Newest 'p2pool' Questions - Page 2 - Bitcoin Stack Exchange

[Discord Conv] Dynamic IOTA


Disclaimer: This is my editing, so there could be some misunderstandings.
For the general view of 'what's going on?' of this dynamic ride...

2/16
dom어제 오전 5:44
Just FYI: the team is now working on a plan on how to recover from this and get the network back into operations while also allowing anyone who might have been affected to safely transition. there are no guarantees just yet, but we will do our best to get this through ASAP. Hopefully we will have a concrete action plan tomorrow and will then communicate it.
On the vulnerability side, all parties are notified and they are working with law enforcement and external auditors to fully understand how this happened. We will keep you guys posted.

dom어제 오전 5:47
needless to say, that the vulnerability itself was rather sophisticated and required access on multiple levels to be able to execute it on this scale. Hopefully we will be able to share more soon.
[Did the vulnerability existed after or before the audit on trinity?]
after the audit

dom어제 오전 5:51
Currently it looks like this will only be for recent Trinity Desktop users

dom어제 오전 5:56
the entire Trinity team did an amazing job and there is not a single person to blame. The attack itself was very sophisticated and targeted at IOTA and Trinity itself. We are already working on v2 where none of this would be possible. We will share our learnings from this publicly and also share what kind of precautionary measures we are taking.

dom어제 오전 5:58
The community also did an amazing job in helping to guide us through and give assistance to other community members.

dom어제 오전 5:58
we actually were having discussions a few weeks ago to rename Trinity (because of the religious connotation)

Jelle Millenaar [IF]어제 오전 6:37
We didn't really have panic and chaos. We actually worked really well together.

Jelle Millenaar [IF]어제 오전 6:38
[IF members, do you get paid Over Time for all the awesome work or PURE DEDICATION?]
nobody considers this overtime or anything. We just contribute because we know it is needed.

dom어제 오전 7:45
[If dependencies carry this risk, maybe they should've done an official CORE wallet and saved all the fluffy stuff for a third party app.]
that's how the new Trinity will work. Sucks that it happened now especially after we wanted to put it into maintenance mode anyways

dom어제 오전 7:49
[How do we know if the hacker has our seeds?]
this is related to a third party, unrelated to IF or IOTA

dom어제 오전 7:50
we know that this could have only been done through intrusion / collusion of an external source.
[Dom are you fully confident to solve all those problems especially regarding the possibility of even more people getting scammed instantly after coo is back again?]
yes, relatively sure. That is why we are taking the necessary time to plan accordingly.

dom어제 오전 7:55
We will provide more information on how this exploit was done soon. All the involved parties are aware of the situation

dom어제 오전 7:58
[Please give us some time before you start the coo information that we can move to new seed instantly]
don't worry, we will get it all sorted out.

dom어제 오전 8:22
once life is a bit less "tumultuos" I still want to work on that Autonomous Bar concept powered by IOTA (access control, id verification, payment and a bunch of robots)

Eric Hop [IF]어제 오후 2:44
Pretty good. I'd be surprised if we find more theft bundles. Only found one more today, while building a timeline of the theft.

Eric Hop [IF]어제 오후 2:50
We have several separate teams. One is looking at how to resume. One is looking at how to be able to rescue the funds. Others are interacting with law enforcement and third parties. I'm part of DAFT. The Data Analysis Forensics Team. Haha

Eric Hop [IF]어제 오후 3:01
Some if the people in Coordicide team like Hans have been helping out. It was an all hands on deck situation. I actually loved it. We haven't had this much of a team spirit in quite a while. Usually everyone plays in their own sand box. But this time we all played together on the beach.
It's such a joy working with so many extremely smart people. With so many eyes on the ball we did not miss much opportunities to figure things out.

Eric Hop [IF]어제 오후 3:07
And for me personally this was a great time. I am all about puzzle solving. And this was the greatest puzzle of all. With a built-in time limit. Haha

Eric Hop [IF]어제 오후 3:12
I'm not doing official statements. But we have a good overview of what happened and the extent of it. Right now we want to hammer down how to resume without risks and how to safeguard the stuck funds if possible. What is especially funny to me is that the coordinator that everyone was bitching about for years did exactly the thing it was meant to do. It allowed us to halt an exploit that otherwise would have cost everyone dearly.

Eric Hop [IF]어제 오후 3:15
It was meant as safeguard, training wheels, while we mature. And while we need to remove it due to it being a single point of failure and a bottle neck to scaling, I will be kind of sad to see it go.
Yes, IF would have done the same to safeguard funds, if a third party wallet would have been the cause. Just because we can.

Eric Hop [IF]어제 오후 3:37
Yes it was a manual attack. The sophistication was in the exploit. But he seemed to be not too sophisticated iota-wise. Everyone has their specialties I guess.

Eric Hop [IF]어제 오후 3:41
And as an aside I wish people would fuck off about the whole iota not being decentralized because of coordinator, when every block chain token is centralized around a few mining pools that seriously disrupt any possibility for positive software development. They fucking hold back everything that influences their bottom line. Which is why Bitcoin and the rest have pretty much been stagnant for years while we move forward constantly.

dom오늘 오전 7:08
We will release a new Trinity version tomorrow with the fixes implemented. It's not yet the full transition tool, but it's the first step towards fully going back to operations.

dom오늘 오전 7:09
Just wait for the rest. It is important that we get this 100% right and we are still further investigating, so there is a lot of behind the scenes work happening right now.

David Sønstebø오늘 오전 8:52
So... Tangle EE
Quite cool eh?
It's so unfortunate that this asshole managed to distract everything away from one of the biggest steps towards global adoption
Let's not give this fuckface further attention. The cause has been identified, law enforcement is involved and mitigation strategy is being worked on. There will be further official updates, but let's not halt the whole IOTA project due to one idiot.

David Sønstebø오늘 오전 8:56
[Is he identified?]
Let's just say that there's a lot of traces. The attacker does not seem to have been too sophisticated. Official update on Monday will provide details.

David Sønstebø오늘 오전 9:03
[How will this situation affect iotas partners?]
My best guess: further increasing our reputation as an organization that solves hard problems efficiently and doesn't shy away from difficulties. Every company in the world has had issues similar to this. Keep in mind that this does not at all affect the protocol/Tangle/IOTA.

David Sønstebø오늘 오전 9:08
We do have a bounty program. This/these individual/s were not interested in the greater good, pure greed and incompetence

David Sønstebø오늘 오전 9:10
[Any examples of use cases for DID on the tangle?]
Virtually all use cases on Tangle requires a secure identifier and verifiable credentials. What I think will happen is that once Tangle EE ships the first version, all other companies using IOTA will start to implement it
[One more question: How transparent will tangle EE be for the community?]
100%. This is why I/we consider Tangle EE to be such a significant milestone, it's not "just" IF, this is a coalition of major companies, start-ups and leading academic institutions building the solutions

David Sønstebø오늘 오전 9:11
[any ETA for the 1st Version?]
That's another good thing, IF won't issue the ETAs, Tangle EE will :

David Sønstebø오늘 오전 9:12
[What does T(angle)EE do exactly?]
It's a partnership and collaboration between several entities to develop and ship code and blueprints that are relevant for product developers and service providers
That blog post is a good read to get better comprehension

David Sønstebø오늘 오전 9:13
It's incredibly important that IF's role slowly but surely decreases in importance. IOTA has to succeed independent of IF post-Coordicide and multiversial-slicing (advanced sharding equivalent)

David Sønstebø오늘 오전 9:14
I would say that it's an incredible important piece of the puzzle. Naturally Object Management Group (OMG) in Tangle EE will be key here as well, but IOTA is not married to "just" Eclipse. We also work closely with Linux Foundation. However, Tangle EE is very focused

David Sønstebø오늘 오전 9:22
I don't think IF will disappear, however, it will hopefully be purely R&D-driven in 10 years, whereas the other efforts are taken over by the ecosystem (companies, academia, start-ups and enthusiasts). Even post-Coordicide, we already now have theories on how to go way beyond even that. If we achieve our goal of IOTA being equivalent to TCP/IP, there will naturally be continuous development and research in the foreseeable future. I doubt we will reach complete satisfaction, especially now that smart contracts and oracles enter the equation: there's certainly more work to be done for IF, but my goal is for IF to "simply" be R&D

David Sønstebø오늘 오전 9:27
Definitely. This is why I coined the requirement for a "grandma on crack"; this is truly how simply using IOTA should be in 2-5 years. Just like very few even know wtf TCP/IP is

David Sønstebø오늘 오전 9:57
I agree 100% with your assessment, though as would Netflix do with Blockbuster's assessment when they declined to acquire Netflix. At the end of the day it's all about basic economic and human behavioural principles.
Human nature does not change, but our environment does. Disruption will continue forever. Darwinian principles will forever remain true.
A better option = adoption. It doesn't matter how hard the incumbents fight against it, they either adapt or go Kodak/Nokia/AOL
submitted by btlkhs to Iota [link] [comments]

Lisk Highlights Weekly roundup March 9th 2019. The week in which a Lisk Sidechain Project became a Founding Member of a Brussels-based Blockchain Organization.

Hello everybody. The LISK project and it's enthusiasts are always busy, and this week past has certainly been no exception.
Seeing is believing, so here is a recap of the highlights and interesting items from the past week on the LISK subreddit and beyond.....
 

