Topic: Blockchain is 360GB and growing, can it be consolidated? (Read 872 times)

Dude, I'm obviously talking about running Electrum server (ElectrumX or electrs). Running either of those allows people to use SPV wallets without trusting the preset server; thus providing more services to the Bitcoin network. I was just showing two examples (Electrum & Lightning) of software that some may define as necessary to achieve 'full functionality'. This was all just to show that a definition of 'full node = full functionality' is super vague, since there is no hard set definition of 'full functionality' itself.

All in all, this whole argument is pointless, since it's obvious (and has always been) that the definition of full node is fully verifying node.

For example, take this page from December 2014:
You're not a 'top guy' just because you run a full (pruned or not) node

no its not about the noob GUI..
its about that since pruning even became an thing entirely. the whole definition has changed.
take cores system requirements of versions pre prune invention date. hard drive specification for running a core full node was excess of blockchain size. now since pruning that 'min system requirement' has altered.


electrum? oh please!(facepalm). everyone refers to electrum as lite wallet/spv.
(just google 'electrum litewallet' or google 'electrum spv' and you will see millions of references)
also
other wallets and nodes that are not fully verifying or not archiving are classed as lite/spv/thin

seriously.. you really thought electrum was a full node?(facepalm)

not every wallet/node should be classed as a full just to give noobs the deceptive and false belief they are helping, when they are not fully helping. its ok to use bitcoin without being a full helper. be ok that you use bitcoin without fully helping the network if you are choosing your lifestyle doesnt require/want to help.


ok. this topic has derailed into debate that pruners and litewallet users want to call them selves full noders just to have some namebadge of importance, like they are the top guys and supporting the network... shame though that reality has shown they actually chosen not to be supporting the network.
hey its ok that you chose not to support the network. just be honest with yourself about your choice and be happy that your not.

anyways yea this topic has derailed into some social debate of redefining old terms to fit noobs sentiment of false level of network support. by trying to class bitcoins blockchain as non-important and insignificant..
such a shame that soo many people think blockchains are insignificant.

im moving on. this topic is dead.

edit:
to answer below.. as he still cant quite grasp it.
"electrum server" is not a stand alone software that works byitself. it requires also using bitcoind.

even when electrum server is running but not doing all the "backbone of the network" stuff. its not a full node
emphasis backbone of the network.
(unless you think theymos is wrong wrong too.. and you as a lite wallet user know better than theymos)

as for previous stuff about nodes that support alternative networks all in one. those are called 'master nodes'

and i never wanted or tried to claim myself as important. im just shocked that some people want to pretend there is a social hierarchy of non-devs where they deserve to be top guy amungst the dev group, while not supporting the backbone of the network.
my premiss is not to be a top guy or a 'buzzword definition election committees director'. im just calling out reality to people that are deceived or have been deceived. its called just spelling out the obvious. it requires no rank system.

maybe mentioning 'backbone of the network' 3 times isnt subtle enough hint about the wiki definition

If you're referring to my account age, maybe consider a person can be in Bitcoin many years before creating a Bitcoin Talk account.

Oh, you're such a self-proclaimed know-it-all but use Bitcoin Core only via GUI? I recently found out that the GUI changed; however never used it except to do the IBD on a fast computer once. All my nodes run on Linux CLI as a service (as they should for best uptime and best serve the network), and in CLI mode it's still default not to prune. Pruning is deliberately for the 'n00bs' that want to run Core on their laptop and similar. So it makes sense that for personal machines which aren't always on and such, people might be more disk space restricted, whereas someone setting up a dedicated Bitcoin node (who uses CLI) most probably has enough disk space allocated for running it in full archival mode.

Even before pruning you could be not fully verifying or fully verifying. That's why the terminology exists longer than the pruned mode itself.

Now tell me: what is 'full' in your opinion? Fully archiving? What about Electrum? How can something that does not have all the bells and whistles be considered full? So in your terms, a full node must at least contain: Core with full archive + Electrum + Lightning; now when we get into LN you have to define how many and how large channels you need to be considered 'full'. I could argue that you're not a 'full node' if you don't offer me to transmit 1BTC to another person through your channels. We can go down this hole further and further; such as requiring you need to offer all services both through Tor and clearnet etc etc.

