Bad Contracts + High Fees= Major Renter Losses

[maol]

My couple cents...

The renter algorithm is not well understood in the public. It would help to have a Renters Handbook just like the Hosters Handbook.

From what I'm seeing, I assume that the devs have put a lot of thought in the hoster and renter price optimizing algos, but there's also some randomness built in (on purpose), leading to the situation that sometimes a very expensive option is chosen instead of always the most economical one.

The Sia network today is way too small to play fully to the design of the renter & hoster optimizers. That's why these high priced instances can have such a big impact in the network. Once it's past a certain threshold, it should just be fine. Until then, everybody using Sia is an early adopter, and has to accept that things can be more expensive than expected, IMHO. I have a feeling that not everybody playing with Sia is aware of how cutting edge the thing is that they're running with their own money! There's a reason I'm only hosting and not renting...

Having said all that, it's obviously possible that the renter price optimizer is buggy, or that the design is not optimal for the short and/or long term! Maybe the devs are even aware of problems in the code right now...

But if not, then the best way I see to figure this out is to perform a thorough review of the spec (if available), code and examples that you came across, and see a) if the design is adequate for the objectives, and b) if the implementation in code is correct.

Anybody up for this? It doesn't have to be the busy devs, the code is quite well readable and available on Github! Or maybe somebody's done that already (internally or externally to the devs) and we're just not aware of that work?

[moorsc0de]

@maol It doesnt take users looking at code to understand a problem in real world application. We are the real world application (hosters and renters).
This is what I rage against when I see it.. is people talking about we have to just TRUST the code or review it. No that isnot how things work in "real world".

Competent developers or technical writers can write exactly how the creation works. You mentioned this and it is missing- and I do not think its missing on purpose. I think
it is missing because the realm of crypto- "believers" do not TEST.. they just BELIEVE everything will be ok...when the people who have already done the testing can see this is miles away
from being even marginally acceptable from what was written years ago to today.

There is never ANYONE talking about an "autobuyer" in the interface pairing a renter and host together, but today that is what the system does. Oh, and by the way there is no manual/guide detailing this.?? Hmmm
That is not an accident. Its by design and on purpose. But even if we give your theory of "IT WORKS" a chance.. the renter contracts file proves that it does not work.

You can clearly see high priced hosts- and also Hosts with no available space being paid in the RENTER CONTRACTS FILE RIGHT NOW. No code needed. That cannot be an accident.
Unless, they never tested this at all- and are simply using the common everyday user to test this to make $$. This is exactly whats happening and there is no word about it. There are people wh run the UI that have no idea what
the renter commands are to even look at the file. There are renters who think they can end a contract whenever they want - you can but it requires forfeiting your allowance. All roads lead to a loss of coins in the scenario with no clarity.

That cannot be an accident that requires code.. its too many dead end roads.

[reinisp]

@moorsc0de
It's the reason why I repeat-
Sia isn't ready,
there are critical parts missing,
the current coin price does not reflect the current state and usability of the system/network.

Every current Sia user (renter and hoster) is taking part in testing.
I plan to use Sia commercially, but it isn't mature for that yet.

[moorsc0de]

@reinisp I have stated this numerous times... for disclaimers and such to be displayed so users are aware that the testing can be expensive if the newbie has no idea what they are getting into. I think this creates a negative setback for SIA in not seeing this. However when it is mentioned the "insiders" either deflect, ignore, or fight you in the slack. IMO the UI is not even ready for testing.. it misses crucial control for renter and host regarding contract visibility and controls.

[maol]

@moorsc0de I dig your frustration. We all want Sia to succeed, and these events leave the project a bad reputation. I totally agree that the Sia's website does not make it clear that people are risking their entire funds when playing with the technology. People may be mislead about the maturity of the stack. This is the area that I think we can and should improve.

Separately, while I agree that we see symptoms, I only trust the code. Maybe it's possible in the code that a renter contracts with a fully loaded host, but there are also other explanations possible (such as (i) it wasn't full when the contract was formed, or (ii) there are/were a few bytes available still, or more). We won't know just from looking at the symptoms. Unfortunately there are no detailed enough logs from what I know, these would be very helpful.

