cwsx

Coding is, and always has been, the easy part of software development

I switched to them a few months ago, I was previously using duckduckgo (and Google before that). As most of you have probably noticed Google search results have seriously dropped in quality the last few years, but especially in 2023. I'm no longer able to get meaningful results for almost any topic, especially if it's technical, the only results are AI generated (?) / obvious SEO spam websites. It takes me multiple different search terms and clicking through multiple results to find anything semi relevant, and even then it's a shallow article maybe summarising what I'm looking for. Unfortunately DDG seems to be going the same way.

Whereas Kagi reminds me of the 'old' google search. The results are meaningful and relevant, not diluted with pages of generic article results. They also offer a lot of great customisation options like being able to block or boost certain sites in results. They have some built in lists for common filler sites. I can't comment on the AI variation but I hear that's progressing well.

I wouldn't call myself a power user of Kagi, but even then I'm getting far better results than other search engines, definitely worth the price per month.

I'm not affiliated with them in any way, just thought I'd share my anecdotal experience.

Amazing, thanks for the reply, I'm finding some good resources afyer a quick search of `cl100k_base`.

If you have any other resources (for anything AI related) please share!

Commenting to follow, curious about the answer.

From what I've found through Google (with no real understanding of llm) 2^16 is the max tokens per minute for fine tuning OpenAI's models via their platform. I don't believe this is the same as the training token count.

Then there's the context token limit, which is 16k for 3.5 turbo, but I don't think that's relevant here.

Though somebody please tell me why I'm wrong, I'm still trying to wrap my head around the training side.

I'm a completely unknown artist with 4 songs on Spotify, mostly released during 2020. In total I'm at 54388 plays, which has earned $42.41. This is across all platforms, though Spotify is 95% of the plays.

I'm not sure if Spotify has dropped their payout per play since 2020 but I'm likely at the lowest payout rate and I'd say it's not terrible (although it's not great). You also get paid more for Spotify premium streams, which afaik was the majority of my streams.

Phillip J Airfry made me smile, love the name.

Do this frequently enough and they'll start inviting you back with free travel + accommodation.

Shout-out Nebula, an alternative YT/creator platform which has no ads or sponsor segments. It's a monthly subscription but fairly cheap, and it gives you access to all videos on the platform (unlike patreon which is for a single creator). The monthly subscription cost is then split between all creators on the platform.

It's not a 1-1 alternative to YT as creators have to opt in, so most (imo low effort) videos/creators won't be on there. It's fantastic for any tech/engineering/history/news though, high quality/effort vids with no bullshit.

Note: I have no vested interest in Nebula, I'm just a user that's happy to support good creators and a platform that's actively opposed to advertising.

If this counts as an ad/spam - let me know and I'll delete this comment.

I don't think I've ever completed a training course on time, only after getting "urgent: 7 days to complete".

I've got severe ADHD so these types of assessments are near impossible due to the slow dialogue and forced wait time. Though most of these courses give you multiple (or unlimited) attempts, so I'll screenshot each slide + wrong answers and brute force until I'm done. At least I can get other stuff done in the meantime.

The problem wasn't necessarily the A/B testing rather than what came next, Cambridge Analytica. This current headline seems to follow the same message, that being able to manipulate a users mental health is beneficial to Meta.

When I lived in Sydney I had a few friends that frequently bought Rikodeine (dihydrocodeine cough syrup), which is also meant to be tracked via pharmacists recording IDs. My friends knew what pharmacies didn't bother to track it, normally if you were a 'repeat' customer and knew the pharmacist. They were able to buy 10~ bottles in a day from 8 - 10 pharmacies, repeating every few days. This was also in Sydney's central cbd so there's a pharmacy every other block.

I assume the situation is similar with pseudoephedrine, though the drug class/restrictions may be different.

Relative to infinity, anything finite is small. I agree with OP, if anything it should be "finite universe".

Normally mine last 2 - 3 days, with maybe a couple days of ramp up. These ones sound very real, exactly like the voice of a friend or relative etc. The thing that makes them especially feel real, the voices always have some 'spatial' element to them, like I'm overhearing a conversation from the other room. They also get quieter (or disappear entirely) if I walk away from the source.