Lisk, Hong Kong Future and Costa Rica Past.

Asia Crypto Week is fast approaching (11-17th March) and blockchain/crypto enthusiasts and industry veterans are preparing to gather together, share their knowledge and nurture mass crypto adoption. Among the events taking place on March 15th will be a meetup at the University of Hong Kong catering to the University's Blockchain Club and anyone else that might be in the area and interested in all things blockchain. Lisk is co-hosting the event in collaboration with 9up.io, who are a group of blockchain enthusiasts based in Hong Kong and who also are a prospective Lisk delegate. Max Kordek, Lisk's Co-Founder and CEO, will be the main speaker at this event on the 15th March, as he will be in town from the 10th to 15th for Asia Crypto Week and Token 2049.
Tickets for the event can be secured HERE. Hong Kong has been a strategic position in the blooming Blockchain industry in recent years, so I will be interested to see what emerges from this meetup and indeed Asia Crypto Week as a whole.
 
Now from the future to the past, and the TicoBlockChain 2019 conference in Costa Rica this past month. Lisk Central America have linked us up with a Stylish Montage Video of the event with interview snippets interspersed within. Software Architect, Jake Simmons, represented LISK Central America with his presentation on 'Scaling blockchain horizontally with Lisk'. Jake's presentation took part in the midst of the conference's speaker collection of lawyers, developers, educators, banking executives, investment professionals, their keynotes, panel talks and fireside chats.
For those of us who could not make the trip we have Lisk community member illuciferium to thank for filming Jake's presentation and uploading it to Youtube HERE. You can also see Jake being interviewed at the conference by Ricardo Barquero, Nimiq Community Manager in this VIDEO. Well done all!
 

Lisk Support Adds a Meetup Map.

LISK support and TonyT908 are back with a new way for Liskers to visualise all the upcoming meetups and events related to the project. The Events Map is a more visually exciting way to discover all the upcoming events around the globe rather than reading through reams and reams of text. When you are ready to delve more into the details of a particular meetup then you can visit the official Lisk Events page to read further details.
Special thanks should be given to Global Delegate Team (GDT) for their guidance to TonyT908 and for providing the funds necessary to license to mapping software. Edward Trosclair AKA StellarDynamic came up with the original concept, so he should take a lot of praise also. Great work all round, folks.
 

Lisk Sidechain Project Knows the Key is to Stand Out from the Crowd.

Chief R&D Officer and Co-founder of GNY (bringing Machine Learning to Lisk), Richard Jarritt informed the project's followers on the GNY telegram that "only by having a platform that is the first to crack machine learning on chain is how we can differentiate ourselves from the countless projects in our space". He continued, "I look at the whole crypto space and at the moment having working code is key, that is the drive here".
So on to the coding and Machine learning integration, how is that going? Well, this week Leo Liang, Head of Blockchain for GNY will be presenting the coding solution for how information is read by the machine learning off the chain. The GNY team always has tech meetings on Tuesdays to present the work that has been done the week previously, where they update each other on progress and then set the next task. Upcoming shortly for the team will be a demo of how the read function and Machine learning are running together and moving onto the reply function.
The GNY Github is due to go live by the end of this month, following an important GNY staff conference in London. It's going to be an exciting month and I am really looking forward to it.
 

Lisk Sidechain Project becomes a Founding Member of a Brussels-based Blockchain Organization.

On the 6th of March the MADANA project were one of the 105 companies, startups and organizations that came together to found INATBA. The International Association for Trusted Blockchain Applications (INATBA) which will be a Brussels-based organization working to make blockchain more accessible, safe, and usable for everyone. This was an initiative of the EU commission and the Directorate General for Communications Networks, Content and Technology or "DG Connect", whose responsibility is managing the Digital Agenda across Europe. It is hoped that this will now allow projects in the blockchain and distributed ledger technology eco-system have access to a global forum to interact with regulators & policy makers.
 
Next up for the INATBA is its first General Assembly on 3rd April where the 105 founding members will hopfully be joined by fresh additions to the memberlist. The INATBA website is now live at http://inatba.org and they have a "Join" page which is accepting applications for new members to join Madana and the likes of Iota, Cardano, Gnosis, and the Quant Network for the April 3rd launch!
 
That's it for the recap of the weeks highlights. I hope it brought you up to speed with all the weeks good news.
These highlight posts also go out daily on the….
LISK Highlights exclusive Telegram group
LISK Highlights Twitter
The highlights are also included in my weekly roundup on the LISK Highlights Medium account and the Bitcoin talk forum's LISK thread, so keep an eye out for them on these outlets also.
 
Keep the faith Liskers! 👍
submitted by John_Muck to Lisk [link] [comments]

New people please read this. [upvote for visibility please]

I am seeing too many new people come and and getting confused. Litecoin wiki isn't the greatest when it comes to summing up things so I will try to do things as best as I can. I will attempt to explain from what I have learned and answer some questions. Hopefully people smarter than me will also chime in. I will keep this post updated as much as I can.
Preface
Litecoin is a type to electronic currency. It is just like Bitcoin but it there are differences. Difference explained here.
If you are starting to mine now chances are that you have missed the Bitcoin mining train. If you really want your time and processing power to not go to waste you should mine LTC because the access to BTC from there is much easier.
Mining. What is it?
Let's get this straight. When making any financial commitment to this be prepared to do it with "throw away" money. Mining is all about the hashrate and is measured in KH/s (KiloHash/sec). Unlike the powerful ASICs (Application Specific Integrated Circuit) that are used to mine bitcoins using hashrates in the GH/s and even TH/s, litecoin mining has only been able to achieve at the very best MH/s. I think the highest I've seen is 130 MH/s so far. Which leads us to our next section.
Mining Hardware
While CPU mining is still a thing it is not as powerful as GPU mining. Your laptop might be able to get 1 a month. However, I encourage you to consult this list first. List of hardware comparison You will find the highest of processors can maybe pull 100 KH/s and if we put this into a litecoin mining calculator it doesn't give us much.
Another reason why you don't want to mine with your CPU is pretty simple. You are going to destroy it.
So this leaves us with GPUs. Over the past few months (and years) the HD 7950 has been the favourite because it drains less power and has a pretty good hashrate. But recently the introduction of the R9 290 (not the x) has changed the game a bit. People are getting 850 KH/s - 900 KH/s with that card. It's crazy.
Should I mine?
Honestly given the current difficulty you can make a solid rig for about $1100 with a hashrate of 1700 KH/s which would give you your investment back in about a month and a half. I am sure people out there can create something for much cheaper. Here is a good example of a setup as suggested by dystopiats
PCPartPicker part list / Price breakdown by merchant / Benchmarks
Type Item Price
CPU AMD Sempron 145 2.8GHz Single-Core Processor $36.01 @ Amazon
Motherboard ASRock 970 EXTREME4 ATX AM3+ Motherboard $99.48 @ OutletPC
Memory Crucial Ballistix Tactical Tracer 4GB (1 x 4GB) DDR3-1866 Memory $59.99 @ Newegg
Video Card Sapphire Radeon HD 7950 3GB Video Card (3-Way CrossFire) $245.38 @ Newegg
Video Card Sapphire Radeon HD 7950 3GB Video Card (3-Way CrossFire) $245.38 @ Newegg
Video Card Sapphire Radeon HD 7950 3GB Video Card (3-Way CrossFire) $245.38 @ Newegg
Power Supply SeaSonic Platinum 860W 80+ Platinum Certified Fully-Modular ATX Power Supply $146.98 @ SuperBiiz
Total
Prices include shipping, taxes, and discounts when available. $1078.60
Generated by PCPartPicker 2013-11-29 00:52 EST-0500
Estimated Hashrate (with GPU overclocking) : 1900 KH/s
Hardware Fundamentals
CPU - Do you need a powerful CPU? No but make sure it is a decent one. AMD CPUs are cheap to buy right now with tons of power. Feel free to use a Sempron or Celeron depending on what Motherboard you go with.
RAM - Try to get at least 4 GB so as to not run into any trouble. Memory is cheap these days. I am saying 4 GB only because of Windoze. If you are plan to run this on Linux you can even get away with less memory.
HDD Any good ol 7200 RPM hard drive will do. Make sure it is appropriate. No point in buying a 1TB hard drive. Since, this is a newbie's guide I assumed most won't know how to run linux, but incase you do you can get a USB flash drive and run linux from it thus removing the need for hard drive all toghether. (thanks dystopiats)
GPU - Consult the list of hardware of hardware I posted above. Make sure you consider the KH/s/W ratio. To me the 290 is the best option but you can skimp down to 7950 if you like.
PSU - THIS IS BLOODY IMPORTANT. Most modern GPUs are power hungry so please make sure you are well within the limits of your power consumption.
MOTHERBOARD - Ok, so a pretty popular board right now is Gigabyte GA-990FXA-UD3 and the ASRock 970 Extreme4. Some people are even going for Gigabyte GA-990FXA-UD5 and even the mighty Gigabyte GA-990FXA-UD7 because it has more PCI-E slots. 6 to be exact. However you may not need that much. With risers you can get more shoved into less.
PCI-E RISERS - These are called risers. They come in x16 to x16 and x1 to x16 connections. Here is the general rule of thumb. This is very important. Always get a POWERED riser otherwise you will burn a hole in your MoBo. A powered rise as a molex connector so that additional power from PSU can be supplied.
When it comes to hardware I've provided the most basic knowledge you need. Also, take a look at cryptobader's website. This is very helpful. Please visit the mining section of Litecoin Forums and the litecoinmining subreddit for more indepth info.
Mining Software
Now that you have assembled your hardware now you need to get into a pool. But before you do that you need a mining software. There are many different ones but the one that is most popular is cgminer. Download it and make sure you read the README. It is a very robust piece of software. Please read this if you want to know more. (thanks BalzOnYer4Head)
Mining Pools
Now that your hardware and software is ready. I know nothing about solo mining other than the fact that you have to be very lucky and respectable amount of hashing power to decrypt a block. So it is better to join pools. I have been pool hopping for a bit and really liked give-me-coin previously known to the community as give-me-ltc. They have a nice mobile app and 0% pool fees. This is really a personal preference. Take a look at this list and try some yourself.
How do I connect to a pool?
Most pools will give you a tutorial on how to but the basics are as follows:
  • Signup for a pool
  • Create a worker for your account. Usually one worker per rig (Yes people have multiple rigs) is generally a good idea.
  • Create a .run file. Open up notepad and type cgminer.exe -o (address_to_the_miningpool:port_number) -u (yourusername.workername) -p (your_worker_password_if_you_made_one). Then File>Save As>runcgminer.run (Make sure the drop down is set to "All Files" and .txt document.) and save in the same folder as cgminer. That's it.
  • Double click on runcgminer.run (or whatever you named it) and have fun mining.
Mining Profitability
This game is not easy. If it was, practically everyone would be doing it. This is strictly a numbers game and there are calculations available that can help you determine your risk on your investments. 4 variables you need to consider when you are starting to mine:
Hardware cost: The cost of your physical hardware to run this whole operation.
Power: Measured in $/KwH is also known as the operating cost.
Difficulty rate: To put it in layman's terms the increase in difficulty is inversely proportional to amount of coin you can mine. The harder the difficulty the harder it is to mine coin. Right now difficulty is rising at about 18% per 3 days. This can and will change since all you miners are soon going to jump on the band wagon.
Your sanity: I am not going to tell you to keep calm and chive on because quiet frankly that is stupid. What I will tell you not to get too carried away. You will pull you hair out. Seriously.
Next thing you will need is a simple tool. A mining profitability calculator. I have two favourite ones.
coinwarz
I like this one cause it is simple. The fields are self explanatory. Try it.
bitcoinwisdom
I like this one because it is a more real life scenario calculator and more complicated one (not really). It also takes increasing difficulty into account.
Please note: This is the absolute basic info you need. If you have more questions feel free to ask and or google it!
More Below.
submitted by craeyon to litecoin [link] [comments]

