Fork me on GitHub
#off-topic
<
2022-01-09
>
vemv05:01:48

> Carreras had the equivalent of $7 million in ether before taxes. https://archive.fo/wHCIe artist Anna Carreras sold 1000 pieces of generative art which was made via Processing guess a motivated hacker could try their luck with Quil, the Clojure wrapper 😄

Aron11:01:05

I am curious how people guarantee or insure that the NFTs will not be changed deleted or otherwise corrupted

Ben Sless15:01:21

Really depends on the nft. When you think about it, the world is full of nfts without block chains - phone numbers, email addresses, domain names. How do they not get changed or deleted?

lilactown16:01:42

https://moxie.org/2022/01/07/web3-first-impressions.html where he described an NFT he made that changed what image it showed depending on your referrer. it got removed from OpenSea

👏 1
🔔 1
Ben Sless16:01:01

"People don’t want to run their own servers, and never will" At the risk of bringing up an inflammatory related topic from the past, some people tried to solve this, but they NFT aspect was the least contentious part

walterl16:01:24

I was just about to point to m0xie's blog post point out just how mutable an NFT can be. What it implies is that an NFT doesn't represent ownership of an actual work, but the URL pointing to it. It's quite bizarre; I just assumed that the NFT itself (on the blockchain) would include at least something like a hash of the work it represents.

Ben Sless17:01:37

An nft is a deed of ownership It's up to the two parties to decide what the deed stands for. It's a reftype 🙂

Ben Sless17:01:34

You can agree that you get the same thing every time you "deref" it, but it doesn't have to be

👍 2
lilactown17:01:38

yeah. and I understand why you can't put an image on the blockchain, for the same reason you can't (and wouldn't want to) put your house in a bank.

Ben Sless18:01:27

There's something for the sentimental value or nominal copyright holder, though. Just like you can display a painting non a gallery which is publicly accessible even though the painting is privately owned, why not an image? (Putting aside technical issues like copying)

lilactown18:01:29

I don't get any personal value from it but I understand that other people do

Ben Sless18:01:47

Same, hope they have fun :)

Aron19:01:09

> phone numbers, email addresses, domain names. How do they not get changed or deleted? if only someone had a talk mentioning exactly this topic... https://github.com/matthiasn/talk-transcripts/blob/master/Hickey_Rich/ValueOfValues.md

Ben Sless19:01:06

Rich and Clojure shaped my view on this subject :)

Aron19:01:20

Then why ask such a misleading rhetorical question?

Ben Sless19:01:59

I don't think that's a misleading question, I think it's leading in another direction. NFTs as immutable values in a global namespace which may be references to something else slices the problem along a different axis

Aron19:01:29

you lost me somewhere, maybe I have fallen between the axes

Ben Sless19:01:57

Maybe I'm less coherent after a week of covid 😅

Aron19:01:19

get well soon 🙂

❤️ 1
seancorfield22:01:22

Since the NFT is the URL, not the thing the URL points to, you can't guarantee anything about what it points to, just the URL itself doesn't change (and is presumably unique?). You can't even guarantee the URL points at anything at all. NFTs seem like a really silly idea to me.

☝️ 3
naomarik23:01:53

I would have used a more derogatory sentiment to describe NFTs. I'm just glad I don't have anyone in my immediate circles that are into this.

💯 1
Ben Sless10:01:53

There are probably several users in this slack who now feel attacked or discouraged.

Aron18:01:19

Can I have a poll please? 🙂 Questions: 1. Making trades that are favorable for us and unfavorable for the person who we are dealing with is just good market sense, agree or disagree? 2. As software tradespeople it's our job to identify productive niches where our skills are best compensated, agree or disagree? 3. Ultimately what matters is that everything has to be legal and done with full consent, and it's the participant's responsibility to deliberately make rational choices when it comes to giving that consent, agree or disagree? https://i.imgur.com/Dg3Tl8X.png

Ben Sless18:01:38

1. disagree. even from a very rationalist perspective, this is a repeated game. you want to not just make a good trade, but for the reciprocating party for have a similar feeling. Otherwise you invite conflict and discord 2. yes, then we rank them according to how we feel about them and see what we're willing to put up with 3. it depends. It's a good lowest common denominator, a good start, but we should aspire for more

Aron10:01:13

seems like there are indeed some people who are "discouraged", but maybe not on the side you were talking about 🙂 ftr. I disagree with all three points. I even if people are more than willing to be mislead, I think it's not ok, even if this means that I have to take a heterodox stance.

Nom Nom Mousse14:01:48

This thread seems to equate properties of NFTs on Ethereum with properties of NFTs in general. Cardano might be the most interesting blockchain for functional programmers.

Aron14:01:50

Can you be a bit more explicit, please?

Nom Nom Mousse15:01:55

I do not know much about it except that Ethereum does not have native support for NFTs while others like Cardano does. I think that means that NFTs on Cardano are not merely pointers from a contract, but rather something you own. But I need to read up on this. Interesting field.

