3,882 karma · joined March 12, 2011
website: thomasboyt.com
I'm still planning to write a longer blog post about it with more details. I think in terms of enterprise usage, there is no equivalent to a batteries-included framework like Spring Boot - especially not one with paid support available for big companies.
I went ahead and used Javalin, which I'm a little bummed to not see represented on the survey - it's meant as a successor to Spark (which I haven't used), and I really like it. It uses similar patterns to Koa/Express in Node, which made it easy to understand. The big downside of not using an integrated ecosystem like Spring Boot, of course, is more glue code, but I actually haven't found it _that_ bad.
That said, because of how fragmented the ecosystem is, it is difficult for me to jump in. For example, my main side project right now is a Vue app that uses SCSS syntax in components, which the Vue extension for Nova does not yet support. This might sound like a small detail, but it's a total dealbreaker - and while, if this was a totally open source editor, I might be interested in contributing to the extension, I can't say that I'm interested in paying $100 and then helping out. It's a very difficult chicken and egg situation.
Anyone who has been able to stay productive, let alone get more productive, in this reality is either in denial or far more optimistic than I'll ever be.
> The cellphone tracking data showed about 10% of Sturgis attendees hailed from within South Dakota, with about 19% from border states and 72% from across the rest of the country, with heavy attendance from Arizona, California, Colorado, Iowa, Minnesota, Nebraska, Washington and Wyoming.
I find this aspect far more upsetting than the stupidity of the event itself. It's one thing when a bunch of locals decide to congregate in an environment they really shouldn't be, but another thing entirely when it's 400,000 people biking across the country.
Ironically - or perhaps not - I saw an NPR article ~a week ago[1] indicating the infection rate in Sturgis and its county, while certainly higher than it was before the rally, wasn't as catastrophic as you'd expect. Apparently, many people who live in Sturgis left town or just tried to avoid contact with anyone that weekend. Can't find any up to date information on whether the infection rate has increased since then with more time/testing, though.
[1] https://www.npr.org/sections/coronavirus-live-updates/2020/0...
For the server instances, I can't imagine 4 CPUs/8 gigs RAM is really needed 24/7, but can imagine it being needed in bursts, and I assume GCP has various "elastic" auto-scaling products for that. Would love to know if your metrics indicate that this is the smallest box you could comfortably run on, or if it's just a best-guess. Could potentially spin down blue-green environments after some set amount of time (once you know a deploy "succeeded"), as well.
For Metabase, I can't find any "system requirements," but I'm not _super_ shocked by the price tag. I've always liked the idea of hosting my own tools instead of relying on some SaaS's free/trial tiers (particularly for things like analytics, logging, or metrics), but a lot of the open source options out there assume you have a pretty hefty box to throw it at. At the same time, I haven't found any better alternative other than just doing ad-hoc analyzing in SQL (using a GUI like Postico).
For (a), they've already banned a lot of adult content (thus the rise of OnlyFans, until they follow suit), and they'll probably ban more if they get enough attention from either media, angry card processors, or both.
For (b), they've already narrowly avoided fucking this up once back in 2017 (https://blog.patreon.com/updating-patreons-fee-structure). They increased fees in 2019 (https://support.patreon.com/hc/en-us/articles/360024952552-P...) but at least grandfathered in past users, preventing their existing userbase from being too much of a flight risk.
I'm not sure the Container Registry on its own will necessarily be attractive to people just looking for commodity-priced container storage, but GH Actions + Container Registry does make for a pretty compelling CI/CD story, I have to admit.
To me, the price to beat is the $0.04-$0.20 a month I pay for Google Container Registry (price depending on whether I remember to go back and delete older images or not). I'm guessing it's not going to get to that level of commodity priced, but I don't see this product being competitive without giving a heck of a lot more space for cheaper.
I've stopped using the Youtube app on my iPad since I have no way of blocking ads there (probably what's going to get me to finally set up a Pi Hole).
Unsurprisingly, this ends up hurting all sorts of people who do not use their legal names online: people who have just chosen other names, people who want to avoid being targeted or stalked, trans people, etc. They've updated this policy to allow for some of these situations (https://www.cbc.ca/news/technology/facebook-real-names-1.336...) but folks still get banned for failing to comply.
Kind of astonished it's quietly been on air for so long, and I wonder if the rebranding means they're actually going to revitalize it in some way or if this is just going to lead to them quietly dropping the more interesting aspects of it.
