chrisdalke

Not only that, but Java's reflection features opened the door for a huge, vibrant modding community that is arguably one of the largest among any PC game.

Very early in Minecraft's development, people were already decompiling/modifying/injecting their own mods, and a lot of frameworks (Bukkit, Spigot, etc.) emerged to provide a common API for modding.

The large modding community arguably had a very positive impact on Minecraft's early success -- Although I don't have any quantitative metrics to reinforce that point, I fondly remember early Minecraft as having a relatively technical community that tinkered with the game as a sandbox for countless custom experiences.

Yeah, for sure this is the type of niche, high-effort work that would be used extensively by the closed-source world and yet struggle to find adequate support.

I would also love this, in the past I’ve used the H2 in-memory database for tests with Spring boot applications.

The Hibernate ORM handles configuring the H2 schema but they’re not fully compatible, so it means I have to be careful to not rely on Postgres-specific features. I generally am not testing database logic, just need a working persistence layer to get the tests running, but an in-memory Postgres implementation would be amazing.

Fusion 360 is free but has a certain clunkiness that is hard to get past if you’ve used other CAD products.

There are many sneaky ways that people get expensive CAD licenses (I use Solidworks) - many organizations offer an educational license if you are a member. For example, check out the Experimental Aircraft Association.

My favorite use of a Raspberry Pi has been to run a Jenkins instance.

I use it for CI/CD on projects, but also for automating other tasks -- You can use Jenkins to wrap any arbitrary script with more higher-level logic and extensibility than a cron job.

For example, I use Jenkins to automate multiplatform builds for some side projects, to periodically ingest data into a database, perform cleanup jobs, etc.

I wish GitHub had a better interface to browse forks, right now it just displays a single list and requires you to search through individual forks to find any meaningful changes.

They also featured "fork this project" prominently for a while leading to many novice Git users forking projects unnecessarily, which pollutes the list of forks.

+1 for the mention of Endoh's work -- Yusuke Endoh's YouTube channel with many of his quines is really fantastic: https://www.youtube.com/c/YusukeEndoh/videos

A quine that prints itself as scrolling text in an AVI file: https://www.youtube.com/watch?v=rMlJy5He-Tk

A self-similar quine that behaves like a fractal zooming out infinitely: https://www.youtube.com/watch?v=6J7u2G52scc

A quine with source code that contains sheet music to a Bach song in comments and plays it: https://www.youtube.com/watch?v=OryXKUgRVOg

Of course, once you buy the 3d printer, filament, metal cutting supplies, various epoxies and waterproofing sealants, cabling, soldering tools, and nuts and bolts...

There's a whole community of DIYers building e-foil boards at https://efoil.builders/. I haven't actually built one because I don't think I would ride it enough to justify the cost.

I do lurk on that forum and check out the projects people are making, there's a lot of really cool work done to 3d print propellers and other structural parts and bring the cost of manufacturing down. Between rapid prototyping and chinese BLDCs / ESCs, small electric vehicles like e-foils are actually reasonably priced to build yourself.

If this is something you do in CMake projects, check out FetchContent_Declare — newer CMake versions have this command which downloads a git repository and caches it.

I much prefer this to manually downloading the files, even for header-only libraries. The command provided a nice declarative way of adding a dependency.

I've built a few ~10-20 hp equivalent electric boat motors and some smaller brushless pod motors like the ones that efoils use -- It's really satisfying engineering work and you have to master so many disciplines along the way: machining/fabrication, CAD, electronics, software, etc... And as with any project of that complexity, you could spend a lifetime just getting interested in a single part of the project.

I've seen a few startups (For example https://www.getoutline.com/) using the GitHub Discussions beta as a defacto support forum. I like that use case and it seems like it will help both developers and users.

For devs, you'll get a forum setup without requiring time/money to set up your own hosting, which is big for small or bootstrapped projects. It will also help move specific questions out of your issues section/ticketing system.

For users, it lowers the friction associated with getting support since many already have a GitHub account.

I've seen the same trend in other recent fonts designed for programming, like JetBrains Mono, with the idea that "more height" on lowercase letters improves legibility at small font sizes. I'm in agreement with your distaste, mostly because I work on high-DPI displays where legibility is not necessarily an issue and fonts that do this look like a "blob" of text.

Just speculating, but this could help in any web project where you're including multiple font weights/styles -- Currently that involves building a CSS font-family from multiple font files which have a lot of redundant information. With a variable font, your font family could pull from a single font file and use the font parameters to tweak different styles.