Ben Sless15:01:43

Also, any kind of token with use on its own is useful in this context. Attendance tokens or convention tickets can also be NFTs

Nom Nom Mousse15:01:44

That is what I'm starting to wrap my head around. Just like a webpage can be anything (not just the crummy things with rotating skulls playing midi from ~25 years ago), NFTs can be anything. They can be code snippets, only that they have the useful properties of ownership and scarcity.

Aron15:01:06

a webpage cannot be 'anything'

Aron15:01:47

anything is mostly things that can't be webpages, funnily enough. An average thing from the things that make up 'anything' is anything but a webpage.

Ben Sless15:01:54

What purpose is this serving?

Aron15:01:31

> I am curious how people guarantee or insure that the NFTs will not be changed deleted or otherwise corrupted

lilactown19:01:03

AFAICT the success and failure of any crypto project atm seems to be how well it works to financialize the thing it represents. the novelty of NFTs isn't actually that it is a durable db that represents ownership - that is easy to do using non-blockchain tech - it's all the financial machinery built around crypto that allows people to trade them, speculate, loan them, etc. there's reasonable questions to ask about how well NFTs work to do this for certain use cases. whether it does or not, the next question is whether treating all these things as financial apparatus is good, actually. IMHO it is not but there are plenty of people who disagree.

Aron03:01:31

There is a lot of disagreement here for sure, mostly because it's so rare that people actually spend time learning how crypto or distributed systems function. As in, how would you write your own distributed ledger from scratch, especially if you don't want to do the easy thing like bitcoin. But NFTs is another level, I just found out yesterday that in the English language there exist a saying that fits very nicely: https://idioms.thefreedictionary.com/sell+a+bill+of+goods

Ben Sless19:01:14

Anyone here has any experience with regression testing services which receive inputs over http and write them to Kafka?

Jon Boone20:01:57

I don't have any experience with that using Kafka specifically, but have with AWS SQS, for example. Are you looking for a review of your plan?

Jon Boone20:01:58

None at all with Clojure component as the mitm…

Ben Sless20:01:21

less a review of plans and more a suggestion for a game plan, approach, experience reports

Jon Boone20:01:08

What is your primary goal in the testing?

Jon Boone20:01:08

Aside from identifying regressions..

Ben Sless20:01:48

Identifying regression, doing a big refactor

Ben Sless20:01:07

In the future running it in CI

Jon Boone20:01:46

What ability to do you have to obtain/create the input data (HTTP req/res)?

Jon Boone20:01:38

Keep in mind that data privacy concerns must also factor into this analysis….

Ben Sless20:01:22

I can capture all incoming traffic no problem

Ben Sless20:01:30

No privacy issues

Jon Boone20:01:45

Ok — and can you easily obtain the resulting Kafka entries?

Jon Boone21:01:21

My recommendation would be:

Jon Boone21:01:29

1. Identify use cases of concern

Ben Sless21:01:35

I can even run this entirely locally, it's mostly annoying

Jon Boone21:01:27

2. For each use case, collect input HTTP req/resps, Kafka request/responses (if desired) and Kafka messages

Jon Boone21:01:48

Then in your testing, feed the HTTP reqs into your code, validate you get expected responses, send expected Kafka requests with appropriate body containing the messages for that use case.

Jon Boone21:01:59

Essentially, integration testing with canned inputs/outputs….

lilactown23:01:50

I'm looking for a scheme impl that runs on an m1 Mac. anyone tried this yet? gnu/mit scheme doesn't officially support the platform but https://kennethfriedman.org/thoughts/2021/mit-scheme-on-apple-silicon/ to modify it to work. I'm looking for something that officially supports it

lilactown23:01:59

hmm looks like gambit scheme might work out of the box

lilactown23:01:46

chicken installed easily from brew. I'll try that for now

Sam Ritchie02:01:14

Yes, that works!

Sam Ritchie02:01:22

The MIT Scheme guide

Sam Ritchie02:01:43

And I believe that fix is in the master branch now

Sam Ritchie02:01:55

So the latest MIT Scheme will work for you @U4YGF4NGM

lilactown03:01:04

I ran into quite a few hiccups w/ chicken scheme get a dev env setup

andrioni15:01:16

Racket also works natively, but it might not be as Scheme as you want/need to

Dimitar Uzunov15:01:23

hmm there is kava <ftp://ftp.gnu.org/pub/gnu/kawa/kawa-3.1.1.tar.gz>

Dimitar Uzunov15:01:31

in lib there is a jar which you can run

Dimitar Uzunov15:01:54

I’m not a scheme user but it should be pretty portable

Dimitar Uzunov15:01:13

@U067TPTUL has made a cool kawa clojure interop demo about a year ago: https://asciinema.org/a/388173