Web Registry DAO / Wika Network Grant Request #682
Conversation
|
|
randombishop
changed the title
|
Hi Nabil, thank you for the application. I appreciate the level of detail, and I think your project has merit. The first three milestones are pretty straightforward. I would just ask you to provide a bit more detail about what exactly you are delivering in milestone 1. They are mostly services, so I assume they will be deployed somewhere, but what can I expect as deliverable (config file, deployment script, ...)? I would also expect some Substrate-related deliverable there. M4 is mainly code review, which we don't usually fund. It is also rather vague ("Add necessary pallets"). M5, as it is, should really be covered by the previous milestones. What I would suggest is that you apply for M1-M3 now, and then, once you had someone review the Substrate part of the project and have a slightly better understanding of what's required for the governance part, apply for a follow-up grant. Would that work for you? |
semuelle
added
the
changes requested
|
Thank you very much Sebastian for the fast and excellent feedback!
Very looking forward to hearing back from you and will update the PR as soon as I receive your directions. Thanks again! |
|
Thanks for the quick reply, Nabil.
If you look at the FAQ I linked above, you will see that we don't usually fund activities around deployment or hosting. The main reason being that a single, centralised deployment of a service is not of much value to the ecosystem if it's not easily reproducible, as it could be shut down the day after approval. However, I'd say that, if the documentation of these is thorough enough that it is reproducible, it is acceptable as a deliverable.
Yes, please add details to the specification.
That makes sense, and we get that often. We are just hesitant to fund activities that are impossible to scope, which is why postponing this to a follow-up grant makes the most sense here.
That is absolutely fine for a grant. |
* Grant Request v1.1 * Update wika_network.md * Update wika_network.md
|
Fantastic, thanks for the guidance Sebastian. |
|
Thanks for the update. I will share it with the rest of the team. Please note that most of the team is out of office this week, so it'll probably take a week before they have a chance to read it. |
semuelle
added
ready for review
|
Great, thank you @semuelle ! |
|
|
||
| Because there are too many Like buttons out there! | ||
|
|
||
| Each one generates data in private databases that are often used to optimize for advertisment or hidden agendas. |
There was a problem hiding this comment.
The issue I see here is that, instead of allowing one company to process my likes, I now allow everyone to process them. That's not exactly an improvement. And wouldn't I have to give a website that integrates your like button access to the accounts in my wallet? That would allow them to link any other data they have of me with all the likes in your registry.
There was a problem hiding this comment.
Very good point!
When a website integrates the plugin, it doesn't solve the data privacy problem as you point out, it even gives them access to more data, but one could argue that it's still a step towards progress.
Why? We think that the direct integration in websites is an intermediate step towards web3 to gain more mass adoption.
First of all, in that setting, the user doesn't have to be logged in anymore, or can use an anonymous account.
Secondly, we expect the widget to be first integrated by actors who actually want to promote web3 rather than very powerful web2 companies, and by doing so, will give us a user base ready to upgrade to the browser extension version, or even polkadot wallets that provide wika functionality in a seamless and protected fashion.
The ideal setting is when the browser is equipped with an extension that hides the actual address from the websites and only injects the widget and the like functionality. In that case the website will only be able to record the like action without an actual handle on the address.
My point is that by providing a web3 compliant infrastructure for registering web pages and liking, the world should progressively evolve into using it in the ideal way.
Somehow, this question applies to web3 in general: as I've been connecting my Metamask and Polkadot extension to so many websites already, it's true that I didn't share my personal credentials, but I also shared much more data. I also added my email address to opensea to receive offers, so now not only opensea has my personal info, but it also has all my wallet activity all the way to Uniswap and Aave. Not ideal from a data privacy standpoint at all, but i still believe that using opensea is a positive step towards web3 evolution.
This is actually the first time I think about the problem the way you pose it and this is just my intuitive immediate reaction. I will share this with the rest of the team and discuss further about what other ways we could come up to address it. And also please help me with any useful links on the subject or ways you think the infrastructure could be better designed. We are definitely here to learn!
| | 1. | list_url_by_liker(address) | List the URLs liked by input address. | | ||
| | 2. | list_url_by_owner(address) | List the URLs owned by input address. | | ||
| | 3. | search(query) | Search URLs matching the query using domain, title and meta data | | ||
| | 4. | recommend(address) | Generate a list of recommendations for a user | |
There was a problem hiding this comment.
