Hardfork Early, Hardfork Often

in #steemit7 years ago (edited)

arrow-1773970_1280.png


Following a few articles on Steemit related to the upcoming Hardfork 0.17, specifically about how early this hardfork is scheduled after the last one, I decided to share my views related to the frequency of the platform changes.

If you're a veteran on the Steemit platform, then you remember the last hardfork, 0.16, was in December, last year. It changed the rules of the game at a deep level, modifying significantly the economical model. The next hardfork, 0.17 is set to take place anytime now, but I suspect it will be in March, second half.

That will probably make the timeframe between the two frameworks around 3 months. I consider this to be a reasonable interval for such a change.

Here's why:

  • we are still in beta. Let's not forget that we're all just paid beta testers and the product is supposed to change while is in beta, following users' feedback.
  • the "release early, release often" paradigm, which fostered the development of the Linux platform and, subsequently, of the entire Open Source system, is a real proof that this building model works.
  • the bigger the audience in the platform, the more difficult the adoption of any new feature will be. So, let's try to wreck some stuff while the damage is still relatively low.

Caveats

Steemit is a much more complicated system than just an Open Source app, though, and that's because of the underlying economical system. We're not just a social media platform, we're an economy, with a real cryptocurrency, which has some value attached, which is transferred in exchanges, which carries interest if it's blocked in certain assets, etc.

People are much more careful with their money than they are with their social media statuses. If they can't upload images to a blog post they may feel frustrated for a while, but if they feel their financial rewards are at risk (no matter how entitled they really are to those rewards, that's a different discussion) they will freak out.

The endless debate around the reward pool, curation rewards, algorithms and so on is a direct result of this natural reaction towards money. Not all users are closely tied to money, some of them - mostly whales - have a more relaxed way of acting, but the majority is heavily influenced by the price of Steem and by the changes in the economic model.

So, while I still maintain my position about "hardfork early, hardfork often" I do think there are some caveats that should be taken seriously. Some of them are already implemented, to various degrees, but some of them are not. I truly think the Steemit founders should have a look at these, and provided they agree, provided they have the time and the tools, to implement them.

It will make the entire experience more valuable to everybody.

1. Create a unified channel for feedback

Right now the feedback is scattered around Steemit articles, like this one. I think that each hardfork should have a specific place where the feedback should be provided and some sort of a voting process. These votes should not be active, but consultative, I don't think that decisions should be taken by a voting democracy (history proved that most of the time the majority in the democracy have no idea about what it's good for it) but the voting process should provide a clear image about the expectations of the users.

2. Provide clear simulations for the financials

Like I said, the financial component of Steemit is probably the most sensitive one. In HF 0.16 @dan provided a few visuals about how the new redistribution of tokens will work and that helped. A lot. I do think there should be a place, or a communication procedure that should state what are the monetary implications of a measure, to the extent the founders are aware of. It will help everybody calibrate their expectations.

3. Give fast feedback to the proposals

I think it's time for the Steemit team to hire a customer support person. If I would be them, which I'm obviously not, I will probably start with an announcement on Steemit. I'm sure there are many people here which could embrace this job opportunity. The main tasks of this person would be: confirm that a proposal has been received, seen or processed by the responsible persons in the Steemit team, pass on important information to and from the Steemit team, and keep an active eye on the community, identifying trends and communicating them to the relevant persons in the team. I think the @steemit account now holds enough SP to allow for a decent compensation to a good person taking care of this position.

4. Create, invent, manifest, or otherwise make possible the apparition of a community manager in Steemit

One of the most obvious and most ignored mistakes when it comes to the problems of Steemit is that code will fix anything. Or, since Steemit is also a financial ecosystem, that a new redistribution algorithm will fix anything. This never happens. Never. Only people can change things. And if you don't take care of people, people won't take care of you, as an ecosystem. Steemit needs a person capable to create some solid social glue around the scattered interests here, to understand and manage trolls, to separate legitimate unrest from attempt to game the platform for bigger incentives and so on and so forth. I don't know how Steemit will hire, or otherwise incentivize a person to take on this job, but, after a year of being out there, dozens of thousands of user accounts created and a lot of money paid out, I think it's just about time to do it.

