snakegi (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 02, 2015, 02:40:20 PM Last edit: June 11, 2015, 04:33:35 PM by snakegi |
|
THIS IS AN IDEA LOOKING FOR APPROVAL, NOT A FINISHED PROJECT
Its quite hard to target bitcoin related traffic. Of course you can use a crapload of ad tools like AdWords, Facebook Ads or even the specific ones like CoinAd or CoinURL If you used one of them before i'm sure you can list several reasons they don't pay off. This leads us back to this forum. Since ever, Bitcointalk has been the main source on information about bitcoin and all related stuff.
So, here we go:
BitcoinTalk Crawler This tool should help people who want to run ad campaigns (like Signature Campaigns or Post Campaigns) at Bitcointalk.
How? First of all we need a small marketplace to connect those who want to pay for a campaign and those who will work at the campaigns. After that, the bigger problem the project aims to solve is how people can prove they're actually working on a campaign, and looking by the other side, how can i know that i'm getting what i payed for. This issue can be resolved by a crawler-like tool, that will monitor registered workers, their posts and signatures from time to time, and generate a series of reports that can be managed by both workers and advertisers.
I am one of the CEOs at MountainCode LDA Portugal, company that is responsible for useCryptos exchange, and we have the know how that is needed to accomplish this project in a short period of time, but i need your feedback to know if it viable and if there are enough people interested in the project.
To measure that, please send a tip of any value (the minimum transaction is 0.0001 BTC i guess?) to this address and post here: 1MRAg4F3CqNNjGZWBbFvC2PEsqkqwzQCYj
The goal is to reach 100 supporters between advertisers and workers so i can move the project to the demo phase.
The initial supporters will have free premium accounts if the project is fulfilled.
Tips returned.
If the project doesn't pass the acceptation phase, i will return all transactions, so please DO NOT SEND FROM EXCHANGES OR POOLS
What happens next? Then we will make a demo project, with some of the features we talked about to test if we need any changes on the concept and raise funds for the real thing, that can be a huge project for everyone.
Please, speak out, i want to hear what you guys think.
Kind regards, Egidio
|
|
|
|
btc-facebook
Legendary
Offline
Activity: 1862
Merit: 1015
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 02, 2015, 03:45:51 PM |
|
Did you project just like bitmixer & bit-x signature post counter ? Maybe you can just show us the design or some flowchart how it works...
|
|
|
|
tspacepilot
Legendary
Offline
Activity: 1456
Merit: 1081
I may write code in exchange for bitcoins.
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 02, 2015, 03:54:51 PM |
|
Did you project just like bitmixer & bit-x signature post counter ? Maybe you can just show us the design or some flowchart how it works...
Right, if he's just making a counter robot, he has to compete with marcotheminer's program. If he's doing something more, it's completely clear what it is. Also, a lot of folks have pointed out that in order to keep spam out of the forums, you still need a human to check the posts---like dadice campaign does.
|
|
|
|
snakegi (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 02, 2015, 03:55:33 PM |
|
Did you project just like bitmixer & bit-x signature post counter ? Maybe you can just show us the design or some flowchart how it works...
The post counter would be one of the tools, important but not the main one i think. I mean, you can have 1000 posts last month and you claim to have used the signature for that period, but how can you prove that effectively? You would need something looking at profile from time to time and check if the signature is correct for example. The project intend to make all of this more easy to manage, i'm not saying is something new or innovative, but would be totally different if you could just register your user, pick a campaign and go posting. I'll make some flowcharts to explain the modules i though about. Thanks for your suggestion.
|
|
|
|
snakegi (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 02, 2015, 04:00:59 PM |
|
Did you project just like bitmixer & bit-x signature post counter ? Maybe you can just show us the design or some flowchart how it works...
Right, if he's just making a counter robot, he has to compete with marcotheminer's program. If he's doing something more, it's completely clear what it is. Also, a lot of folks have pointed out that in order to keep spam out of the forums, you still need a human to check the posts---like dadice campaign does. As i told, its not only the counter, but of course the counter is part of the project. A spam check would be a great touch to the project, i can think of two ways that this could work. 1- Semi automated spam check, that group posts that look alike, and count them as one for the campaigns. 2- Allow the advertiser to check the posts that are participating in his campaign and mark the posts that he won't pay (because break the rules of his campaign).
|
|
|
|
DiamondCardz
Legendary
Offline
Activity: 1134
Merit: 1118
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 02, 2015, 04:30:08 PM |
|
I was thinking of something just like this about a week ago, and I actually mentioned it on a thread and why exactly it wouldn't work. This is the principal issue that post-based automatic campaigns have: spam. It's very simple. That's why signature campaign managers exist, they check quality, etc. If your counter-argument to that is "someone can just check for quality every once in a while", then your bot is useless regardless.
