Let's create a foundation for competition among BTS developers

[jakub]

I would learn the codebase and document it for 3-4m
bts per month but i doubt any other senior developer would becausethey would look at chart first and extrapolate.

No one should be expected to document the codebase except the original author. Anything else would be highly misleading to future readers of the comments/documentation.

As far as documenting the code-base is concerned,  this has already been done to a large extent in the form of doxygen docs.
Therefore, IMO the main point is to create a tutorial that will guide you step by step how to use the code-base and create new smart-contracts and new operation types.

Something similar to this section of the Ethereum tutorial describing how some basic smart-contracts can be created and deployed:
https://ethereum.gitbooks.io/frontier-guide/content/contract_tutorials.html

[monsterer]

I would learn the codebase and document it for 3-4m
bts per month but i doubt any other senior developer would becausethey would look at chart first and extrapolate.

No one should be expected to document the codebase except the original author. Anything else would be highly misleading to future readers of the comments/documentation.

[Samupaha]

But maybe we don't really need classic "blockchain programming" to be able to implement & deploy our ideas expressed in BSIPs.
Maybe all that's needed is the ability to construct new smart-contracts, create new types of operations and thoroughly test them.

Just like Android - it took the most talented coders to create the platform. But building apps on it is not so demanding.
Maybe with Graphene this contrast is not so extreme but still I guess the most difficult task of designing & building the platform has been done. Now we need coders who will just make use of it.

At first we need just somebody who can build the app, Cryptonomex can check and test the code when it's ready.

[cass]

wouldn't make much sense to me

it for 3-4m bts per month

[cube]

I would learn the codebase and document it for 3-4m
bts per month but i doubt any other senior developer would becausethey would look at chart first and extrapolate.

3-4M = Mille or/ Million?

3500000 BTS = 11893.99 USD --> for learning the codebase?
(You should not forget the advantage you'll get through this (knowledge, contacts, etc).. IMO a way to much)

Did he mean 3 to 4 months?

[cass]

I would learn the codebase and document it for 3-4m
bts per month but i doubt any other senior developer would becausethey would look at chart first and extrapolate.

3-4M = Mille or/ Million?

3500000 BTS = 11893.99 USD --> for learning the codebase?
(You should not forget the advantage you'll get through this (knowledge, contacts, etc).. IMO a way to much)

[cube]

I would learn the codebase and document it for 3-4m
bts per month but i doubt any other senior developer would becausethey would look at chart first and extrapolate.

My estimation is about there too.

[jsidhu]

I would learn the codebase and document it for 3-4m
bts per month but i doubt any other senior developer would becausethey would look at chart first and extrapolate.

[jakub]

Just like Android - it took the most talented coders to create the platform. But building apps on it is not so demanding.
Maybe with Graphene this contrast is not so extreme but still I guess the most difficult task of designing & building the platform has been done. Now we need coders who will just make use of it.

There is a chasm between the difficulty of building a smartphone app and the difficulty of deploying a P2P currency system change; principally that a bug in an app doesn't lose people money, whereas in blockchain projects, it very well can do.

That only confirms my assumption that the most crucial and difficult area is testing, not the coding itself.
That''s why I added creating a test network as an integral part of this proposal. This way the entire community can help with that.

[IOHKCharles]

This is an interesting idea. Has anyone written up something on a potential Bitshares Foundation?

[monsterer]

Just like Android - it took the most talented coders to create the platform. But building apps on it is not so demanding.
Maybe with Graphene this contrast is not so extreme but still I guess the most difficult task of designing & building the platform has been done. Now we need coders who will just make use of it.

There is a chasm between the difficulty of building a smartphone app and the difficulty of deploying a P2P currency system change; principally that a bug in an app doesn't lose people money, whereas in blockchain projects, it very well can do.

[jakub]

Just throwing out ideas here. Aside from the C++ learning curve, my perspective is more or less aligned with monsterer's. However I haven't seriously looked at the graphene C++ code, only the API commands and I find the API very complicated and not very well structured, with inconsistent arguments for similar API calls for example. I wouldn't surprise me it will be a steep learning curve for all but the most gifted C++ coders.

