Since Clippy 2.0 is out in the real world now, you can pick another legend to revive. I went with AIM, replacing your AIM friends with AIs. You should do Stumbleupon with AI generated websites or bring back MSN :)
I thought this immediately also. I already have ollama set up to run llm tasks locally. I don't want to duplicate that but it would be fun to try this front end.
Looks like it's a special font provided by https://github.com/jdan/98.css (Which has come a long way in the past couple of years, despite still being 0.1.x)
I should probably 1.0 it and call it a day, it's pretty much done. In the back of my mind I've thought of making the markup more amenable to LLMs like Sonnet (maybe with tailwind style utility classes)
Microsoft uses Clippy for the paperclip emoji in the Fluent Emoji set. The trademark is why the open source version of Fluent Emoji doesn't use Clippy's likeness.
Actually this says the trademark is pending them proving they are using it and they aren’t so they keep filing extensions. Also it’s limited in scope to word processing (at there is lots of back and forth with the trademark office about that)
Super cool. Serious 90s vibes. I also tried to make a super clippy here. https://chatbotkit.com/examples/super-clippy I think I match the color shema perfectly but does not have the same feeling as the original.
> A JavaScript implementation of the Jinja templating language
A guess without looking into the code: Jinja templating is used to define how to prompt the model (i.e. system first, then this specific character / token, then user, then if it's a tool prepend this and append that, etc.)
I think it’s legitimate to ask why these dependencies are necessary. LLMs have created whole new classes of vulnerabilities, and things like a GitHub client (which downloads arbitrary data/code) and a templating engine (which executes it) expose an even larger attack surface.
If someone’s going to get RCE on my machine, I don’t want it to be through the silly Clippy LLM UI, you know?
The real answer is that some of us (the Electron maintainers) have been playing with local LLMs in desktop apps and right now, node-llama-cpp is by far the easiest way to experiment - but it's also not meant for desktop apps and hence has _a lot_ of dependencies.
In general, pruning libraries in Electron isn't as easy as it should be - it's probably something for us to work on.
I think this is explained on the linked project page:
This project isn't trying to be your best chat bot. I'd like you to enjoy a weird mix of nostalgia for 1990s technology paired with one the most magical technologies we can run on our computers in 2025.
You might be looking for the more minimalist Grumpy which is hand-hewn from a pure silicon monocrystal.
You sure wouldn't want them spying on you, stealing your data, chewing up your resources for shady profit schemes, or making your machine unbootable. Better to leave that to the experts at Microsoft and FAANG since all those features come preinstalled nowadays.
Snark aside, given the context, this really seems like a baseless attack on independent open source developers, who represent a significant potion of this site's subject matter and target audience. Genuine question: why do you feel that this warning is appropriate here but not the dozens of other solo github projects that make it to the HN front page every week?
Review the guidelines; this type of slander is well afoul of them. Go ahead and get the last word in if that's important to you, as I won't be wasting any more of my time on this thread or your destructive attitude. Comparing different opinions is acceptable but mean-spirited emotionally driven personal jabs are not.
There's a bit of gatekeeping from multiple participants in this subthread, but I think it was fair enough of raydiak to push back on a bit of alarmism in the root comment.
Regardless, your comments were needlessly hostile and not in keeping with the style of discussion the guidelines ask of us.
I'm late to this, but we can do without this kind of comment on Hacker News, as it falls under the guidelines about generic tangents and tangential annoyances. It led to a hostile exchange down-thread, which is exactly the kind of thing we're trying to avoid here, but is what happens when people take threads away from the main topic.
Given we're a technology-focused site, Hacker News readers can be reasonably assumed to be technically proficient, and aware of the importance of taking normal security measures.
Question for the devs in here...something I've been thinking about a lot recently. So I see that OP linked out to a public github repo...but when downloading the actual bundle, what's a quick way for me to determine that what I'm installing on my mac is actually the same as what's in the public repo? It's always seemed like a loophole to me ready for (potential) exploitation.
>> Ship project.
>> Link out Github repo on the static site somewhere
>> Gain trust instantly as users presume the public repo is what's used behind the scenes
Disclaimer: I'm a web dev and don't know a single thing about native MacOS software
I sign my binaries on macOS with Apple codesign and notarize - and with Microsoft's Azure trusted signing for Windows. Both operating systems will actually show you a lot of warning dialogs before running anything unsigned. It's far from perfect - but I do wish we'd get more into the habit of signing binaries, even if open source.
