I cannot describe how disappointing it is to be switching to this insane time limit window based pricing. I absolutely abhor that I'll be subjected to 5 hour chunks of time where I'll be limited at some point in that window of time, and be told I'll have to wait. And then there is a weekly limit.
That's not how my creative energy works. I have time that I want to solve problems, and I want to solve them. I don't want a cooldown timer applied to solving a problem. Not to mention the anxiety of realizing that while I sleep I could have burned tokens in that time.
I'm incredibly disappointed when I sat down to my hobbyist programming time and realized copilot was suddenly and dramatically changed in a way that is incredibly disheartening.
Meter my token usage DON'T tell me when I can use them! ARGH.
> I'm incredibly disappointed when I sat down to my hobbyist programming time and realized copilot was suddenly and dramatically changed in a way that is incredibly disheartening.
Guess it’s time to rediscover the lost art of programming without an LLM.
Ahhhh yes thank you for this link! I'm working on a project and my agent referred me to the factbook. It said all this great savory information was made available on the factbook website. I thought great but when I went to check it was all very basic. Truly I thought my agent was hallucinating!
It wasn't! Note to self: also check archive.org in case there is an internet archive for any sites an agent might reference.
I checked out this repo -- it has the information useful to my project. Thanks for sharing this!
I wonder if this will wind up being true. Yes it’s cheaper to produce an app but most normies I know don’t really want to produce an app. No instead they want to consume a curated app. If anything we’ve moved the value proposition from “it exists” to “it exists and is good, especially compared to the competition.”
App creators will be competing and copying each other. The software that can support change will probably win in the market. Probably…?
What the OP says someone else will do it cheaper not the normies will do themselves. We already have tons of free and cheap software for the sake of it without making any commercial sense, now we will have way more it.
I believe we are having this discussion from a purely technical perspective and not from a business one. Let's take slack for example. Assuming a company can perfectly clone it, why would they? Yes they would skip paying for it, but they would have to maintain it, starting from its infrastructure all they way up to its UI. Will they think of new features? Will they follow industry developments in sound / streaming technology? Will they keep integrating other tools into it? I am sure they would rather pay to have somebody else do it for them, someone like slack itself.
Also, assuming a company has the capital to burn through enough tokens to create something so big and complex, why spend it on an internal tool? Shouldn't they be spinning slack-sized apps to expand their existing market share or try to disrupt new markets?
Hi! I recently (2 weeks) chose this software to invest time into in order to make music/sfx for video games. Do you personally use this software to create music yourself? Just curious!
Not really. My only released music is a single album at https://pauldavismusic.bandcamp.com/album/suspended-generati... which was made almost entirely with VCV Rack (Ardour was used for some pretty minimal editing). I've also used it for a short podcast series ("Audio Developer Chats") that is currently offline. I do try to talk to musicians and engineers almost every single day about what they're doing, however.
Hobbyist game dev here. Getting into audio and music effects has been fun but I constantly feel overwhelmed. I chose Ardour as my DAW (digital audio workstation) and have been excitedly working on learning. I also bought the book “ Writing Interactive Music for Video Games: A Composer's Guide” which has been very helpful at understanding high level vocabulary.
It’s a lot of work. I slightly enjoy it but boooooy is getting into audio and music pretty challenging. It’ll be good if I ever need to know what I’m talking about when working with others… in the future where I can dedicate myself full time to game dev… One day one day…
I don’t really have a point here. If anyone has any resources, tips, or recommendations on this subject let me know.
At the risk of sending you down a giant rabbit hole, the book Designing Sound is all about making programmatic sounds with Pure Data, and open source low-code programming environment available for all platforms. From what I've read, the book is considered a classic in the video game sound world. It's really good. Combine that with the Cipriani book on PD and learning Ardour would give you a very good learning path.
I recently dated someone in her 30s with a DMA, and she she told me how proud she was to know me because I was the first non-musician friend she's ever had. It's a very deep and insular world.
Pablo Casals famously replied when asked why he was still practicing in his 70s that he "felt like we was making progress", so don't let yourself feel inadequate.
Stuff like that sends me down rabbit holes and when I finally come up for air, I say, "Gee, now I see how people can build their entire career around this!"
A light weight journaling of your learnings as you go along would probably be real beneficial to many (such as myself who has zero knowledge on the subject of DAWs and creating music effects for games). And since you said it’s very challenging maybe writing about it in small bite-sized learnings might make the process easier? Going from “I must learn all this stuff!!” to “let’s see what audio gems we pick up in our adventure today”.
Good point w/ journaling and maybe even preparing stuff to share. Right now it feels like trying to eat a whale whole. …That’s the beauty of doing something in a new (to me) domain though. Even if I don’t share, the practice of formulating how I would explain to someone is beneficial.
In the Axiom Verge noclip documentary, the dev talks about how everyone is surprised about how he really made the whole game himself (except localization + marketing (and I imagine final testing but...)).