To clarify, I'm not saying that graphene codebase is difficult to learn in particular, I'm saying that blockchain programming in general is an incredibly specialised and complicated niche that you cannot expect even experienced c++ programmers to just pick up and run with, without prior learning, knowledge and experience in the field.

But maybe we don't really need classic "blockchain programming" to be able to implement & deploy our ideas expressed in BSIPs.
Maybe all that's needed is the ability to construct new smart-contracts, create new types of operations and thoroughly test them.

Just like Android - it took the most talented coders to create the platform. But building apps on it is not so demanding.
Maybe with Graphene this contrast is not so extreme but still I guess the most difficult task of designing & building the platform has been done. Now we need coders who will just make use of it.

[monsterer]

Just throwing out ideas here. Aside from the C++ learning curve, my perspective is more or less aligned with monsterer's. However I haven't seriously looked at the graphene C++ code, only the API commands and I find the API very complicated and not very well structured, with inconsistent arguments for similar API calls for example. I wouldn't surprise me it will be a steep learning curve for all but the most gifted C++ coders.

To clarify, I'm not saying that graphene codebase is difficult to learn in particular, I'm saying that blockchain programming in general is an incredibly specialised and complicated niche that you cannot expect even experienced c++ programmers to just pick up and run with, without prior learning, knowledge and experience in the field.

[jakub]

If learning graphene is really that difficult maybe the best we can do is take a small step approach and focus on small chunks of the API at a time. Document how the various API calls are used to achieve an existing end result, and perhaps from that context it would be possible to extrapolate how to use the API to accomplish other things.

I think we should make a clear distinction between:
(1) making use of the API (e.g. for building a mobile wallet or a bridge/gateway or a trading bot)
(2) adding new features (or modifying existing ones) to the BitShares code-base - I guess this is mainly done by creating new smart-contracts and new types of operations.

I guess (1) is (or will be) covered by xeroc and it seems that reasonably experienced coders can already handle this area, even without advanced C++ skills.

The problem I'm trying to address in this thread is (2).
Monsterer suggests (2) is extremely difficult and takes years of experience with blockchain technology to be able to master it.
But I recall BM saying during Mumble hangouts that the coding itself is relatively easy, the difficult part is testing and making sure there are no unexpected side-effects.

I guess we need to wait for the devs to shed some light on this.

[Thom]

You're asking all the right questions Jakub. Perhaps it is unrealistic to expect a relatively short learning curve for an 5 - 10 year C++ coder without blockchain experience to be productive in coding BSIPs. I think that's what monsterer's perspective is, and it sounds contrary to Stan's.

Even a programmer like Andreaus Antonopolis doesn't claim to be an expert, tho he has written a book targeted to bitcoin programmers. He is up front in saying his skills as a C++ programmer are not as productive as other bitcoin programmers and he feels speaking, teaching and evangelizing are tasks he is better suited for.

If learning graphene is really that difficult maybe the best we can do is take a small step approach and focus on small chunks of the API at a time. Document how the various API calls are used to achieve an existing end result, and perhaps from that context it would be possible to extrapolate how to use the API to accomplish other things.

Just throwing out ideas here. Aside from the C++ learning curve, my perspective is more or less aligned with monsterer's. However I haven't seriously looked at the graphene C++ code, only the API commands and I find the API very complicated and not very well structured, with inconsistent arguments for similar API calls for example. I wouldn't surprise me it will be a steep learning curve for all but the most gifted C++ coders.

[jakub]

The thing is you can't teach people that stuff, it's too time consuming and costly. You can only give them the tools so they can use them and build something themselves. As long as we have a few tutorials, examples, videos, etc, devs will only need to see them in order to build something themselves.

Yes, but the big question is this: how we can have "a few tutorials, examples, videos" if CNX does not have the manpower to do it and we don't have the know-how required to do it.
Seems like a classic chicken-and-egg problem.

[Akado]

I'd suggest this instead: one of our community members (someone who has a good understanding of BitShares) teams up with an external C++ expert and they write a tutorial together in close cooperation with CNX.

What you're saying is like saying you're going to make a tutorial to turn a piano salesman into a world class pianist.

OK, I'm in no position to argue with that as I've never done any blockchain development and never looked into Graphene's source code.
@Stan and @xeroc seem to be more optimistic about this but none of them is an actual Graphene developer. BM's remarks have very vague so far.

