When I worked at Microsoft, I started blocking every Friday afternoon for "20% time" (even though it was only ~10% of my working week). My (really good) manager was supportive and my argument was that, to be most effective in my role, I needed to spend dedicated time learning.
My single piece of advice is that, if you're going to do this, be committed to it. I had a block on my calendar. I had an office (in those days) and closed my office door. I didn't respond to emails and I declined meeting requests. Consistency was key and, once everyone knew that it was my learning time, it was respected (in part because I respected it). One afternoon, my boss knocked walked in, looked at me, realized what time it was, turned around and walked back out.
After Microsoft, I worked at Google where "20% time" (had been a thing but no longer really was in the 2010s) and, once again, Friday afternoons were blocked on my calendar and I used them diligently for learning. My Google managers were consistently supportive and respected my commitment to the time.
One advantage to Friday afternoons is that they're generally very slack time. People are either leaving early or working little, emails are fewer and, in a global organization, (for US West Coast), much of the rest of the World has already entered the weekend.
I think there's no issue with permitting writers to use AI tools to participate in NaNoWriMo.
I've participated a three (or four?) times in the annual NaNoWriMo and completed twice.
When I first participated, I attended a group to help understand how to approach the project. In the group there were several aspiring writers. The group continued after the year's project began and many of the people who most wanted to be writers were already struggling to meet the daily writing quota (so that they'd reach the goal after 30 days).
I think any tools that people use that help them complete NaNoWriMo are fine. People must live with their own decisions and, if the tools write the majority of their submission, that's their decision.
Other people using AI tools doesn't impact my ability to complete the year's NaNoWriMo. It doesn't affect me in any way.
The group leader told us about a tool that she used that would begin erasing characters if she stopped typing for longer than about 5 seconds during her daily write. That's a tool I wouldn't ever use :-)
One person using generative AI doesn't impact anyone else's ability to complete the challenge, but it does seem weird to officially condone using AI and partner with an AI company. If the goal of the challenge is to write 50,000 words in 30 days, and you use an AI to write most of that... then you didn't meet that goal. Like, at all. We should be honest that telling an AI to write a story and writing one yourself are very different things. Anyone who writes knows that having an idea is the easy part, and most of the work happens in turning the idea into the story. Writing to meet an aggressive quota like in NaNoWriMo takes work and discipline. I used to like NaNoWriMo for promoting writing as a creative exercise and motivating people to write and finish a project, but the main criticism has always been that pushing people to set potentially unrealistic goals can also be discouraging or encourage bad practices. Saying "AI is fine if it helps you meet the quota" kind of doubles-down on NaNoWriMo as a word factory and misses the point of its own challenge.
And you could argue the competition is about creating meaning and orchestrating a story rather than individual words.
Even if you use AI to actually produce the individual words and still edit the text afterwards to match your meaning, first of all it might be more work than just writing it yourself, and second, it's basically like having a partner to throw around ideas with.
So sure, you might be cheating yourself, and at the end I don't think AI is going to produce anything of such enormous original value that it would in itself threaten the value of, say, literature.
At the end of the day the results are what count to me. If you're using AI in a disciplined way to help you learn to write better, then all the best to you.
The goal of the competition is to show that disciplined and consistent work can turn an idea into a finished first-draft of a story. You are competing against no one but yourself (and the clock), and if you cheat by using AI to write for you, then you are cheating no one but yourself.
> first of all it might be more work than just writing it yourself
I mean, that sounds like a great reason to not use AI, and if true, kind of defeats the purpose of condoning its use in the first place.
You're making this weird assumption that the only use of AI is to literally write the story for you, but 1) that's wrong and 2) nobody is actually suggesting or encouraging that for NaNoWriMo.
I was intrigued by this and looked through the README for examples of Afrim being used and learn more about the problem it's solving and how but didn't find these.
This video linked from the HN post is an example, although I did't understand yet how it works. They type on a latin keyboard but get some other symbols in result.
Noob me has no idea how editing is done. Delete and re-enter?
Have you ever used WordPerfect's reveal codes mode? It'd show both the actual input stream (characters you typed) and the formatted (rendered) output at the same time. (Here's the top hit showing the feature on Windows. https://youtu.be/LQOYYi2IHIY I used the DOS version, back when dinosaurs roamed the earth; I loved it.)
Should/Could there be a "reveal codes" mode for text input?
I've never used an IME, so please disregard this notion if it's not even wrong. :)
PROJECT="..."
ACCOUNT="..."
# Add this value to your Sheet(s)
EMAIL="${ACCOUNT}@${PROJECT}.iam.gserviceaccount.com"
export GOOGLE_APPLICATION_CREDENTIALS=${PWD}/${ACCOUNT}.json
Interesting, so this approach still works through an explicitly created service account?
