Interval (https://www.threeten.org/threeten-extra/apidocs/org.threeten...) isn't built into java.time, however, it is in the popular threeten-extra library. The docs say "An interval represents the time on the time-line between two Instants." The main difference being that Interval is anchored to the timeline while Duration and Period are not.
You could be thinking of the Claude tool search tool [0][1], which lets it do that on-demand loading. Skills (like another comment mentioned) have a progressive disclosure mechanism built-in. Agent Skills [2] appear to be the common format across most of the tools, but I have no idea if this actually has tool provider buy-in.
I had the same experience with JobScan. I felt that it helped me on my resume overhaul with some good clear suggestions, but didn't work well with editing and replacement, especially with my resume hosted in Google docs. I cancelled my subscription during the free trial.
Thankfully, this is another tool in the current arms war of job searchers versus corporate recruiting.
I've been watching a YouTube channel named Climate Town that recently did a video talking about natural gas, and leakage was a focus point in the video. It's like a documentary-style comedy channel, and I quite enjoy both the content and the format. It reminds me a bit of Jon Stewart and John Oliver.
One aspect that I find discouraging is the significant opportunity cost associated with the time and effort required during the job search process. Instead of using this time for skills development, exploring new technologies, or excelling in a current role, it's consumed by:
- Tailoring resumes to suit ATS systems (now with AI/other automation)
- Leetcode and study for live interviews
- Identifying suitable positions on job boards for cold applications
- Probing your network
- Responding to unsolicited contacts
- Chasing ghost recruiters
It's disheartening and frustrating to remain unemployed for an extended period, especially when I possess relevant experience and expertise, yet continually face rejections due to the challenges of navigating the recruitment pipeline.
The ATS stuff in particular is a real pain, and I wonder how much of that is companies overtuning their application process instead of adapting to the realities of a remote workforce. It feels bizarre to be ghosted or summarily rejected by 97% of companies and yet very easily pass through multiple interview rounds with the remaining 3% who actually read my resume.
The consequence of this is cynicism and the use of AI on both ends. Applicants are using AI assisted tools to apply because of the sheer quantity of jobs they have to apply to to get anywhere and recruiters are using AI assisted tools to screen because of the sheer quantity of candidates they have. It's robots talking to robots and everyone is miserable.
They wrote an interesting blog some time back about how "random shuffle" isn't necessarily what people want, and how their algorithm works (https://engineering.atspotify.com/2014/02/how-to-shuffle-son... ). That was a decade ago, so maybe their approach has changed or that it does not perform well under certain conditions (like the one you mention). It works well for me on most playlists on the order of 10s.
I agree with you that it seems to work fine on playlists of less than 50 or 100 songs.
The problem seems to be that on larger playlists they will only use 50-100 of the tracks to shuffle through. Most times I'm listening to music I just want to put on a shuffle of all my favorites and listen. It's been that way since I got my first CD changer. Maybe that's a super unusual use case, but it's my primary one, and I get really tired of hearing the same songs repeatedly over a week. YMMV, my wife for example likes listening to the same songs every day.
As I mentioned above: I copied my Spotify playlists to YTMusic and am doing the same "shuffle my liked songs" and I'm literally hearing songs Spotify hasn't played for me in years. Usually the algorithm complaint in music players is that they are using random rather than shuffle, but even in that case I'd think that 2K songs over 2-3 years, I'd be hearing SOME of those songs that YTMusic is playing but Spotify is not. The cynic in me figured that they were prioritizing the songs by the ones that made them the most money, or from artists that paid for placement. But something about their shuffle is just totally off.
Yes, it looks like there is some artificial placement. This may be driven by malice (some sort of paid or more lucrative placement, like you said), but also by stupidity (algo prioritizing songs already in the client cache, to save some egress bandwidth perhaps?).
So I started clearing the Spotify client cache more often, and it looks to me there is more diversity, at least on the auto-generated "recommended songs" playlists. But still, no hard proof of this.
The "recommended songs" playlists seem to have more diversity, but also seem to be fairly short (they'll repeat in a couple hours it feels like; I rarely listen to them when I'm working because they'll start repeating, and I don't usually listen to music for a large fraction of the day, so I'm guessing 1-2 hours).
My best guess is that they are assuming no playlist is more than 50-100 songs, and are limiting the shuffle to that number, so that a shuffle doesn't consume too many resources (memory, database hits, CPU cycles). Maybe someone, possibly in the distant past had a large playlist that caused service problems. And because of that they clamped WAY down to prevent it.
The threads feature is pretty close to useful, but it still buries information and it isn't clear to other users in a channel when a thread becomes "live" again unless they explicitly follow the thread or make a comment. Both of those actions require excessive clicking on the desktop client (at least for Ubuntu).
The most frustrating part for me is that keyboard navigation for threads is nonexistent, and it forces you to weave in and out with your mouse.
I thought this was going to be similar to https://play.kotlinlang.org/koans/overview . I used that as a quick way when doing other things to slowly learn parts of the language before diving in, was hoping this would be similar.
Period (https://docs.oracle.com/en/java/javase/21/docs/api/java.base...) is a date-based measure of time. 2 years, 3 months, and 4 days.
Interval (https://www.threeten.org/threeten-extra/apidocs/org.threeten...) isn't built into java.time, however, it is in the popular threeten-extra library. The docs say "An interval represents the time on the time-line between two Instants." The main difference being that Interval is anchored to the timeline while Duration and Period are not.
reply