Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Welcome to Wikifunctions

wikifunctions.org

Post reply on HN
Print view View original

171–179 of 179 posts

Re: Welcome to Wikifunctions

#171

Post by vrandecic »

yosito wrote: post #34

I couldn't find a function for which the implementation is visible. Which leaves me quite confused about what exactly the nature of this wiki is.

Sine the wiki hasn't been opened for editing yet, there is indeed only one function that has visible implementations (which was added for testing purposed): https://www.wikifunctions.org/view/en/Z10000 - all others are built-in and then invisible.

The two implementations are in Python and JavaScript (and rather trivial): https://www.wikifunctions.org/view/en/Z10004 https://www.wikifunctions.org/view/en/Z10005

Re: Welcome to Wikifunctions

#172

Post by tasogare »

brimwats wrote: post #164
lanstin wrote: post #149

Earlier quoted context omitted.

Fascinating. I had no idea people are still working on this sweetly optimistic project, like the architects writing a catalog of data taxonomies but for the unfenced territory of humanities knowledge. Sort of a modern Principia Mathematica with the Gödel of GPT4 waiting to blow it away. Still 42 for Douglas Adams manually assigned token is a nice touch.

much of the AI projects would not really be possible without linked data, so it seems a bit silly to call it optimistic. the two are connected

[dead]

Re: Welcome to Wikifunctions

#173

Post by lanstin »

vrandecic wrote: post #166
lanstin wrote: post #149

Earlier quoted context omitted.

Fascinating. I had no idea people are still working on this sweetly optimistic project, like the architects writing a catalog of data taxonomies but for the unfenced territory of humanities knowledge. Sort of a modern Principia Mathematica with the Gödel of GPT4 waiting to blow it away. Still 42 for Douglas Adams manually assigned token is a nice touch.

I have a half hour YouTube video on that topic, maybe you'll find it interesting, but likely too long: https://www.youtube.com/watch?v=WqYBx2gB6vA

It was interesting. I don’t think cheap matters much in a world:where compute is still getting rapidly cheaper. Being certain is good, but the throwmtons of data i to massive models keeps beating everything else. I do think the models need to get some self doubt built in some how but I don’t think the data catalog will be more than post NOC fact checking. Fun times tho.

Re: Welcome to Wikifunctions

#174

Post by ckoerner »

Levitz wrote: post #123
abathologist wrote: post #5

Earlier quoted context omitted.

That page seems totally reasonable and admirable to me. It’s literally just advocating for including the wide and varied history of algorithm development, in order to encourage readers from many different backgrounds to find ways to connect with the important topic.

>It’s literally just advocating for including the wide and varied history of algorithm development, in order to encourage readers from many different backgrounds to find ways to connect with the important topic. I'm at a loss of words for what I feel towards respecting (even promoting!) the idea that a piece of knowledge ought to become interesting or valuable if there is ethnic relationship to it. It reeks of suprem…

Representation matters. Many people are not represented in the worlds knowledge, much less the creation of it.

Re: Welcome to Wikifunctions

#175

Post by beepbooptheory »

Levitz wrote: post #155
beepbooptheory wrote: post #145

Earlier quoted context omitted.

[flagged]

Feigning ignorance? I am explicit in pointing how minorities tend to get a special deal when it comes to supremacist talk. I just don't condone it. I don't believe that glorification based on race is something you can keep to certain races. I am simply not racist about racism.

Gotcha, well good luck with that I guess.

Re: Welcome to Wikifunctions

#176

Post by vrandecic »

vrandecic wrote: post #168
andersa wrote: post #44

This has to be one of the most confusing documentation sites I have ever seen. Maybe I just don't get it, since I never edited Wikipedia. Opening some of the pages at random, like https://www.wikifunctions.org/view/en/Z828 https://www.wikifunctions.org/view/en/Z110 There's a "try this function" section, which I couldn't get to take any sort of input on a few tries. It always ended up with a field that said "no result…

Thanks, that's really helpful to read! The page is still in a very early stage. You are right, the current object section is just confusing, we will remove it. Maybe a better start to understand what Wikifunctions does - but it is a very simple example too - is this function: https://www.wikifunctions.org/wiki/Z10000 The other functions that you found randomly are currently not well supported. We are working on it, t…

As planned, with our first scheduled weekly deployment, the JSON object is not being displayed anymore.

Re: Welcome to Wikifunctions

#177

Post by olsonjeffery »

lanstin wrote: post #148
olsonjeffery wrote: post #85

Earlier quoted context omitted.

https://erlang.org/pipermail/erlang-questions/2011-May/05876...

I remember being intrigued by this idea. I doubt it would work but it’s interesting to understand why. Seems the main hold up is that each function might have to point to other function as dependencies, so it’s not going to be linear but really a graph. And then you are immediately in the position of I like this code but I am user Log package N, but they use Log package M so I have to copy and edit. . Vs _ and how th…

it’s eminently workable. it’s every “modern” package system, but at the finest absolute level of granularity.

Re: Welcome to Wikifunctions

#178

Post by lanstin »

lanstin wrote: post #173
vrandecic wrote: post #166

Earlier quoted context omitted.

I have a half hour YouTube video on that topic, maybe you'll find it interesting, but likely too long: https://www.youtube.com/watch?v=WqYBx2gB6vA

It was interesting. I don’t think cheap matters much in a world:where compute is still getting rapidly cheaper. Being certain is good, but the throwmtons of data i to massive models keeps beating everything else. I do think the models need to get some self doubt built in some how but I don’t think the data catalog will be more than post NOC fact checking. Fun times tho.

Edited: It was interesting. I don’t think cheap matters much in a world where compute is still getting rapidly cheaper. Being certain is good, but the throw tons of data into massive models keeps beating everything else. I do think the models need to get some self doubt built in some how but I don’t think the data catalog will be more than post hoc fact checking. Fun times tho. (My mom was an editor, she'd be horrified.)

Re: Welcome to Wikifunctions

#179

Post by lanstin »

olsonjeffery wrote: post #177
lanstin wrote: post #148

Earlier quoted context omitted.

I remember being intrigued by this idea. I doubt it would work but it’s interesting to understand why. Seems the main hold up is that each function might have to point to other function as dependencies, so it’s not going to be linear but really a graph. And then you are immediately in the position of I like this code but I am user Log package N, but they use Log package M so I have to copy and edit. . Vs _ and how th…

it’s eminently workable. it’s every “modern” package system, but at the finest absolute level of granularity.

Wouldn't you end up with 50 "get an URL" functions for different types of concurrency frameworks, different retry strategies, different SSL implementations, etc.?
Post reply on HN