Topic: Slimcoin | First Proof of Burn currency | Decentralized Web - page 22. (Read 137085 times)

It would perhaps make sense if we planned and tested an "attack" scenario, playing around with the difficulty and trying to maximize the influence of one participant on the blockchain (in the sense of trying to get the maximum number of blocks in a row). Obviously, with only two or three testnet nodes, that would be highly artificial but would give the opportunity to test for the "worst case".

Yes, that's also what I observed - once the cap is reached, you have to restart the client to begin to stake again after the next PoW blocks, which is obviously unacceptable. IMO this points to that the staking thread isn't started again after being stopped. I can look at it again once I start the testnet client next time, but it would be the best option if someone skilled in C++ could look at what is wrong.

I suspect the bchain.info operator (Markus Tervooren) is hosting the Slimcoin blockchain for his own reasons. If any Slimcoin hodlers wish to show concrete support, there's an LTC donation addy. The chainz explorer is hosted on a straight commercial basis, the hosting fee is 8.90€ per month and, as of today, there are 4 days left on the meter.

It's just whitespace - P4Titan's IDE automatically reformatted every file he edited from 4-space tab indentation to 2-space tab indentation, removed all the spaces between if, for and while tokens and the following parenthesis, e.g. if (foo) to if(foo). To me, that smacks of the typical posturing dogma of "I'm a real programmer" that ironically betrays just the opposite. I've reverted his IDE's auto-applied changes to the codebase to bring it back into conformity to the original coding style/format (ditto for the QtGui code).

I realised it was a pointless exercise and turned it off - the network conditions for testnet do not in any way replicate the network conditions for mainnet and so it is impossible to draw any reliable, well-founded conclusions. As regards testing a different staking regime on mainnet, well that option is wide open to any node owner whether they be blackhat, whitehat or greyhat.

As far as I could tell, the hard cap client seemed to simply stop all staking permanently. Unfortunately I don't yet have a good understanding of staking behaviour, so I wasn't able to conclude anything useful. Instead, I returned to using a standard client (compiled from master) and it's been running happily for a couple of months now. I had to stop-and-restart it a couple of times when it froze up during the recent network instability but the restarted client synced up straight away and, now that the network has settled down somewhat, the client is running fairly economically (for Slimcoin, that is) using between 10-15% of its allocated thread. The last PoS minting was on 2019/10/19:  and the last burn minting was on the same date.

Cheers

Graham

Well, I agree. In my opinion, the top priorities for "core" development are:
1) Finalize tests for hard cap (see last entry at "develop" branch) and plan the coming soft fork. Improving SLM's security is top priority.
2) Prepare the code style for porting SLM to new codebase.
3) Port SLM to a newer codebase (I would choose first PPC 0.7 as a base, directly to 0.8 is imo a jump a bit too big because that's already modern Bitcoin code which needs lots of changes.).
The token mechanism for proof-of-donation could be developed in parallel to point 2.

I would strongly recommend a LAMP setup (Linux, Apache, MySQL/MariaDB) which is pretty standard for a VPS. A small VPS with about 1 GB RAM (even 512 MB should work) is enough. You also need Python 3, and a Slimcoin full node. (You can feed more than one Slimweb node with a single Slimcoin node, but I didn't try that.)

You need to install first Apache Jena Fuseki 2. Documentation is here: https://jena.apache.org/documentation/fuseki2/ Basically, it's just unpacking the compressed file and include "fuseki-server" into your PATH. Create a directory for your database. Before you start the scripts from the next step you need to run fuseki with the following command:


(with DATABASE being the directory you just created for the DB and /NAME being a name used by your server, I used "pub_slmchain" and I think that's hardcoded somewhere in the scripts, so you should use that, too, for now).

Then you have to download https://github.com/d5000/acme-minitools . There is a readme file. The important script is block2rdf.py. This script converts the original blockchain data to RDF so it can be queried by the gateway and other applications. You can start it with "-pub" mode so it only stores the relevant data for Slimweb publications.

I'll try to write a better tutorial these days (I think I did that some months ago here in the forum, but I have to search that post).

There is a short overview here: https://github.com/slimcoin-project/Slimcoin/wiki/Slimweb-comparison

Slimweb doesn't store the complete website data on the blockchain like Steem, and does not use NMC's namecoin domains.

The concept doesn't allow server side scripting, so no "standard" CMS like Wordpress are possible. However, in theory a similarly comfortable editing functionality could be reached with techniques from ZeroNet (but that has to be ported). For now, the best way to edit Slimweb pages comfortably is maybe something like [ulr=https://jekyllrb.com/]Jekyll[/quote].

