Dear Blurt Community & Followers

in blurt •  last year  (edited)

blurt-witnesses-vibe.jpg


Dear Blurt Community & Followers,

I would like to apologize to you, I got caught up in the vibe and the hustle of Blurt witnesses since yesterday and I forgot that some of you who follow me were not familiar with the DPoS (Delegated Proof of Stake) system of Blurt and may be worried about recent events. This is a breach of my duties as a witness, I should have made a post much earlier to reassure you.

I would therefore like to reassure you, the Blurt blockchain by itself does not suffer much and the Blurt blockchain lives its life outside of all this turmoil.

This is for a simple reason, the witnesses! And I hope that at the end of this post you will have understood the importance that witnesses have in securing the blockchain as well as the importance of your vote and will go to bring your support in recognition to the one(s) in whom you believe and trust by a vote(s).

The blockchain is secured by the first 20 witnesses (the top 20), nothing can happen until a supermajority, 16 out of the top 20 witnesses + 1 backup witness (outside the top 20), is obtained.

So even if a person (as it is now) puts all the energy he wants to push the witness of his choice to the first position it does not change something, it just brings personal pride to have done it.

We are neither on STEEM nor on HIVE, the will of a person cannot control the Blurt blockchain because here, each additional vote for a witness decreases the value of the voting power he will receive.

If you vote 1 witness, he will receive 100% of the value of your BLURT POWER.
If you vote 2 witnesses, each one will receive 50% of the value of your BLURT POWER
...
If you vote 10 witnesses, each one will receive 10% of the value of your BLURT POWER

it's really a good protection and we can say thank you to @megadrive for the idea, @saboin to code it, and all the top 20 witnesses (including @fervi) to approve it.

Don't take everything you hear lately seriously and see it more as an aggressive marketing campaign. When in doubt, don't hesitate to ask someone you trust and read different points of view.

As a dev, knowing a little bit about the heart of this blockchain, I will only tell you this, when some people will have understood and assimilated that they can create their own interface directly connected to the blockchain to present its content in their image, making it popular if they know how then they will play with it and will let the community blow away from their agitation.

For my part, despite what some people like to think, I find the work done on the Blurt blockchain more than respectable, just thinking of the work involved in removing the SBD (STEEM Backed Dollars) from the core of the blockchain knowing all the changes it implies I am impressed (sorry I'm a dev then it mean something for me but probably not for you hahaha), not to mention all the tools for dApp developers that had to be adapted for Blurt

I think many of you don't realize how lucky we are to have, among others, @saboin as a core dev, I wish he would take time (which he doesn't have like any good dev) to share his work through posts so that you too can see it too.

I'm still far from knowing everyone and their skills, but part of the reason I'm here on Blurt is seeing their motivation which makes me think this blockchain is worth it.

As for the witnesses, what a mix, each one has its own character, its own personality, don't believe that everything is smooth and without interest, in any case, one thing is sure, they all want to go in the same direction and take their responsibility towards you seriously.

That's why if you could show them your support by voting for the one(s) you trust and believe in then it would be a very nice signal from you especially since many did not sleep much last night.

you can vote here: https://blurtwallet.com/~witnesses

And if you don't know any of them then I invite you to have a look at some profiles

@saboin - core dev
@randula - the organizer of the Blurt Power Up Day #bpud (are you ready for October 1st?)
@outofthematrix - for its neutrality
@world-travel-pro - idem
@techcoderx - like this, I wanted hahaha
@casualfriday - because he is 57th and I would like to see him higher in the ranking

Regards,
@nalexadre

Original background photo of this post by William White on Unsplash

Authors get paid when people like you upvote their post.
If you enjoyed what you read here, create your account today and start earning FREE BLURT!
Sort Order:  

I just gave you my first Witness Vote on Blurt. I've been here for many months, but I haven't voted for witnesses yet because I haven't been able to get a real picture and observe over a longer period of time how any of the witnesses behave during conflicts or questions. I am also still not sure on what other basis I should cast my vote, other than communication and negotiation skills, as I have no technical expertise equal to those of developers and techs.

I found your questions to fervi and how you engaged in the topic refreshing and diplomatic at the same time.

Thanks a lot for your nice comment and your support of my witness.

Yes, it's not really easy, you have those who bring value on visible actions, but also some who are less expressed and who work in the shadow but who, by their great competence bring a lot, and that non-technical expertise will have difficulty in seeing it, it's necessary to know where to look. For example, when I take a look at the posts of Saboin this does not reflect the huge role he plays every day in the Blurt infrastructure (both in terms of server administration and dev).

what are your thoughts on burning the inactive exchange accounts ?


Posted from https://blurtlatam.com

Even if I have BLURTS in it (Bittrex) I am for burning them because there is no guarantee that if one day Bittrex lists BLURT it will give me my share (although I trust Bittrex more than some of the other exchanges that have already proven in the past that they can't be trusted). I think that having them is more of a handicap for people potentially wanting to buy some BLURT and that it can be a dissuasive factor considering the amount of BLURT involved.

