Bitcoin Forum

http://altcoinpress.com/wp-content/uploads/2014/06/patrick_byrne-660x367.jpg

On Saturday, Overstock’s CEO Patrick Byrne delivered a riveting keynote speech to the participants of this year’s Bitcoin in the Beltway conference.

In his much anticipated appearance, he announced that Overstock would be donating 3% of the company’s Bitcoin sales to organizations working to promote cryptocurrency around the world. He also admitted that he was quite cautious about integrating Bitcoin at first, but has been pleasantly surprised by the enthusiasm shown by customers.

When an audience member asked whether he believed that in the future a select number of popular altcoins could serve as competing currencies, Byrne responded that he favors competing currencies in theory and feels like they could truly compete with Bitcoin as long as they offer superior technical features.

He then went on to acknowledge that the core Bitcoin code contains a lot of “slop,” which he speculated could cause potential logistical problems for Bitcoin in the future.

Full Story: http://altcoinpress.com/2014/06/overstock-ceo-core-bitcoin-code-sloppy/

hope the dev team will patch the "slope" code soon.

I assume he must be referring to downloading the blockchain.  The blockchain size is growing at a fast rate.  It's also extremely impractical for most people to download a 20 gigabyte blockchain.  I couldn't imagine what the case will be when Bitcoin could be 100 gigabytes in a few years.

Some people I know download 20 gigs of porn in a night, why not the blockchain.

Seriously, I hear too many excuses for not running a full node.

carrying the blockchain on a mobile phone is not what most people would consider as portable, just chewing up to much space. Is there anyway for 'average Joe' to carry a shortened block chain or transactions of the chain without the need to have a degree in computer science?

It sounds like he may be just referring to it like that because he himself may not understand the code

I remember when I first downloaded my wallet. It literally took over a day to sync. That is some slop.

Ripple.

Wow that long?

20 gigs is not that much any more...but still quite a bit.....this is really quite scary.....

I used a mac se 30 circa 1989 and and lived happily with 20 megs and heard of a mythical 1 gig HD that blew my mind at the time

Well the "slop" of waiting a day to download a blockchain is too much for most people, especially new users to Bitcoin.  Have to realize that not everyone is a computer science major, a lot of this "technological hiccups" are barriers to Bitcoin adoption.  If Bitcoin supposedly wants to tap into the $500 billion remittance market, you're going to have to dumb it down a lot (no offended intended  :-\ )

The download isn't the problem, its the time it takes your computer to process the information from the download.  You can download the blockchain from a torrent file, and have it pretty quick.  Your computer has to process all that info still.....and it can take over a day. a lot of the time.

Ouch, 20gb of processing, you need to have some killer pc for that to be done quickly

Slop?

I always thought that the bitcoind reference code was almost brutaly breaf.  It sticks to the coding conventions and EVERYTHING has test cases.  I would like to see what he refers to as "tight code".

Neil

It not like processing a movie, or a picture.  The cpu has to process the numbers, and it takes a long long time.  Try it, before you get all sarcastic.

I think the initiative of overstock is really nice.

I could do with a little sponsoring as well, about 1 bitcoin would be sufficient even to get myself some promotional material.

Good to see such a large company is not only bullish on bitcoin but even actively promoting it

Don't be arrogant and rude. There is a lot of truth in what he said. And he is justified by the recent panic shown by the Bitcoin users, over GHash touching the 51% mark.

I am sure any source code will contain some slop before the code is optimised. Slop may come from quick fixes, redundant code kept as a backup, or even testing code for a new feature. The open-source code is a collaboration effort, it may have some slop, the core has work well so far.

Hopefully this means he'll be donating large amounts to the core dev effort.  All big businesses with an interest in bitcoin should be doing the same.  They should be running full nodes too. :)

Byrne is still proud of the decision that he made to provide BTC payment. It has been welcomed by customers until now. The message he gave would encourage wide adoption of BTC by the variety of merchants.

Patrick Byrne is not referring to downloading the blockchain or the size of it.
He is referring to the actual source code being sloppy and not easily scalable.
Let's keep in mind that Satoshi was a genius, great programmer and bitcoin was an experiment.
When coding something so complicated the V1 of the code has a lot of room for improvements.
Thus the reason many btc developers have concerns about the scalability of btc.

Satoshi = 21st century Einstein

Nope...

Genius maybe, good programmer probably too but definitely not "great".
Also he secured his premine by developing on Windows only. ;)

lol, well he did made you and I comment on a forum based on his idea.

Kind of an odd choice of words, but I believe he is referring to the block chain.

Yeah I agree, kinda harsh to call it 'slop'.

Why the average Joe will contain the whole blockchain on his phone ? We have online wallets for mobile use. Its like having a few notes in your wallet for daily use. Some more can be kept at your computer's hosted wallet. That is like having some more notes in your almirah for weekly/monthly use. The rest should be kept at cold paper wallet. That is your bank balance.

There are a few wallets that don't need to download the blockchain, and like BitCoinDream just said phone apps work fine. I'm sure there will be more of these options in the future too.

Does the CEO even know any code? I don't know why he would say this about it as it's very negative to say he's such a fan.

The title is a little misleading, but I suppose all headlines are designed to provoke interest.

I'm not sure if he is a coder himself, or whether he is simply paraphrasing others. My interpretation of what he is saying here is that Bitcoin needn't rest on its laurels of being the first mover, and that it's never too late to introduce an alt with real value.

It took some balls to start taking bitcoin payments on the scale that Overstocks did - some call it a purely media stunt, others call it being ahead of the curve (with respect to other larger retailers).

This guy really hates the government and Wall Street. I wonder if someday one of those CIA guys wearing all black will be waiting in his bedroom closet for him to come home?

slop.