In the end, your definition is completely wrong (and has always been! full refers to the validation of the blockchain...), and I never heard anyone use your definition in almost 10 years now that I speak to people about BTC. It would never work anyway; because the definition of what is 'everything' (which seems to be your definition of 'full') always changes e.g. with the introduction of LN. There may be people considering you not running 'everything' if you're not running the sidechains as well, for example.

i understand that certain self-assumed 'community' electorates who feel empowered that they are the voice of all, think that full node is an invented term after pruning became a thing and refers to nodes that only verify.. just to ego boost that their less than offering is still felt as full offering. although they lack responding by mentioning the term of what a node is called that does do all the jobs(true full).

but by the looks of their forum activity numbers they have not been around since the days of proper full nodes terminology creation.. you know the days before full nodes even had the ability to prune and where by reality was that archiving by default and un-optionable was actually part of being a full node.

i understand in recent years the paradigm has changed where pruning is now the default and archiving is now optional.. . but that does not change what full meant right from the start(when full nodes meant full job).

as for quoting wiki.. you might want to read a little more then your prefered snippet

again even the wiki is a few month out of date because now the default is that prune is ticked and to now archive, the option needs to be unticked. so as you can see things have changed where definitions have moved the goal post RECENTLY

it only became efficient to not archive SINCE pruning became an option. because full nodes before pruning was even a thing, was default store entire blockchain because there was no choice within a full node not to archive.

so yea. your basically saying the default definition of storing the blockchain has been re-defined since pruned was invented. and so you now want to say and pretend its always been a less=full scenario even when real history shows that 'full' was in its original terminology full=full

..
with all that said. if you want to be a pruned node. thats ok for you. just be honest that you are a pruned node and happy being one. no need for deception. no need to boost your level of involvement while not actually being able to perform the higher level of involvement. just be honest and happy with your choice to be limited part of the networks security and decentralisation. then be happy to learn what that limited level of network support your node is or is not helping with. and not assume you are fully supporting the network in full, by simply redefining your version of 'full'

Here's the thing. In Bitcoin wiki and pretty much anywhere else, full node is used interchangeably with or even defined as fully verifying node.
You're actually stuck on believing 'full' means 'full features'; but that's not the definition. 'Full' in this context is more a short-form for 'fully verifying' instead of meaning 'all features'. By your definition, I could say, if you don't run Electrum and Lightning on your node and don't have a giant txindex, you're not providing 'full functionality' to the network. So as you see, such a definition would be very vague: what is everything? This is not clear-cut. But anyway: full = fully verifying. It's defined like that, after all.

Didn't you know franky reindexes his blockchain folder every day? No really, it makes no sense. Verified once, it's verified, that's it LOL; similar as if someone hashed the same thing 1000x and expected the result to change after some while.

---

Replying to the underlined: actually this is exactly what you do during the IBD of a full node (pruned or not - no difference). You fully verify every single utxo from the beginning to the current state. After it's verified by you, you're good. There is no benefit in repeating it later on. That's why the old blocks can be discarded after verification. Of course, the big downside of not archiving all blocks is you can't seed to new nodes. But the direct benefit (security, privacy, ...) to you as a user is exactly the same if you prune or not.

Paradoxically, pruning and validating only the current blocks may be sufficient for security, but only if it doesn't become too common.

It is better not to provoke the evolution of the protocol towards increasing scalability, which would not only threaten decentralization, but also security.
There can always be a bug that validators cannot detect and that will not be disclosed until a full chain audit by inquisitive, independent researchers.
As was the case with the Stellar coin once
https://coinmetrics.substack.com/p/coin-metrics-state-of-the-network-37f

ok there we have it.
it looks like tromp took a secret vote and elected bitcoin full nodes EG 'core' as just a litewallet, thats job is nothing to do with decentralising the network by offering IBD, getdata to peers.
(facepalm)