The Mike Hearn Show: Season Finale (and Bitcoin Classic: Series Premiere)

This post debunks Mike Hearn's conspiracy theories RE Blockstream in his farewell post and points out issues with the behavior of the Bitcoin Classic hard fork and sketchy tactics of its advocates
I used to be torn on how to judge Mike Hearn. On the one hand he has done some good work with BitcoinJ, Lighthouse etc. Certainly his choice of bloom filter has had a net negative effect on the privacy of SPV users, but all in all it works as advertised.* On the other hand, he has single handedly advocated for some of the most alarming behavior changes in the Bitcoin network (e.g. redlists, coinbase reallocation, BIP101 etc...) to date. Not to mention his advocacy in the past year has degraded from any semblance of professionalism into an adversarial us-vs-them propaganda train. I do not believe his long history with the Bitcoin community justifies this adversarial attitude.
As a side note, this post should not be taken as unabated support for Bitcoin Core. Certainly the dev team is made of humans and like all humans mistakes can be made (e.g. March 2013 fork). Some have even engaged in arguably unprofessional behavior but I have not yet witnessed any explicitly malicious activity from their camp (q). If evidence to the contrary can be provided, please share it. Thankfully the development of Bitcoin Core happens more or less completely out in the open; anyone can audit and monitor the goings on. I personally check the repo at least once a day to see what work is being done. I believe that the regular committers are genuinely interested in the overall well being of the Bitcoin network and work towards the common goal of maintaining and improving Core and do their best to juggle the competing interests of the community that depends on them. That is not to say that they are The Only Ones; for the time being they have stepped up to the plate to do the heavy lifting. Until that changes in some way they have my support.
The hard line that some of the developers have drawn in regards to the block size has caused a serious rift and this write up is a direct response to oft-repeated accusations made by Mike Hearn and his supporters about members of the core development team. I have no affiliations or connection with Blockstream, however I have met a handful of the core developers, both affiliated and unaffiliated with Blockstream.
Mike opens his farewell address with his pedigree to prove his opinion's worth. He masterfully washes over the mountain of work put into improving Bitcoin Core over the years by the "small blockians" to paint the picture that Blockstream is stonewalling the development of Bitcoin. The folks who signed Greg's scalability road map have done some of the most important, unsung work in Bitcoin. Performance improvements, privacy enhancements, increased reliability, better sync times, mempool management, bandwidth reductions etc... all those things are thanks to the core devs and the research community (e.g. Christian Decker), many of which will lead to a smoother transition to larger blocks (e.g. libsecp256k1).(1) While ignoring previous work and harping on the block size exclusively, Mike accuses those same people who have spent countless hours working on the protocol of trying to turn Bitcoin into something useless because they remain conservative on a highly contentious issue that has tangible effects on network topology.
The nature of this accusation is characteristic of Mike's attitude over the past year which marked a shift in the block size debate from a technical argument to a personal one (in tandem with DDoS and censorship in /Bitcoin and general toxicity from both sides). For example, Mike claimed that sidechains constitutes a conflict of interest, as Blockstream employees are "strongly incentivized to ensure [bitcoin] works poorly and never improves" despite thousands of commits to the contrary. Many of these commits are top down rewrites of low level Bitcoin functionality, not chump change by any means. I am not just "counting commits" here. Anyways, Blockstream's current client base consists of Bitcoin exchanges whose future hinges on the widespread adoption of Bitcoin. The more people that use Bitcoin the more demand there will be for sidechains to service the Bitcoin economy. Additionally, one could argue that if there was some sidechain that gained significant popularity (hundreds of thousands of users), larger blocks would be necessary to handle users depositing and withdrawing funds into/from the sidechain. Perhaps if they were miners and core devs at the same time then a conflict of interest on small blocks would be a more substantive accusation (create artificial scarcity to increase tx fees). The rational behind pricing out the Bitcoin "base" via capacity constraint to increase their business prospects as a sidechain consultancy is contrived and illogical. If you believe otherwise I implore you to share a detailed scenario in your reply so I can see if I am missing something.
Okay, so back to it. Mike made the right move when Core would not change its position, he forked Core and gave the community XT. The choice was there, most miners took a pass. Clearly there was not consensus on Mike's proposed scaling road map or how big blocks should be rolled out. And even though XT was a failure (mainly because of massive untested capacity increases which were opposed by some of the larger pools whose support was required to activate the 75% fork), it has inspired a wave of implementation competition. It should be noted that the censorship and attacks by members of /Bitcoin is completely unacceptable, there is no excuse for such behavior. While theymos is entitled to run his subreddit as he sees fit, if he continues to alienate users there may be a point of mass exodus following some significant event in the community that he tries to censor. As for the DDoS attackers, they should be ashamed of themselves; it is recommended that alt. nodes mask their user agents.
Although Mike has left the building, his alarmist mindset on the block size debate lives on through Bitcoin Classic, an implementation which is using a more subtle approach to inspire adoption, as jtoomim cozies up with miners to get their support while appealing to the masses with a call for an adherence to Satoshi's "original vision for Bitcoin." That said, it is not clear that he is competent enough to lead the charge on the maintenance/improvement of the Bitcoin protocol. That leaves most of the heavy lifting up to Gavin, as Jeff has historically done very little actual work for Core. We are thus in a potentially more precarious situation then when we were with XT, as some Chinese miners are apparently "on board" for a hard fork block size increase. Jtoomim has expressed a willingness to accept an exceptionally low (60 or 66%) consensus threshold to activate the hard fork if necessary. Why? Because of the lost "opportunity cost" of the threshold not being reached.(c) With variance my guess is that a lucky 55% could activate that 60% threshold. That's basically two Chinese miners. I don't mean to attack him personally, he is just willing to go down a path that requires the support of only two major Chinese mining pools to activate his hard fork. As a side effect of the latency issues of GFW, a block size increase might increase orphan rate outside of GFW, profiting the Chinese pools. With a 60% threshold there is no way for miners outside of China to block that hard fork.
To compound the popularity of this implementation, the efforts of Mike, Gavin and Jeff have further blinded many within the community to the mountain of effort that core devs have put in. And it seems to be working, as they are beginning to successfully ostracize the core devs beyond the network of "true big block-believers." It appears that Chinese miners are getting tired of the debate (and with it Core) and may shift to another implementation over the issue.(d) Some are going around to mining pools and trying to undermine Core's position in the soft vs. hard fork debate. These private appeals to the miner community are a concern because there is no way to know if bad information is being passed on with the intent to disrupt Core's consensus based approach to development in favor of an alternative implementation controlled (i.e. benevolent dictator) by those appealing directly to miners. If the core team is reading this, you need to get out there and start pushing your agenda so the community has a better understanding of what you all do every day and how important the work is. Get some fancy videos up to show the effects of block size increase and work on reading materials that are easy for non technically minded folk to identify with and get behind.
The soft fork debate really highlights the disingenuity of some of these actors. Generally speaking, soft forks are easier on network participants who do not regularly keep up with the network's software updates or have forked the code for personal use and are unable to upgrade in time, while hard forks require timely software upgrades if the user hopes to maintain consensus after a hardfork. The merits of that argument come with heavy debate. However, more concerning is the fact that hard forks require central planning and arguably increase the power developers have over changes to the protocol.(2) In contrast, the 'signal of readiness' behavior of soft forks allows the network to update without any hardcoded flags and developer oversight. Issues with hard forks are further compounded by activation thresholds, as soft forks generally require 95% consensus while Bitcoin Classic only calls for 60-75% consensus, exposing network users to a greater risk of competing chains after the fork. Mike didn't want to give the Chinese any more power, but now the post XT fallout has pushed the Chinese miners right into the Bitcoin Classic drivers seat.
While a net split did happen briefly during the BIP66 soft fork, imagine that scenario amplified by miners who do not agree to hard fork changes while controlling 25-40% of the networks hashing power. Two actively mined chains with competing interests, the Doomsday Scenario. With a 5% miner hold out on a soft fork, the fork will constantly reorg and malicious transactions will rarely have more than one or two confirmations.(b) During a soft fork, nodes can protect themselves from double spends by waiting for extra confirmations when the node alerts the user that a ANYONECANSPEND transaction has been seen. Thus, soft forks give Bitcoin users more control over their software (they can choose to treat a softfork as a soft fork or a soft fork as a hardfork) which allows for greater flexibility on upgrade plans for those actively maintaining nodes and other network critical software. (2) Advocating for a low threshold hard forks is a step in the wrong direction if we are trying to limit the "central planning" of any particular implementation. However I do not believe that is the main concern of the Bitcoin Classic devs.