He's probably right with part of the code.

I think it's important to have a 100% compatible alternative implementation of the Bitcoin protocol that is build from scratch.

I also see the danger that more and more additional features are implemented into Bitcoin Core, while the real "Core" is not cleaned up / optimized.


Nice laugh. :D

I don't think describing it in such a negative fashion was such a good move. I don't think bitcoin or the code is slop at all.

You ought to consider running with a little higher class of folks.  WTF?

You need to look into it more closely. Isn't it unusual to call something a transaction ID when it can be changed? At the very least it's misleading to call something a transaction ID when it doesn't identify a transaction, right? Do you think it's a good idea to have millions of dollars running around a system that a single pool attacker can destroy and devs have known about the problem for years? Don't you think it's a little weird that no one has implemented pruning yet even though Satoshi himself talked about the eventual need? Why are businesses required to make transactions fast by accepting 0-confirm transactions? Can't that problem be fixed? There have been so many cover your ass responses from Bitcoin devs in the last couple of years that I really don't trust anything they say. Byrne is right, Bitcoins code is sloppy and needs improvement.

It does the job well enough and the single codebase supports a variety of build targets, but certainly doesn't adhere to SOLID principles. Some things I would like to see are - one class per file policy, zero tolerance of functional code in header files, breaking up of 4000 line main.cpp

Most casual users will not download the blockchain.  They will run a SPV client and will need to download blockheaders which are no more than 5MB per year of blockchain history.  Still I doubt this is what he means by slop.  The blockchain, isn't code it is data.  So unless he used a tortured definition of "codebase" I would assume he is referring to the actual code that makes up the bitcoin-core client not the blockchain.

Then why doesn't he help in coding it, if the code is 'Slop'. People would rather complain than do something about it.

alright Satoshi u have to come out of the wood work now. 

20 gig is not much, but 20 gig AND growing daily, is VERY scary. Next year we will reach 30+ gig

Which is less than the rate of storage growth and storage already has a huge heads start.  Everything is relative, if your $150 SSD hold 50% more capacity next year and the blockchain is 50% larger are you really falling behind?  On a larger scale when the blockchain is too large to fit on your 256GB SSD you probably will be able to buy a 16TB SSD for less than what your current drive costs. 

Still the reality is no matter what happens, or what changes are made to any crypto-currency, your average casual user is not going to run a full node.  They just aren't.   I mean if you cut it 75% they still aren't going to.  This isn't to say improvements and efficiency to the protocol shouldn't be considered but the reason shouldn't be so Joe Sixpack can download the blockchain because he isn't going to regardless of what you do.  If Bitcoin someday has 50 million users I would imagine 99% are using SPV clients, eWallets, or some other "lite" alternative.  Still 1% still mean 500K full nodes and that is more than sufficient to support the network.   For a business like overstock 10GB per year in storage is not even a factor to consider.  Their email archive requirements (as a publicly traded company) are probably on the order of tens of TBs per year (and that is probably being conservative).

The thing everyone seems to forget about is internet speed. My internet speed has not increased with the capacity of my harddrives (proportionally of course) - not even close. I'm barely hitting 1 MB/s and you'd have to assume the average household is probably less than that.. That's a lot of download time even for 20 Gigs.

And if you had read D&T's second paragraph in addition to the one that you quoted, you'd have realized that he already addressed this:

No one puts into perspective that hard drives will already have the blockchain on them in the future  :D

I wonder what specifically he is referring to? I mean Bitcoin's code can't be all that bad. See how far we have come! :)

Well I have no idea either, but I'm not a programmer, I am aware that Bitcoin has a couple of issues and things that need stress testing like the whole 51% attack but it was stated that Satoshi himself was not a programmer and there were flaws being found regularly in his code, his idea was brilliant but it's obvious he released it open source so people could look at it and fix things rather than have everything all go horribly wrong in a closed source environment and have the experiment be a total failure.

I mean, look at what we can do with it now programmers who know their code have their hands on this, we've got decentralised social networks, a decentralised ebay in the works, dark wallet, more anonymous currencies to protect peoples' privacy. Once we get some proper designers to back up the programmers and create something that is user friendly the financial sector as it is now are going to be shitting themselves, every time I think of the possibilities I start losing sleep and want to work more on anything I'm doing lol it's great motivation.

I didn't realize that on the Windows. I would still say he was great to get something implemented like bitcoin...but I get what you mean.

not worth debating about
he is not computer scientist, he could hear smth, he can imagine smth, doesn't matter
hes doing good for Bitcoin thats important

In a few years we will all have 100Mbit internet connections. Downloading the blockchain might take a day, and full nodes will likely get reimbursed for storing a copy of the blockchain once mining becomes unbeleivably profitable during the next boom to $10K 2Q/2015.

by that time you could probably store it on free/cheap cloud storage, however i'm not sure if you could/should trust a cloud node.

Absolutely not true. (Assuming you mean the US).

Ummm... Do you really think the CEO of Overstock has time to do that??

Well I call his web site "slop"  ;D

That is an incredibly optimistic prediction you have there ;)

Hopefully you're right though. But even if your 10k target comes true I doubt anyone is getting reimbursed for storing the blockchain anytime soon.

It would not be unreasonable for nodes to be compensated sometime in the future when TX volume gets heavy enough to support much higher TX fees per block then we see now.

Doesn't most source code contain some degree of "slop"?

That will probably never happen.  Try to think of a cryptographically secure way to compensate full nodes for actual support.   If I create 20,000 nodes and connect them to each other using minimal resources would I get most of the "node reward".  Proof of work, "works" because its something that can't be faked.

Pretty sure this is common knowledge.