True. That's another thing that should be updated once the codebase has been ported to PPC 0.6+ (in PPC, the mandatory fee was also reduced drastically). It needs probably a hard fork, however (see hardfork wishlist).

True.
Yes, it is possible, but at this moment only via a gateway, which can lead into problems if a gateway disappears. It is then possible that bots find it. However, there is a problem with Google (Slimweb pages constitute currently "mixed content") and so they won't rank very high. The best long-term solution would be an own search engine, however.

There's no additional anonymization technique. If he's afraid to get his IP detected, the most safe way would be to publish the page from another computer (Internet cafe etc.), creating the needed hashes/transactions offline, for example if he's living in a repressive country.

One idea that was brought up by Graham was to use it to store scientific publications. But anyway: It is potentially a very easy and simple way to publish all kinds of stuff online, potentially simpler than with rented webspace, and it can be made (if we integrate additional tools, which already exist, only they have to be bundled) similarly simple to platforms like Jimdo.

Do you mean something like the Semantic Web? That's also an idea that was circulating.

Yep. The features you mention (also the categories like "opinions" and "facts") can be achieved with Semantic Web techniques. We would need a vocabulary (rules) for that. We had a discusssion to integrate the Slimweb with the Solid project (developed by Tim Berners-Lee) which would provide a basic ruleset. It could then also integrate features from decentralized social networks.

I think, however, that voting wouldn't be very useful. It adds lots of complexity, and you can't avoid that somebody publishes something which violates the rules. You can, however, provide tools like a search engine or catalogue that only will show pages abiding to the rules (that can be automated). The rules can be voted, but on an external platform.

Yes, that can also be achieved with Semantic Web vocabulary.

Agree. One of the biggest advantages of Web2Web/Slimweb is that it can combine websites and payments (e.g. for electronic commerce or donations) in a very simple way, because an address is already provided by the page. So you can be sure you really pay to a specific publisher.

To create a roadmap we need to activate a discussion. So we need people ready to discuss things. It can be just me and you, but it's a work that has to be done. And once it's done and we have concepts ready that make SLM different from any other coin we need to put it on bitcointalk SLM thread and everywhere we are speaking about SLM as first things to mention. Because there is so many coins that an investor doesn't have time to dedicate to understand whether a coin has any novelty or not. If it's not mentioned as first thing where he is reading he'll just pass to another coin.

I can host some nodes. What are the minimum specs to host a node? Where can I find setup instructions?

I'd like to ask some things:
- How Slimweb concept is different from steemed for instance? And from NMC?
- Right now we are speaking about just on page publishing, right? Wordpress publishing is out of our reach at the moment for instance, right?
- The price of publishing is the price of transaction? In this case should the price of SLM go up could it be that the price of publishing would become too high?
- If the file published on Bittorrent disappears the published pages disappears as well right?
- Is it possible to point a traditional domain name to a Slimweb page? If so is there any chance that the Google bots will able to find it at some point?
- Being BTC technology not completely anonymous how the Slimweb publisher can be sure he stays anonymous?

I think we need to use blockchain and specifically Slimweb for something peculiar, for instance being blockchain immutable it can be used as registry office (registering the marriages, births, deaths, ownership of properties).

Another idea could be to set up rules for Slimweb pages publishing in order them to be intrinsically easily searchable.

The main challenge in Internet is traffic. If we find some area that is not covered by traditional internet and which represents a need for many people Slimweb may become useful.


Edit: maybe we can create the internet of pages so there is no domain name involved and there is a maximum decentralisation. Each publisher may chose a category/sub-category/sub-sub-category for his page, so the pages are easily found (for instance inputting category tag on the page itself).  The readers may report the pages that are not in the correct category. The new categories/sub-categories/sub-sub-categories may be decided on the base of the voting inside of SLM ecosystem. Or on the opposite the publishers can create their own sub-sub-....-sub-sub-categories with the goal his page to be among few other pages in the same category in order it to be easily found by who needs them. Or the publisher in the same category may decide to split their category into sub-categories in order their pages to have less competition in their sub-category.
We can create a main distinction between "opinions" and "facts" to be the two root categories of the Slimweb.

Edit1: Maybe we can think about internet of objects instead of internet of pages. Each object: text, video, audio, code, file, tweet and many others that maybe are not invented yet would have one or more than one identification category array. So the internet of objects will be a kind of multidimensional matrix. Everybody can create objects or assembly them from the objects of the others. Such constructs can be quite complex, so the structure resulting from the assembly activity can be more sophisticated than a site or a platform used in traditional internet. The publisher may split its own objects into smaller pieces and give to each piece its own categorisation array. SLM will be used to pay the publishing or to tip the publishers in order them to go on with their business if the readers do like their products. The search results of a possible search engine could be a number of category arrays to chose from, each one of those category arrays containing a limited number (10?) of objects.