That's why we need CNX developers' opinion on this - they are the only people who know the answer to this important question:
How much time and effort will it take to train an experienced C++ developer (without any deep understanding of the blockchain concept) to become an independent contributor capable of handling any of the tasks listed in our BSIPs?

The thing is you can't teach people that stuff, it's too time consuming and costly. You can only give them the tools so they can use them and build something themselves. As long as we have a few tutorials, examples, videos, etc, devs will only need to see them in order to build something themselves.

[jakub]

I'd suggest this instead: one of our community members (someone who has a good understanding of BitShares) teams up with an external C++ expert and they write a tutorial together in close cooperation with CNX.

What you're saying is like saying you're going to make a tutorial to turn a piano salesman into a world class pianist.

OK, I'm in no position to argue with that as I've never done any blockchain development and never looked into Graphene's source code.
@Stan and @xeroc seem to be more optimistic about this but none of them is an actual Graphene developer. BM's remarks have very vague so far.

That's why we need CNX developers' opinion on this - they are the only people who know the answer to this important question:
How much time and effort will it take to train an experienced C++ developer (without any deep understanding of the blockchain concept) to become an independent contributor capable of handling any of the tasks listed in our BSIPs?

[gamey]

Regardless of how complicated the chosen software stack for BTS may or may not be, tutorials for at least one "smart contract" that does something useful is needed. Otherwise having to learn the code base to figure out how to get that far is going to make an obstruction no one is EVER going to try to tackle. There are simply better options. 

When reverse engineering something, it is always better to start from a top down type approach.  By not having this sort of documention, you are forcing a bottom up approach.  Good luck there.


edit - I suppose maybe tutorial is not the right word. An example would suffice, but even then enticing people to Bitshares is tough.  The example should use FBAs I suppose.

Edit 2 - I used smart contracts because I don't know what BitShares calls this vague concept of a feature voted in by the blockchain or how that works. (Granted, I have also not read existing documentation.)

[monsterer]

I'd suggest this instead: one of our community members (someone who has a good understanding of BitShares) teams up with an external C++ expert and they write a tutorial together in close cooperation with CNX.

What you're saying is like saying you're going to make a tutorial to turn a piano salesman into a world class pianist.

[jakub]

Yes, @xeroc (if he wants) plus one C++ talent familiar with or interested in (passionate about) blockchain tech, working together, both paid.  That's my preference.  The C++ coder?  I think we should pitch it like a scholarship and get a boatload of applicants to sort through, I would trust xeroc plus one core dev to do this, if we don't already have someone among us who is willing and able.  I can help find candidates if y'all are at a loss for where to look.

Maybe if we openly declare how much we are willing to spend to have a comprehensive tutorial, then this will attract talents.

[lovejoy]

So what do you propose?
If it is the case that for an "uninitiated c++ developer" it will take years to acquire enough knowledge to implement & deploy any of our BSIPs, then we are screwed indeed.

Instead of throwing away money trying to teach something so fantastically complicated, you should recruit talent with the necessary skills already under their belt. Advertise on bitcointalk.org, that's about the only place I know where these guys hang out.

So you suggest to advertise on bitcointalk.org for a blockchain guy to come to our community, extract knowledge from CNX and write it down for us?
It might happen but it's not very likely to me.

I'd suggest this instead: one of our community members (someone who has a good understanding of BitShares) teams up with an external C++ expert and they write a tutorial together in close cooperation with CNX.
And this tutorial should not assume deep prior knowledge about the blockchain technology (only general understanding of the concept). This is how I see it.

Yes, @xeroc (if he wants) plus one C++ talent familiar with or interested in (passionate about) blockchain tech, working together, both paid.  That's my preference.  The C++ coder?  I think we should pitch it like a scholarship and get a boatload of applicants to sort through, I would trust xeroc plus one core dev to do this, if we don't already have someone among us who is willing and able.  I can help find candidates if y'all are at a loss for where to look.

[jakub]

So what do you propose?
If it is the case that for an "uninitiated c++ developer" it will take years to acquire enough knowledge to implement & deploy any of our BSIPs, then we are screwed indeed.

