I feel like going to the source of an algorithm/implementation is a super power, as illustrated by this example.
Having used Graphviz for fairly complex visualizations I was initially shocked that someone would rewrite it themselves. Then I saw the breakdown of the algorithm and realized it may not be as complicated as I first thought.
All that being said, as a general rule it is hard to know what the hidden complexities may be until you are finished implementing the algorithm.
I haven’t explicitly recreated a template for an academic venue, but I have recreated a custom template to match and existing PDF. It was pretty straightforward to recreate it as the language and “standard library” (if you could call it that?) is well designed and has excellent documentation.
I love Typst! Currently rewriting my CV in Typst and it has been an excellent experience. One small hindrance is the inability to have multiple bibliographies.
In the past I have also used it to generates quotes (in terms of finances, invoice etc.). It was neat because all of the logic for adding up the subtotal was written in the language (and was fairly easy to understand). I can imagine trying to do that logic in LaTeX…
Agreed, to make it even more interesting Browser Company discontinued Arc earlier this year. So not only did they do all of the things OP listed, but also didn't have a current product when acquired.
That's very interesting. I downloaded arc because I saw it in some twitter screenshot and I thought the UI was neat, when I could have actually been looking for Zed instead of Arc.
Ironic that I wasn't familiar with this company or their products before today, and having read about both Arc and Dia, including reading this blog post you've linked, the product that makes me want to try it is the one they've stopped developing...
It's still worth trying and using. The developers consider it a "finished product" and I don't disagree. It does lots of small things well* that many browsers (even the self-confessed clones like Zen) don't do out of the box, if at all. Maybe in two years the browser will no longer be distributed or receive Chromium updates, but it exists and works fine now.
* For example, I get a lot of value from renaming my tabs and even replacing their favicons with emojis of my choice. Zen appears to have limited support for this.
Damn, I really appreciate the decision to do this in a new product. Arc is the best browser I've ever used, and I'd hate to see AI features forced upon me. Thanks Browser Company!
I agree. I think Arc was the biggest innovation in browser UI since Chrome.
I think you will eventually have to switch because it will lack behind given that it's not their priority anymore. Zen browser seems like viable alternative but I haven't used it enough yet to know how well polished it is.
It is impressive what a single person with a vision can achieve hacking away on Firefox, especially considering Mozilla's track record in recent years.
A bus factor of 1 is still a bit red flag on something as involved as browser maintenance. Hopefully a community can emerge around the project.
Yeah, this is the unfortunate part about products kept alive in maintenance mode in a rapidly evolving space.
I guess you could argue (as TBC did) it’s actually not rapidly evolving, and that gives it staying power. But eventually someone will reach parity and eventually eclipse the original product.
Hopefully Zen does that. I’m just tired of moving the same data to the effectively the same product run by a different team for no good reason.
Interesting to think about if finding a database dump of messages 30 years from now would bring back the same nostalgia. I would think that the tangible aspect would have a more profound impact.
I also think the format of letters lends itself a bit better to being re-read. A single message talking about the "last little while" rather than atomic thoughts and reactions. More context about the snapshot of life it describes. Like a page from a diary that was shared with someone.
I have been enamored with developing a second brain and other productivity hacks, but have recently been turned off to them, because I believe the benefits are over-promised. Similar to OP, I haven’t been able to achieve the clarity of mind and creative thoughts that are promised by a second brain.
While I do think that deleting the whole thing is extreme, I can imagine that there is a level of catharsis experienced by that.
Lately I have subscribed to Oliver Burkeman’s (author of “4,000 Weeks”) line of thinking where life, and subsequently thoughts, are more meant to be experience rather than optimized. For me, I have seen a negative drop in “life enjoyment” when I have tried to capture everything, and have yet to realize the results and even stick with it consistently (which may be the reason for not seeing the positives).
That's an idea... Make a fork, add a file at taxpayers/${SSN}.yaml describing your return in terms of income/deductions/circumstances, make sure it lints successfully, and then submit a PR for the IRS to review. If it's merged, CI/CD initiates a bank payment/withdrawal. If you get audited, resolve the conflicts and update the PR.
I've stumbled across it a few times way back when I was looking at MRP stuff for an ERP I worked on. I remember it being a rare gem of a website. I'd throw away 99% of web pages if the ones remaining were all like this one. Thanks for the reminder!
You need to renew the license every 10 years, but as long as you renew you don't need to take a test (which is maybe what you mean by "lasts a lifetime").
Yeah depends on the country/jurisdiction - in Canada, an amateur radio certification is valid for life, and doesn't require any sort of re-testing or paid renewal or anything. Pretty nice. One of the few times the government has really done something right, IMO :)
In the US, it lasts a lifetime but you have to tell them you're still alive every ten years. This is free, but if you don't file the form, your license is no longer valid.
Having used Graphviz for fairly complex visualizations I was initially shocked that someone would rewrite it themselves. Then I saw the breakdown of the algorithm and realized it may not be as complicated as I first thought.
All that being said, as a general rule it is hard to know what the hidden complexities may be until you are finished implementing the algorithm.