Nxt Core Dev Q&A 20-03-2016

Bas Wisselink [18:00]
Welcome to the first Slack Q&A with the Core Devs.

[18:00]
Anyone can post their questions in here and they will do their best to answer

[18:00]
I would like to request of the devs that they say which question they are answering in their answer to avoid confusion

[18:01]
The Q&A will last for an hour :simple_smile:

[18:01]
OK, shoot :simple_smile:

Cryptkeeper [18:01]
Hi guys! I've read about 1.8 and wondered where we lost the 2FA feature. It seemed to me that Petko was quite far with the development for this.

jean-luc [18:02]
it is not a simple feature, and it required account transfer to be implemented too

[18:02]
which turned out to be really complicated

[18:02]
so, after 2.0, if really needed

Cryptkeeper [18:02]
Account Transfer == Panic Button?

jean-luc [18:03]
yes, in case an attacker got your password

Cryptkeeper [18:03]
OK, ack!

bcdev [18:04]
Are you working on the project full-time, or do you have some other major sink of time [dayjob, etc]?

Lionel Jeannerat [18:04]
hi

jean-luc [18:04]
I am full time
1

Lyaffe [18:05]
I'm still keeping my day job for 2-3 days a week

jeff mcdonald [18:05]
joined #general

Lionel Jeannerat [18:06]
Hi

Martis [18:06]
There was some roadmap for 1.8. Is 1.9 planned, and if yes, what features planed for 1.9?

jeff mcdonald [18:06]
hi everyone

jean-luc [18:06]
to do an FNXT asset distribution, we would need a hardfork, so this would be 1.9

Martis [18:07]
so, 1.9 only for FNXT stuff?

Lionel Jeannerat [18:07]
What is the goal of the road map ? Much nxters don't understand the vision behind the announcements ?

jean-luc [18:08]
there may be a few minor fixes that require hard fork too, to go into 1.9, other than that, don't know yet

altsheets [18:08]
Thanks for (re)implementing the http://localhost:7876&modal=send_money_modal ... etc. - it was one of my wishes; great, thx.
Will make it easy to link from any webpage into the local wallet.

nxter [18:08]
Hi There. Is it planned somewhere in the development roadmap to enhance the asset exchange and add new features to it?

Lionel Jeannerat [18:09]
We were leader in the assets exchange but we lose our big advantage

jean-luc [18:09]
anything that requires more work or a hard fork, after 2.0

[18:09]
for UI only enhancements, if riker has time...

bummer gaycoin [18:09]
joined #general

Websioux [18:09]
hi

Lyaffe [18:10]
We do have capacity for some UI features in case they make sense for 1.9. For 1.8 we are already introducing the desktop wallet which is a major enhancement which will probably breed more feature requests and ideas.

nxtswe [18:10]
Hi!
There has been a few misunderstandings regarding the dividend feature and that it does not show up in the transaction list. Is there anything planned to make this more user friendly?

jean-luc [18:11]
I suggested adding a more prominent "view ledger" link on the dashboard, under recent transactions

[18:11]
or even replacing the recent transactions with account ledger, or fit both

bummer gaycoin [18:11]
great progress on nxt

nxtswe [18:11]
sounds like a good idea to replace transactions with account ledger, that would solve it I think. thanx.

Lyaffe [18:12]
The account ledger is not simple to read

jean-luc [18:12]
I think one reason against that was that unconfirmed transactions would not show in account ledger

Lyaffe [18:12]
The transaction list is more important for most users

jean-luc [18:12]
and people expect to see them on top on the dashboard

Killakem TVE [18:12]
Hi @jean-luc, @riker! I would first like to say, Thank You, for all the time and effort you guys dedicate to this project!! 2.0 is a very interesting concept indeed, I can fully understand the benefits and the effect it will have on nxt, which will of course be very positive. The main concern from people I have spoken with is assets, what protections will be in place to ensure the Asset markets don't have excessive downwards pressure from people selling assets for nxt.

altsheets [18:13]
> account ledger is not simple to read
what about designing ICONS for each type of transaction on the account ledger?