The other form of voices I get, which are much more frequent + longer lasting, sound like my internal monologue. They can be different voices, but always sound like I'm 'thinking' them (rather than overhearing). These tend to last 2~ weeks after an episode, and come back when my anxiety or insomnia gets worse. Luckily these are pretty ignorable, if not a helpful sign that I'm nearing an episode. I'd compare them to anxious/intrusive thoughts (though a bit more intense).

I haven't had an episode in almost a year and have learnt to easily manage the intrusive thoughts (/possible start of an episode). This hasn't even impacted my lifestyle all that much, if anything it's taught me to manage my work/life balance.

Adding to this, I've experienced a few psychotic episodes but am not (for now) diagnosed schizophrenic.

Even during my first episode, which was by far my worst while having the least experience, I was aware that what I was experiencing (hearing voices/feeling watched) was likely me experiencing psychosis. While I still 'went along' with the episode, I was doing everything in a way that minimised harm to myself and others. I eventually called an ambulance, thinking that either I'm going through psychosis (and need an ambulance), or what I was hearing was real (and having ambulance/police on scene would help).

Once the ambulance arrived I had 'confirmation' that I was going through psychosis, and could then 'ignore' the voices to the point that I wouldn't act on whatever I was hearing. My symptoms continued for a couple weeks, but at no point did I even consider these to be real (despite them feeling very, very real).

Going forward, I've had a couple other episodes, all of which I've noticed early on and been able to act accordingly (reaching out to friends or professionals). While some of these episodes were equivalent in intensity to my first episode, none of them had me acting on the symptoms (beyond calling my psych and saying I'm having an episode).

I was briefly on anti-psychotics, but as I'm yet undiagnosed, my psych wasn't comfortable with me being on them long term. Instead I've been able to manage it through exercise, sleep management, mindfulness, and better management of my anxiety.

That said, I'm only discussing my own experiences with psychosis/potential of having another episode. I'm not medicated but don't worry about it (potential harm to myself or others) in my day to day, though I make sure to mention it to those that could be affected (flatmates or friends/family).

Thanks for taking the time to explain, very informative! I've been meaning to look into the mechanisms of Ritalin for a while buuuut ADHD.

Strattera is really interesting, I've heard of it but I don't think it's available in my country. I'll ask my psych about it. We also don't have Vyvanse here, which my psych thinks would suit me better, as I'm currently on 162mg/day (of Concerta).

Have you tried Strattera? How does it compare to Ritalin for you?

Not sure about the pharmacology, but I want to say methylphenidate (Ritalin)?

I have severe ADHD, so I'm taking a pretty high dose daily, and I haven't noticed diminishing effects nor rebound. My dose hasn't changed in years either.

We're so poorly multi-threaded, even addressing climate change has been horribly slow...

When you fall off a horse, do you get back on? Personally I head straight home and refuse to even look at a horse ever again.

I'm not sure what the solution is, but I do feel IP protection is important.

If we were to get rid of it entirely then your success in the market largely depends on what resources you already have. Large companies have the upper hand as they already have the user base, along with the marketing budget to sink you (not that the current situation differs all that much).

If your argument is "thats the way it should be, let the market decide who succeeds" then that's a whole different conversation.

Either way I don't see an obvious solution that doesn't benefit the very companies you say we need to stop, though I agree on the sentiment.

If only I could, I could use some time off...

I've managed to break venv, npm and composer (php).

I don't use that as a reason to choose what I'll use in my projects, that's decided by the PTSD incurred from 7 years of php.

I absolutely would not opt to self-host, but unfortunately that's not my department. I agree about the performance issues, most of our issues were performance related and did seem quite random (though unsure if that was our self hosted instance or a Gitlab issue, sounds like it could be both).

I will admit I'm probably overly harsh on the support staff, and misrepresenting the support issues we had. I wasn't directly involved in most of it so I'm parroting what I've heard from coworkers that were more involved, which isn't the whole story. Though the times I was directly involved (in support requests) the experience mostly matched yours, with a couple (albeit rare) cases of slow replies.

In terms of self hosting I 100% agree, and anyone who is thinking of setting up their own self hosted instance should take note of your comment.