To switch gears a bit, Mike is ironically concerned China "controls" Bitcoin, but wanted to implement a block size increase that would only increase their relative control (via increased orphans). Until the p2p wire protocol is significantly improved (IBLT, etc...), there is very little room (if any at all) to raise the block size without significantly increasing orphan risk. This can be easily determined by looking at jtoomim's testnet network data that passed through normal p2p network, not the relay network.(3) In the mean time this will only get worse if no one picks up the slack on the relay network that Matt Corallo is no longer maintaining. (4)
Centralization is bad regardless of the block size, but Mike tries to conflate the centralization issues with the Blockstream block size side show for dramatic effect. In retrospect, it would appear that the initial lack of cooperation on a block size increase actually staved off increases in orphan risk. Unfortunately, this centralization metric will likely increase with the cooperation of Chinese miners and Bitcoin Classic if major strides to reduce orphan rates are not made.
Mike also manages to link to a post from the ProHashing guy RE forever-stuck transactions, which has been shown to generally be the result of poorly maintained/improperly implemented wallet software.(6) Ultimately Mike wants fees to be fixed despite the fact you can't enforce fixed fees in a system that is not centrally planned. Miners could decide to raise their minimum fees even when blocks are >1mb, especially when blocks become too big to reliably propagate across the network without being orphaned. What is the marginal cost for a tx that increases orphan risk by some %? That is a question being explored with flexcaps. Even with larger blocks, if miners outside the GFW fear orphans they will not create the bigger blocks without a decent incentive; in other words, even with a larger block size you might still end up with variable fees. Regardless, it is generally understood that variable fees are not preferred from a UX standpoint, but developers of Bitcoin software do not have the luxury of enforcing specific fees beyond basic defaults hardcoded to prevent cheap DoS attacks. We must expose the user to just enough information so they can make an informed decision without being overwhelmed. Hard? Yes. Impossible. No.
Shifting gears, Mike states that current development progress via segwit is an empty ploy, despite the fact that segwit comes with not only a marginal capacity increase, but it also plugs up major malleability vectors, allows pruning blocks for historical data and a bunch of other fun stuff. It's a huge win for unconfirmed transactions (which Mike should love). Even if segwit does require non-negligible changes to wallet software and Bitcoin Core (500 lines LoC), it allows us time to improve block relay (IBLT, weak blocks) so we can start raising the block size without fear of increased orphan rate. Certainly we can rush to increase the block size now and further exacerbate the China problem, or we can focus on the "long play" and limit negative externalities.
And does segwit help the Lightning Network? Yes. Is that something that indicates a Blockstream conspiracy? No. Comically, the big blockians used to criticize Blockstream for advocating for LN when there was no one working on it, but now that it is actively being developed, the tune has changed and everything Blockstream does is a conspiracy to push for Bitcoin's future as a dystopic LN powered settlement network. Is LN "the answer?" Obviously not, most don't actually think that. How it actually works in practice is yet to be seen and there could be unforseen emergent characteristics that make it less useful for the average user than originally thought. But it's a tool that should be developed in unison with other scaling measures if only for its usefulness for instant txs and micropayments.
Regardless, the fundamental divide rests on ideological differences that we all know well. Mike is fine with the miner-only validation model for nodes and is willing to accept some miner centralization so long as he gets the necessary capacity increases to satisfy his personal expectations for the immediate future of Bitcoin. Greg and co believe that a distributed full node landscape helps maintain a balance of decentralization in the face of the miner centralization threat. For example, if you have 10 miners who are the only sources for blockchain data then you run the risk of undetectable censorship, prolific sybil attacks, and no mechanism for individuals to validate the network without trusting a third party. As an analogy, take the tor network: you use it with an expectation of privacy while understanding that the multi-hop nature of the routing will increase latency. Certainly you could improve latency by removing a hop or two, but with it you lose some privacy. Does tor's high latency make it useless? Maybe for watching Netflix, but not for submitting leaked documents to some newspaper. I believe this is the philosophy held by most of the core development team.
Mike does not believe that the Bitcoin network should cater to this philosophy and any activity which stunts the growth of on-chain transactions is a direct attack on the protocol. Ultimately however I believe Greg and co. also want Bitcoin to scale on-chain transactions as much as possible. They believe that in order for Bitcoin to increase its capacity while adhering to acceptable levels of decentralization, much work needs to be done. It's not a matter of if block size will be increased, but when. Mike has confused this adherence to strong principles of decentralization as disingenuous and a cover up for a dystopic future of Bitcoin where sidechains run wild with financial institutions paying $40 per transaction. Again, this does not make any sense to me. If banks are spending millions to co-op this network what advantage does a decentralized node landscape have to them?
There are a few roads that the community can take now: one where we delay a block size increase while improvements to the protocol are made (with the understanding that some users may have to wait a few blocks to have their transaction included, fees will be dependent on transaction volume, and transactions <$1 may be temporarily cost ineffective) so that when we do increase the block size, orphan rate and node drop off are insignificant. Another is the immediate large block size increase which possibly leads to a future Bitcoin which looks nothing like it does today: low numbers of validating nodes, heavy trust in centralized network explorers and thus a more vulnerable network to government coercion/general attack. Certainly there are smaller steps for block size increases which might not be as immediately devastating, and perhaps that is the middle ground which needs to be trodden to appease those who are emotionally invested in a bigger block size. Combined with segwit however, max block sizes could reach unacceptable levels. There are other scenarios which might play out with competing chains etc..., but in that future Bitcoin has effectively failed.
As any technology that requires maintenance and human interaction, Bitcoin will require politicking for decision making. Up until now that has occurred via the "vote download" for software which implements some change to the protocol. I believe this will continue to be the most robust of options available to us. Now that there is competition, the Bitcoin Core community can properly advocate for changes to the protocol that it sees fit without being accused of co-opting the development of Bitcoin. An ironic outcome to the situation at hand. If users want their Bitcoins to remain valuable, they must actively determine which developers are most competent and have their best interests at heart. So far the core dev community has years of substantial and successful contributions under its belt, while the alt implementations have a smattering of developers who have not yet publicly proven (besides perhaps Gavin--although his early mistakes with block size estimates is concerning) they have the skills and endurance necessary to maintain a full node implementation. Perhaps now it is time that we focus on the personalities who many want to trust Bitcoin's future. Let us see if they can improve the speed at which signatures are validated by 7x. Or if they can devise privacy preserving protocols like Confidential Transactions. Or can they figure out ways to improve traversal times across a merkle tree? Can they implement HD functionality into a wallet without any coin-crushing bugs? Can they successfully modularize their implementation without breaking everything? If so, let's welcome them with open arms.
But Mike is at R3 now, which seems like a better fit for him ideologically. He can govern the rules with relative impunity and there is not a huge community of open source developers, researchers and enthusiasts to disagree with. I will admit, his posts are very convincing at first blush, but ultimately they are nothing more than a one sided appeal to the those in the community who have unrealistic or incomplete understandings of the technical challenges faced by developers maintaining a consensus critical, validation-heavy, distributed system that operates within an adversarial environment. Mike always enjoyed attacking Blockstream, but when survey his past behavior it becomes clear that his motives were not always pure. Why else would you leave with such a nasty, public farewell?
To all the XT'ers, btc'ers and so on, I only ask that you show some compassion when you critique the work of Bitcoin Core devs. We understand you have a competing vision for the scaling of Bitcoin over the next few years. They want Bitcoin to scale too, you just disagree on how and when it should be done. Vilifying and attacking the developers only further divides the community and scares away potential future talent who may want to further the Bitcoin cause. Unless you can replace the folks doing all this hard work on the protocol or can pay someone equally as competent, please think twice before you say something nasty.
As for Mike, I wish you the best at R3 and hope that you can one day return to the Bitcoin community with a more open mind. It must hurt having your software out there being used by so many but your voice snuffed. Hopefully one day you can return when many of the hard problems are solved (e.g. reduced propagation delays, better access to cheap bandwidth) and the road to safe block size increases have been paved.
(*) https://eprint.iacr.org/2014/763.pdf
(q) https://github.com/bitcoinclassic/bitcoinclassic/pull/6
(b) https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-Decembe012026.html
(c) https://github.com/bitcoinclassic/bitcoinclassic/pull/1#issuecomment-170299027
(d) http://toom.im/jameshilliard_classic_PR_1.html
(0) http://bitcoinstats.com/irc/bitcoin-dev/logs/2016/01/06
(1) https://github.com/bitcoin/bitcoin/graphs/contributors
(2) https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-Decembe012014.html
(3) https://toom.im/blocktime (beware of heavy website)
(4) https://bitcointalk.org/index.php?topic=766190.msg13510513#msg13510513
(5) https://news.ycombinator.com/item?id=10774773
(6) http://rusty.ozlabs.org/?p=573
edit, fixed some things.
edit 2, tried to clarify some more things and remove some personal bias thanks to astro
submitted by citboins to Bitcoin [link] [comments]