The HN audience loves to discuss the bloat in web design -- If you can reduce 500kb of font-<heavy, bold, italic, light> files into a single file that interpolates between font vectors dynamically, that goes a long way to avoiding excessive resources. And you also get the benefit of being able to tweak the parameters of each font style without changing out static files.

For example, with Recursive the entire static font family come in at 12.8mb, while the dynamic font family in a single file is 1.45mb. Obviously you wouldn't load up every variant and Recursive is a very large font, but I could certainly imagine replacing ~8-10 font files with a single variable font.

The Google Fonts page doesn't really do justice to how dynamic this font is -- https://www.recursive.design/ has a live demo that shows more of the options, maybe the link should be changed to their demo.

There are four dimensions and a few other flags that the font can be adjusted by in CSS: natural-width to fixed-width, Linear to "Casual", Font weight, Font slant, and cursive variants.

By tweaking any of the variable font parameters it you can get anything in the range of a "serious" monospace font to a more casual font useful for UI mockups. I don't spend a huge amount of time in the font design world but this is one of the first variable fonts I've seen that feels less like a gimmicky demo and more like a seriously versatile tool, really great work!

Yeah, I see your point, AI-generated CGI is still CGI!

I don't think they are using conventional CGI techniques to "remake" his mouth/lips and lower face, if that is what you are suggesting. Seems like the dialog replacement is using a technology called VDR: https://www.fxguide.com/quicktakes/cannyai-vdr-face-replacem.... There's a demo video on that page and some more details about the technique.

I made a plugin called Focus.me - https://chrome.google.com/webstore/detail/focusme/jdkgogfonl... - Although it doesn't automate a workflow, I use it to stay productive during a workday. Focus.me blocks a list of websites in fixed time increments: The press of a button adds time to a timer, blocking sites until the timer expires. I find myself instinctively flipping between Youtube, HN, reddit etc. in a loop, and the plugin helps break this cycle and get the ball rolling quicker when I need to focus.

I made the plugin because I wasn't satisfied with other productivity/website blocking plugins which often contain URL tracking and other telemetry. Focus.me is completely local and doesn't record any information of any sort.

Thank you for collecting that list, I stand corrected. I've heard that original statement echoed several times in the RC community -- But I think sometimes people are separating out incidents from gas RC aircraft, RC helicopters, commercial operation etc. in a way that might be a little disingenuous.

I agree, I think there's a much smaller pushback from the more dedicated community about the method for tracking and the limits around DIY aircraft -- If it was a plug-and-play transmitter that you had to pay for and register, I think that would be fine. It could be integrated into the RC receivers on your plane, powered via 5v servo cable, or part of your transmitter. But this odd system involving a proprietary cloud and "phone-home" style telemetry doesn't sit well, especially because it's specified in a way that makes it impossible for any DIY aircraft to comply.

- Flite Test (A large RC community) has collected a lot of stories & comments but suggests reaching out to a local government representative: https://ftca.flitetest.com/ - If you are a member of a local AMA (Model aircraft association), communicate with the AMA leadership about your concerns since so far they've been relatively passive and haven't voiced any concerns.

I, like many others, put a comment on the FAA proposal, but that comment period is over and the pandemic has slowed down a lot of the public progress of this legislation. It's hard to answer your question, since it feels like the FAA is proceeded as planned despite tens of tousands of concerned comments.

DJI, who makes the majority of consumer drones, has already implemented broadcast-based radio ID, but that's not what the FAA proposal suggests and is not the part of the regulation that is being criticized.

In fact, the FAA has explicitly stated that they don't want to use a radio-broadcast system like ADS-B (the equivalent for general aviation). The solution being proposed involves a proprietary internet-based system that has privacy/security implications among other issues -- The system would send the location of any user to a third-party company, would require a subscription, etc.

The solution does very little to alleviate the concerns you are discussing and many more reasonable proposals have been suggested which DO involve radio ID. My favorite solutions involve a combination of a radio beacon broadcasting GPS and registration # in the style of ADS-B for larger drones/aircraft, and for smaller aircraft an app-based system where users "check in" with a flight plan. But keep in mind that the outrage is not over any regulation by the FAA, it's that their proposal completely disregards existing industry standards like those put out by DJI and creates an expensive, hard to enforce system that favors large corporate interests.

Let's keep in mind that there are millions of drones in the US (1.1 million registered with the FAA, and many more that have not been registered because they are <250g or custom-built.) and have not had a single incident involving injury or death. Every single incident involving a drone has involved the concern of violation of someone's privacy or violation of an airspace, and even in those incidents there has been a lot of hysteria involved in the media response. In my opinion this absolutely does not justify preemptively building an expansive surveillance infrastructure around technology that has, so far, proven to be very safe.