Lyaffe [18:14]
@killakem: I think that making predictions about asset prices a year from now is not something we can base design decisions upon. NXT 2.0 also brings some advantages to asset issuer in the form of being able to trade the same asset on multiple child chains.

jean-luc [18:15]
also, such selling pressure will exist mostly before the snapshot, but after that, markets should come back to normal

Lyaffe [18:15]
@altsheets: its all about priorities

jean-luc [18:16]
and we have the idea of doing a continuous snapshot, over a period of even few months, to minimize market impact

Lionel Jeannerat [18:17]
When will be the snapshot?

Cryptkeeper [18:17]
I have another question about the 1.8 changes: one of the new features is a "Simple server side plugin framework". Sounds interesting, could you elaborate what can be done with it?
1

jean-luc [18:18]
the poll suggested 1-4 months from now, but of course we need to get 1.8 out first and then plan the hard fork

[18:18]
so the start may be more like 2 months from now at least, since we need to allow people time before hard fork

[18:19]
the plugin framework is for doing simple things in java, without having to patch and recompile NRS every time

[18:19]
you can write your own class that does something when e.g. an account balance changes, and there are hooks to register this class with the NRS
1

Lionel Jeannerat [18:20]
I have a important question: how can the community help?
1

Killakem TVE [18:21]
Thanks guys! @riker Thats a useful feature indeed, we could have a sidechain for TVE and trade the TVE asset on our sidechain as well!

Lyaffe [18:21]
The community can help a lot, Wiki documentation, support site, promotion with anyone who is willing to talk to you (edited)

Lionel Jeannerat [18:21]
I have the feeling that the community is disconnected with the vision of the devs, how can we participate to the vision of future of Nxt

josenxt [18:21]
Hi Devs, Thanks a lot for your job :smiley:
Do you already know how often the prunning process will take place in 2.0? I mean if it will take place everytime the blockchain reaches a certain size, or the 1st Monday of every month, or...

Lionel Jeannerat [18:22]
Much people ask me "what is the future of Nxt" and I can't anymore answer

[18:22]
Because I don't understand the roadmap

jean-luc [18:23]
pruning will be configurable, just like it is now for prunable messages and cloud data, a node can decide to store everything forever

[18:23]
what the defaults should be, it is way to early to say
1

Cryptkeeper [18:24]
LOL - when the hard  disk is full, it's too late to prune :grinning: (edited)

jean-luc [18:25]
one of the goals of 2.0, other than scalability, was to allow others to use the Nxt platform without necessarily using the NXT token itself, but still be a part of the same system, rather than starting their own cloned copy

nxter [18:26]
Will aliases be global to all childchains?

Tosch [18:26]
joined #general

Lyaffe [18:26]
@ludom: I don't have a crystal ball, look at the blockchain usage trend https://www.google.com/trends/explore#q=blockchain - usage of blockchain applications will explode soon. We need to keep in the forefront and we will find our niche

Google Trends - Web Search interest - Worldwide, 2004 - present
Explore Google Search trends with Google Trends.

Lionel Jeannerat [18:27]
And what is our niche ?

jean-luc [18:27]
aliases, I don't think they should be global

[18:28]
but by default, when you are using the Nxt child chain, you will be seeing the aliases that exist on it

[18:28]
to query the value of an alias on another chain, you would need something like a prefix, e.g. BTC.aliasname

[18:29]
since nothing internally depends on aliases, I am not thinking about that now

Lyaffe [18:29]
@ludom: I think that it can come from many different fronts AE, MS, Voting, phasing we have the best technology out there

Lionel Jeannerat [18:29]
But we are not leader

[18:30]
Because the fees are to high

Lyaffe [18:30]
But we already voted about fees before 1.7 and the community decision was to leave them on 1 nxt (edited)

Lionel Jeannerat [18:31]
Yes

[18:31]
I came with a project of dedicated "voting system"

Lyaffe [18:32]
Once we improve scalability we can reduce fees for the child chains

Lionel Jeannerat [18:32]
ok

come-from-beyond [18:32]
joined #general

Tosch [18:32]
Hello guys! Thank you for the Q&A I am a bit late now and I hope I don't repeat questions. Actually I have many questions regarding childchains. And I will wait until more specs arrive. But kind of urgent for me to know how to continue with for example the mynxt.info website. How do you imagine such a service makes sense later? Taking the blockexplorer. Will I be able to display ALL childchains?