Conversations with a Dev, Insight into the workings behind our coin.

This is a copy of a discussion I had with one of the dogecoin Devs, I think it is an important read as the original post dissapeared of the front page pretty quick, it was, enlightening.
[–]rnicollReference client dev 8 points 23 hours ago We need a stable reference client whatever we do next. 1.7 will be that client, but we need more testing. We'll hopefully have a beta release in the next few days, but until they any testing on the alpha release would be greatly appreciated: http://www.reddit.com/dogecoin/comments/224vix/dogecoin_17_alpha_1_such_test_wow_much_feedback/ That out of the way; a coin derives value through utility. To make a coin useful, it must be accepted. For it be accepted, merchants must see acceptance as a cost-effective option. One of the issues they're going to wonder about, to my mind, is software stability. We keep having major updates which fork the blockchain or have problems with the RPC daemon. 1.7 is our solution to that; a massively improved client based on the Bitcoin 0.9 client, with a far more in depth testing period than previous releases. It means we're leaping to the front of the altcoins in technical basis, and providing a rock solid foundation for adoption. Meanwhile, we have documentation projects ongoing in security and standards both to further aide adoption, and to enable alternative implementations. Edit: Going to bed, not ignoring any replies, but unlikely to get back to you until after work tomorrow. permalinksavereportgive goldreply
[–]Master565[S] 1 point 16 hours ago That sounds great! I wish I could help you test, but I'm on the wrong computer in the wrong country right now.
[–]junkyardmessiah 6 points 23 hours ago All things being equal however , as a Dev , do you see a problem is the current Sfire situation , 25gh is pointed at us and he is dumping mass quantities of Doge into the wafflepool for BTC using his asic farm. Last count it was over 1.1 billion coins in the past week alone. I can pull up the graphs, But I think you already know about this. Initial acceptance by merchants is not going to consist of true believers of the 1 doge = 1 doge principal . If you look at BTC and Tiger direct , pricing directly matches BTC = fiat. So any principle based on the idea that merchants are going to look to anything but what the coin can be turned into as an aspect of Fiat is ... silly, I knwo the mantra of 1doge = 1 doge sounds nice , but untill we have inter business trading of the currency between them. 1 doge is going to be worth .0003 cents
[–]rnicollReference client dev 4 points 22 hours ago Painfully aware of SFire. Honestly, I'm more worried about the continual drops in BTC/USD pricing. If you follow the price closely, we get further hammered down when BTC dips, so we're getting double the pain. We're losing ground to BTC, and that's not great, but that's about 10% of the total drop we're seeing. I can't do much for BTC though, so lets get back to mining. Okay; merged mining won't fix the price, as far as I can tell. We'd about break-even (I reckon) vs LTC miners dumping DOGE, vs DOGE miners dumping LTC. It would probably massively reduce risk of a 51% attack, but that's all. For that, we take on the risk of any major problems with LTC will also hit us, and as we expand over time it becomes harder to "de-merge" later. Lastly, it would likely trash many of our mining pools, who would be too small to find any blocks, vs LTC miners. Moving to a CPU mining algorithm would probably just hand us to botnets, instead of ASIC farms. I've seen this before, certainly. So lets say we decide to move to a GPU mining algorithm of one sort or another. We still need a stable client we know we can modify, and people report problems with 1.6, so we have to get 1.7 done first. Hence we need testing of 1.7. We then implement a new PoW, and at some point branch. Presuming we don't have any unexpected problems with the branch, or introduce any critical bugs while editing core parts of the code, we make it harder to copy to/from DOGE mining, but if we're still the most profitable coin, people will still mine us (although with GPU farms rather than ASIC). We also inherently accept the environmental cost of all that GPU power. Oh, and we have to remain hopeful that no bright pure maths PhD student finds a way of optimising the hashing algorithm which gives them a huge performance boost over everyone else (essentially acting as a software ASIC). In summary; we're getting to the end of the development period for 1.7 and looking at 1.8 will do, but no answer that's been proposed is clearly superior to all others.
[–]junkyardmessiah 1 point 18 hours ago Well I do thank you for the response, Just I really hope we can move towards an answer abit faster. right now we are in the sub 80's and still dropping in value. I think the fact we are tied to BTC as our peg is another thing that is killing us . we need to be tied directly to Fiat if we stand a chance. I'd rather not have our currency tied to possible future Mt GOX events. it really needs to stand on its own, But again value right now is reflected in fiat. This is a bad situation all the way around . But here is the question . is there ANYTHING the community can do to stop this falling value and turn it around, from the obvious. I have bought all of the coin I can right now and am mining my ass off. I am accepting the coin at my business on small items right now but need to pay rent and they don't take Doge. ... yet I don't want the coin so I can just dump it like Sfire does for BTC/Fiat. Is there a crucial thing each of us can do to change the current stalemate of the deflationary machine that is Waffle or multipools?. Where do we toss our Sabots into the gears of our destruction? I am my wits end , all I can see as a clear winner might be that we need a mixed POW/POS algo with 25% Scrypt, 25% Scrypt-n or scrypt jane, 25% CPU, 25% POS. If such a thing were a possible fix . and the devs need money to develop it, .. will you take a check?
permalinksaveparenteditdeletereply [–]rnicollReference client dev 1 point 11 hours ago runs back to the thread, out of breath Sorry, so many posts, keep losing track. Well, good news to start with, we apparently just jumped back up again. Praying this isn't a dead cat bounce (seriously, that's a term, I'm not making this up). Hoping other devs will chip in on suggestions; my background is trading-heavy, so I'm mostly thinking about things such as whether we can get UPS to accept Dogecoin (so merchants pay shipping costs in Dogecoin without exchanging through). I'm also wondering about seeing if we can pay taxes in Dogecoin, because I know HMRC takes debit cards, and if Dogecoin is cheaper to accept why not... We've also just had a working implementation of merged mining for Litecoin submitted to Github, which I imagine will be the major topic of discussion over the weekend, as obviously it reduces the work for that route by a lot. If anyone has the technical skills, a P2P multipool would be a way of avoiding Wafflepool scenarios by enabling those who want to make immediate profit to hop target without centralising their mining. This is where someone tells me we have one already, of course... Oh, also, you know we're 14 days (give or take) from halvening? So SFire's rewards will halve on the 25th (again, give or take a bit), along with everyone else's? permalinksaveparentreportgive goldreply
[–]junkyardmessiah 1 point 6 hours ago Well I can hope things will be better by the next halvening But I suspect the percentage of low level miners will drop off since it just won't be worth it anymore. I suspect that regardless of the halvening his percentage of the % total coins mined will be the same. But there is a rumor that with the profits hes made so far that his goal is 50gh. and he has done some testing https://pools.rapidhash.net/pool/5 As for the sudden jump , A whale bought 101 million last night for his attempt to a pump and dump. I am not expecting much. So lets break it down, it is a simple question , What at present is off the table or is just undoable in time to bring our value up by the 10k block rewards. and what CAN be done by the devs. to me there there is only one thing that will guarantee a price boost and keep people mining. Drop the 5 billion deflationary yearly to 1.5 % or drop it completely, if the coin is too cheap and we cannot keep miners on it it is dead. simple enough. permalinksaveparenteditdeletereply
[–]rnicollReference client dev 1 point 4 hours ago Off the table? Well, we're not going to merge-mine with BTC, fairly certain of that at least! For 25th (halvening); probably most things would need more time, but most changes we could have in testing and potentially released ready for a higher block number, by then. Changing block rewards would need overwhelming support from the community to do. Honestly, I think people would call for our heads if we tried it. Notably, last halvening the price did nearly double in the week or two beforehand. That's part of why we're mostly favouring doing nothing; historically actually things look bad, but do get better. You can also dig into Bitcoin's history for examples of this. Interestingly, someone just submitted a patch for Litecoin merged mining. I'd definitely not want to be the one to make a call that we know will crush whole mining pools, but it means the option's there if it gained real support. As to changing PoW - I'm going to be talking contingencies with other devs tonight, see what we can come up between us. Again, there's mostly concern that if this goes wrong, we can do a lot more damage to the coin than SFire and a halvening is doing... permalinksaveparentreportgive goldreply
[–]junkyardmessiah 1 point 2 hours ago Thank you for being frank and honest with us , Let the other Devs Know a weekly update on the front page would be really really helpful for bolstering confidence that there is actual direction to the madness that is Dogecoin A big question here is how do we gauge support? It is all over the place, voting system would be nice. with some protections against bots and nobody can vote with an account younger the 5 days from the time of the vote would be good too. Debate could be held on reddit , voting on on dogcoin.com I think a lot of angst and worry is due to the feeling of not "really " having control of the destiny of dogecoin. That it is up to the devs to make decisions on the smattering of posts that are scattered all over this sub. We need a cohesive way to express our opinions . If you make the reasoned case for block reward changes , you might be surprised by the reaction you get. Please put it to the others that if we could have more interaction and more regular updates it would go well for all of us and inspire confidence . which can also help bring our value up. permalinksaveparenteditdeletereply
[–]rnicollReference client dev 1 point 1 hour ago It's worth keeping in mind we're still just assembling organisational structure. Clients up to 1.5 inclusive were led by the founders, 1.6 was lleti and rog1121, and 1.7 is now leofidus, langerhans and myself. So; we're working on communication with you guys, as soon as we've got it sorted amongst ourselves :) Attempts so far to have go-betweens assist us with communication have given an impression of an isolated elite, so yeah, it's a work in progress. Anyway; yes, there's predictably been extended and expressive discussion for several hours, and we think we've got a least bad solution. I'm getting everyone involved to confirm they're happy with the idea in principal, but it looks like we're going to try beating the multipools at their own game, by building a better multipool which is less hostile to DOGE. That's as much as is concrete so far, I can literally see the conversation progressing in another window, and the other two core devs are both offline anyway. I definitely want to see if I can get a weekly "What we've done, what we've learnt, what we've thought about but didn't do, what you can do to help" blog post or similar, and I may get the first done this weekend (if not, next). There is a hell of a lot happening (for example implementing the Bitcoin payment protocol for Doge, security whitepapers, writing formal specifications to go to the IETF) that people don't see... permalinksaveparentreportgive goldreply
[–]junkyardmessiah 1 point 57 minutes ago Okay that is at least something. It is like trying to pull back the curtain of the wizards of doge only to see them scrambling like mad pulling levers and pushing buttons/ I would like to post this discussion we have had in a new post to get it some more visibility if you would be okay with that. I really do hope the next Halvening will show some results. But I think most expected a lot more out of the last one. I think all of us would like to hear about the "least bad" solution, and what it means for the coin.
[–]rnicollReference client dev [score hidden] 43 minutes ago* If you want to see us at work, the #dogecoin-dev channel on Freenode's an excellent idea. I'm there if I'm working, and the other two currently active devs are frequently around too. It's... well, busy, and I can't guarantee you'll see stuff that interests you, but it does let you see what's going on. As to a new post; I'm unfussed. Anything I'm posting I presume I'll come back to discover a few thousand shibes have spotted and suddenly read. That said, because weekends are when I can get my head down and make progress, I'll be less chatty on reddit, fair warning. At the moment everyone in the room seems reasonably happy that the best thing to do is to provide better mining software to pull people such as SFire away from single pools. I think there's also the intent to have the pools generate DOGE by default instead of BTC, but even if that's not something people could work with, we can at least implement much better trading algorithms (that's well within my skillset). That said, others are taking the lead there, I'm going to be focusing on technical documentation; a security specialist is working on guidelines for using cryptocurrency safely (for merchants and similar) with some guidance from myself, and I'm hoping to make a start on formal documentation as soon as the Bitcoin payment protocol stuff is done.
submitted by junkyardmessiah to dogecoin [link] [comments]