IIRC correctly, Clippy’s most famous feature was interrupting you to offer advice. The advice was usually basic/useless/annoying, hence Clippy’s reputation, but a powerful LLM could actually make the original concept work. It would not be simply a chatbot that responds to text, but rather would observe your screen, understand it through a vision model, and give appropriate advice. Things like “did you know there’s an easier way to do what you’re doing”. I don’t think the necessary trust exists yet to do this using public LLM APIs, nor does the hardware to do it locally, but crack either of those and I could see ClipGPT being genuinely useful.
We are probably getting closer to that with the newer multimodal LLMs, but you'd almost need to take a screenshot on intervals fed directly to the LLM to provide a sort of chronological context to help it understand what the user is trying to do and gauge the users intentions.
As you say though, I don't know how many people would be comfortable having screenshots of their computer sent arbitrarily to a non-local LLM.
> As you say though, I don't know how many people would be comfortable having screenshots of their computer sent arbitrarily to a non-local LLM.
Of the technical, hang-out-on-HN crowd? Ya, probably not many.
Of the other 99.99% of computer users? The majority of them wouldn't even think about it, let alone care. To quote a phrase, ”the user is going to pick dancing pigs over security every time”.
Even without the non-chalent attitude towards security, the majority of the population has been so conditioned that everything they do on a computer is already being sent to 1) Apple, 2) Google, 3) Microsoft, or 4) their employer, that they're burnt-out of caring.
All that is to say that if you can make a widely-available real-time LLM assistant that appeals to non-technical users, please invite me to your private-island-celebrity-filled-yacht-parties.
I think we're well into the paradigm of "hidden employee activity monitoring software" already taking periodic screenshots and sending it to an LLM somewhere, which then generates aggregate performance metrics and dashboards for managers. I've heard of multiple companies working on this for $bigcorp environments, customer service/call center workstation PCs, etc.
Models with native video understanding would do the trick - Advanced Voice Mode on the ChatGPT iOS/Android app lets you use your camera, works pretty well; there's also https://aistudio.google.com/live (AFAIK there are no open-source models with similar capabilities)
The way I remember it a lot of software had "help" documentation with full text search in the late 1980s and early 1990s but the common denominator was that it didn't work in the sense that you got useful answers less than 10% of the time. Until Google came along, users got trained to avoid full text search facilities.
The full text facility attached to Clippy really was helpful, getting useful answers around 50% of the time. I thought the whole point of making him an engaging cartoon character was to overcome the prejudice mid-1990s users had towards full-text search in help.
It looks like you're one of the 1% of humans who still write letters themselves! Dear me, imagine that, what do you think this is, the 90s or something?! Would you like to join the other 99% of humans and doomscroll and shytpost while I write that letter for you?
i added a minutely scrot cronjob about a year ago and haven't used it once. remembering "that website i was on last week" is apparently not a real problem I was having
> Things like “did you know there’s an easier way to do what you’re doing”
That could come off just as patronizing as the original Clippy. If it said things like "Would you like me to generate you a letter for X?" it would be miles ahead of the original.
Actually this is a good way to find product ideas. I placed a query in Grok to find posts about what people want, similar to this. Then it performs multiple searches on X including embedding search, and suggested people want stuff like tamagotchi, ICQ etc. back.
I feel like these are all great examples of things people think they want. Making a post about it is one thing, actually buying or using a product, I think the majority of nostalgic people will quickly remember why they don't actually want it in their adult life.
I see this a lot in vintage computing. What we want is the feelings we had back then, the context, the growing possibilities, youth, the 90s, whatever. What we get is a semi-working physical object that we can never quite fix enough to relive those experiences. But we keep acquiring and fixing and tinkering anyway hoping this time will be different while our hearts become torn between past and present.
299 comments
[ 4.1 ms ] story [ 308 ms ] threadhttps://www.youtube.com/watch?v=tu_Pzuwy-JY
I hope that one day a non-Electron app (to minimize resource usage when idle) will also appear!
Although there is a CSS rule for manipulating how fonts are anti-aliased, it was never standardized, and Firefox doesn't implement the vital no-smoothing option: https://developer.mozilla.org/en-US/docs/Web/CSS/font-smooth
Maybe with enough retro revivals it will receive attention.
https://tsdr.uspto.gov/#caseNumber=90782096&caseSearchType=U...