And he was like "people say 'even the music?' The music was the easiest part!!!'
It does make me feel like once you get into the right headspace and figure out how most of the tooling works all of this becomes quite smooth.
I've been making music now for 8 months and have absolutely fallen in love with it. I'd love to write music for a game, what kind of game are you creating?
Sounds like you’re ahead of me then! The game I’m working on off & on is a 32v32 moba. I want the gameplay of Heroes of the Storm but with the match size of a battlefield server.
I wish I could dedicate myself full time to the effort (and in fact I’m working with my financial planner to figure out a plan) but this is a side project that is prioritized behind a few things. I wouldn’t feel too comfortable collaborating right now since I would be unpredictable. I’ll keep you in mind for when that changes though!
I remember when Ubuntu decided to reroute apt installations into SNAP installs. So you install a package via apt and there was logic to see if they should disregard your command and install a SNAP instead. Do they still do that?
Yes. I know its more than firefox, but I don't have the full list. On 24.04:
me@comp:~$ apt info firefox | head -n 5
WARNING: apt does not have a stable CLI interface. Use with caution in scripts.
Package: firefox
Version: 1:1snap1-0ubuntu7
Priority: optional
Section: web
Origin: Ubuntu
me@comp:~$
This is the same company who creates internal systems that encourage wringing out every drop of effort no matter how many piss bottles litter their work environment. When a faceless program uses gamification and comparison estimations to keep their employed serfs always working, constantly fed a sense of being behind. The stress of it all without the minimal of a “good job!”
You’re telling me THAT style of company isn’t capable of achieving this goal for another 2 months? If the company is going to use reprehensible practices at least use it to achieve good quicker.
To me this feels like releasing a press announcement to generate good PR and waiting until everyone forgets before not actually doing the thing… That’s my cynical take.
I’m so close to the switch myself for silly reasons. I don’t like windows due to their creepy business practices and negative design patterns in their OS so I’m very bias against it. Forcing copilot is just the latest in their creepy practices…
For more details on why I came close to switching: I use my win desktop as a host for ai services such as Comfy UI for stable diffusion generation since it is a beefy platform; for example, I generate reference stuff for Krita (digital painting software) illustrations on my drawing tablet. I remember the process to configure windows as being strange, GUI bound (NOT windows strong suit), and just annoying due to my aforementioned bias. Valve has done great work with running games on linux which is the only reason I keep that OS and I’d rather set up services on linux.
This comment serves as a reminder to myself that I should just go ahead grab my windows license keys for archival purposes and flash a better OS on that system.
Krita is among the main reasons why I am so impressed with the KDE project. Not only do they deliver a very good desktop environment, but they also deliver some genuine flagship apps for it
I call it silly because essentially I’m complaining that I don’t like setting up a service on an OS I don’t normally set up services for. My quibbles with that process can in part be addressed by crafting a terminal based workflow to configure the host and enable the service on my desktop, skipping the GUI completely. E.g. I’m sure I can do Task Scheduler shenanigans through powershell. More experience would help sand down the rough parts I experienced.
Now the product decisions behind the OS giving me the icks… The terminal can’t (completely) help with that ^_^
Eh those applications can both be run in linux without issue.
They are likely to have more issues related to getting the drawing tablet configured correctly.
The rest is just having to start from scratch and lose the decades of windows experience and intuition which can make things painful as that type of thing cant be replaced without time.
Having just yesterday installed a fresh Mint distro on a newly received PC that came with Windows pre-installed, I can tell you that this is merely an hour of work, and most of the time will be spent downloading and burning the .iso on the USB key.
Do you have reference for the krita+comfyui setup? I have a drawing tablet and always wanted to augment drawings using AI but never got around to deploying a stack for that. I have a 3090 that should be enough for it, I just need a reference for the setup.
Running ComfyUI or _any_ AI stuff on Linux is a night and day difference in terms of ease of use and performance when compared against that of Windows users. Python on Windows is suffering
The pain I’ve experienced running ComfyUI on windows is from (1) pytorch and the complexities of managing it through pip when python’s platform concept doesn't encompass CUDA versions, (2) dependency conflicts between custom nodes (some of which also involve #1 because they pin a specific pytorch version as a dependency), and (3) gratuitous breakage in ComfyUI updates.
It works great until it doesn't. Wait until your dependencies require you to change torch versions, or a comfy update requires dependency changes, or you have to change CUDA versions.
Solving these problems (and avoiding them in the first place) is a lot easier to manage on a Linux install.
Copilot is already great and it's only getting better. I can get so much more work done and the same amount of time or get the same amount of work done in a lot less time.
It's funny how many people were afraid of AI. Technophobes abound in this world
Funny, I disagree. I think copilot truly sucks compared to the other options. But you can uninstall copilot, so I don’t see why it bothers people at all.