[sarcasm]
guess thats case closed then..
core devs dont need to code any parts of core relating to archiving the blockchain or providing peers with blocks older than the latest 288blocks

guess tromps community decided. bitcoin does not rely on blockchains. death to the un-needed blockchain i guess.. tromp has decided that blockchains are not important.

praise be tromp, the election committee director, the master of what core is responsible for, praise be
[/sarcasm]

ill leave tromp with one question.
if full means only certain features. then what terminology is there for a node that offers not just verification, but the full bitcoin offering of features such as verification AND archiving AND full peers services....

here is the funny thing. 'full node' was a term that existed before pruning existed. so pretending that pruning is full, is a false idea and a rewrite of history

lets await tromp to take another secret poll and come up with a new terminology
(isnt it weird that in my first post on this topic i said that too many people like to argue buzzwords, rewriting reality and thus making the buzzwords meaningless in their definition)


if you have the blockchain. (important) you can check utxo to find its block, then its blockhash, and compare against the block contents. find the transaction in question, see it exists. then check the entire chain of blockhashes match up to what they suppose to be, and yep. you can see if there were edits or not.
EG edit one transaction from 2010 edits that blocks hash and all blockhashes there after meaning the latest blocks hash wont be the same as your edited latest blocks hash

you cant do all this if you prune, unless you continually prompt a third party to re-send you all the data again
(isnt it weird that one of the first rebuttals to me was that certain things didnt need explaining because it was assumed they knew the basics already.. guess not)

anyways, seems this topic is going no where fast where buzzwords dont mean what they intend because half a job is a full job, as voted.  and blockchains are not important and not part of the protocol that fullnodes are involved in. as voted by the established community (facepalm)

bored now. moving on

Why should you verify an already-verified set of outputs?

That has nothing to do with pruning. Alright, let's take it in another way:  What if a hacker replaced your UTXO with his UTXO while you're running a full node? Once you had verified the outputs you'd have no way of knowing if your verified outputs changed.

The only difference is that full node does the IBD once.

Yes, pooya, along with most others in this discussion, as well as the vast majority of bitcoin experts,
is stuck on the notion that full node is short for fully verifying node.

You are one of a small minority of people that have miraculously overlooked this established terminology.

its not about being forced to.

lets use a mcdonald employee analogy. maybe something more understandable for some
its about not pretending you deserve a 5star gold badge, when you only offer a 1 star service. its about being honest with yourself and others that if you choose to only offer a 1 star service, accept that fact that you are a 1 star service member. be ok that you are a one star member. and stop trying to pretend your 5star contributor.

it doesnt matter if you make 1000 more posts saying the 1star training is all that is needed to be 5star.. reality is there are more things then the 1 star offering involved to be given the 5star badge.
yes 5star contributors also do 1star tasks... along with many other tasks. but that does not make a 1 star member a 5 star member just because "but 5star guys does my job too'. the 1star member still needs to do more then a 1 star task to be promoted. ignoring all the other jobs a 5star member does, pretending they are insignificant. does not earn you a promotion
just because your a burger flipper and then you see the manager also flips burgers. does not then mean you get to call yourself a manager because you both flip burgers. the manager does other tasks too. which is why he is the manager


its about stop deceiving others that your a full service 5 star contributor. its about realising what you are not contributing to, and thus accepting that lesser position and the hazards to you and others  for not being able to do it all.

its simply that you can be a pruner. but call yourself a pruner not a 'full node'.
and no, dont demote a full node description to being 'just verification'. because a full node is not just about verification. there is alot more too it. and if you are still unsure of all the features.. learn

there are many many many things that involve being a full node. not just that at some point in time you verified something.. because months later that set cant be trusted
it has no UTXOset hash to compare to other peers UTXOset to verify the set has not been edited

nodes with nothing more then the lastest 288 blocks and a utxo set cannot in their true heart be 100% sure that their utxo set will never be compromised. they cannot also truly know for sure if it ever was. nor can they help other nodes.
maybe people need to revise and study and research what blockchains offer that spreadsheet accounting doesnt

its not about forcing people to be full nodes. its about getting pruners to accept their limited ability due to their own choice to limit what they do.

sorry but pruners do not get a 5star gold badge. they have not earned the promotion.