- A JavaScript implementation of the Jinja templating language
- A full GitHub API client
- A library that takes a string and tells you if it's a valid npm package name
- A useless shim for the JavaScript Math module
And 119 other libraries? This thing would have taken up 10% of the maximum disk space available on a Windows 95 FAT16 volume.
A guess without looking into the code: Jinja templating is used to define how to prompt the model (i.e. system first, then this specific character / token, then user, then if it's a tool prepend this and append that, etc.)
If someone’s going to get RCE on my machine, I don’t want it to be through the silly Clippy LLM UI, you know?
In general, pruning libraries in Electron isn't as easy as it should be - it's probably something for us to work on.
This project isn't trying to be your best chat bot. I'd like you to enjoy a weird mix of nostalgia for 1990s technology paired with one the most magical technologies we can run on our computers in 2025.
You might be looking for the more minimalist Grumpy which is hand-hewn from a pure silicon monocrystal.
Snark aside, given the context, this really seems like a baseless attack on independent open source developers, who represent a significant potion of this site's subject matter and target audience. Genuine question: why do you feel that this warning is appropriate here but not the dozens of other solo github projects that make it to the HN front page every week?
Regardless, your comments were needlessly hostile and not in keeping with the style of discussion the guidelines ask of us.
https://news.ycombinator.com/newsguidelines.html
https://news.ycombinator.com/newsguidelines.html
Given we're a technology-focused site, Hacker News readers can be reasonably assumed to be technically proficient, and aware of the importance of taking normal security measures.
>> Ship project. >> Link out Github repo on the static site somewhere >> Gain trust instantly as users presume the public repo is what's used behind the scenes
Disclaimer: I'm a web dev and don't know a single thing about native MacOS software
https://en.wikipedia.org/wiki/Reproducible_builds
I sign my binaries on macOS with Apple codesign and notarize - and with Microsoft's Azure trusted signing for Windows. Both operating systems will actually show you a lot of warning dialogs before running anything unsigned. It's far from perfect - but I do wish we'd get more into the habit of signing binaries, even if open source.
https://somethingorotherwhatever.com/tiny-elvis/
As you say though, I don't know how many people would be comfortable having screenshots of their computer sent arbitrarily to a non-local LLM.
Of the technical, hang-out-on-HN crowd? Ya, probably not many.
Of the other 99.99% of computer users? The majority of them wouldn't even think about it, let alone care. To quote a phrase, ”the user is going to pick dancing pigs over security every time”.
Even without the non-chalent attitude towards security, the majority of the population has been so conditioned that everything they do on a computer is already being sent to 1) Apple, 2) Google, 3) Microsoft, or 4) their employer, that they're burnt-out of caring.
All that is to say that if you can make a widely-available real-time LLM assistant that appeals to non-technical users, please invite me to your private-island-celebrity-filled-yacht-parties.
shudders.
"It's time to work, Dave"
The full text facility attached to Clippy really was helpful, getting useful answers around 50% of the time. I thought the whole point of making him an engaging cartoon character was to overcome the prejudice mid-1990s users had towards full-text search in help.
Would you like help?
* Get help with writing the letter
* Just type the letter without help
|_| Don't show me this tip again
It looks like you are writing a comment on Hacker News.
Would you like help with:
- Commas? There shouldn't be one behind "responds to text"
- Capitalization? You've missed a D in "did you know..."
- Punctuation? You've missed a question mark behind "what you’re doing". It goes inside the quotes, of course!
[] Don't ever suggest anything like this ever again.
Pretty soon I won’t even need biological memory.
Wait, are you really looking this up? You don't even know how to do this? Are you kidding me?
Gosh, it's been an hour and you still haven't fixed this bug? Are you retarded or something? You don't deserve this job.
That said, if we could automate it, it might free up more of my brain for productivity…
That could come off just as patronizing as the original Clippy. If it said things like "Would you like me to generate you a letter for X?" it would be miles ahead of the original.
https://fabulous.systems/posts/2024/06/if-i-ever-get-a-dog-i...
https://youtube.com/watch?v=6umxhkdKzSY&t=79s
After all it was requested almost daily over at x.com
https://x.com/search?q=ai%20bring%20clippy%20back&src=typed_...