How do you plan on generating recommendations? Do you store any other data than addresses and liked URLs?
There was a problem hiding this comment.
The chain only stores addresses and liked URLs, but as specified in the ETL, Elastic Search will store one document per URL with title and other meta information pulled from the web page.
We plan on combining collaborative filtering (who liked same pages and what other pages they liked) combined with content based ML (with the data pulled by the ETL) to build a first version of the recommended function.
Please note that we really plan to make it the simplest thing that could possibly work as a proof of concept. Performance optimizations and scalability concerns will not be part of this initial scope.
semuelle
left a comment
semuelle
left a comment
There was a problem hiding this comment.
Thanks for the replies. Happy to proceed.
Noc2
left a comment
Noc2
left a comment
There was a problem hiding this comment.
Thanks for the application. I have a few questions/comments: Regarding the indexer, how do you make sure it’s decentralized or how do you connect the off-chain world with the on-chain world? Also did you take a look at other indexing solutions in our ecosystem: https://github.com/w3f/Grants-Program/blob/master/docs/polkadot_stack.md#wrench-tools-apis-and-languages Furthermore, could you potentially remove the part about Sora and the bridge from the image. We generally don’t want to give the impression that we fund specific deployments or specific projects in our ecosystem. We only care about the development of open source software that is useful for our ecosystem. Finally, have you considered developing smart contracts instead of pallets? This way you could for example potentially deploy on a parachain directly without the need for the bridge.
* Remove reference to DEX
Thanks for the excellent questions @Noc2 and looking forward to hearing back from you! |
Noc2
left a comment
Noc2
left a comment
There was a problem hiding this comment.
Thanks for the update. I’m willing to go ahead with the application.
mmagician
left a comment
mmagician
left a comment
There was a problem hiding this comment.
I'm curious to see Elasticsearch as the system to query your database acquired via ETL! I've worked with it in the past and have been quite happy with the usability. Also, for an ETL solution you might want to check you https://github.com/p2p-org/polkadot-profit-transformer
who are just finishing their grant - their solution (while complex) is fairly complete, as far as I can tell.
|
Congratulations and welcome to the Web3 Foundation Grants Program! Please refer to our Milestone Delivery repository for instructions on how to submit milestones and invoices, our FAQ for frequently asked questions and the support section of our README for more ways to find answers to your questions. |
|
Fantastic news, thank you W3F team, we sincerely appreciated your transparency and the quality of your feedback! |
|
Hi @semuelle @mmagician @Noc2!
I know this timeline is drifting from the initial grant request plan, but we had other workloads coming in at the same time, plus holidays season. Please let me know if you have any questions or concerns, and thank you very much for your participation! |
|
Thanks for the update, @randombishop. We are glad to hear things are progressing. Don't worry about small delays; as long as we know that the project is active, we have no reason to be concerned. |
|
Hey W3F team! |
|
Hello hello W3F team! |
|
That’s great. Thanks for pinging us here. It might take a little bit longer at the moment, since we currently have a lot of outstanding evaluations and part of the team is sick/ooo. |
|
No worries, get better! |
|
Hey team, Thanks again for giving us the opportunity to work together and the smoothness of the process so far with milestones 1 and 2. Posting the project update here for general communication:
Cheers! |
|
@randombishop likewise, thanks for your deliveries, the good words and the update. I'm happy to hear that you're satisfied with our program.
That's not a problem, looking forward to M3! 👍
Maybe we can ask our grants PR team. I'll send them an email and cc you:) |
|
Checking in @randombishop, are you still planning to deliver milestone 3? If so, could you please submit an amendment to update the timeline? |
hey @alxs ! Yes, though we didn't make enough progress on it lately, we're still planning to deliver it. @cwilker will make an amendment with a new timeline and take the lead for the last milestone. |
|
@redben @cwilker As discussed yesterday, noting here that you guys are committed to other obligations and won't have time to work on this grant in the near future. We will terminate this grant now; but feel free to re-submit an application for milestone 3 in the future if you'd like to re-activate this project. |
|
Just for reference, termination PR: #1568 (merged) |
8 participants
Project Abstract
Our mission is to build a public and non-intrusive blockchain registry to empower authors, search engines and recommendation systems that respect privacy and are 100% transparent.
https://www.wika.network
For which grant level are you applying?
Application Checklist
project_name.md) and updated.