You are viewing a single comment's thread from:

RE: Core development proposal year 7

in #corelast month

Hi ! Thanks for your long comment, I appreciate you taking the time :)

I'll reply point by point:

There is zero reason to set your profile to private.

I've set the profile to private because I am currently working on security issues and want to hide my work. It's been public since day one and will be back so once I'm done. But also if you are using the activity tracker as a way to measure work I would recommend against it. it's a flawed metric that is both easily gameable and a bad representation of activity. For instance you can work for two weeks and commit 200 times, this will count as one activity unlike github where it would color the whole week. Plus for the longest time I didn't realize people cared for it so I didn't bother with it. for instance I never set my commit email for for the gitlab one so all my commits wouldn't' be counted in the activity tracker. Therefore I don't want people to get the wrong idea from it without the context.

Here's what it looks like:

image.png

You can easily see when I started catering to it vs just working and not minding it.

You're asking for over $100k USD a year for this proposal, but you haven't cleared up whether you're working 8 hours a day, 40 hour weeks for Hive.
I highly suspect that you have a job already and Hive is a nice little side income for you, which is fine. But if you're going to ask for a full-time salary at a time when Hive is at an all-time low, that's poor taste in my opinion.

If I did have another job I wouldn't be asking for 100k, I am painfully aware that it is a big ask especially these days. It would be much easier for me to get a proposal going with a lesser amount. But this is my only salary and this is a full time job, 8h a day 40h a week. I mentioned it in another comment. The price for this proposal is driven by my cost of living and I am already more than below market rate.

You contradicted yourself. You said it's misguided to focus pressure on spending, but then you point out the economics of software development thanks to AI are changing, meaning the cost of developing software is crashing and not as expense or exclusive as it once one. I don't think it's misguided, I think it's sensible we're focusing on spending, the Hive token price is collapsing.

Not really, I'm saying it's better to focus on making more rather than spending less time (aka money).
Realistically AI allows us to go so much faster but it shifts the weight from writing code to the actual management and product which is other jobs core devs do, it's not just code. And that still takes time. Think ideating, thinking of design specs, reviewing the work (which became a big burden given the amount that is pumped out), testing on actual data (some of the hivemind tests take 30m and requires a human to check if the results actually make any sense).

Basically what I'm saying is that sure AI let's me go much faster but there is still a lot of work that requires me being behind the computer and working. And let's say I go part time and then work 10h per week, I would go fast probably faster than pre-AI but that's still a whole lot of time where I'm not behind a computer building things resulting in less output than what we could have had.

Content focus, not ambitious enough, could do in one month

I'm kind of summarizing the rest of your points, on the content focus, I would disagree. Content and being a decentralized social media, one of the few that still exists has value, I just think we failed to capture it for now but it's not over. We've all seen the success farcaster had and the mass exodus twitter (now X) created. People hunger for better social media, it's just that ours, quite frankly, sucks. Because we are doing blogging when this is an outdated way of displaying things and in the era where everyone is doing incredibly fine tuned algorithm feeds (see tiktok) we have a super basic feed.

I think our other fast/feeless avenues are strong too but they are being leveraged by other initiatives (magi and lots of game dapps). And lots of other chain offer this too now and much better, so I'm making sure our differentiating factor isn't going away. And also a lot of our user base still relies on this social layer.

Also that's not saying my only focus is on the social layer, it's more that it's where I'm focusing my efforts in the short term.

On the whole "ah claude could easily do this in one month" yeah, once the feature is set and done, it's very easy to replicate. This is because you're not seeing the discussions with the rest of the core team, the tests, the failed designs etc. it's like saying "making a pokemon clone can be done easily" yeah because you're skipping all the game design steps. That's the hard part, not the coding (not anymore anyways).

On the proposal not being ambitious enough and wanting bigger features, that's fair enough honestly. The tricky thing is those big features don't happen in a vacuum, we work as a team and they require a ton of discussion. I can say "I'm gonna do light accounts !" but that's a big multi-project initiative that probably needs a hard fork and a bunch of moving parts, and I really don't want to promise something that might not see the light of day because we later decide not to implement it. But yeah you're right, I'll edit the proposal to address your comments.