.3102/000283

[9] Lin, Z. (2023). Ten simple rules for crafting effective prompts for large language models. ResearchGate. https://www.researchgate.net/publication/371123456_Ten_Simple_Rules_for_Crafting_Effective_Prompts_for_Large_Language_Models

[10] Hewing, L., & Leinhos, M. (2024). The Prompt Canvas: A literature-based practitioner guide for creating effective prompts in large language models (Version 1) [Preprint]. arXiv. https://arxiv.org/abs/2412.05127

[11] Do, H., Tan, Y. S., Zhang, Y., & Salesforce Research. (2025). What makes a good natural language prompt? (Version 1) [Preprint]. arXiv. https://arxiv.org/abs/2506.06950

FAQ

You may also be interested in these articles