When all this will be in place, I think that we will all be eager to see the next hardfork, and not keep balancing all day long between "meh" and "this is outrageous".

image source - Pixabay


I'm a serial entrepreneur, blogger and ultrarunner. You can find me mainly on my blog at Dragos Roua where I write about productivity, business, relationships and running. Here on Steemit you may stay updated by following me @dragosroua.


Dragos Roua


You can also vote for me as a Steemit witness here:
https://steemit.com/~witnesses

Sort:  

You're welcome. I see that many Steemians have the tendency to forget that and treat the platform as set in stone and release ready.

In HF 0.16 @dan actually created some graphs showing the new inflation curve distribution and it was much easier to digest.

Great artical, I agree with the ability to earn money changes people's emotional investment, there is an economy involved in this experiment.

That's what makes the whole thing so different and difficult to manage.

It also makes this platform absolutely fascinating. We could try all sorts of social and economic experiments that would be impossible to do in the fleshy world.

Most significantly, point 4: People. Especially since the "wrapping" for the Steem blockchain is a social network. People create communities... and code simply follows the needs and wants of the community. This is not a "new" issue, of course, Working in IT 20 years ago, I was "Usability and Human Factors," which made my department "most hated" with the developers.

I used to work in Q&A at a software company and also took a lot of heat from devs.

Great post. I'm a big fan of deploy early, deploy often. Now is the time to test stuff out, break things, figure out what works and what doesn't. If we don't work through the systemic risks now, things will implode pretty badly in the future.

We will remember these days as the "good old days" when we had a hardfork every 3 months :))

Really great points, especially #1. I like to stay up to date on this and I do miss posts as they're scattered and I have limited time to be here. #3 kind of feeds into this too.

#2 is a no brainer, this is really important.

#4 is interesting and I'm not sure, but I think it could be a really good move. I've read plenty in the last days about Steemit being too directionless and that this makes mass adoption hard. I know it doesn't directly apply, but another maxim of programming is that a component should do one thing and do it well. There's some that think that, though I'm not so sure. I like the idea of different clients having different focuses (e.g. the instasteem idea which seems to be in closed production currently).

A lot of great posts have come out of these discussions, it's been interesting. I'd love to see something from @ned or @dan in response soon 🙃

Yeap, me too. That would be nice.

We want to scale the engineering organization first, and will add support and community resources as the market expands. In my opinion, we aren't there yet.

Thanks for the note, and your perspective here 😋 Actually far too often it's the engineering team that is lagging behind so perhaps a good move. 👍

Probably the issue is more about the current user base, but I guess the sub text is: patience!

Can i ask, as an engineer, do you see it as important that the userbase in fact does not grow too much too soon at this point? 🤔 I've heard some people throwing around that idea and it makes a certain amount of sense.

@sneak, this looks like exactly the type of approach that I specifically outlined as not working, which is "code will fix everything". I will just restate my opinion: code is important, but only if you have real people using it.

When you have 10k+ people posting every month on your platform, but you think this is not enough to add a customer support/community manager person to the team, then funny things may happen and funny things usually spiral very fast out of control.

Of course, I may be wrong and we can safely agree to disagree.

I didn't state or believe that "code will fix everything".

I said that in my opinion, it is too early to add support resources yet. We don't even have a mobile app yet.

There is a mobile app, I'm using it right now to post this comment. It's not made by Steemit Inc but it's widely used.

And you are paying -$7000/day to users, or -$210.000/month. That's a lot of money spent for a community "not big enough to justify support resources yet."

Anyway, like I said, we can safely agree to disagree.Its your business, I just own a tiny, tiny part of it...

That payment amount is not based in any way on the size of the community.

I didn't imply that. I was just saying you're already burning an incredible amount of money just to have (a certain number of) people posting. It would make sense to spend at least 2-3% of this just to see why, what and how those people are posting and to extract business intelligence out of it. Since you're already paying A LOT of money just to have them posting...

Great ideas! I hope these get implemented!

We shall see.

for how much longer we will be in beta ?

That's a very good question. Obviously, I have no idea.

oh well ....lets see...I am hanging in here :)

Coin Marketplace

STEEM 0.20
TRX 0.14
JST 0.030
BTC 67917.78
ETH 3516.77
USDT 1.00
SBD 2.72