In that video he asserts through the closed door that Omegle hasn't done anything "for the children" and then in the article they have one measly line about how Omegle actually has been productive on that front. The text now on the Omegle site seems to support that they did what they could as well. Of course they're not going to get a good conversation with him when that's how they're going to frame it compared to reality. Whether Omegle was doing enough or should exist to begin with are different arguments but the premise of "Omegle is doing nothing" appears very wrong and I imagine offensive to creator/employees.
Like siblings, I switched and never looked back. I used !g with DDG often, more as the years went on tbh, but with Kagi the only time I use !g is when the search results end in one page, which is rare and almost always the Google results are noisy trash. Maybe 5% of searches I check Google and 10% of those actually return something useful that Kagi didn't. On balance with how much better the other 95% of queries are, first time I'm excited to search for things.
My biggest complaint with Kagi is sometimes when I search for something the top result is a pinned domain that maybe hit the right keywords but mentally to me it was unrelated to the search I was performing. That's a pretty good problem to have at the end of the day.
> with Kagi the only time I use !g is when the search results end in one page, which is rare and almost always the Google results are noisy trash
That might be because Google won't admit when it doesn't have results responsive to your query. It will just fill the results page with nonsense instead of saying "I couldn't find any results".
I had the privilege and good fortune to introduce a diverse group of highschoolers to Scratch as an intro to programming around 2013/2014, it was a lot of fun! A few of them seemed pretty hooked and were doing some wild things pretty quickly, I imagine at least a couple pursued it further. Reminded me a bit of like ~2005 Flash community.
I'm not familiar with the summarization or NLP space really but I remember ~2011-2015 I signed up for a couple of daily email services that summarized a number of news articles and the summaries were fantastic. I don't even remember what they were called, they eventually sold out with ads and worse formatting/summaries to make money I guess. I often use them as an example of 1) why LLMs are a bit old news for the summary use case and 2) how various LLM use cases will probably also be ruined because for a lot of people tools like that seem novel and useful but all I can see is onboarding to more advertising.
So to someone who is actually knowledgeable in this space, are LLMs really that much better than whatever we had 10 years ago? Is this tech the key to some features we truly didn't have before?
Not an NLP expert but the biggest difference in my experience is guided focus, so to speak. When summarizing something huge like the US Code, for example, you can tell the LLM to focus on specific topics and anything adjacent to them so that it ignores irrelevant details (which is usually >99.9% of the text in my use case). The word relationships encoded in the LLM are really good at identifying important adjacent topics and entities.
LLMs are also really good at the harder NLP problems like coreference resolution, dependency parsing, and relations which makes a huge difference when using recursive summarization on complex documents where something like "the Commisioner" might be defined at the beginning and used throughout a 100,000 token document. When instructed, the LLM can track the definitions in memory itself and even modify it live by calling OpenAI functions.
I use OpenAI function calling most of the time I use the OpenAI API since that's the easiest way to get structured data and implement retry logic.
The simplest implementation is "retrieve_definition(word_to_lookup, word_to_replace)" with some number of tokens at the beginning of the prompt dedicated to definitions. You can use a separate LLM call with a long list of words (without their definitions) to do the actual selection since sometimes there might be ambiguity, which the LLM can usually figure out itself (it can also include both definitions when it's too uncertain if instructed).
A more complex variant does multiple passes: first pass identifies ambiguous words in each chunk, second pass identifies their definitions, third pass does actual summarization using the output of the previous passes to craft the prompt.
Glancing at some comments here mentioning the overnight charging feature working for them, am I in the minority by charging randomly whenever it needs it? My charger isn't even in the bedroom. Curious how my random on/off habits compare to overnighters in long term battery health.
I try to take it off around 80% so I was interested in this feature and confused when the update didn't include it. I don't like that it's in my "pros" column for upgrading when it could be available to everyone, would rather not reward them for that.