Of course, as other commenters have been pointed out, it's hard to sell companies on paying something they "have" for free. Which is why companies will shell out big money for the O'Reilly Learning Platform or MSDN ($500/year/seat for the former!) but not even think about spending money on a community resource.
Of course, we all know that asking startups to actually fund open source (whether code, documentation, or learning resources) is like pulling teeth. I'd say maybe the collapse of some larger open source project(s) could convince companies to start actually giving a shit about whether or not the maintainers of the technology their entire businesses rely on have enough money to continue developing said technology, but it's more likely the companies would just say "well, let's put 10 engineers on this problem for a month to replace this dead open source technology with something new and shiny" and not recognize how much more they've had to spend. Or, y'know, they'll go with paid support of a Microsoft product or something.
I am curious, as a bit of a devops outsider, how it fits into the toolset of a modern application. From what I can tell, a lot of the aspects of Ansible are made redundant in a world of _hosted_ containers - you don't need Ansible if you're, say, deploying container images to EKS/ECS/whatever - but if you were to self-host your Docker hosts (or... whatever the Kubernetes is equivalent is called), Ansible still seems to me to be the best tool around for standing up and maintaining those hosts.
I understand why Mozilla wants to make more money new ways. I think everyone does. I was going to get their VPN as soon as it became available on Mac. I also completely understand that they can't monetize Firefox, at least not in its current state. Users would obviously rebel against any kind of proprietary features that live behind a commercial license, and I don't think they can make any compelling hosted services that integrate with Firefox either. But I don't understand why Mozilla seems to be trying to speed along its death.
That said, I probably still would avoid a cheapo player like this that does not have an adjustable tone arm. For one, it's more likely to have issues skipping when you can't adjust it properly. The second concern - which I'll admit I have no data to back up, and since this is an opinion from the audiophile community, may warrant some skepticism - is that cheap players with unbalanced tone arms are infamous for supposedly wearing down records more quickly over time by applying too much pressure.
The author's note about the stereo output being unbalanced, by the way, is a relatively common side effect of an unbalanced tone arm (though there's various other reasons it can occur).
It does seem useful to have a _standard_ for type definitions - RBS as the equivalent to a .d.ts file - as that allows for different type checking implementations to use the same system under the hood. This was a big problem for Flow, and why it lost the fight as soon as TypeScript's definitely-typed repository started gaining momentum - users wanted to use the type-checker that they knew had definitions for the libraries they used.
On the other hand, RBS as hand-written seems rather dangerous, to me. Nothing wrong with using them to define previously-untyped external code, as long as you know the caveats, but I think you really want to have definitions generated from your code. Sorbet cleverly (and unsurprisingly, given it's Ruby) used a DSL for definitions in code, which had the (excellent) additional boost of runtime checking, so you actually could know whether your types were accurate - by far the biggest pain-point of erased-type systems like TypeScript.
Given that Ruby 3 was supposed to "support type checking," I'm surprised that it does not seem to have syntax for type definitions in code, and instead will focus on external type checking. I might be missing a piece of the full puzzle not covered in the blog post, however.
This is somewhat obvious, but take care in designing social interactions on your app, and introduce easy reporting systems and user-controlled blocking systems (e.g., even if you don't respond to a user's report in a timely manner, a user should still be able to block communications from another user using a button in the app). You may want to allow users to choose to only allow messages from users who have been on the site for a certain amount of time, or have some kind of additional level of verification. Also consider introducing active, paid moderators who actively respond to reports as they come in, if your budget allows.
Allowing users access to open text fields on the internet is inherently dangerous. I have tried to avoid it in all of my apps (for example, I've made games that use randomly-generated user names rather than allow users to input their own, and I only added user comments to an app I was building _after_ I'd built a user reporting system and ensured I got an immediate alert on my phone if a user sent a report). It is your responsibility to try to design your app, and moderate your community, to mitigate these risks.
On the other hand, I absolutely loathe every single UI that has ever come out of AWS, and there is absolutely no chance I give the editor they've built here a shot without hearing lots of positive reactions first.
edit: actually going to pull back slightly on this comparison, since apparently Honeycomb is a little closer to something like Access, where there is a database _in_ the app, whereas Retool is built around connecting to an external data source (e.g. a SQL database). This surprised me because I assumed the entire reason you'd want an AWS version of this kind of tool is to integrate with the broader AWS ecosystem, like frictionless hookup to a DynamoDB or whatever. It may have this but just not spotlight it on the marketing pages?