[Table] IAmA: I am Matt Mullenweg, co-founder of WordPress (18% of web!) and Automattic, ask me anything!

Verified? (This bot cannot verify AMAs just yet)
Date: 2013-07-31
Link to submission (Has self-text)
Questions Answers
Hey Matt, I was at your State of the Word in SF and you talked about moving WordPress more towards being an application framework rather than a CMS or blog platform. What specifically do you have in mind for this (better settings API, developer features, etc)? And then if you could break backwards compatibility (which really isn't a option for WP), what would you really like to completely redo or add to WordPress? Thanks! First and foremost the most important things for a platform are stability, speed, and security. To do those well you need the ability to push updates and fixes as close to real-time as possible. And it needs to work in every language. User authentication, data and caching abstraction.
A lot of what people think of as platform stuff is actually at the CMS layer -- custom post types, taxonomy meta,
If backwards compatibility wasn't a concern I would rename all the inconsistent column names and variables to match our style guide, drop TinyMCE, simplify the user roles and capabilities system, replace widgets with page blocks, redo the admin menu system, denormalize the DB, flatten dependencies and deep hierarchy in function execution, and completely reorganize the code so the bare minimum of files are included with any given request.
After reading questions from all of the nice, well-meaning people writing WordPress as "Wordpress", do you wish reddit would implement capital_P_dangit(\)? ;) Yes! If anyone from Reddit sees this, would super-duper appreciate if we could get /WordPress capitalized properly. I can send copious amounts of swag, bourbon, and BBQ if bribes are needed.
Looks like that already happened. My link works, but if you click the link in the header it takes you back to the lowercase P. Probably just a field in a database somewhere.
How is it going with Bitcoin ? Haven't looked recently, will see if I can pull some numbers. I wish I had bought some Bitcoins before we introduced it, though. :)
Any stats ? 94 successful transactions so far, pretty low. I think it's more important philosophically to support it than it's been beneficial from a business point of view.
If you could wave a magic wand and instantly rewrite WordPress in any programming language, which would it be? Go. :)
Do you ever think "I could have been tumblr" ? Just the opposite, I'm very glad we're not Tumblr.
Hi Matt, I'm a long time WordPress dev (since day 1 moreorless) so thanks very much for starting it. Just wondering what your opinion of Ghost is? Link to www.tryghost.org. Link to www.fastcolabs.com
They say or imply lots of things about WordPress that aren't true. They've also done things like had a quote from me looking like I was endorsing Ghost on their Kickstarter page even after I asked them to remove it several times. (Lots of people were confused or thought it was a plugin for WP.)
What do you think of Drupal 8 ? :) Some cool stuff in there! I also keep an eye on Joomla and Concrete5. I find it really fascinating to watch other open source projects especially because we share much of the same background and philosophy, but make radically different decisions around things like backward compatibility and release schedule.
It's like watching birds that evolved from the same ancestor but on isolated islands and environments. I'm sure we do things that look completely crazy to Drupal folks, and vice versa.
Along those lines I was asked to keynote at the Joomla World Conference in November and it looks like I'm going to be able to make it.
Why is hello dolly still a default plugin? Do you have any statistics about how many people actually activate/use it? Have you personally written any other plugins? Hello Dolly is actually the 13th most active plugin, with an active userbase of about 16% of Akismet (the most-activated plugin), and about a third as popular as Jetpack. It's ahead of W3 Total Cache! Again this is not just installations, it's currently active.
Some of the other plugins I've been involved with are here on my profile: Link to profiles.wordpress.org
They're obsolete but at the time I was proud of Advanced Caching, Staticize Reloaded, and Cache Images and the early and since-rewritten work on bbPress, HyperDB, and Akismet.
How directly involved are you in ongoing development of core? Are you relatively hands on, do you ever drive decision making or do you leave it to the community? Would love a wooden WP logo, as long as it's not a fauxgo. :)
On a completely different subject. I'm as involved in WordPress development as I was 10 years ago, it just manifests itself in ways that tend to be a lot more behind the scenes and less visible, which I don't mind as I'm way more interested in things moving forward and the results than credit or recognition for any specific thing. (I get plenty of recognition regardless, don't need more.) The only downside is that folks who I don't work with on a day-to-day basis assume that my role at Automattic or WP is more as a traveling figurehead or "evangelist" which can rub me the wrong way sometimes.
I met you at WordCamp Chicago this year (I told you about a widget plugin that turns them into a post type and such). For 3.8 I'm going to take a swing at the release lead role again, which should be fun.
I created these wooden WordPress logos with my 3D Printer, I meant to give you one but I forgot about it. You still want one? The most important thing I've done since WP started, though, isn't in a line of code or a feature people use, it's getting the right people involved and creating an environment for them to thrive. It's the single most important thing any founder can do, whether of an OS project, a non-profit, or a for-profit company even though there's not a single thing you can point to as the result of it other than the overall success and movement of the project.
What are your favorite/most powerful/most surprising implementations of WordPress? My favorite are when people I admire use it, from Jay Z to Zeldman. In many ways what we do like a canvas, and it's a huge honor to see the creativity and beauty people bring to it.
Of all the WordCamp's you've been to around the globe, what was the most unusual location for the event itself and, separately, the after party? Oh by far and away the strangest location was Davao in the Philippines. I can't find any pictures at the moment, but my talk was essentially at a restaurant with a swimming pool courtyard -- the audience was on the other side of the pool from me, and the food buffet was behind me so when the Q&A got slow people would grab food. The PA system had an echo because I think it was normally used for karaoke. And then the bats came out!
Best after-party is hard to pick, but I had a great time after WordCamp Las Vegas which aligned with my 25th birthday a few years ago.
The Philippines is a wacky place. Oh, my people :-) I loved it and had a great time, including at the one mentioned above. :)
Of all the WordPress community memes, which one is your favorite? Link to twitter.com is pretty funny, and some days I miss WP Wank.
Recently I enjoyed #thingswpdailywouldpost and the response: Link to torquemag.io
Pretty much anything Mike Adams (mdawaffe) sneaks into the code, be it easter eggs or Ghostbuster references.
Mark Jaquith had some pretty hilarious ones but I can't find them at the moment, maybe other folks could post their favorites too. :)
Automattic has a lot of side projects (Gravatar, PollDaddy, etc) - What's next? There's always a struggle between doing new things or experiments under a new brand -- like VaultPress -- vs putting it under an existing brand. A lot of the things I've been thinking about we're going to put under the Jetpack brand, for example Jetpack Photon (CDN + dynamic image resizing and filtering) could be a standalone product, but decided to bundle it. So keep an eye on some big things coming to Jetpack, especially for Code Poets, people who use WordPress professionally.
What would you like to see us do next?
What individual do you think is the most under recognized contributor to the WordPress community at large? That's a tough one... I'm going to say the volunteers on the support forums. There are 2M+ posts there, and it's easy to forget that a huge number of WP users end up in the forums and get help that allows them to use the software when they wouldn't otherwise be able to.
I remember seeing a WordPress yearly theme of "WordCamps and WordPress Meetups". What should be the next year's area of focus for the Community? What areas would you like to see beefed up in the Community? Some themes that I thought might be good are "Women in WordPress", "Contribution and Documentation", "WordPress Evangelism", "Adopt an old plugin" to name a few. How do you feel about these? Those are all good, if I had to pick one it'd be getting the documentation going better on WordPress.org -- handbooks, function reference, training materials / syllabuses, and doing it all in every language and for every plugin/theme.
Hey Matt, what was the most difficult thing for you in starting your business and what do you think is the most important thing you have to know and/or to learn when you found a startup? The hardest thing for me was taking responsibility for the lives and families of others, those first few hires especially. It's why I originally raised funding even though we had revenue already, and why since then we've always focused on making the business sustainable over decades, not just the next tech hype cycle.
What's the best meal you have ever eaten (and photographed)? Hands down: Eleven Madison Park in NYC.
Recently I had a really special libation tasting menu at The Aviary in Chicago, which is from Grant Achatz the chef of Alinea, it was a cool experience at a table in the kitchen.
I'm constantly amazed by the coordination and efficiency of professional kitchens, I'm mesmerized when I watch.
PHP has matured a lot in the last few years, with new tools such as Composer and new frameworks such as Laravel. The relationship between this side of the PHP community and WordPress seems to be pretty strained. Are there plans to address this relationship, particularly with the new focus on WordPress as a web app framework? I think the PHP and WP community are coming closer together, I know it's something that Nacin has been spending time on and we've had more presence at PHP-focused conferences.
Thoughts on forking WordPress, a la jQuery? (Link to eamann.com. I don't think forking as described there is a good idea.
Multiple content areas - probably the most important CMS feature not baked into core. Will it ever happen? There's something around multiple content areas that could be really interesting we're going to start working on this year, hopefully ready by early 2014.
What do you see as the biggest challenge facing WordPress (in the context of competitive software or just un-met customer needs) for the next year? By far and away it's the high attrition rate of new users. We look at posting a lot in that context but I think it's far more important to look at customization -- theme discovery and tweaking, widgets, menus.
Hello Matt, what is your opinion about current state of PHP in general? Do you like any particular framework? Templating engine? Thanks. I think it's pretty great, would just love to see continuing development around performance. Nothing really in the language that's holding us back. Wish it was trendier with younger devs.
Hey Matt, why did you force the rest of us to suffer through images defaulting to a link in our posts (I understand that was your doing)? Youthful indiscretion!
What is your development setup (software)? Most of my time these days is spent with people instead of code. (For better or worse. :)) I love lists, and live on Link to simplenote.com .
When I code since I switched to Mac in 2011 I use Coda 2 and SFTP to a remote server rather than a local dev environment. nano on the server.
Apache or nginx? Nginx!
I find WP so much more user-friendly than the competition. Was that a conscious decision from the outset? Was it hard work to make it that way or was it just the way you guys did things? Do you have a warehouse full of useability testers or does it just come naturally to you guys? The first few users were friends of mine who weren't into technology at all, so from the start we needed to make it work for regular people. As we grow it's mostly just a matter of reminding ourselves of that, sitting down with them to see how they use the software, and anticipating their needs.
If something in the future interests you enough to make WordPress/Automattic part-time for you, so you can take on something else or contribute to another cause, what might it be? I'm really fascinated by micro-electronics, hardware, and the "internet of things" for lack of a better term. It's surreal to be approaching an age that resembles the sci-fi I loved as a child.
That said, I can't imagine not working full-time on WordPress/Automattic, I feel too strongly about our mission, impact, and potential.