Making a crawler for that isn't hard, it's just a bad idea.
|
BA Computer Science, University of Oxford Dissertation was about threat modelling on distributed ledgers.
|
|
|
marcotheminer
Legendary
Offline
Activity: 2072
Merit: 1049
┴puoʎǝq ʞool┴
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 02, 2015, 04:33:52 PM |
|
I have a signature campaign bot released already and features that you've mentioned are being implemented (for example a marketplace where campaigns are listed). The other things you mentioned are already included in the bot.
Will link it here as soon as I can.
|
|
|
|
snakegi (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 02, 2015, 04:48:27 PM |
|
I was thinking of something just like this about a week ago, and I actually mentioned it on a thread and why exactly it wouldn't work. This is the principal issue that post-based automatic campaigns have: spam. It's very simple. That's why signature campaign managers exist, they check quality, etc. If your counter-argument to that is "someone can just check for quality every once in a while", then your bot is useless regardless.
Making a crawler for that isn't hard, it's just a bad idea.
Spam is and will always be a problem, not only for posting in forums, but ad clicking, surfing, email and even if you are handling folders in the streets, but the point is, the better you can manage, better will be your campaign...
|
|
|
|
snakegi (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 02, 2015, 04:50:25 PM |
|
I have a signature campaign bot released already and features that you've mentioned are being implemented (for example a marketplace where campaigns are listed). The other things you mentioned are already included in the bot.
Will link it here as soon as I can.
Nice, the main point i've came up with this project is, i wanted to use something like that and never heard of anything good... Will be waiting for your link.
|
|
|
|
DiamondCardz
Legendary
Offline
Activity: 1134
Merit: 1118
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 02, 2015, 04:52:19 PM |
|
I was thinking of something just like this about a week ago, and I actually mentioned it on a thread and why exactly it wouldn't work. This is the principal issue that post-based automatic campaigns have: spam. It's very simple. That's why signature campaign managers exist, they check quality, etc. If your counter-argument to that is "someone can just check for quality every once in a while", then your bot is useless regardless.
Making a crawler for that isn't hard, it's just a bad idea.
Spam is and will always be a problem, not only for posting in forums, but ad clicking, surfing, email and even if you are handling folders in the streets, but the point is, the better you can manage, better will be your campaign... But a crawler is going to save you next to no time. Campaign managers will still have to check for spam - that takes a lot of time - and then deduct posts, make payments keeping deductions in mind, perhaps add in bonuses depending on whatever contests they're running... Basically the crawler is just going to track post count, and that's something that you can do with a calculator by taking the starting post count, ending post count and any deductions in about 10 seconds.
|
BA Computer Science, University of Oxford Dissertation was about threat modelling on distributed ledgers.
|
|
|
snakegi (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 02, 2015, 05:10:37 PM |
|
I was thinking of something just like this about a week ago, and I actually mentioned it on a thread and why exactly it wouldn't work. This is the principal issue that post-based automatic campaigns have: spam. It's very simple. That's why signature campaign managers exist, they check quality, etc. If your counter-argument to that is "someone can just check for quality every once in a while", then your bot is useless regardless.
Making a crawler for that isn't hard, it's just a bad idea.
Spam is and will always be a problem, not only for posting in forums, but ad clicking, surfing, email and even if you are handling folders in the streets, but the point is, the better you can manage, better will be your campaign... But a crawler is going to save you next to no time. Campaign managers will still have to check for spam - that takes a lot of time - and then deduct posts, make payments keeping deductions in mind, perhaps add in bonuses depending on whatever contests they're running... Basically the crawler is just going to track post count, and that's something that you can do with a calculator by taking the starting post count, ending post count and any deductions in about 10 seconds. Let me explain better. Lets say you want to run a signature campaign of post/week. -First of all you would have a place to publish your campaign and find workers (users), and to do so, you would have deposited the amount you want to spend in that campaign, so the system would also serve as escrow. -The users would register at the system and they would be allowed to work on your campaign (only one campaign at the time of course). -The system from time to time, will track that user's profile every hour to make sure the worker didn't changed the signature. -The system from time to time, will track that user's posts and try to group them as spam or not, and provide statistics for the advertiser and the worker. -At the end of each payment cycle the system would ask the advertiser to validate the data collected, grouping the posts that are definitely spam(equal posts), and flag as "probably" spam the posts that look alike, so the validation work is minimized. The system could also make the payouts since we could easily connect it to usecryptos api or something... Like i said, the point is to reduce most of the work of the advertiser and the worker. Its not possible to avoid all the spamming even if you are a human checking everything, but the better the tool better will be your campaigns...
|
|
|
|
DiamondCardz
Legendary
Offline
Activity: 1134
Merit: 1118
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 02, 2015, 05:23:13 PM |
|
-First of all you would have a place to publish your campaign and find workers (users), and to do so, you would have deposited the amount you want to spend in that campaign, so the system would also serve as escrow.