I have no objections to that. I have already created a page for feature requests/proposals in the Wiki, but yes, some space to organize and assign priorities to the tasks that have to be done would be a great addition. If you want, you can create a page of this kind there.

My "Slimweb" tools are just some additions to Graham's original "ACME" concept. It uses the same codebase, only it provides a tool for easier publishing ("slimweb-publisher"), the Gateway at slimco.in (as an easy tool to access to hashed pages) and a little update to the code which transforms the blockchain data into a RDF database which can easily be queried (as SPARQL endpoints). To this code (a blocknotify script) I added some options to be able to re-sync the chain easier in the case the Slimcoin node or the "RDF node" crash.

Everybody can, in theory, verify the content comparing the hash stored on the blockchain with the HTML source of the displayed pages.

I outlined some ideas for new Slimweb features here: https://github.com/slimcoin-project/Slimcoin/wiki/The-Slimweb:--A-decentralized-web-based-on-Slimcoin on "Future plans"

However maybe the most urgent thing is that we would need more people hosting a "Slimweb" RDF node. Currently, as I'm the only one (Graham had a test node but I don't know if it's currently online), every time my client crashes, the node stops adding pages to the RDF database, and if the server crashes completely, Slimweb is offline. With more nodes, people could select the node they want in the Gateway, and if one doesn't work, another one would, surely.

There was an user who wanted to develop something of this kind, but unfortunately I haven't heard any news from that project.

Edit: Just saw that Graham has added the hard cap code to the "develop" branch. Thanks! I hope we can finalize the tests in the next weeks and thus implement the hard cap still this year (the most complicated thing being the soft fork).

It's seems like their business model has changed and they are basing on donations right now. I think we'd better donate something to them at some point to keep them working.

If your idea could be implemented it would be great. I'd like us to make a list of the goals we'd need to reach and to create a kind of roadmap (without binding timetables), it would help us to set up tasks for any additional developer we might find and to find investors to cover the expenses of each task/step.


Great tool, I think as you guys have worked in that direction we need to develop it further. Maybe connecting Slimweb to the project of hashed pages of Graham in order to be absolutely sure the page that the visitor is seeing is what the published wanted him to see.  
And what if we develop Slimcoin to use it as an unbiased search engine or a kind of web directory?

Good news, so somebody probably paid this block explorer again (it was the main SLM explorer in 2017 but then it was discontinued). I'll link it in the OP again. Thanks 

Regarding the problem of the coding style: I don't know if Graham has tried that, but wouldn't it be easier to port the original SLM changes (PoB and DCrypt, mainly) to a newer codebase (at least PPC 0.6 or 0.7) than vice-versa? In this case, the coding style could be reviewed and adapted.

@gjhiggins: I've just fired up my testnet node again. Is yours still online? Are you testing the "hard cap client" (with my small addition which prevents more than 13 PoS blocks in a row) or the original "improved PoS client" from iguanodon1?

Ah, one thing: Due to the mentioned VPS problems currently the Slimweb gateway (http://slimco.in/gateway.html) isn't working. I doubt however many people were using it lately

That's great! By the way I've just discovered that there is another explorer not mentioned in OP: https://bchain.info/slm/

No need, the chainz node is now synched.

Cheers

Graham

I thought so, is there any way to contact them asking to reboot their wallet? (I was not able to find any info on how to get in touch with them on their site)

It has been a bit bumpy, both my develop and master clients discovered they'd lost touch with the network, suddenly reporting that they'd found themselves several hundred blocks behind. I stopped and started them both, they synced up without problems, seeing stake and burn rewards. chainz client seems to have suffered the same problem, a simple restart might well fix it.

From the Heztner-hosted box' `getpeerinfo`:


Cheers

Graham

That's why you were advising me about people who made a lot of effort before me. Now I understand a bit better the situation.
My idea was the following: I've bought a quantity of SLM as I said in order to offer a share of the market to a possible interested developer. How does it sound to you?

No, I meant muf18 who spent weeks of his time searching out and talking to various devs, trying to find any interest in taking on the legacy Slimcoin codebase, unfortunately without success.

Cheers

Graham

Is it some kind of network outrage?

Explorer shows 4 hours 19 min the last block

From here it seems like the last block was generated 3 hours 51 minutes ago.