Page 2 of 3

Re: Hardfork 2016 Value Size Limit

Posted: Sun Feb 07, 2016 12:45 am
by biolizard89
johnc wrote:
Not having to depend on "import" for space so much would also simplify the implementation of Namecoin services.
If data is allowed on a name doesn't mean that people will use it for spam.
Just look the namesquatters, they just post a bitmessage address...

Services like Onename were not scalable, we can't have everyone's id on chain, eventually it becomes expensive.

However if Namecoin only allows to store a tiny amount of data what is the point? Bitcoin does that already!

Well, to store all data used on a regular DNS service. You can point to dns servers or just include all the info onchain. Making impossible-to-shutdown domains a reality, which is the point of Namecoin.

Unlike the onename case, not everyone needs a domain name, so there will always be enought namecoins for domains, as unlike bitcoin, we don't expect everyone to register domains, but just to be able to access them, read them.

Bitcoin=/=Namecoin
It's worth pointing out that more research would be useful on how exactly identity use cases and DNS use cases differ on scalability. But the research I've done on the DNS use cases suggest that DNS is not too far from the upper limit of what we can handle. And that research assumes that people won't be doing things like on-chain dynamic DNS. My research is fairly preliminary (it makes up a lot of numbers, which I think are within ballpark but are not backed by empirical evidence), but the solution is to do more research, not to discount research as a useful tool here.

If anyone thinks that storing 4 KiB GPG keys in the blockchain is sufficiently scalable that we should change our consensus rules to facilitate it, I think the burden of evidence should be on them.

Re: Hardfork 2016 Value Size Limit

Posted: Sun Feb 07, 2016 2:24 am
by johnc
How many people know what GPG/PGP is?, to begin with...

Most likely only root authorities like institutions and developers (ex: Satoshi) will need to have (and maintain) a key onchain. Maybe journalists like Assange?.

Re: Hardfork 2016 Value Size Limit

Posted: Sun Feb 07, 2016 2:36 am
by biolizard89
johnc wrote:How many people know what GPG/PGP is?, to begin with...
100% of people who use email should be using GPG. That's the usage level that should be assumed if we're going to evaluate scalability concerns.
johnc wrote:Most likely only root authorities like institutions and developers (ex: Satoshi) will need to have (and maintain) a key onchain. Maybe journalists like Assange?.
All of the users you mention would be just fine with a GPG fingerprint, which already works fine without a value size increase. Taking the example of Assange, WikiLeaks already maintains a Tor hidden service, which could easily serve the full GPG key (I assume they already do this?).

I also wonder whether ECC GPG keys would work. Anyone know whether ECC GPG keys would fit into a Namecoin value? Or perhaps something post-quantum? (I'm not saying there are reasons to do this, just that they're more sane than RSA GPG keys.)

Re: Hardfork 2016 Value Size Limit

Posted: Sun Feb 07, 2016 8:16 pm
by johnc
Most likely 99% of the digital signature users in general will use their government (or company) issued ID card to sign. Because that's what will be recognized in their countries. :roll:

Example here.

Re: Hardfork 2016 Value Size Limit

Posted: Sun Feb 07, 2016 10:18 pm
by phelix
biolizard89 wrote:
phelix wrote:Reopening previous discussions is bad practice but I can't help it:
(from https://forum.namecoin.info/viewtopic.php?p=6574#p6574)
biolizard89 wrote:
[...]

Assuming that a single large-value name is more expensive than multiple small-value names, then users will only use long-valued names for use cases such as the 4k GPG key.

5k is reasonable in my opinion. But I don't feel particularly strongly.
What made you change your mind by a factor of 10?
Been quite a while since that discussion, and I haven't read through that thread recently. However, my weak opinion at the time was based on some assumptions, which may not be true. Also, since that time, I did calculations of blockchain size, UTXO size, block size, etc., which suggest to me that supporting use cases like 4k GPG keys is not obviously a good idea.
I completely agree that GPG keys can stay off chain. But the question at hand is about 1k.