again to try bringing the discussion back on topic to the topic creators question
having a whole network of nodes with just 288 latest blocks and then a dataset of unspent values. puts the network at big risk. blockchains were invented for a good reason, doing something that spreadsheet balance accounting cant
it doesnt matter if at some point in the past you checked the balances were real. once verified if you are only storing a balance sheet of what you believe is still unspent. that sheet can become compromised and you have nothing on your system to check it against.
a dataset of unspents should never be used as "same security as blockchains" because it lacks the evidence of origins to ensure no compromise happened. a dataset of unspents should only be used as a quick index to find the real unspent on the actual blockchain. to then double verify the unspent is real when new transactions appear.

If someone is able to maliciously add a UTXO that is not in the UTXO set, they have potentially (and likely) added malware to your computer. You would not be able to trust any output your computer provides. If your computer is infected with malware, validating the entire blockchain (again) is not going to show you that you have received a valid transaction.

seems pooya is stuck on the 'only verification'=full features..


actually although you think you have a full verified UTXO set of all unspents from block 0 to say block 712k.. can you verify that today? right now! without requesting a new IBD to compare it to???
(answer is no)

heres the thing. you might have been online for 12 years and stayed synced and decided to prune say 6 months ago. you might have just come online 6 months ago and IBD and then pruned that month.
but here is the question for you.
what if your computer glitched 4 months ago or a hacker got in and added his own false utxo to your utxo set.

can you now with just your UTXO set verify today that no one slipped in a utxo into your set.
no. you cant
can you prove that a tx in your utxo set is today valid, when some chain re-org happens.. no you cant

your UTXO set is only as good as the minute you verify it. anything after that, the utxo set by itself cant be fully trusted. loss of data, virus, hacks. and boom your utxoset is compromised

yet having the blockdata to support it allows you to seek the utxo from the set(like reading an index of a book). and then compare it to the blockchain.
a utxo set is only useful to save having to find the tx via reading all blocks every time a transaction appears.

its just like an index of a book to save you having to read the whole book every time just to find a reference.
a utxo set is not good for verified accounting. its not good for security. its not good for proving the transaction has a real utxo.. because your utxo set can be compromised at any time and there is no way you can tell

again there is a reason why blockchains become popular compared to spreadsheet balance accounting

I guess you can call a pruned node whatever you want. However, a pruned node provides all the same security that a full node provides, and that is what is important to the end-user. The same is true if a non-pruned node disregards signature data after it has validated the signature of a transaction.

Unfortunately, we don't live in a utopia, and users will prioritize their limited resources to things that are important to them. If someone does not have the resources to store all blocks, or to send all block/transaction data to the internet 8+ times, they should not be forced into doing so.

Even if a centralized service were to "sell" the ability to download the entire blockchain for new nodes, a node would be able to validate the entire blockchain, and know if they are receiving the correct version of the blockchain because they would be validating all blocks, including the POW, and would be able to validate the POW of new blocks they receive from other nodes. It is not possible to fake POW of old blocks.

A pruned node keeping 5 GB of the most recent block data would be storing a minimum of 1,250 blocks if they kept the block and signature data. This works out to approximately 8.5 days worth of blocks with no difficulty growth. A reorg anywhere near that size (even 1% of that size) would signal very serious problems, that far outweigh any potential issue about not having enough nodes.

I made no such assumption. Many merchants run their own nodes (full or pruned) and many go through third parties but almost all those who need some sort of "taint analysis" rely on a third party (like chainanalysis) because this is despite their own preference like an exchange that is forced by the regulators to implement such nonsense.
Implementing such a system themselves is a burden and would cost more than using the third party not to mention that the responsibility would be with that third party not the user which is favorible.