To have my complete thought it is necessary to go back further, personally, I think that the copy of the STEEM wallet was a mistake and that it would have been better to have a system of CLAIM WALLET with an expiration date, like that we would not have found ourselves with this dilemma 2 years later because of the huge amount of BLURT in ghost account.

And you what your thoughts on burning them?

account sovereignty is sacrosanct

the whole point of crypto is account sovereignty


Posted from https://blurtlatam.com

I can understand your point of view, I think our vision differs in that you consider the account of an exchange as any other user whereas I see it more as a part of the exchange itself on the blockchain, just a bridge to its own database where the exchange user accounts reside (on which it dictates its own rules centrally)

Besides, my dev part tells me that after 2 years and considering the work on their database needed to be able to distribute the BLURT to their real owner, I don't think they will ever do it.

After a conciliation can be to ask the exchanges themselves to send the BLURT of their account to the @null account if they have no intention one day to airdrop their BLURT to their owner.

also, burning those tokens will reduce the reward pool


Posted from https://blurtlatam.com

considering the work on their database needed to be able to distribute the BLURT to their real owner, I don't think they will ever do it.

agreed and since those tokens are out of circulation, i don't see any "crisis"

megadrive said he wants to burn the 75 million tokens and then recreate them on the ION network

this would increase AVAILABLE SUPPLY and potentially drive the token price lower

even if you don't give a shit about account sovereignty


Posted from https://blurtlatam.com

For my part, I would not say a crisis but rather a dissuasion weapon that can discourage people from investing/developing on Blurt, just because something is sleeping doesn't mean it can't wake up at any time.

About the ION network, I'm not for it but I have not yet studied the subject enough nor heard enough arguments to make my position definitive.

Beautiful words dev dude

Thanks :)

Thanks for the useful information. Now I have the opportunity to know who to vote for as a witness. I do not have much experience in the blockchain, so this kind of information is important and interesting for me. Good luck to you and all the best.


Posted from https://blurtlatam.com

Thank you for your nice comment :)

Technically as long as everything is controlled by the BF you have no choice but to either accept or be thrown out.


Posted from https://blurtlatam.com

I don't know how it came about that patience ran out on those involved, but don't you see any chance of resuming talks and after talking to nalexadre about your app idea? I would think that those who are against resuming talks will find reasons, those who are for it will find ways. Both sides.

With the last hardfork Blurt Core team is not in complete control of the witness ranking. There are witnesses who are in top 20 without the help of Blurt Core or Ctime & his friends and this is possible only because of the recent hardfork.

And yes, Blurt Core controls Blurt Official Discord and will remove you from it for writing the FUD post but they can't stop you from being a witness.

I think you need to calm down and think before you take any action. You are a good developer but you act in haste.

If you speak as a developer if you take the block_api methods for the direct reading of the blocks of the blockchain and the broadcast methods for writing I don't see too much where the BF is involved in it.

If you speak as a witness I find the top 20 very varied, it is not to know each witness individually to think that all witnesses are sheep of the BF, there are @outofthematrix, @world-travel-pro to name very few who take a neutral position, you who is against... take each name of the top 20 one by one and try to put it in a box, you will quickly end up with a lot of boxes hahaha

The last change in the voting system of the witnesses makes the entry ticket in the top 20 much more accessible than on STEEM or HIVE, for me, it is and will remain a courageous act of @megadrive giving up a power and a very good step towards a much more decentralized Blurt than its sisters.

Always love to read your posts. And thank you for mentioning me.

Thank you for your comment, glad that the BPUD is so close, a good moment of fun and good mood will do good after days like these

Just a couple of corrections:

It’s 16+1 of 20+1 for the majority to approve a hardfork. If 16 of the top 20 are voting for a hardfork and 1 witness below the top 20, then it’ll pass.

@tekraze and I don’t code hardforks since we don’t do C++. It’s usually @jga, and sometimes @baabeetaa. @baabeetaa was the one who coded Blurt in the beginning and he also did the first 3 hardforks. Then it was mostly @jga and some contributions from other devs as well. @baabeetaa was actually one of the three founders of Blurt.

Thanks for the corrections 👍 It's getting late here and I often forget the story of the backup witness hahaha.

For @tekraze I knew it, as I know he is a great help for developers (he helped me several times) but for you I was sure you were doing C++.

I learned the basics so that I can read the code, but C++ has a steep learning curve and I won’t be proficient in it any time soon.

I started learning Python in 2020. Then I learned JavaScript later on when I started working on Blurt. I also did a bit of Go. I wrote the image server in Go in an afternoon. It’s just something I threw together so Blurt would have an image server. I was planning on adding error handling to it later, but I never got around to it. Ironically that’s the only one that has never crashed once even though it doesn’t have any error handling.

I failed in C++ twice while doing my post graduation and now I know I'm not the only one finding it difficult. 😂

for my part, I can read and understand in part code in C++ if I have a box of paracetamol near me hahaha

I've heard a lot of good things about Go and it's crossed my path a few times, but I never got into it.

I wrote the image server in Go

Starting from the git repo of Steem Imagehoster that you have adapted or starting from zero?

