Gun: Is gun dead?

[obv-mikhail]

There seem to have been very few updates on gun in the past two years. I thought there were full time developers paid to work on gun? Is that not the case? https://techcrunch.com/2018/05/23/gun-raises-more-than-1-5m-for-its-decentralized-database-system/

[obv-mikhail]

@Dletta GunJS is a company and Mark has employees as visible on LinkedIn. He also talked about hiring software engineers in a YouTube video.

[amark]

@obv-mikhail huh? You haven't been around to see...the past 2 years we've gone from:

• tens of thousands of ops/sec to tens of millions of ops/sec.

• 1.5 rewrites from scratch.

• prototype projects only to production scale with millions of users, top 300 globally ranked sites running us.

Closing.

[spiffytech]

Which top-300 sites are using Gun?

[obv-mikhail]

@amark Gun looks very impressive - you have done an amazing job with it. But, you haven't answered the main question I had - are there paid developers working on Gun (or are there plans to hire any)? Or is this a one person project right now?

[amark]

@spiffytech Internet Archive https://news.ycombinator.com/item?id=17685682

@obv-mikhail why does it matter? This is OSS not greed.

If it is that important, it is easy to google our VC funding and see who is on the team.

[obv-mikhail]

@amark I want to choose between FeathersJS and Gun for a non P2P project. Gun overall fits my needs better but I can't find any proper examples of how events sent to Gun can be parsed for purposes of rejecting / accepting them based on some token. There have been issues opened here regarding authorization since two years ago, but there still exist no good examples (that I can find) of how non P2P authorization could be implemented.

If Gun doesn't have an active community and isn't actively developed, it would be easier for me to go with Feathers. So, I wanted to find out whether Gun is still active or not.

[amark]

@obv-mikhail Our community is about as big as FeathersJS.

What gives you the impression GUN is dead?

https://github.com/zrrrzzt/bullet-catcher

[obv-mikhail]

@amark I got the impression because there are no other recent contributors to Gun besides for you.

I have seen bullet catcher and it is unhelpful. I understand how to intercept messages. I don't understand how I would extract useful information from them.

For example, I don't understand how I would extract the path ("messages") for a set from a "msg":

gun.get("messages").set({});

Also, I don't understand how I would look into existing data in Gun before rejecting / accepting a message.

[biocrypto730]

@obv-mikhail

You know there's a theory backed by some pretty reputable historians that the world isn't pushed forward by teams of scientists, but by one or two individuals in each generation with the stubbornness / innovative minds to see what other people can't. A salaried employee working on a team is prone to tremendous groupthink and is chained by promises to investors. I would go as far as to say that VC funding only hurts innovation because oftentimes it bogs you down into previous presentations you've already given, instead of looking at other places in tech for inspiration. It's a pretty well known fact in the Valley that working by yourself allows you to innovate and change directions much more quickly than in a team.

You should know that this is one of those "Chaos" patterns that exists at every element of the universe. As a guy with a bio background myself this theory is called punctuated evolution in biology. These are periods of rapid change where one group of species undergo rapid changes compared to the rest of its history, often caused by one new gene enabling a population burst. I'm sure you would also find this "singular influencing stimulus" pattern in the areas of fluid dynamics, politics, star constructions, and just about every other studied topic out there. You might also know names like Albert Einstein, Watson and Crick, and Steve Jobs better than other scientists and researchers.

In short you could not be more wrong about this library. It solves a problem that a lot of people won't appreciate this year but likely will sometime the next decade. Gun does what it's supposed to do; Its DB structure and philosophy aren't for a traditionally monetized project, and it's a radical change from what the web looks like now compared to what it's going to look like.

So if you only know how to design secure systems using a server-generated token or something you were taught in school or at another job, you'll find yourself bending backwards to use this library (although bullet-catcher makes it really easy). If you know what proof-of-work or cryptocurrency is, or been in those spaces as long as I have, combing through this library is basically like watching porn.