Martis [18:32]
can child chain creators decide on child chain fees?

stephane gaudreault [18:32]
joined #general

Cryptkeeper [18:33]
I needed a while to understand the concept of Nxt 2.0, but now I think it's a great idea! :+1:
IMHO one aspect isn't yet fully explained: how the child chains get their tx included into the fNXT blocks and who must pay whom for this in which currency?

jean-luc [18:34]
transaction senders pay fees in native child chain currency, then anyone who wants to step in (although by default it may be the block forger) can take those fees, and pay in fNXT to the forger instead

[18:34]
the min fee will be defined in fNXT

josenxt [18:35]
How many transactions per second could NRS2.0 support?

[18:35]
Would there be room for even more optimization in tps?
1

jean-luc [18:36]
@tosh, it should be possible to display all child chains, why not

Tosch [18:37]
cool. I thought maybe you are thinking about "private" vs "public" chains but there is no such thing?

Cryptkeeper [18:37]
So there are forgers on the child chain and forgers on the fNXT chain? And the child chain forgers take the tx for "their" currency and includes it in fNXT chain  for fNXT (or whatever it's called then)? (edited)

jean-luc [18:37]
we can't have private chains, in the sense of content not visible, because all nodes need to validate all transactions

[18:38]
yes, although I wouldn't call them forgers, besides anyone could do it for the child chains, regardless of balance, as long as they are willing to pay that fNXT amount in exchange for the child token amount

Lyaffe [18:39]
@josenxt: the scalability depends on the usage pattern. Our best case scenario are not many very active child chains. In this scenario we can reach great scaling improvement perhaps 50:1 over the current scaling. I think it's too early to tell. We can later improve this further for sure.
2

Lionel Jeannerat [18:39]
What about concurrence between childchains? If we reach the limit of TPS, some childchains will died because their transaction will not be validated?

jean-luc [18:40]
those chains will have to pay more, to compete for inclusion in blocks

Websioux [18:41]
Big picture : shouldn't we try to become the reference of crypto framework for web designers ? or are we trying to find a hit with a sudden lucky feature ?

Cryptkeeper [18:41]
So if there is not much incentive to trade in "native currency" for fNXT, I guess the only "child chain forger" would be the creator of the child chain or some "operators".

Tosch [18:41]
There will be different sort of childchains right? Some will have simple send, some will have the Asset Exchange, DGS etc. Would it be possible to integrate new functionality like solidity from ethereum in such a childchain? Would it be possible to have a childchain with own rules as long as the transaction that comes into the block with fNXT is valid?

nxt3d.net [18:42]
there is some news about the DGS ? thank you

Fernando Ruiz-Tapiador [18:42]
joined #general

jean-luc [18:42]
yes, if you create a child chain and want to keep it running even if its token doesn't have real market value, such a creator would have to keep paying the fees himself

Lyaffe [18:43]
@websioux: web designers ?

Lionel Jeannerat [18:43]
 Pinned a message. See all pinned items in this channel.
jean-luc
those chains will have to pay more, to compete for inclusion in blocks
Today at 18:40

jean-luc [18:43]
childchain with own rules, yes, but they must be a subset of what is possible

Lyaffe [18:43]
@nxt3d_net: what are you missing ?

Cryptkeeper [18:43]
OK, I guess you have all the edge cases covered. But my head is spinning! :stuck_out_tongue_winking_eye:

Websioux [18:43]
@riker any business though the web

Lionel Jeannerat [18:43]
But if there is not enough TPS for all the chain, it'll be a problem

jean-luc [18:43]
e.g., with no shuffling allowed, or with only some accounts permitted to issue assets

Websioux [18:44]
I should have say web business designer

Lyaffe [18:44]
@websioux: I don't think we need to specialize on a specific feature. We need to provide the best tools for decentralized apps dev.