What are the hardware specs of the Blurt image server? I would also be curious to know how much space the storage of the images takes?

It would be nice if one day we could have a decentralized image storage solution to beat STEEM and HIVE in this field, too many people still consider the images of the posts as without added value for the blockchain because not decentralized.

And once again I thank you for the work provided which contributes to give more value to the Blurt blockchain 👍

I would also be curious to know how much space the storage of the images takes?

As of a few seconds ago, the images add up to about 520 GB of space.

Thanks, far below what I expected :) Another one, I forgot to ask you hahaha

What is the maximum accepted weight you put for the images? Or maybe you do a resizing on the server side?

Just to add to my earlier comment: the main reason I switched back to the original Blurt image servers with the separate uploader/proxy/server is because the Steem/Hive Imagehoster stores local copies when it proxies remote images. The pro of doing that is that, if the remote image gets deleted, it will still show up on the blog since there is a local copy of it. The con of doing that is the enormous storage space used and the cost of storage.

I opted for the space/cost savings. Our proxy only proxies images; it doesn't save a local copy.

Yes Hive had to change its server because of this (168TB RAID drive now), but the advantage as a user is huge. When I first got on Steem I started by using the Steepshot Steem dApp (which stored the images itself) which has since closed, I lost all the images from my posts from that time hahaha

Ooh i had a video chat with the Steepshot team once

Starting from the git repo of Steem Imagehoster that you have adapted or starting from zero?

The Steem/Hive Imagehoster is an all-in-one server/uploader/proxy. On Blurt we use three separate programs for those functions. Blurt did use the Steem/Hive Imagehoster for a while, but when I came on board, I rewrote the uploader, tweaked the proxy, and built a server from scratch in Go that just fetches the images from our S3-compatible storage bucket and serves it to the client.

So to answer the question simply, I wrote the image server from scratch in Go in an afternoon and launched it so that there would be something. At the time, the Blurt infrastructure was in shambles because a lot of it was run by a dev who had left Blurt and turned off his servers. This meant that any image that was uploaded through the blurt.blog UI didn't show up on the blogs, image upload didn't work, there was a lot of stuff that didn't work. I should make a post someday to rehash this history and tell the story of how and why I got recruited to the core team.

It runs in a Docker container that is setup to restart if it crashes, so I wasn't too concerned about error handling since it would just restart automatically if something went wrong. And to my surprise, it didn't crash even once.

The proxy, on the other hand, is written in JavaScript, has error handling and crashes every few days. We don't notice it because it also runs in a Docker container that is setup to auto-restart when it crashes.

The image uploader code is here (I'm going to update the README.md file over the weekend).

The image proxy code is here (this one also needs an updated README.md).

The image server code is here (I should copy that to the Blurt repo and update README.md on this one, too).

Over the weekend, I plan on updating the REAMDE.md files for all of the components of the Blurt infrastructure with instructions on how to deploy. I plan on making some posts that will layout the infrastructure that we run as the Blurt core team.

Thanks a lot for this very interesting feedback 👍

For my part in one project for a customer I use imgproxy in a Docker container, and I am quite satisfied with it.

You seem to know docker really well! For my part, I have always had trouble getting into it and am far from being a specialist hahaha

I should make a post someday

I would be interested to read it ;)

Thanks for the links :)

  ·  last year (edited)

That image proxy looks very interesting. I've been wanting to rewrite some of our infrastructure in Go for the performance gains. Since all our infra is proxied through Cloudflare CDN, I'm not sure there would be that much difference by having more performant image proxy and server, but I'd like to test the JS version against the Go version to see how much difference it makes.

I'll bookmark that image proxy and look at it later in case I ever get around to rewriting the image proxy in Go.

I think @jacobgadikian was using that specific proxy before I came on the Blurt core team. It saves copies of remote images in the S3 bucket, so that would take up a lot of space.

When I copied the images from Jacob, a lot of the space taken was from the proxy saving remote images. I prefer just a plain proxy that doesn't save anything for the reasons I mentioned in my other comments.

I think Go is a good way to explore for performance gains 👍

What are the hardware specs of the Blurt image server? I would also be curious to know how much space the storage of the images takes?

That'd be interesting to know.

Is IPFS used on the image server as well?

Personally I would prefer Arweave for immutable or Filecoin if we need GDPR compliance to take down images.

Cc: @tekraze, @saboin

We use S3-compatible storage bucket, the same as what they use for Steem and Hive.

We had plans to switch over to uploading on IPFS as some point, but we never went through with it.

As @tekraze said in his comment, IPFS is slow and unreliable. We would have to run IPFS nodes as well, which would cost a lot more than using an S3-compatible storage bucket.

No, it's a storage bucket

A lot of blockchain tech uses IPFS, is there any reason for not doing so with Blurt?

IPFS is slow as well not reliable, it was used at start but then was removed.

I approve this message! Well written and well explained! Two Big Thumbs Up! Re-Blurted!

Thanks

Thanks for mentioning me brother!! 🙏🍀 100% upvote from me!!! ❤️🌞


Posted from https://blurtlatam.com