The rally is over

[johnyj]

We'll see 30 or lower very soon

Don't mistake me, I regard this fork accident a very positive feeback to the community, now everyone will take this chance to see lot's of interesting aspect in bitcoin world, which are not that clear or almost hidden before

[dave111223]

Once Mt. Gox opens BTC deposits then the fun will start, big money to be made and lost IMHO.  I'd imagine a MASSIVE sell off followed by a big buy back...hoping I can buy back in at $5   Get ready

[checkers6676]

hoping I can buy back in at $5  

hahaha

[Ban Curtain]

We'll see 30 or lower very soon

Don't mistake me, I regard this fork accident a very positive feeback to the community, now everyone will take this chance to see lot's of interesting aspect in bitcoin world, which are not that clear or almost hidden before

Just another "expert's forecast" from expert who hopes to make prices lower

[thefiniteidea]

If anything, the price will go up.

This has been incredibly well handled.

Anyone remember the equities bot glitch that caused the Dow to plummet $1000 in one session?

Shit happens all the time everywhere. We're only human. It's how you react to problems that show how stable you are, not how few problems you have to begin with. It only took one iceberg to sink the unsinkable Titanic. Bitcoin is only going to be stronger and more prepared moving forward after this dust settles.

This whole thing has been very positive in my view.

[johnyj]

If anything, the price will go up.

This has been incredibly well handled.

Anyone remember the equities bot glitch that caused the Dow to plummet $1000 in one session?

Shit happens all the time everywhere. We're only human. It's how you react to problems that show how stable you are, not how few problems you have to begin with. It only took one iceberg to sink the unsinkable Titanic. Bitcoin is only going to be stronger and more prepared moving forward after this dust settles.

This whole thing has been very positive in my view.

Ture, the response of the community is very fast and efficient, and this is largely due to the majority of pools can be coordinated quickly.

But there are still several questions to be asked, I'm going to sleep now, take them tomorrow 

[dave111223]

Gentleman start your engines, blockchain 0.7 is now main and mt gox will soon be open....sure all the hardcore Bitcoin users will know bitcoin has been given a test and passed... as such we will be buying back in....

But i think there will be quite a few more causal traders and who will waking up to stories of "Bitcoin is broken"...

I have to admit when I saw that big red "Critical: Current Bitcoin network fork" message at the top of the forum and messages on all the exchanges/blockchaininfo about client bugs etc...I had to change my underpants.

[RodeoX]

Bold statements show only ignorance. I mean if I say "Bitcoin will break $50 today!" does that counteract the OP's prediction and cause the price to become $40?

Well, I just tossed a coin and it looks like the OP was right. I stand corrected. 

[johnyj]

Wow, the market is so bullish and it just ignored any negative news for now, but we have seen an example that how a hard fork will quickly destroy people's confidence in bitcoin

Imagine that core devs insisted to let 60% of miners /mtgox /merchants running on 0.8 and urge the rest of miners upgrade, then the hard fork would happen for real, and all the transactions happened on 0.7 chain will become a huge topic to debate for the coming months and the price of bitcoin will at least cut by half, and maybe never recover, since by then both chains would claim they are the genuine bitcoin

Anyway, the community fell back to 0.7 chain immediately, this showed they can act professionally in a crisis and maybe that is why the price is stablizing now

[thefiniteidea]

...the community fell back to 0.7 chain immediately, this showed they can act professionally in a crisis and maybe that is why the price is stablizing now

Correct.

I'm obviously one who see's this as positive news, not negative. Something bad may have happened, but bad things happen all the time and there is nothing we can do about it. No realistic amount of testing would have caught this weakness. It's how the community handled this situation that proves how strong it is. Barely any press on it, and the exchange rate continues right where it left off.

However, we do seriously need to focus on the max block size issue now and ensure we create a flexible, scaleable, and balanced system. If we can get past this issue, then we're over the hump.

If not, then yeah. The rally is over.

[notme]

...the community fell back to 0.7 chain immediately, this showed they can act professionally in a crisis and maybe that is why the price is stablizing now

Correct.

I'm obviously one who see's this as positive news, not negative. Something bad may have happened, but bad things happen all the time and there is nothing we can do about it. No realistic amount of testing would have caught this weakness. It's how the community handled this situation that proves how strong it is. Barely any press on it, and the exchange rate continues right where it left off.

However, we do seriously need to focus on the max block size issue now and ensure we create a flexible, scaleable, and balanced system. If we can get past this issue, then we're over the hump.

If not, then yeah. The rally is over.

A test case with a maximum size blocks made up of minimum size transactions would have caught this.  We certainly need a test case like that going forward, especially if we are going to raise the block size limit.