Citation to the Remote ID proposal, for those interested. The comment period is closed, but the proposal received over 50k comments: https://www.regulations.gov/docket?D=FAA-2019-1100

The amount of hysteria over drones (Particularly incidents like this) is really concerning to me, because it plays right into the hands of regulators and lobbyists in the US trying to justify unnecessary Remote ID legislation under the guise of "safety".

The recently released proposed rules for Remote ID raised a huge amount of concern in the model aircraft world, because it proposed restrictions that would effectively kill the hobby:

- All drones/RC aircraft must have a GPS-based tracking ("Full remote ID") on the aircraft, connected to a ground station and relayed to a subscription third-party service. These "Full remote ID" aircraft can be flown past line of site.

- Aircraft not constructed by a single manufacturer (For example, handmade RC planes) are not allowed to use full remote ID, and instead must use "limited" remote ID in which GPS can be on the ground station, but the aircraft must disable itself past 400ft from the ground station.

- Aircraft with limited remote ID can only be flown in a dedicated "FRIA" flying site.

- FRIA sites are "legacy" and after a short period, no new sites will be created. The stated intention of the FAA is to phase these sites out completely, meaning after a certain period all planes must comply with full remote ID, effectively killing hand-built model aircraft.

The end result of this regulation is _not_ safer airways: It's an airspace <400ft owned by commercial entities such as Amazon who are able to comply with the requirements of the Remote ID tracking and single-manufacturer requirements. It's an example of regulatory capture that will genuinely make the world worse for us by increasing noise pollution and drone traffic at low heights. Meanwhile, the model aircraft and drone hobby will have a large barrier to entry that pushes away the innovation of millions of engineers who got their start as kids flying affordable DIY planes.

The UI is bad, but the most frustrating aspect of YouTube is the continuous struggle between creators and the automatic copyright strike system. Nearly every single large YouTuber I've subscribed to has had some experience with a false copyright claim interfering with their platform at one point.

It feels like YouTube actively doesn't want creators to grow a community on their platform.

Nice job, the algorithm is really effective! I tried this in multiple cities where it ended up recommending very similar routes to some of my established running routes, and even some new combinations :)

Love that your algorithm tries to exclude out-and-backs, I think that goes a long way to make the routes feel organic and appeal to runners. This is really polished and I wish you the best!

One feature suggestion: I wonder if you could have a flag to generate routes starting in parking lots/park entrances in a radius around the starting location. This could go a long way to recommend some common long-run routes where you need to travel a brief distance via car or public transportation but then can get 10-20 miles entirely in the woods.

Hermit Thrush is an amazing brewery, and I stock up every time I’m in the area. I strongly agree with your comment, if you look outside of IPAs there are lots of really creative breweries in New England doing great things.

I still don't see Instagram used as a replacement for text messaging like Snapchat is for many people (again, just my experience in the college-age demographic). Snapchat themselves don't seem to want to be a messaging app though, probably because the path to ad revenue and real profitability is in an algorithmic feed.

well said -- There's a fundamental difference in TikTok, Instagram, Facebook, etc and Snapchat. Snapchat is heavily focused on private (one person, small private group) content in contrast with the other networks which mostly focus on public content.

Anecdotally, in my age group (College / new grads in Northeastern USA) Snapchat occupies a slightly different segment from TikTok (rant incoming...)

TikTok, Instagram, and Facebook are all places where people share more public / curated content, whether that's a silly video you made, a meme, or a little more curated photo of something exciting you're doing. Regardless of the content, there's a big focus on reputation and superficial interactions (likes, comments, etc.)

In contrast, Snapchat is still extremely popular for private communication (group chats, private messaging). For many college-age people, Snapchat is closer to an alternative to text messaging -- people heavily use the editing and multimedia features in creative ways that you couldn't do with normal SMS (For example, drawing over an image, circling something, overlaying another image, typing out a sentence over an image). Snapchat still has a consistent user base here because they've continually focused on improving the chat features. One example of this phenomenon: Whenever I met new people in college, we exchanged 1) Phone numbers 2) Snapchat accounts, since those are the primary methods for private communication for many. My snapchat friends list is nearly 1:1 with my phone contact book for anyone I met socially.

Instagram competes with Snapchat (and is arguably winning) on any feature that focuses on public sharing, such as the "stories" feature -- And for good reason, since Instagram as a platform has been designed from the start around public sharing. Similarly, TikTok will definitely beat Snapchat on the curated public video content, because that's just never what Snapchat was good at.

Only time will tell whether the smaller user base that is hyper-loyal to Snapchat for private messaging is large enough to justify their insane valuation, but I don't think Instagram or TikTok have really captured Snapchat's core value in private messaging.