Finally, renting is fully automated and hidden from users on purpose. Renters do not need more control regarding contract visibility or selecting hosts based on financials. That's what the renter module is for. (However I do expect further settings around location and speed of access i.e. optimize very fast data access for small files over very cheap but slow access to large files. I don't think that possible today?)

My own personal conclusion of all that is that I'm slowly phasing out my hosting business for now (just letting existing contracts end gracefully over the next couple of months, I've already started reducing my maxduration), and will check back on the status in a year or two. I wish Sia all the best, but it's just not where I need it to be for my needs today.

[moorsc0de]

@maol Interesting I have already done what your doing...
shrugs

[moorsc0de]

@maol
@bryan @2xjO9M3P @maol fair warning- seems people such as myself who speak up are getting targeted by the insider group on sia. You may be next.

XURIOUS,
BEARDO
ABEC
SIANIDE
and even TAEK

What some of them are doing is purposely attacking those who present valid questions- so they can ban you and then go in and demote your posts to reduce your credibility. They have done it to me. I had a 10 reputation now it is -3.
They have also gone as far to censor me on reddit with posts they have never asked me to take down.. not 1 of the tema has ever contacted me directly about any of the issues either.
This is blatant. and misuse of the forum.. and a misguiding of the people..

Just passing on for your knowledge so you can pass it on to the people you know. This is no way to run a platform... all they have to do is address the issues... but they rather attack and hide.
If what posted did not have merit no one would read it, but seems there is an audience and when people seek to silence others who are attempting to help and deliver truth.

They have gone in to purposely down vote the credible things I type- and in cases where I am just voicing an opinion.. or repeating what was posted in slack to make my posts and profile lack credibility.
How is that for a Sia crew.. Sure I have said some things out of anger because I am targeted, but a group that takes personal attack on a person that is simply trying to make them better is obviously hiding something. Good luck all- this is just the beginning.

[beardo]

In the spirit of full transparency and to provide context, you can see the post that moorsc0de is referring to: https://forum.sia.tech/topic/1415/sia-coin-burn

[moorsc0de]

@beardo indeed and you can see the same people targeting my comments too... go figure.

[maol]

@moorsc0de @beardo thanks guys, interesting thread, wouldn't have come across it otherwise. Now if y'all could please remember to fight about the facts and leave out all personal attacks and remarks, that would really help keep the forum a friendly place and welcoming to newcomers.

We may disagree on what's the right way to do this or the right way to say that, but we should do that in a friendly and respectful manner, without attacking each other personally.

[bryan]

@moorsc0de I will not censor anyone's opinion what so ever. I'm a protector of opinions. I haven't been instructed to anything other than be a part of the community. In fact I'm only a moderator because the post delay was hindering my help replies.

As far as your rep, it's unfortunately subject to abuse by any one who might disagree with you. I have several disagreements with the project as well but I think it's more about how you present them.

[moorsc0de]

@bryan Thats greatly appreciated... have over 140 posts in the forum- helping- analysis- and sometimes being critical of things.
So people claiming I just throw FUD are not aware of the full circle of things I do.

I would like to keep convo centered on the platform because that is what brought us all here. But by me pointing out issues.. is my no means FUD (as many have said)- esp when everyone else can do and see the same. Just not quite sure "angle" on the personal attacks and being banned from slack, when it comes to some of these key areas of issue with the Sia contracts and renting transparency. If someone would just say - yes we are aware..and working on XYZ.. cool.. but thats not what you get..and for Sia to be launched with this many problems and proclaiming to be a competitor to Amazon and Dropbox.. its highly misleading (at this stage).

I am not so sure Nebulous is behind the miseducation, but they could mitigate a lot of the back and forth- and negative energy by just keeping their documentation updated - accurate- and orderly.
Utilizing applications with dated user manuals 3-4 yrs old that do not speak to the current operation of what we have today is a major problem and I dont see it helping them in the long term or helping investors. Which is sorely irresponsible. Someone has to sound the alarm.. (or is it too late)?