Instead of throwing away money trying to teach something so fantastically complicated, you should recruit talent with the necessary skills already under their belt. Advertise on bitcointalk.org, that's about the only place I know where these guys hang out.

So you suggest to advertise on bitcointalk.org for a blockchain guy to come to our community, extract knowledge from CNX and write it down for us?
It might happen but it's not very likely to me.

I'd suggest this instead: one of our community members (someone who has a good understanding of BitShares) teams up with an external C++ expert and they write a tutorial together in close cooperation with CNX.
And this tutorial should not assume deep prior knowledge about the blockchain technology (only general understanding of the concept). This is how I see it.

[cass]

i personally really like gitbooks..  there were no direct complains about.

But tbh ..it's not really on my subject, so a trust fabi's advise on this ..
(i f wanted we could port it to gitbooks as well i guess)

[jakub]

There is ample documentation and examples within the code and a community of developers ready to answer any questions those developers may have.

You see @Stan how it works in practice.
We don't even seem to be able to estimate the level of expertise / experience required as nobody from CNX (except for a non-technical person like you) bothers to comment here.

Honestly, I cannot image this know-how take-over happening unless you guys actively step in.
We need your guidance in this area. We need your opinions.

[monsterer]

So what do you propose?
If it is the case that for an "uninitiated c++ developer" it will take years to acquire enough knowledge to implement & deploy any of our BSIPs, then we are screwed indeed.

Instead of throwing away money trying to teach something so fantastically complicated, you should recruit talent with the necessary skills already under their belt. Advertise on bitcointalk.org, that's about the only place I know where these guys hang out.

[jakub]

I have given some thought to the matter of Graphene tutorial and here are my conclusions:

(1) Our situation is quite similar to Android - we don't need to teach people a new programming language, we only need to teach them how to use a known language in a very specific environment.

Sorry, this is nonsense. You propose to teach an uninitiated c++ developer something which takes years to master. This is not possible within the budget.

So what do you propose?
If it is the case that for an "uninitiated c++ developer" it will take years to acquire enough knowledge to implement & deploy any of our BSIPs, then we are screwed indeed.

[monsterer]

I have given some thought to the matter of Graphene tutorial and here are my conclusions:

(1) Our situation is quite similar to Android - we don't need to teach people a new programming language, we only need to teach them how to use a known language in a very specific environment.

Sorry, this is nonsense. You propose to teach an uninitiated c++ developer something which takes years to master. This is not possible within the budget.

[jakub]

This is what Ethereum currently has:
https://ethereum.gitbooks.io/frontier-guide/content/

Very simple but appears to be quite practical.
It has a search utility which is another important feature of a good tutorial.

Nice find .. i can definitely improve our stuff on that!

I wish we could ask CNX to translate those Ethereum examples into their Graphene equivalents:
https://ethereum.gitbooks.io/frontier-guide/content/contract_tutorials.html
This would be a good starting point.

Generally I think it's a good idea to write the tutorial by comparing Ethereum's approach to ours.
In a friendly way, not trying to prove our superiority, but rather use Ethereum as a background.

[xeroc]

For those that haven't noticed .. there is a tutorial section for bitshares already:
http://docs.bitshares.eu/bitshares/tutorials/index.html

[xeroc]

I think sphinx is way for powerful than gitbook .. but moving from rest to markdown syntax using pandoc is easy to ... if you are not happy with the current docs ..

sphinx can compile to PDF and ODF as well, btw
and it can create flow charts in svg using gravphizs
See:
https://raw.githubusercontent.com/BitSharesEurope/docs.bitshares.eu/7b641f495fd5a6e3247785f9631b074750bd8359/source/bitshares/user/transactions-proposed.rst
image: http://docs.bitshares.eu/bitshares/user/transactions-proposed.html

[jakub]

This is what Ethereum currently has:
https://ethereum.gitbooks.io/frontier-guide/content/

Very simple but appears to be quite practical.
It has a search utility which is another important feature of a good tutorial.

Nice find .. i can definitely improve our stuff on that!

they are using gitbooks, we used as well for the past BitAssets WP IRRC

@cass
Was it any good? Anybody complained about it?

[cass]

This is what Ethereum currently has:
https://ethereum.gitbooks.io/frontier-guide/content/

Very simple but appears to be quite practical.
It has a search utility which is another important feature of a good tutorial.