Do you think it's possible to make it work without a GCP Service Account at all?
I began note-taking at school when I would make extensive hand-written notes (using colored pens) to study for examinations.
I continued this practice at college and credit my success at school and college in large part to the confidence I created in the "memory" that my written notes gave me.
I continue to feel that I have a very poor memory and so note-taking has become a lifelong skill|tool.
Hand-written notes are best for me to actually remember (without recourse to my notes) but I almost entirely swype|type notes today because of the benefits of accessing them anytime (any device), searching etc.
I try to keep my note-taking DRY and prefer to update|correct older notes rather than recreate content.
1. I write technical blogs partly to share but equally to document for my own purposes; I have a blog called "snippets" (co-opted from a Google practice) where I document tips-and-tricks from coding (e.g. how to set up Protocol Buffers proto_paths when using Visual Studio code)
2. I keep arbitrary notes in Google Keep. My shopping list was migrated to Keep too and, I write trip packing lists and the humans' names at the dog park (keyed by their dog's name, of course) there too.
3. Google used to have (!?) an excellent practice that everyone in a meeting recorded meeting notes in real-time in Google Docs. I would comment-ping people with to do items.
4. Until recently, I used Google Docs for record my programming notes. Every day, I would add a new H3 header for the date with "Actions", "To Do", "Miscellany" sections. I have 3 years of one project's development recorded this way and it served me well. However, as the document grew in size, it became unwieldy and I was forced to split it into two documents (before|after) which made it less useful. I've begun (though it's also imperfect) using Markdown and a Hugo static site to record the same content. Now, each day is a dated Markdown file. This scales better. I prefer using Markdown. It's just as accessible but it's not as good for searching.
5. I contribute to Stack overflow and I've a GitHub repo for every answer I've ever worked on. When I write code or explore a technology, just as everywhere else, I document the solution (as a README) and commit it to the repo. This is mostly for my benefit because it allows me to quickly find technologies that I've worked on.
My coding velocity is slowed by my sometimes obsessive desire to make notes but, note-taking has paid me so many dividends over the years. For example, I rarely feel anxiety that I know I've used an obscure tool, programming language or some such but am unable to find a record of this; reading my extensive notes gives me a thorough play-by-play of what to do.
I worked on PC support at an investment bank in London in the early 90's and developed a Notes app to replace our paper-based tickets.
It took me very little time to develop and really revolutionized our team's ability to manage the incoming flow of tickets for our nearly 1000 users.
We soon gave access to the system to delegated principals in each of the departments who were then able to coordinate their department's tickets in real-time too. They loved it.
> "separate server pools solely dedicated to handling the scraping Google and Microsoft do from Reddit every day"
I assume Google and Microsoft are HTML "scraping", not using the API and not paying. While it's in Reddit's self-interest to be indexed by the search engines, the article mentions these companies also use Reddit's corpus to feed their AIs, is this not monetizable?
It is reasonable for Reddit to want to monetize its content particularly when it loses ad revenue to 3rd-party apps but, could its approach not lead developers to follow Google's and Microsoft's lead and try to recreate (some of their API experience) by scraping Reddit?
I have a hacky a solution to update Google Domains Dynamic DNS record but would value a more functional API to update DNS records. Had I known about this before today, I may have migrated my domains to porkbun for this reason alone.
We previously built probably an equally hacky DDNS client using Python, an archive of it is hosted on GitHub under PorkbunDomains. Unfortunately we haven't been able to keep up with dev for it.
I know there's support for our API on the current GitHub hosted version of ddclient, however I haven't had a chance to get it working and a tutorial out, but it's an option.
For no frills dns and dynamic dns update via api, try hurricane electrics dns service: dns.he.net. Not a registrar but free dns hosting. Happy customer for more than a decade now.
I enjoy sci-fi series, interesting spacecraft, compelling aliens and am a sucker for machine-based intelligence.
This series combines all these and re-engaged in sci-fi. I went on to read most of Asher's other books including "Agent Cormac" and "Spatterjay" series.
My single piece of advice is that, if you're going to do this, be committed to it. I had a block on my calendar. I had an office (in those days) and closed my office door. I didn't respond to emails and I declined meeting requests. Consistency was key and, once everyone knew that it was my learning time, it was respected (in part because I respected it). One afternoon, my boss knocked walked in, looked at me, realized what time it was, turned around and walked back out.
After Microsoft, I worked at Google where "20% time" (had been a thing but no longer really was in the 2010s) and, once again, Friday afternoons were blocked on my calendar and I used them diligently for learning. My Google managers were consistently supportive and respected my commitment to the time.
One advantage to Friday afternoons is that they're generally very slack time. People are either leaving early or working little, emails are fewer and, in a global organization, (for US West Coast), much of the rest of the World has already entered the weekend.