I always thought the companies I worked for would implement chaos testing shortly after this talk/blog released. However; only last year did we do anything even approaching chaos testing. I think this goes to show that the adage “the future is already here just unevenly distributed” carries some truth in some contexts!
I think the companies I worked for were prioritizing working on no issue deployments (built from a series of documented and undocumented manual processes!) rather than making services resilient through chaos testing. As a younger dev this priority struck me as heresy (come on guys, follow the herd!); as a more mature dev I understand time & effort are scarce resources and the daily toil tax needs to be paid to make forward progress… it’s tough living in a non-ideal world!
Chaos testing rarely uncovers anything significant or actionable beyond things you can suss out yourself with a thorough review but has the added potential for customer harm if you don't have all your ducks in a row. It also neatly requires, as a prerequisite, for you to have your ducks in a row.
I think that's why most companies don't do it. A lot of tedium and the main benefit was actually getting your ducks in a row.
I think it is more of a social technology for keeping your ducks in a row. Developers won’t be able to gamble that something “never happens” if we induce it weekly.
Much of the value from Chaos testing can be gotten much more simply with good rolling CI. Many of the problems that Chaos engineering solved are now considered table stakes, directly implemented into our frameworks and tested well by saidsame CI.
A significant problem with early 'Web Scale' deployments was out of date or stale configuration values. You would specify that your application connects to backend1.example.com for payments and backend2.example.com for search. A common bug in early libraries was that the connection was established once at startup, and then never again. When the backend1 service was long lived, this just worked for months or years at a time - TCP is very reliable, especially if you have sane values on keepalives and retries. Chaos Monkey helped find this class of bug. A more advanced but quite similar class of bug: You configured a DNS name, which was evaluated once at startup, and again didn't update, Your server for backend1 had a stable address for years at a time, but suddenly you needed to failover to your backup or move it to new hardware. At the time of chaos monkey, I had people fight me on this - They believed that doing a DNS lookup every five minutes for your important backends was unacceptable overhead.
The other part is - Modern deployment strategies make these old problems untenable to begin with. If you're deploying on kubernetes, you don't have an option here - Your pods are getting rebuilt with new IP addresses regularly. If you're connecting to a service IP, then that IP is explicitly a LB - It is defined as stable. These concepts are not complex, but they are edge boundaries, and we have better and more explicit contracts because we've realized the need and you "just do" deploy this way now.
Those are just Chaos Monkey problems, though - Latency Monkey is huge, but solves a much less common problem. Conformity Monkey is mostly solved by compliance tools; You don't build, you buy it. Doctor Monkey is just healthchecks - K8s (and other deployment frameworks) has those built in.
In short, Chaos Monkey isn't necessary because we've injected the chaos and learned to control most of what that was doing, and people have adopted the other tools - They're just not standalone, they're built in.
It's a great way of thinking about resiliency and fault tolerance, but it's also definitely on the very mature end of the systems engineering spectrum.
If you know things will break when you start making non-deterministic configuration changes, you aren't ready for chaos engineering. Most companies never get out of this state.
What’s the difference between scraping and malicious scraping? Does google engage in scraping or malicious scraping? Do the AI companies engage in scraping or malicious scraping?
Note that I am not defending the merits of Google's lawsuit, but they did describe in this very post what they believe distinguishes their scraping versus SerpApi.
> Stealthy scrapers like SerpApi override those directives and give sites no choice at all. SerpApi uses shady back doors — like cloaking themselves, bombarding websites with massive networks of bots and giving their crawlers fake and constantly changing names — circumventing our security measures to take websites’ content wholesale. [...] SerpApi deceptively takes content that Google licenses from others (like images that appear in Knowledge Panels, real-time data in Search features and much more), and then resells it for a fee. In doing so, it willfully disregards the rights and directives of websites and providers whose content appears in Search.
To me this seems... interesting, for sure. I think that Google already set a bad precedent by pulling content from the web directly into its results, and an even worse one by paying websites with user-generated content for said content (while those sites didn't pay the users that actually made the user-generated content, as an additional bitchslap.)
But it seems like at the very least Google is suggesting that SerpApi is effectively trying to "steal" the work Google did, rather than do the same work themselves. Though I wonder if this is really Google pulling up the ladder behind them a bit, given how privileged of a position they are in with regards to web scraping.
It's a tough case. I think that something does need to ultimately be done about "malicious" web scraping that ignores robots.txt, but traditionally that sort of thing did not violate any laws, and I feel somewhat skeptical that it will be found to violate the law today. I mean, didn't LinkedIn try this same thing?
That's not how my creative energy works. I have time that I want to solve problems, and I want to solve them. I don't want a cooldown timer applied to solving a problem. Not to mention the anxiety of realizing that while I sleep I could have burned tokens in that time.
I'm incredibly disappointed when I sat down to my hobbyist programming time and realized copilot was suddenly and dramatically changed in a way that is incredibly disheartening.
Meter my token usage DON'T tell me when I can use them! ARGH.
reply