Nice find .. i can definitely improve our stuff on that!

they are using gitbooks, we used as well for the past BitAssets WP IRRC

[xeroc]

This is what Ethereum currently has:
https://ethereum.gitbooks.io/frontier-guide/content/

Very simple but appears to be quite practical.
It has a search utility which is another important feature of a good tutorial.

Nice find .. i can definitely improve our stuff on that!

[xeroc]

Except for
  "- create working examples of "hello world" smart contracts and operations,"
that is exactly what I am doing already for quite some time

In your worker proposal you state very vaguely:

Lots of Documentation (e.g. docs.bitshares.eu), Whitepapers, BSIPs,

@xeroc , is there any place where I can find what areas you intend to address in the documentation covered by your worker proposal?
And also what areas you definitely don't intend to address?

This would help us greatly in avoiding overlapping.

There is nothing that I do NOT want to have in there .. (except maybe politics) ..

There is also a poll on what I should work on next:
https://bitsharestalk.org/index.php/topic,19967.new.html#new
I tackled some of them already

I know that ideally we would like to have everything united in one place.
However, please be more specific what your plans and capabilities are for the next 2-3 months, as otherwise it's hard for me to specify goals for a person attempting to create a Graphene development tutorial.
Telling him/her "just work with xeroc and fill in whatever you can" is not specific enough IMO.

My aim is to create a tutorial which will enable an experienced C++ developer to code, test and deploy something as complex as BSIP #6
What parts of this know-how will you cover? What areas are you not able to cover?

EDIT: Judging from the poll, it looks to me that your documentation is aimed at those two use-cases:
- how to use BitShares
- how to create third-party apps that interface with BitShares (e.g. mobile wallets, gateways, bridges)
And what you don't seem to cover is anything that introduces changes affecting the BitShares protocol (i.e. requiring hard-forks).
Is it a fair definition of the scope of your docs?

(If so, how does your intention to write "hello world" contracts and operations fit into this?)

I'd like to keep my work more user centric and only document stuff that is useful for integrating with bitshares.

Hence, you are free to design your c++ turorials as you like .. i will only integrate it into the existing structure but we can talk about that detail later on ...

[jakub]

This is what Ethereum currently has:
https://ethereum.gitbooks.io/frontier-guide/content/

Very simple but appears to be quite practical.
It has a search utility which is another important feature of a good tutorial.

[jakub]

I have given some thought to the matter of Graphene tutorial and here are my conclusions:

(1) Our situation is quite similar to Android - we don't need to teach people a new programming language, we only need to teach them how to use a known language in a very specific environment.
In Android's case this familiar programming language is Java, in our case it's C++. When coding natively in Android, you need to be aware of and familiar with the Android Activity Cycle. When coding within Graphene, I guess you need to be aware of being within a blockchain OS.
As we have this nice analogy between Android/Java and Graphene/C++, and Android's tutorials are already mature and rich with comprehensive examples, I propose to make Android's tutorials a quality benchmark we aim to achieve in the long run.

(2) We have these high quality examples:
http://developer.android.com/training/index.html
http://www.vogella.com/tutorials/Android/article.html
http://www.codelearn.org/android-tutorial#tutorial
https://www.objc.io/issues/11-android/android_101_for_ios_developers/

Does anybody know of software tools aimed at creating and maintaining tutorials like the ones listed above?
I know xeroc is using something called ReStructureText for docs.bitshares.eu but it looks a bit rough to me in terms of UX.
Maybe it can be customized with CSS but maybe there are more flexible or more advanced tools?

Also, in the future we might need to have a comments/questions section, as this tutorial does:
http://www.codelearn.org/android-tutorial/android-introduction
So I guess we might need some more complex tool-set than ReStructureText can offer.

EDIT: Please suggest examples of your favorite tutorials so that we can define our expectations a bit better.

[jakub]

Except for
  "- create working examples of "hello world" smart contracts and operations,"
that is exactly what I am doing already for quite some time

In your worker proposal you state very vaguely:

Lots of Documentation (e.g. docs.bitshares.eu), Whitepapers, BSIPs,

@xeroc , is there any place where I can find what areas you intend to address in the documentation covered by your worker proposal?
And also what areas you definitely don't intend to address?