The world of lithium-ion battery chemistry is quite complicated, but in my experience, the top contributors to battery wear-out are, in decreasing order of severity:
- Charge cycles (including partial cycles, e.g. 100% -> 50% is half a charge cycle)
- Depth of discharge (deeper discharge cycles are worse)
- Extended periods in hot conditions (over 40 degrees Celsius)
- Extended periods of being held at 100% state of charge
In essence, your charging behavior dings the first two items: higher number of charge cycles, and deep discharge cycles. If your goal was battery longevity, keeping the phone plugged in as much of the time as possible would be the best strategy. Bonus points if your phone supports a "stop at 80%" mode, but still the winner even without this feature.
I think another huge killer of lithium batteries is charging them when below freezing. I bet most phones probably wouldn’t allow it though - or at least would limit it.
I left out things that a typical battery management system will prevent, like over/under charging, or operating outside safe temperature limits. But yes, charging a typical cell below 0 degrees Celsius will immediately and permanently damage it. Here's a ref with some details on why:
My US state has been legal for some years now and it's a similar story. Part of me wants the market to collapse a bit as it's kind of annoying how many businesses exist, in my city this one corner had 3 shops in a touristy part of town and no one ever in them. I would hate to see it collapse into like 2-3 mega brands owned by some global conglomerate that also owns mega beer/tobacco though.
As someone who smoked/vaped daily for several years, it wouldn't bother me and if anything I would get excited like "where is that?" But after several months of voluntarily not using I actually did get to a point where I dislike the smell of it. Cigarettes, perfume, cologne, even strong laundry detergent still smell worse to me, but yeah weed is not a great smell.
It really hit me when I went to Manhattan for the first time since covid and the ratio of pee to cannabis stench had flipped rather strong to cannabis.
It's been like 10 years since I looked into the options but at the time I was getting my first cards I was doing the math on various points cards to compare to cash back and what I found was, unless you were really gaming the system, the best you could hope for is roughly 1-2% returns. And the system, whether you're gaming it or not, also requires you to then play in the system so any given card might work out to like 5% if you spend those points on air travel... with a whole host of restrictions including but not limited to a specific airline. I don't have the time in my life to game the system so I can fly marginally cheaper with a specific airline and therefore restricted set of airports and scheduling options.
Travel cards haven’t been that restrictive in the time I’ve been using them (since around 2016). Chase Sapphire Preferred/Reserve and Amex Gold/Platinum don’t have any restrictions on what earns points, it just needs to code as travel (flights, rides, or hotels) or restaurants and while there’s some benefits from using the Chase/Amex portals to book even that’s not required. Similarly points can used towards anything that codes as travel.
Of course there are airline/hotel specific cards and those may be more restrictive but I’ve never seen the point in those.
I used to have a sapphire preferred, which promised 1.25x point value or something when spent on travel. When I went to book via the chase portal, everything was magically .25x more expensive when compared to the airlines page.
I can’t say I’ve observed that. Checking just now the prices I’m seeing in the Chase portal are identical to those listed on Expedia (which the Chase portal is built on). Some flights are slightly cheaper on airliner sites though.
hmm, maybe it was just jetblue specific, or something about the time I was booking. But it left a bad taste in my mouth, I decided to cancel the card at that point.
The one points card I have actually is a Marriott card which obviously only works with Marriott, but Marriott is everywhere and doesn't restrict scheduling options so it was an easy choice at the time. I used to travel weekly for work and the hotel was on me for reimbursement whereas the airline was on company card, and Marriott was the preferred booking option by my company so it made a lot of sense to get the card. The math at the time for that was like 5-7% when spending the points at Marriott plus a free night for what was an $85/year card. Useful for personal travel.
> Software engineers who think they are better without AI may quickly be replaced with "lesser" software engineers who get things done much quicker.
Using AI to program is just ctrl-c ctrl-v software engineering, very clearly not the point of programming competitions. High level tools as you mention don't make you a better programmer, they make you a better software engineer.
I think the discussion you're raising is a good one it's just not relevant to a programming competition.
Pretty subjective at the end of the day. Personally I find them roughly equivalent quality in mapping/traffic but Google has turned into a pile of sponsored content. Apple frequently has the exact locations of businesses within a building wrong or confusing but I'll take some walking around a building over a recommendation for Subway when I search for "camera store". Anyway I submit corrections to Apple Maps and they accept those corrections within a couple of days 90% of the time and when I need that thing in the future it is correct.