That's exactly my argument. When your client downloads and verifies everything from block 0 that makes it a full verifying node. Whether your client discards the blockchain later (and stores less blocks) doesn't change that.

verification of transactions is only one part of what a node does.
just because its the one part you only care about or the only part you want. does not mean its the only thing a node does
a full node does many many many other things.
and needs to perform and offer ALL those MANY things to be classed as full.

assuming a node just has to verify to be "full" is not full

if you think that being "full" is only about verification. then you need to redefined your assumption of what being a full node actually involves.

the real challenge seems you be certain peoples grasp of the concept of full, as oppose to 'less than, but pretends to be full'.

getting back to a technical discussion of this topic after the myriad of social drama discussion about less=full.

having the full blockchain means you can contribute to the network to offer(along with other things) not just Initial block download to peers, not just 'getdata' of random blocks, not just ability to sync up more than the last 2 days.
also theres are things like having the blockdata distributed so there are many copies in many countries to avoid any country 'pulling the plug'. having the data distributed to allow more choice of peer connections. having the data distributed so peers dont rely on a small cluster of attackable nodes

if the whole network (as the topic creators thoughts mention) only had the last month of transactions and a (UTXOSET) list of unspent transactions. where there was no taint or proof of that values initial creation from its original coin reward. then that puts the network at risk of UTXO virus attacks to rewrite sets. it puts the whole true accountability of all true coin at risk. along with other problems.

there is a reason that blockchains have become a success compared to spreadsheet accounting of 'live balance'. no one can just slip in a transaction with an old date and pretend its valid simply because they managed to send everyone their edited utxoset

its ok you you personally want to prune for any personal reason. just dont then pretend to be a full contributor to the network and dont tell others they are fully contributing when you talk them into pruning. just be honest and tell them you and them will be offering less then full network contribution and less than fulll features for you and them you take advantage of

It's really a challenge to have a conversation with you, franky. I confess.

Running a pruned node means you don't have to trust anyone! As I said, whether you run a node who does keep the chain or not, you validate every single transaction you receive from your peers. The system works trustless-ly. What is this you dislike pruned nodes so much?

Huh? Don't pruned and full nodes differentiate only in the size of the storage they require? They both make a full verification.

(my last post on this discussion of less or full, as its a shame that i even have to debate such a basic concept to people that rebutt that basics dont need to be mentioned because it should be assumed they know)

to pooya:

you are right merchants need extra code to sieve through the block data to show the taint path. but its impossible to use their extra code to read the blocks to get the taint path of new incoming transactions unless... wait for it. they have the blocks of previous transactions to check the path.
you will be surprised that any decent merchant that cares enough about taint and customer transactions will actually want to be a full node. and have extra code ontop to do more beautiful things with the block data for their own independent purposes.(emphasis fully independent)

much like any decent business will have its own spreadsheet software and tax accountant software, instead of relying on 3rd party accountants

so try not to assume that all merchants are lite wallets and use third parties

after all the whole trust a third party thing you assume all merchants do, is not what bitcoin is about.
and its a shame that you think that users should trust third parties and use them because a decision to prune means they have to use third parties.
i think you shot yourself in the foot by admitting that those that do prune should and do trust third parties. it kinda rips apart your whole argument about 'full' if you are saying pruned nodes then use third parties for things a real full node can offer within the merchants own system..
.. doesnt really sound much of a rebuttal for your version of 'full' if you admit that they need to use third parties.
...
you did indirectly say selected features. because by your own admission you are not archiving so you have selectively chosen that certain features and services and options of a node are not to be used, hense no longer full
if i was to number all the features and the total was say 1037. defining FULL node. and your node wasnt offering all of them. even say 1036. then thats not full. thats 'less than'

by your own admission you say:
"you think I disagree with "pruned nodes contribute less to the network", but I'm not"
but then you pretend pruned nodes dont offer less because you believe pruned nodes are full.
less is not full.. less is less

if you went to a bar fo a pint of beer. there are 14 barman. the first you approach poured a pint. and then spilled 95% of it, but still gave you the glass of the remaining 5% froth. he can prove he poured a full pint into the glass. but is only offering you the 5% froth. would you accept that the glass is still a full pint. would you accept the barman offered you a full pint service.