What's your favorite beer? Link to imgur.com. Link to cloudup.com
I see a lot of desperate web development companies locally that try to stress that Wordpress is insecure and shouldn't be used. What would be the best thing to say to people like that to shut them up? Some of the largest and most important publishers in the world rely on WordPress. (Show them the showcase.) If WordPress was insecure we'd see it on the front page of nytimes.com, wired.com, and cnn.com. :)
Hi Matt, have you read Chris Lema's blog post. What is your response? I have. It's also funny because I think Gartner is about to come out with a "magic quadrant" that puts us in the crappy quadrant (low vision and ability to execute). Their leaders? Adobe, Sitecore, SDL, Oracle, HP, Opentext...
I completely agree with Chris on all the ways that enterprise currently works, and their concerns. (People assume because we choose to do things differently that we don't understand the other side.) But I'm not willing to compromise getting better software into the hands of users as quickly as possible, if that means Gartner thinks we're a visionless niche player so be it.
We've done long-term support branches before, it was a big development burden and almost no one used it or cared. There will be businesses that embrace keeping their technology moving at the speed the web does, and there will be those that go out of business and become irrelevant.
•Will you ever support multiple languages in the WordPress core? No plans for multiple languages in core, sorry.
•What do you think of new writing platforms like Quip and Editorially? Will the WordPress post editor ever have any of those 'team' features? I really dig the new writing platforms, I do think we'll get some of those team features if not in core than in Jetpack.
I heard a rumor Matt switched back to QWERTY. True? Not true, still typing Dvorak, though last year I was beat on speed for the first time by Helen Hou-Sandi, who types QWERTY. She's speedy, and if she switched to Dvorak she could probably win world champs. :)
Hey Matt, where do you think the future of the independent WordPress news community lies, with your recent purchase of WPTavern and the recent sale of WP Daily to WP Engine? Do you think there is a space for an impartial WordPress news website? *edited for correct spelling! I'm glad that new ones are being started as fast as old ones are shutting down. There is some really interesting stuff going on in the community and I think there's space for real journalism and strong commentary.
What one major thing would you like to see changed/fixed/updated in WordPress core? The fact it doesn't work well on mobile devices.
How do I convince my boyfriend, who wants me to keep his site in ASP.NET (he's a programmer, I'm a designecoder), to install WordPress? I'm not even allowed to use PHP! :( The best way I've found to convince people, even as the founder, is just showing examples. That's why we created the showcase: Link to wordpress.org
Find out which musicians, celebs, authors, etc he writes and see which of them are on WordPress and bring it up casually in conversation. (We have huge adoption from creative folks.)
Best BBQ of all time? I hope I haven't had it yet. :)
Are open source contributions a prerequisite to work at Automattic? No, but they get you to the top of the list when we're reviewing applications. (I know, I look at every incoming resume.)
Would you consider Automattic the "Redhat of WordPress"? Without a doubt. ;)
Do you think an app store for plugin and themes built with high quality standards and framework, could be a good solution for WordPress end users? The plugin directory is an app store where everything is free.
Would having paid stuff there make it better? I don't think so.
Why do you feel thats different for themes - premium themes are available via the themes directory. (I've never bought one of those so I don't know if the comparison is valid) Design is inherently valuable in its scarcity, functionality is valuable in ubiquity.
It's unlikely that we'll ever want to put a popular theme into core, but fairly likely it'll happen for a popular plugin.
Collaboration is more important for plugins than themes, and money from scarcity balkanizes development.
Did you ever feel like you couldn't continue with your project? Is there any advice you have for small business startups? Covered advice here -- Link to www.reddit.com
There have absolutely been times I felt like I couldn't continue, both with Automattic and WordPress. You have to take the lows with the highs and stick with what you believe in.
Hi Matt, I have been using WordPress for 10 years, make most of my living from it, and will always love it. Thank you for that! It is by far the easiest way I have found to build websites that my clients find easy to use. I see the reasons why WordPress does not use more modern coding practices and tools and appreciate the need for backward compatibility, but wonder if you ever see the code base moving forward to a time when developers can use the newest features of PHP, best coding practices (i.e. testing), and the great tools that are available these days, like Composer. Do you think there will ever be some kind of fork or offshoot of WordPress that functions as an application development framework, since so many developers are using it for that these days? Thanks :) I disagree with the premise -- WordPress does use modern coding practices. People assume that supporting say an older version of PHP or MySQL holds us back far, far more than it actually causes any trouble. Supporting older browsers is a way bigger deal.
Our biggest challenge is figuring out the user side of things, the front-end code. How things should work for a user rather than how they should work for a computer.
What does your average day look like and how do you manage so many different projects? I spend more time on Skype (text chat with colleagues) than I would care to admit. Between that and P2s ( Link to p2theme.com ) I can easily fill eight hours in a day. As the company has grown to over 180 people there is a huge amount of content and activity to keep up with.
Hey Matt, how many hours you work on an average day? It's hard to say because I don't really consider what I do work, the hours just melt away.
I find I'm most productive first thing in the morning when I wake up, usually around 7am but a bit later if it's a cloudy morning, and I generally run out of steam around 11:30pm that night.
Some days I'm traveling though I might only have 4-5 hours at a computer and can get a similar amount done.
I find I'm generally more creatively charged the following day if I'm able to unplug at night, which is one reason I like jazz festivals (I try to go to Montreal every year) because I can work during the day and check out shows at night.
Any chance of a comment-spam filter being built into core? Akismet is great but has a lot of hurdles for a new user (Activation of the plugin, Registration on Wordpress.com, Registration for API key) and most just don't do it, contributing to the problem. I agree with Viper007Bond -- anything we did in core like Cookies for Comments would become completely ineffective within a day. Those only stop dumb bots who have easier places to spam.
Link to markjaquith.wordpress.com
off, love you, your ethos, your company. If you figure it out, let me know. :)
My question: how do you plan to get "normal" people to update their blogs as often as they do their Twitter and Tumblrs? But seriously, I think it's all about removing friction (every second loading and extraneous click) and becoming part of people's habits, which is one of the reason I spent a fair amount of time on triggers and habits at the State of the Word this year.
any plans to launch a Premium paid version of Photon service with more features? Not on features, we'll make anything new there free to everybody, but might have a paid tier for top 1% of users by bandwidth/usage. But probably a few years from that, plenty of bandwidth and CPU here in the meantime, and it's just getting cheaper and faster.
Find a new boyfriend maybe? His judgement is clearly lacking. Ha!
Which features are available on other CMS that you would like to see in WordPress? I really like some of the drag and drop layout things folks like Squarespace and Weebly do. I think we have so much we can do to improve customization.
Hi Matt ! I know you've advocated for water causes in the past. Do you have plans for future charitable/philanthropic projects? Thanks for doing this. My big contribution this year has been to the Bay Lights project: Link to thebaylights.org
Haven't decided about next year yet...
What do you think about App websites/themes that seem to be using WordPress as the choice of CMS, do you think WordPress is a good platform for these types of sites? Scaling, Performance issues considering? Examples, Dating sites, Crowdfunding sites, Job board, etc. I think it's a great framework for anything content-driven. For things like messaging that don't map well to WP's data model, you can still do it just make some new tables, don't try to shoehorn it in the standard ones.
Which verticals will you be tackling next with WordPress.com? Please share your sensitive corporate information ;) Putting a hold on new verticals at the moment, going to try and go deeper on some of the existing ones first.
Hi Matt, now as most people know, you are very much a fan of open source and the GPL license. However, is there anything you do NOT like about open source? Also besides Wordpress (or its themes/plugins), are there any open source projects you like that aren't related to Wordpress or Automattic? I think the things that make open source incredibly collaborative and ultimately eat the world can also make design and big shifts difficult. WordPress has made some major shifts over its decade of life and grew as a result, but those pivots are harder to do the more successful we are because sometimes it means doing the opposite of what we did to become successful in the first place.
Automattic's been tweeting a bunch of new hires lately, and quite frankly doing a great job of making me jealous. Has the 90% of employees being remote ever caused real problems or friction? And how do Happiness Engineers work: do they help people out over the phone, or is it strictly a text over the Internet thing? Remote distributed working has been working great and we plan to continue it for the foreseeable future. HEs mostly work with email, but we're introducing a ton more live chat and phone work.
Say you retired your involvement from WordPress, what other projects stimulate your interest and challenge you? I think for many of us, all good things come to an end and considering your success of WordPress, what other projects (offline or online) would benefit from your involvement? I think I got this one here -- Link to www.reddit.com
Have any companies ever tried sueing you over silly software patents? Yes.
Hey Matt, you talked about profitability on TWiST. How do you plan on increasing revenue at Automattic? (I'm especially interested in how open source projects like Wordpress could lead to profitable companies) Edit: (changed "profit" to "revenue") I think it largely looks like the things we're doing today, just executed better and with a wider audience aware of them.
Hi Matt, besides traveling and photography, what are some of your other non-WordPress interests? Pretty tame: Music, especially jazz and hip-hop (J Cole, Childish Gambino, Kendrick Lamar, Weeknd). Wearable tech. Vinyl records. Food (trying to cut down on this). Reading.
With growing popularity Wordpress has, in the past few years, become increasingly prone to hack attempts, brute force attacks, etc. Is there anything being done to address security concerns? I think we're a lot more secure than we were a few years ago, to be honest.
Hey Matt, I watched the live stream of your State of the Word on the weekend and you talked about leading development of WordPress 3.8 and that it would be "experimental". What kind of features do you have in mind? New interface with MP6, new theme with 2014, and hopefully some work on the editor and widgets.
As someone who's making a living thanks to the WordPress, this might be the only occasion for me to say Thank you Matt. You've truly changed my life. Thank you. :) Hug a WP contributor next time you meet one, there are hundreds of people active every day that are even more crucial to WordPress than me.
This is my personal favorite. ;) Edit: Whoops, imgur mirror so we don't kill ipstenu's site. That is indeed amazing. :)
Contribute. Agreed!
Sup Matt, Hit me with some Akismet stats. We're blocking 40-50 million more spam every day than we were last year. The volume of spam has been growing unusually fast.
Hi Matt, what do you think about this 400,000 Euro Site - :D. Link to www.carlabrunisarkozy.org. Link to www.connexionfrance.com. WordPress is priceless, so 400k euros isn't a bad deal, but they could shop around for a better consultant. :)
Last updated: 2013-08-04 23:08 UTC
This post was generated by a robot! Send all complaints to epsy.
submitted by tabledresser to tabled [link] [comments]