Every morning we begin stand-up with a ritual, our whole team simultaneously muttering something arcane whilst manipulating the inner workings of Teams, hoping to appease the MS gods.

Shortly after the ritual is complete you hear a cacophony of "fucking teams wouldn't find my headset/camera", and the day begins.

The company I'm currently at uses a self hosted instance, which due to our "security"* constraints we're grossly behind in updates. Not to say that's Gitlabs fault at all, but the amount of issues we have near daily is hilarious (if it didn't simultanously crush productivity).

That said, I recall issues with the Gitlab Enterprise Support / whoever we contact about our license etc, mostly to do with slow or poor communication. Requests would sit for a month before we got a reply, often dismissive or generally unhelpful, though we _were_ on an outdated version so I don't blame them.

I vaguely recall an email about support ending on self hosted instances? I can't recall the details, but I know it triggered an internal investigation into moving away from Gitlab. EDIT: Pretty sure I'm remembering self hosted Jira, a quick Google search shows Jira EOL but no Gitlab.

All of that said, I largely blame my company for the failures here. I wouldn't expect any company to support self hosted, outdated versions. The support issues were annoying, but I'm also not sure how much I can blame Gitlab for that. I'm also struggling to remember the details, so take this as one mans hazy annecdote.

* film industry, security by obscurity in the worst ways, leads to incredibly outdated and neglected tech (we only _just_ transitioned to Python3, and that's only the core services)

I would be suggesting that if those zoomers ran a business that could die with poor file system management, but that's hardly what we're talking about is it?

I am suggesting that business owners with substantial cash holdings should be well aware of fractional reserve banking, or at least the notion that money in the bank is not guaranteed.

Do you seriously think any of these companies weren't aware of the $250k insurance limit?

I'm genuinely curious what you mean, as your example sounds like a straw man, but I'm hoping there's something I'm missing.

Do people really not know how fractional reserve banking works? I'm not saying the general public, but those with >$250K sitting in a bank.

I agree on the disconnect, but I don't think it's all relevant to this situation. Almost every party involved should have been well aware of this fairly basic concept.

It's unfortunate, and I only see it getting worse, though I'm hoping there's enough of us to influence the velocity. The sentiment seems to be growing, so I'm trying to be optimistic, but it's going to be difficult when up against the resources of "big tech" (for lack of a better term).

For me it ultimately boils down to my personal morals . I am not comfortable supporting these types of companies, even if my contribution makes minimal difference (on either side). I refuse to contribute because I wouldn't be comfortable with _myself_. I hope this decision will be beneficial to more than just myself, but if it isn't I'll at least be happy knowing I did what I could to support what I believe in.

(side note, I didn't mean this to be so subjectively idealistic, but even if you completely disagree with my opinion on tech, I think the introspection and focus of attention would benefit all)

Disliking the industry does not mean you can't find a space where you're contributing to something you believe in.

In my opinion, the industry has been downhill since FB blew up and started abusing its power, though I wasn't around for the dotcom/Microsoft era so that's just my experience.

I absolutely loathe the vast majority of large tech companies, and that's only gotten worse (especially in the last 5 years). I outright refuse to work for companies (that I percieve) that are primarily involved in data collection/privacy violations, as I'm sure a lot of developers would refuse to work on weapons systems.

That said, I'm growing more and more jaded, and wondering how best to focus my efforts. This has lead me to take jobs at smaller and smaller companies, heavily impacting my income, but at least I'm not actively working on projects I despise. I've also found myself spending a lot more time on personal or contributing to open source projects that push back on said projects I despise (largely privacy related).

Not to say everyone is in a position to do that, but I do think we should be more conscious of what we're helping to build. Focus your efforts on projects you believe in, and be willing to push back on those you do not. This is not going to solve anything, but as the crowd grows larger we'll hopefully create enough noise for others to take note.

(this is all very subjective, take my comment with a grain of salt)

I'm not from the US but I started working part-time as a developer at 16, and full-time at 17. This was 10+ years ago, but I know more teenagers now doing something similar than there were when I was growing up (for non-typical teenage jobs).

I agree about there being no virtue in experiencing pain, but I think there's value in it. Neither form of personal enlightenment is inherently better, they are different experiences that lead different yet similar paths, though both provide an understanding the other cannot