Re: Hardfork 2016 Value Size Limit

Posted: Tue Feb 09, 2016 1:21 pm
by biolizard89
phelix wrote:
biolizard89 wrote:
phelix wrote:Reopening previous discussions is bad practice but I can't help it:
(from https://forum.namecoin.info/viewtopic.php?p=6574#p6574)
biolizard89 wrote:
[...]

Assuming that a single large-value name is more expensive than multiple small-value names, then users will only use long-valued names for use cases such as the 4k GPG key.

5k is reasonable in my opinion. But I don't feel particularly strongly.
What made you change your mind by a factor of 10?
Been quite a while since that discussion, and I haven't read through that thread recently. However, my weak opinion at the time was based on some assumptions, which may not be true. Also, since that time, I did calculations of blockchain size, UTXO size, block size, etc., which suggest to me that supporting use cases like 4k GPG keys is not obviously a good idea.
I completely agree that GPG keys can stay off chain. But the question at hand is about 1k.
My 4k GPG key comments were in the context of the 5k argument you quoted, and stuff johnc said.

My general take is that if there's a specific use case that needs 1000 byte value size, I'd like to see documentation of the use case, including analysis of how it would affect unspent name output set size, block size, and relay bandwidth usage if Namecoin became the predominant solution to that use case.

(I also doubt that such analysis will be finished in time for the BIP 9 hardfork, but that's not a big deal if we can do the increase without touching consensus rules.)

Re: Hardfork 2016 Value Size Limit

Posted: Tue Feb 09, 2016 1:23 pm
by biolizard89
johnc wrote:Most likely 99% of the digital signature users in general will use their government (or company) issued ID card to sign. Because that's what will be recognized in their countries. :roll:

Example here.
I'm not sure what you're responding to. Are you saying 99% of the public shouldn't use GPG? That seems like an odd argument to be making.

Re: Hardfork 2016 Value Size Limit

Posted: Wed Feb 10, 2016 6:58 pm
by gigabytecoin
Baby steps...

Why not increase the limit to just the originally proposed 1,000 bytes, essentially doubling the current size... and write up a document on how to easily increase it to 2,000 bytes in the future if required due to popular demand?

Re: Hardfork 2016 Value Size Limit

Posted: Thu Feb 11, 2016 9:00 pm
by johnc
biolizard89 wrote:
johnc wrote:Most likely 99% of the digital signature users in general will use their government (or company) issued ID card to sign. Because that's what will be recognized in their countries. :roll:

Example here.
I'm not sure what you're responding to. Are you saying 99% of the public shouldn't use GPG? That seems like an odd argument to be making.
Jesus Christ!, i'm making a prediction...

I said, that most people will never use GPG directly, they will use a closed scheme promoted by governments or companies, that may be based on GPG or PGP technology indirectly, but will have real id data verified by an approved entity like the police or facebook, that will give assurances that you are giving me your real name.

So 99% of the people don't have the technical knowledge or the will to setting up their own GPG/PGP or Namecoin identity. Because, for starters, there are no applications where you could use it. I cannot pay my taxes with it, i cannot request a new passport online with it, i cannot sign up to paypal or a exchange with it...

Obviously, for authenticating yourself in the context of a random internet forum, like this one, PGP is the only way to assure identity, if you want to remain anonymous.

So if you are scared of your government hijacking your id and using it for tracking you, then by al means use it. But most of the people are much more naive, as we are all sharing all our data with google 24/7.

Re: Hardfork 2016 Value Size Limit

Posted: Fri Feb 12, 2016 2:09 am
by biolizard89
gigabytecoin wrote:Baby steps...

Why not increase the limit to just the originally proposed 1,000 bytes, essentially doubling the current size... and write up a document on how to easily increase it to 2,000 bytes in the future if required due to popular demand?
Because "why not" is not an adequate way to justify a consensus rules change (or for that matter an isStandard change), particularly when there are concerns that have been raised about the impact on scalability. If you want 1 kilobyte, please provide an explanation about why that's a good decision, and we can have the discussion. Would you expect Bitcoin to increase their block size based on a "why not" argument?