Part of my concept I've posted before. That would be feasible to have. -The users would register at the system and they would be allowed to work on your campaign (only one campaign at the time of course).
Also feasible. -The system from time to time, will track that user's profile every hour to make sure the worker didn't changed the signature.
Useful as well. -The system from time to time, will track that user's posts and try to group them as spam or not, and provide statistics for the advertiser and the worker.
Terrible, terrible, terrible idea. You are not honestly going to trust a bot/crawler to judge whether posts are spam, are you? You're just going to create a lot of needlessly pissed-off people, and there's also the fact some signature campaigns will want different things to others. -At the end of each payment cycle the system would ask the advertiser to validate the data collected, grouping the posts that are definitely spam(equal posts), and flag as "probably" spam the posts that look alike, so the validation work is minimized.
You assume this will result in much of a difference in workload. Not really. They'll still have to review every single post, pretty much. First few points would be nice to make but the spam stuff should still be left to a manager, and the payments.
|
BA Computer Science, University of Oxford Dissertation was about threat modelling on distributed ledgers.
|
|
|
snakegi (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 02, 2015, 05:40:23 PM |
|
-First of all you would have a place to publish your campaign and find workers (users), and to do so, you would have deposited the amount you want to spend in that campaign, so the system would also serve as escrow.
Part of my concept I've posted before. That would be feasible to have. -The users would register at the system and they would be allowed to work on your campaign (only one campaign at the time of course).
Also feasible. -The system from time to time, will track that user's profile every hour to make sure the worker didn't changed the signature.
Useful as well. -The system from time to time, will track that user's posts and try to group them as spam or not, and provide statistics for the advertiser and the worker.
Terrible, terrible, terrible idea. You are not honestly going to trust a bot/crawler to judge whether posts are spam, are you? You're just going to create a lot of needlessly pissed-off people, and there's also the fact some signature campaigns will want different things to others. -At the end of each payment cycle the system would ask the advertiser to validate the data collected, grouping the posts that are definitely spam(equal posts), and flag as "probably" spam the posts that look alike, so the validation work is minimized.
You assume this will result in much of a difference in workload. Not really. They'll still have to review every single post, pretty much. First few points would be nice to make but the spam stuff should still be left to a manager, and the payments. Sorry you didn't posted your full concept so i'm not familiar with what you thought before, please post in here if you want, my interest in this project is also as a user/advertiser... And i agree with you. You can't rely on a bot deciding what is spam or not, but a bot can surely help by making text analysis between the posts and flag the suspicious ones...
|
|
|
|
DiamondCardz
Legendary
Offline
Activity: 1134
Merit: 1118
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 02, 2015, 05:43:22 PM |
|
Sorry you didn't posted your full concept so i'm not familiar with you though before, please post in here if you want, my interest in this project is also as a user/advertiser...
And i agree with you. You can't rely on a bot deciding what is spam or not, but a bot can surely help by making text analysis between the posts and flag the suspicious ones...
Considering how long ago it was now, it's probably buried behind something like 200 posts in my post history. It wasn't a fully-fledged concept but rather an idea though consisting of a place where managers, advertisers and workers can meet, advertisers setting the price, rules, etc., and allocating a manager to their campaign, and then workers coming to the manager who then works with them as usual.
|
BA Computer Science, University of Oxford Dissertation was about threat modelling on distributed ledgers.
|
|
|
snakegi (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 02, 2015, 06:38:45 PM |
|
Considering how long ago it was now, it's probably buried behind something like 200 posts in my post history. It wasn't a fully-fledged concept but rather an idea though consisting of a place where managers, advertisers and workers can meet, advertisers setting the price, rules, etc., and allocating a manager to their campaign, and then workers coming to the manager who then works with them as usual.
I see, well i really think something like this would be nice for people who use this kind of ads and those who work on that, but since it don't have enough people to hold the project, there's no point (at least for us) in developing something like this now... But i would still use it if someone published a tool like this...
|
|
|
|
snakegi (OP)
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
June 08, 2015, 12:27:46 PM |
|
Despite someone sent a tip, it seems like there is not enough support for this project, so i guess i'll be closing this post and returning the transactions by the end of this week. If you know about any related projects or posts, please post here so we can get things together and maybe help someone that might interested.
Thanks everyone!
|
|
|
|
|