Promote transparency and authentic student voice with AI writing tools. Discover how TextJam helps teachers guide responsible AI use.

Thank you for reading my latest article Future Of Writing And AI: How TextJam Is Taking On Word And Google Docs. Here at LinkedIn and at Forbes I regularly write about management and technology trends...

Promote transparency and authentic student voice with AI writing tools. Discover how TextJam helps teachers guide responsible AI use.

My project for the textjam, "akrete," is now at the point where the hard work begins. I successfully ported all the easy logic from Iron Arachne's Typescript to akrete's Python. The only hiccup I ran into was in the random number generator class's "next()" method, where in the Typescript version I was using unsigned bitwise operators to handle some of the logic. Python only has signed bitwise operators. It was easy enough to work around, and I knew I'd hit something like that when porting. Ah well, it's done. Also, I discovered that my go-to CLI library for Python, argparse, has long ago been superceded by a library called Click. That was a pleasant discovery. Click is much nicer to work with than argparse. I'll be using that going forward. Anyway. Akrete now generates random planets, including names and statistics, and outputs the results to the terminal. I even found and fixed a couple bugs in the statistics generation that I'll need to go back and fix in the Typescript. As it turns out, I was making a few silly hard-coded decisions about orbital distance and planetary temperature. The calculations for planetary temperature are still quite strange, but since they're mostly in line with what I want the output to be, I'm leaving that alone for now. The hard part is next. I have to add ASCII images of the planet to the output. Effectively, I'll be generating very low resolution pixel art, using ASCII characters. For the Iron Arachne version of the planet generator I use GLSL shaders to handle the planet images. This will be very different, but maybe some of the logic will translate. Who knows; if this works, maybe I'll backport the ASCII image generation to Iron Arachne, to use in utility scripts for testing purposes. I've been doing that more and more lately, especially while working on the region map generator. ...oh yeah, I haven't mentioned the region map generator before. Ah well, just ignore that sentence for now. Assuming I can figure out how image generation will work, Akrete should be done in time to submit to the jam. Once it's submitted, I'll add a link to the code base on my Works pages.

I decided to participate in Jack Harrhy's "game jam-like" for text-based projects, textjam. It's my first time participating in any kind of a jam. The requirements are pretty permissive. We just have to create something that outputs only to a terminal; no graphics allowed. Also, LLMs are strongly discouraged, and only allowed if you describe exactly how and where you used them. Oh, and it should really be something created entirely for the jam, and not be part of some larger extant project. And, of course, it should be fun and entertaining. So while non-games are allowed, it should really be a game, or at least some kind of toy. The hard part here was figuring out what to build. My first idea was writing a planet generator kind of like what I have on Iron Arachne. The trick there is that the planet images would have to be built from ASCII characters. That is definitely going to be difficult, though not impossible. Other options I thought about included things like a language generator, a simple roguelike, an ASCII clone of Legend of Zelda... there are a ton of possibilities. I was tempted to pursue a bigger project, but given my inexperience in writing games, I would probably underestimate how much time it takes. The planet generator idea is pretty safe, even if it does have a difficult component. It doesn't require animation, collision detection, or interactivity. If I decided to make a sci-fi text-based game in the future, it could also prove useful. A lot of the math for planet statistics I already have figured out as a result of the Iron Arachne version. I could probably expand those, too, and make the descriptions more interesting. So, having decided on the project, the next decision to make was what language to use for writing it. The language has to allow for rapid iteration. It should be able to run on any of the major operating systems. And it should be fun to use. I briefly considered using Rust, but that's too heavy for this. The same is true for C, C++, and so on. Java's out because I hate Java. My short list ended up being Javascript, Python, Go, and Crystal. I really was leaning towards Crystal, but in the end, I decided making the result cross-platform was more work than it was worth. Javascript was too familiar, since Iron Arachne was already built in it. I eliminated Go from consideration because, while it checked all the boxes, the iteration loop was just a smidge too slow. So, the language I decided on was Python. The name needed to be meaningful, but also as unique as possible. "Planet-gen" was the first thing that popped into my head, but I discarded that as way too boring. Then I looked up terms for the creation or birth of planets. The scientific process for that is "accretion," and since I like the idea of verbs as CLI commands, I changed that to "akrete." So, that's the project's name. I created the git repository for it on Codeberg. Now I have until June 7 to build this and submit it. I don't know if it'll be done in a few hours, a few days, or not until the full month is over. We'll see!

The newly launched TextJam introduces a collaborative writing platform, offering innovative features for writers to creatively work with AI and each other.

Promote transparency and authentic student voice with an AI writing tool. Discover how TextJam helps teachers guide responsible AI use. The post How to Promote Transparent AI Writing in Your Classroom...

本周，TextJam 正式发布，这是一款颠覆传统人机协作模式的多人编辑工具。其核心创新在于赋予用户对 AI 生成内容的精细控制权，允许用户明确指示 AI 保留或修改特定部分，从而显著提升写作效率与质量。 TextJam 的交互设计融入了从传统笔尖输入到多点触控手势的创新，并能智能调整文本大小，重新定义了文字处理器的可能性。该工具旨在让用户以更快的速度完成更高水平的写作。 TextJam 的推出标志着写作工具领域的一次重要进步，尤其是在 AI 辅助创作方面，为用户提供了前所未有的灵活性和控制力。 查看消息来源