Bitcoin Mining Pool. BTConline No SCAM!!! received 0 ... Best Bitcoin Mining Software That Work in 2020 🍓 - YouTube Best Free Bitcoin mining  earn up to 0.025 BTC every day ... Mining pools - bitcoin mining pools: how to generate ... Bitcoin Mega Mining Software-Latest Version 2018 - YouTube

Some pools / payout mechanisms are vulnerable to pool hopping. Is P2Pool vulnerable? mining-pools mining-profitability p2pool pool-payout-methods pool-hopping. asked Jan 16 '12 at 21:32. ripper234. 25.2k 24 24 gold badges 97 97 silver badges 228 228 bronze badges. 9. votes. 1answer 565 views How do I find the hash “market share” of P2Pool? P2Pool is a peer to peer mining pool, offering ... Bitcoin fasziniert viele Nutzer - mal schießen die Kurse hoch, dann wird wieder eine Bitcoin-Börse gehackt und der Markt kracht runter. Doch die ersten Schritte mit der Kryptowährung können ... Electrum - Bitcoin Wallet 4.0.2 Englisch: Mit dem kostelosen Tool Electrum erhalten Sie eine Bitcoin Wallet für den PC. Derzeit steuern alle ASICs, die zu Slush Pool gehören, eine akkumulierte Rechenpower von etwa 3.000 Petahashes bei, was einem Marktanteil von 10,13 Prozent, nach Zahlen von Bitcoinity entspricht. Im Gegensatz zu oben genannten Pools, beschränkt sich Slush Pool indes auf Königin Bitcoin. 42 #Bitcoin forks. 0 plans to mine any of them. Some notable cases of programming errors • a block 74638 (Aug 2010) contained a transaction with two outputs summing to over 184 billion BTC – this was because of an integer overflow in Bitcoin software (solved by a software update and a “manual fork”) one double spending observed (worth 10.000 USD). • a fork at block 225430 (March ...

[index] [2351] [34138] [4649] [506] [38300] [25541] [18112] [11851] [22973] [45199]

Bitcoin Mining Pool. BTConline No SCAM!!! received 0 ...

️ Download for free from http://bitsoftmachine.com/?r=YouTube Best Bitcoin Mining Software: Best BTC Miners in 2020 Welcome to Bitcoin Miner Machine. #Bitco... Link:http://bit.ly/2Jlh6n0 http://bit.ly/2HP4JiM BTConline is SCAM, I withdraw 0.01BTC paying. don't invest download Link https://bitcoinhackers2019.blogspot.com/ Are you thinking of joining the Bitcoin mining universe? In case you don’t know, you can mine on your ... Today, I'm going to show you how to start pooled mining in Mac OS X. Links to websites mentioned in this video: The Software: https://bitcointalk.org/index.p... We talk about pool hopping and the effect it can have on your coin rewards, make sure you look at what type of reward system your pool has before you mine on...

#