This would help us greatly in avoiding overlapping.

There is nothing that I do NOT want to have in there .. (except maybe politics) ..

There is also a poll on what I should work on next:
https://bitsharestalk.org/index.php/topic,19967.new.html#new
I tackled some of them already

I know that ideally we would like to have everything united in one place.
However, please be more specific what your plans and capabilities are for the next 2-3 months, as otherwise it's hard for me to specify goals for a person attempting to create a Graphene development tutorial.
Telling him/her "just work with xeroc and fill in whatever you can" is not specific enough IMO.

My aim is to create a tutorial which will enable an experienced C++ developer to code, test and deploy something as complex as BSIP #6
What parts of this know-how will you cover? What areas are you not able to cover?

EDIT: Judging from the poll, it looks to me that your documentation is aimed at those two use-cases:
- how to use BitShares
- how to create third-party apps that interface with BitShares (e.g. mobile wallets, gateways, bridges)
And what you don't seem to cover is anything that introduces changes affecting the BitShares protocol (i.e. requiring hard-forks).
Is it a fair definition of the scope of your docs?

(If so, how does your intention to write "hello world" contracts and operations fit into this?)

[xeroc]

Except for
  "- create working examples of "hello world" smart contracts and operations,"
that is exactly what I am doing already for quite some time

In your worker proposal you state very vaguely:

Lots of Documentation (e.g. docs.bitshares.eu), Whitepapers, BSIPs,

@xeroc , is there any place where I can find what areas you intend to address in the documentation covered by your worker proposal?
And also what areas you definitely don't intend to address?

This would help us greatly in avoiding overlapping.

There is nothing that I do NOT want to have in there .. (except maybe politics) ..

There is also a poll on what I should work on next:
https://bitsharestalk.org/index.php/topic,19967.new.html#new
I tackled some of them already

[Empirical1.2]

Hiring some latvian coder from freelancer.com, or some such other generic source will be a waste of time; they will simply not have the expertise to do the job properly even if they do know c++ inside out. Simply put, blockchain programming is an extremely specialised niche within a niche and you must expect to pay a premium to get someone who knows what they are doing.

Moving forward, I suggest you fund CNX to produce a series of competency tests which test candidate programmers for their knowledge of low level, blockchain oriented systems, design methodologies and general c++. CNX would need to be in charge of the hiring process, because they know their systems inside out.

Completely agreed here ...

KenCode seems to have done pretty well sourcing help and expertise for his BitShares mobile wallet project at a competitive price.

https://bitsharestalk.org/index.php/topic,20776.0.html

[cass]

Hiring some latvian coder from freelancer.com, or some such other generic source will be a waste of time; they will simply not have the expertise to do the job properly even if they do know c++ inside out. Simply put, blockchain programming is an extremely specialised niche within a niche and you must expect to pay a premium to get someone who knows what they are doing.

Moving forward, I suggest you fund CNX to produce a series of competency tests which test candidate programmers for their knowledge of low level, blockchain oriented systems, design methodologies and general c++. CNX would need to be in charge of the hiring process, because they know their systems inside out.

Completely agreed here ...

[monsterer]

Hiring some latvian coder from freelancer.com, or some such other generic source will be a waste of time; they will simply not have the expertise to do the job properly even if they do know c++ inside out. Simply put, blockchain programming is an extremely specialised niche within a niche and you must expect to pay a premium to get someone who knows what they are doing.

Moving forward, I suggest you fund CNX to produce a series of competency tests which test candidate programmers for their knowledge of low level, blockchain oriented systems, design methodologies and general c++. CNX would need to be in charge of the hiring process, because they know their systems inside out.

[jakub]

Except for
  "- create working examples of "hello world" smart contracts and operations,"
that is exactly what I am doing already for quite some time

In your worker proposal you state very vaguely:

Lots of Documentation (e.g. docs.bitshares.eu), Whitepapers, BSIPs,

@xeroc , is there any place where I can find what areas you intend to address in the documentation covered by your worker proposal?
And also what areas you definitely don't intend to address?

This would help us greatly in avoiding overlapping.

[pc]

I'm currently working in a full-time contract and am not available for a worker position.