[RodeoX]

ya know, I'm not really concerned about any loss of confidence among people who don't understand bitcoin. It only take a few who do understand that bitcoin is not "broken" to buy the price back up.

[Technomage]

The market seems unphased and very mature. There has been plenty of opportunity to sell if you want to sell, bulls are in control now. The consensus about the bug seems to be that even though it was serious, it was handled in a very professional way which in itself increases confidence.

[zkay]

The market seems unphased and very mature. There has been plenty of opportunity to sell if you want to sell, bulls are in control now. The consensus about the bug seems to be that even though it was serious, it was handled in a very professional way which in itself increases confidence.

I just doubled down. Let's hope you're right.

[humanitee]

The market seems unphased and very mature. There has been plenty of opportunity to sell if you want to sell, bulls are in control now. The consensus about the bug seems to be that even though it was serious, it was handled in a very professional way which in itself increases confidence.

I just doubled down. Let's hope you're right.

Record amounts of fiat sloshing around Gox.

I think you made the right decision

[Ichthyo]

No realistic amount of testing would have caught this weakness. It's how the community handled this situation that proves how strong it is.

A test case with a maximum size blocks made up of minimum size transactions would have caught this.  We certainly need a test case like that going forward, especially if we are going to raise the block size limit.

It's not that simple. If I recall correct, it was about using a huge number of inputs in a certain pattern, not the overall size, not the overall number of transactions. Plus it shows up only on some BerkelyDB library versions.

Bottom line is: any automated feature-based unit and integration testing is necessarily focussed on how the developer intends things to be, and thus will never be able to catch all the cases which arise from connecting things in unexpected and never intended ways.


Thus, well it is a good idea to increase the unit and integration test coverage, which is (as I understand it) rather just "average level" for the reference implementation. However, improving, developing and rehearsing mechanisms of crisis reaction is similarly important, if not even more important than test coverage. Since we will certainly encounter all the really interesting defects in real world usage, not in the testbed.

[Qoheleth]

Bottom line is: any automated feature-based unit and integration testing is necessarily focussed on how the developer intends things to be, and thus will never be able to catch all the cases which arise from connecting things in unexpected and never intended ways.

That's why, ideally, you have an independent QA team - who has no expectations of what was in the developers' heads - doing the integration tests.

Of course, in a crowdfunded project like this (as compared to a start-up or corporate R&D project), an independent QA team may be an unaffordable extravagance, barring a community fundraising effort.

[Richy_T]

I may be missing something but it seems to me that when a new version is released, *nothing* it does should be considered valid until there is a high degree of confidence that it is performing as expected. Rollback to 0.7 should have been the default and anyone upgrading to 0.8 should have been made aware that should it generate problems, any output would be regarded as faulty and ignored. If the blockchain continues to match then each block is accepted as valid one at a time. If the blockchain agrees for a long enough amount of time and no issues show up, at some point (Though I know there is no central authority but surely consensus is possible) the new version is accepted as the default and should any discrepancies show up, it is the fork with the new version which is taken as valid and people running outdated versions who are regarded as invalid.

I'm not sure. Maybe it is too much centralization. But hopefully something like the above could apply going forward. Though for preference, maybe some kind of fork resistance could be built in to the protocol.

[Qoheleth]

I may be missing something but it seems to me that when a new version is released, *nothing* it does should be considered valid until there is a high degree of confidence that it is performing as expected. Rollback to 0.7 should have been the default and anyone upgrading to 0.8 should have been made aware that should it generate problems, any output would be regarded as faulty and ignored. If the blockchain continues to match then each block is accepted as valid one at a time. If the blockchain agrees for a long enough amount of time and no issues show up, at some point (Though I know there is no central authority but surely consensus is possible) the new version is accepted as the default and should any discrepancies show up, it is the fork with the new version which is taken as valid and people running outdated versions who are regarded as invalid.

LukeJR was already doing something like this. Apparently Eligius runs two Bitcoin clients, and in the case of a contradiction between them goes with the more "mature" version.

Unfortunately it's hard to incorporate such logic into the client itself - miners have to do it manually right now.

[BTC Books]

Got to agree that the rally is over. Was fun while it lasted but things are going to die down for a couple of months now.

Don't think so...

I'm pretty convinced that there won't be a weekend dip.  I think the people who typically sell into the weekend's low liquidity (never could figure that one out...) already sold into the fork.  And now there's all that cash floating around at Gox.

If it goes up on the weekend, next week should be stellar.  And if next week turns out like I think it might, the fork will be forgotten by Wednesday.

We'll see though.