one of the very first programming projects I took on was to figure out how to scrobble the records that I was playing. It was my first exposure to so many things: Ruby, FFIs, audio processing, audio fingerprinting (I think I used echo nest ?). Ended up going to local meetups to ask for advice.
last.fm is one of those services that is from the pinnacle of the open web.
The biggest issue in the real estate world is dealing with local state, county, and municipality governments and each one's unique ordinances, etc. While I think LLMs will be great at pulling data from a variety of disparate sources (ie, records databases), or even automating interacting with them, they are not going to be great at solving the problem whose solution is "this is the person in the county court clerks office you need to talk to."
Matt Gallagher and CocoaWithLove are major highlights from the early days of my journey in learning iOS development. Awesome to see he is still publishing such high quality information!
Same. Matt Gallagher's posts on Swift concurrency and Cocoa memory ownership are still the clearest writing on those topics anywhere. This post is still the right answer to a Stack Overflow question I haven't asked yet.
I haven't been to SmugMug in years (decade?) but wow, holy early 2000s big bold brash neon web design. Then there's this huge scrolling ticker in 72pt text at the top, "free festival passes for the next 50 subscribers" that has been there ... for a few days now. Gives total "You may already be a winner!" vibes.
Which is basically a "pro-ish-plus" version of Flickr from the same owners as far as I know. I've been a Pro user of Flickr for a long time but probably hard to justify at this point which probably means that it's even harder to justify for the average consumer. Interviewewed them back in the day when they were a prominent AWS customer.
Yeah, they should buy Flickr and abandon the whole social media aspect and just turn it into an exact copy of SmugMug, but interleave the pricing tiers.
It would really be crazy if they did that, but they claimed that limiting the number of photos users could upload, instead of limiting the quality, somehow made it more like a social media platform.
I know a twelve year old kid who is proactively using LLMs to build websites for lawn-mowing businesses, calling them up, asking them if they want it for $200, and closing deals in seconds.
I know it sounds far-fetched, but he does all the work up-front before even contacting them, using logos and info from Facebook or Google. He's cleared several thousand dollars so far.
I get that the owners aren't going to be the proactive ones who have the awareness, time, or vision for doing this, all your points are valid. However, AI has definitely changed the calculus here--I'm glad I'm not a web dev anymore.
Once he creates the website, does he also host it and handle the billing for his clients? Is he using a website builder like Square space or hosting on AWS?
The hurdle is more than just building the site, a lot of really small non-technical businesses don't want the trouble of handling the billing and maintenance of the site.
It might just be static site hosting for businesses that want a real website, and not just a Facebook page. Static site hosting is so simple, I believe a 12 year old could do it.
I recently stood up a personal and a business site using Claude + Astro + GitHub + Cloudflare Pages, and apart from the Claude subscription, it’s all free.
Definitely not skills that are going to be in the typical restaurant owner’s wheelhouse (not hard to learn, just not likely to care) so you’d need to figure out how to host per-business to avoid hosting everything under one account and running over the free tier. But there’s very little management or payment necessary until you get quite a bit of traffic, which is probably not likely for your average suburban sandwich shop.
So, he's non-technical. He hasn't written a line of code. I don't know the details of how he hosts or deploys the sites, but I'd likely guarantee that he asked whatever AI he uses and it just walked him through the process of getting one hosted, then he has replicated that.
I mean, for a 12 year old, $200 and not having to do any more work in the future might be a good deal, and for a business, a $200 one-time probably seems like a steal. I agree that there might be long-term issues for them if they don't know how to maintain them, but what are they going to do, sue the 12 year old?
Music / mp3 blogs were one of the heights of the free and open internet. RSS, Hype Machine [0] (an aggregator), blog rolls, back links, etc, all allowed for both discovery and taste-making in an organic way that still let individuality shine through. New artists could gain visibility just by emailing a couple MP3’s. One could find a subset of blogs that matched one’s general taste and discover new artists every week.
Today’s world of algorithms and an endless sea of new music really pales in comparison. It’s completely soulless.
HypeMachine was a core memory of my adolescence. I reminisce on the way the site encouraged music discovery through the aggregation of all the little niche blogs, among those pigeons and planes. Memories of Sitting in my room with old headphones while sorting by most recent. Unfortunately nothing has ever replaced it for me and unfortunately it has faded slowly as Spotify and SoundCloud gained more prominence. I agree, music discovery has become soulless…
This is really cool! I really liked the architecture explanation.
Once you get solid rankings for the different LLMs, I think a huge feature of a system like this would be to allow LLMs to pilot user decks to evaluate changes to the deck.
I'm guessing the costs of that would be pretty big, but if decent piloting is ever enabled by the cheaper models, it could be a huge change to how users evaluate their deck construction.
Especially for formats like Commander where cooperation and coordination amongst players can't be evaluated through pure simulation, and the singleton nature makes specific card changes very difficult to evaluate as testing requires many, many games.
For anyone wanting a quick breakdown of the current situation: the Sig Sauer P320 is a striker-fired handgun, which means the firing pin is spring loaded and retained by a sear. Other handguns are hammer-fired, where the trigger (or slide actuation) cocks the hammer. Other popular striker-fired guns include the Glock and Smith and Wesson M&P series. Frequently, striker-fired pistols come without safeties, but optionally add them.
The P320 was popular as it was designed as a modular system, allowing a single FCU (firing control unit, basically a trigger and striker assembly) to be independent and swappable with other parts of the handgun: grip, slide, barrel, etc. This allows for a single platform to serve multiple needs: concealed carry, compact, full-size, or even competition models, as well as be transferrable across calibers. The magazine design also allowed for more rounds to be carried in compact configurations.
The P320 was selected by the US Army [0] as the official replacement for the Beretta M9 as a service-issued sidearm, officially designated the M17 or M18 (in 9mm).
In 2020 SIG SAUER initiated a "voluntary upgrade program" [1] that swapped various components of the trigger to prevent unintended discharge (UD) events that could occur when the pistol dropped in certain orientations. These changes became standard for the M17 and all P320 manufactured after.
Recently, there have been very high-profile cases and investigations around UD events, the most recent being by an event in the Air Force that led to the death of an airman. In that case the Air Force put a suspension on the firearm during the investigation but eventually arrested the airman responsible, as they determined he had lied about the events [2].
Regardless of the specific failure modes of the weapon, there is a stigma around it, resulting in various law enforcement agencies switching from it or ranges banning the firearm. This has been popularized by incidents caught on video and somewhat viral videos of testing the firearm in a variety of scenarios.
All in all, the P320 is one of the most mass-produced firearms in the world, and I would not be surprised to see Sig Sauer continue to fight in the court of public opinion to defend the reputation of the firearm, in what I would deem a losing strategy.
I understand your speculation on the amount of variance, but I haven't seen any data to support it.
Sig's "recall" was a drop-safety issue, where in certain orientations the weight of the trigger could generate enough momentum to allow an unintentional discharge.
There's plenty of data on the variances of P320 parts being much larger than specified by Sig, and it has been presented in a few court cases. Root causing this issue to tolerances hasn't been done, though.
I think one of the best demonstrations of this, with detail on the amount of travel required for most striker-fired handguns is this video [0]. Lots of detail and relatively methodical.
last.fm is one of those services that is from the pinnacle of the open web.