[18:45]
Specific projects can then specialize on their domain

Martis [18:46]
are there some plans to optimize and speed up blockchain download from scratch, if possible?
2

Websioux [18:46]
ok good if this is the goal, stay general

jean-luc [18:46]
the block size limits are not set in stone, if the hardcoded limits are reached, we can think about changing them, if the practical CPU processing limits are reached, this is already quite a lot of transactions, I wouldn't worry that far in the future

Lionel Jeannerat [18:46]
The address of the childchains will be NXT-XXXX-XXXX-XXXXX ? Or they will have their own code ?

jean-luc [18:46]
the secret phrase to public key to account number mapping doesn't change

[18:47]
so accounts will be global, and will be able to have balance on multiple chains

[18:47]
the NXT- is just a vanity prefix

come-from-beyond [18:47]
Is it possible to create some kind of interface (similar to Java interfaces) that would allow to "attach" any platform as a sidechain? The interface is supposed to work in such a way that its complete implementation would be enough and Nxt core wouldn't need to know the nuances of the attached platform. If such an interface is possible, when could we get the documentation (ETA)?

nxt3d.net [18:47]
@riker images, better listing... the firestorm bug relative to the wallet display into the virtual worlds was fixed and now we can fully use the wallet... but the thing that may interest the metaverse users is the marketplace in first... there is some dev on it? thank's

Lyaffe [18:48]
@nxt3d_net: let's take it offline. PM me on the forum

jean-luc [18:48]
attach an external platform? hm, not sure how exactly this could work...

brangdon [18:48]
Do you think this is adequate to satisfy compliance requirements? If, eg, shuffling is not allowed on a business chain, but it still allowed on other chains the business's node is processing, won't the business still get into trouble if local law makes shuffling illegal? Are the businesses you've talked to happy with this?

jean-luc [18:49]
what we call child chains are really transactions denominated in a different token, getting stored in their separate db schema, and optionally pruned differently, but they are still the basic Nxt transaction types

seccour [18:49]
:squirrel:

come-from-beyond [18:50]
Thx

Lyaffe [18:50]
@brangdon: not every business will be able to use the public chain because of compliance issues

bcdev [18:51]
Are there any plans for a scripting language a'la Bitcoin transaction scripts? (edited)

jean-luc [18:52]
for such a business, a private clone will still be needed, but when based on 2.0 such clone will have the advantage of better scalability even if it runs a single child chain for that business only

Lyaffe [18:52]
@bcdev: I don't think we should invest in scripting unless we find something very simple to implement. There are already several blockchains which specialized in this and we'll never catch up

Martis [18:53]
I'll repeat my question, as it was not answered: are there some plans to optimize and speed up blockchain download from scratch, if possible?

Cryptkeeper [18:53]
 Pinned a message. See all pinned items in this channel.
Lyaffe riker
@bcdev: I don't think we should invest in scripting unless we find something very simple to implement. There are already several blockchains which specialized in this and we'll never catch up
Today at 18:52

jean-luc [18:53]
1.8 is already as optimized as I can make it

altsheets [18:54]
what about offering the peerexplorer bootstrap download as a choice in the installer?

jean-luc [18:54]
for 2.0, of course it should be faster, as there will be fewer data to download, but more importantly much less processing to do

Lyaffe [18:54]
@martis: when we have pruning in 2.0 you'll need to download less data

altsheets [18:54]
that IS fast

Martis [18:54]
ROGER :simple_smile:

jean-luc [18:55]
not in the installer

[18:55]
and we need a replacement for peerexplorer anyway

Bas Wisselink [18:55]
We are already talking to crimi

Dave Pearce [18:56]
Ummm....I may have reached a deal with Crimi to keep PE up.
3

Bas Wisselink [18:56]
Chances are good we can keep PE going :simple_smile:

Dave Pearce [18:56]
And....it's time to start wrapping up this Q+A, guys

[18:57]
Shall we take one more question ?

nxter [18:57]
what are the current estimated eta on 1,8   1.9  and 2.0?

altsheets [18:57]
I have one more question:
Which of the ​*far future*​ ideas of the http://www.nxttechnologytree.com do you personally find most exciting? (edited)

Lyaffe [18:57]
@nxter

new messages
[18:58]
@nxter: 1.8 is probably 2-3 weeks away. 1.9 several month. 2.0 not sure

nxt3d.net [18:59]
thank you for this useful meeting and informations, i hope will be more Dev Q&A in the future :simple_smile:
4

Lyaffe [18:59]
@altsheets: for me I get excited when users are using my code. I don't get attached to specific features
1

Dave Pearce [19:00]
I think there will be,....thanks, everyone
1

Lyaffe [19:00]
thanks everyone

Martis [19:00]
Thank you devs