in reality, you should think that the barman is useless if everytime you ask for a pint you only get 5% froth and go find another barman that can offer a full pint service. a fully trained barman that does contribute to the bar
in reality, you might even complain to someone that the barman is not contributing to the bars success.
in reality, you would NOT go on some site and leave a 5star review for the useless barman saying he is a fully skilled barman.
what if out of the other 13 barman you had the same problem with 10more barman not being fully trained..
and its impossible to get served by the 3 fully trained barmen because they are super busy serving regulars that came before you that know which barmen are fully trained. and also the queue of customers that got a bad service and are now queuing up with what little talent remains in the bar. making your experience miserable and slow, making you not want to have a pint altogether and making you not even want to be a fully trained barman in a future career as you see the pressure the 3 fully trained guys are under

as for your last statement that you think the scenario should be 'better to have 3k full 11k pruned rather than just 3k full'
actually no. because in the 3k full 11k pruned. those 3k are under more intense data pressure contributing to leachers. meaning the other peers within the 3k full nodes are getting slower speeds amongst themselves
and those leachers are not then contributing to help new full nodes get their IBD either.
the real scenario is to grow the full node population.


because yet again it needs to be mentioned. by trying to deceive people that being a prune is the same network contribution as being full, can and has made some full noders downgrade. thus instead of having 10k full nodes. its now ~3k.. because. wait for it.. people(like yourself) think that there is no harm and no loss of contribution to the network by being a pruner

sorry to inform you again. pruned nodes dont fully contribute to the network, they actually put more pressure on the ones that do.

so lets just stop the flip floppy changing the answer just to have a reason to debate.
make the decision
a. pruners dont help contribute towards the network and being pruned lessens what you can do/offer
b. remain in the disbelief that being pruned is full ability and full service offering to the peer network

i dont need the answer. the answer is for you to settle your own mind with.
less is not full

anyways, enjoy your 5% froth. just dont tell people your drinking a full pint. your a lightweight, cheap date

Wrong example because just having historical blocks isn't going to tell them the "taint" they also have to have implemented a taint analysis code which is not something merchants ever do. Instead they simply use a centralized third party.

I never said "only a selected feature" I said "full verification" makes a node full node not "full storage" and my reply here doesn't contradict that.

Your confusion and defensive behavior seems to be because you think I disagree with "pruned nodes contribute less to the network", but I'm not. I'm saying that you can't say they are not full nodes just because they don't store historical blocks.

Most important thing about pruned nodes is that (using the numbers you posted earlier) it is so much better to have a total of 14k nodes (3000 full archival nodes and 11k full pruned nodes) than to have 3000 full nodes in total and nothing else contributing to network.

yes blockchain data.
this is not some science website that is asking for a thesis on particle manipulation related to the suns radiation.
of course when discussing bitcoin nodes/peers and the data they receive relating to the blockchain, that data means blockchain data(whole or part is no specific).

this whole topic is about blockchain data.
it does not mean data involving the spectral range of ultraviolet radiation from the sun. or any other data imaginable. it means on this website about bitcoin. in this category of technical discussion about bitcoin. in this topic of bitcoins blockchain.. yes data=blockchain data
1+1+1=3

now stop the social drama tactics.

if you are trying to play chicken(2) and egg(1)
the answer is egg.
step 0 duck+turkey create an egg(1) egg hatches and its a chicken(2) chicken lays an egg(1) another chicken comes out.. and so on. .. egg then chicken(rinse and repeat)

so to get data(2) someone needs to have data(1)... 1 comes before 2
pools create the data(1) peers get data(2) peer stores data(1) other peers get data(2)

if your pretending to be an egg farm but none of your chickens can produce eggs to feed customers. your not an egg farm.
just because you received an egg and verified for your own purposes that it is indeed a chicken by letting it hatch in your farm. but then not allowing that chicken to then produce eggs for others. does not make you a full egg farm

As n0nce pointed out, it's not about poking. I barely understand what you write and you should consider it justified if one does not understand a part of your text. Besides simple grammar mistakes such as improper usage of punctuation etc., some of your sentences don't even make any sense at all.

It seems like you're writing your posts in a rush while you want to shut me up.

And again, I don't understand what you're trying to mean in here:
If I don't understand that to get data, other people need to have the data to give what? Then you conclude that others need to have that “data”. Should I assume the data is the block chain? Then, of which process is this the step one?