In any case, I wouldn't be the best candidate for the job. I'm only vaguely familiar with the Graphene codebase. Most of my previous work has been on the old client and the fc library, not on the actual Graphene code.

[xeroc]

Before we go searching for talented C++ coders outside of the BitShares community, might it not be wise to inquire as to which of our local bts community talents could be persuaded to put time into such an effort?  @xeroc is already well along in his efforts.  What about @svk?  Might a sufficiently well paid worker proposal be able to persuade some of our highly talented friends to focus on developing this 'foundation'?

If we're going to spend bts for such efforts, if possible I'd prefer to see it go towards people whose reputations and interests are already firmly aligned with the success of BitShares.

Thoughts?

Thanks, I'd love to dig into C++ BUT I am quite sure that it will not be very efficient until I get more experienced with it even though I know the API and parts of the code already.
It may be more cost-efficient to hire an experienced C++/BOOST developer for this.
Also, there is @pc, who is very firm with C++ and already did quite some work for the core base. Why not offer a well-paid job to him? @pc: would you be interested in getting hired again by the blockchain?

[abit]

Before we go searching for talented C++ coders outside of the BitShares community, might it not be wise to inquire as to which of our local bts community talents could be persuaded to put time into such an effort?  @xeroc is already well along in his efforts.  What about @svk?  Might a sufficiently well paid worker proposal be able to persuade some of our highly talented friends to focus on developing this 'foundation'?

If we're going to spend bts for such efforts, if possible I'd prefer to see it go towards people whose reputations and interests are already firmly aligned with the success of BitShares.

Thoughts?

Please contact @hybridd and @kuro112 .
I am not sure how good they are with c++, but they could possibly help.

Hybridd has been providing support for sharebits in the freebie section and kuro has written tutorials in the past (with no expectations of compensation ) -
https://bitsharestalk.org/index.php/topic,19612.msg251784.html

Regards
Xypher

People from Freebie have being showing passions and efforts for a long time. Imo it's a good chance for them now.

[Xypher]

Before we go searching for talented C++ coders outside of the BitShares community, might it not be wise to inquire as to which of our local bts community talents could be persuaded to put time into such an effort?  @xeroc is already well along in his efforts.  What about @svk?  Might a sufficiently well paid worker proposal be able to persuade some of our highly talented friends to focus on developing this 'foundation'?

If we're going to spend bts for such efforts, if possible I'd prefer to see it go towards people whose reputations and interests are already firmly aligned with the success of BitShares.

Thoughts?

Please contact @hybridd and @kuro112 .
I am not sure how good they are with c++, but they could possibly help.

Hybridd has been providing support for sharebits in the freebie section and kuro has written tutorials in the past (with no expectations of compensation ) -
https://bitsharestalk.org/index.php/topic,19612.msg251784.html

Regards
Xypher

[lovejoy]

Before we go searching for talented C++ coders outside of the BitShares community, might it not be wise to inquire as to which of our local bts community talents could be persuaded to put time into such an effort?  @xeroc is already well along in his efforts.  What about @svk?  Might a sufficiently well paid worker proposal be able to persuade some of our highly talented friends to focus on developing this 'foundation'?

If we're going to spend bts for such efforts, if possible I'd prefer to see it go towards people whose reputations and interests are already firmly aligned with the success of BitShares.

Thoughts?

[abit]

I would spend more time on the code, not sure how much time I can put in though, nor any result is able to be guaranteed right now.

[xeroc]

5. What subjects will the documentation by xeroc cover as opposed to the documentation aimed at developers?

Let's let Xeroc answer that.  He is free to focus where he feels he can do the most good.

Ideally, I would like to see it grow beyond just my inputs .. Like a community project.
Sources for docs.bitshares.eu are publicly available, are written in ReStructureText (similar to markdown) and pull requests are welcome.

[Stan]

This is a great example of seeing a problem and doing something to fix it!

I can answer a few of the questions, until BM has a chance to weigh in:

1. How big is the Graphene source code? (in terms of lines of code) How much time do you expect an experienced C++ developer (not familiar with the blockchain concept) would require to become productive with Graphene?

I'll leave the size statistics to someone familiar withe the GitHub library.  The Graphene Devs invented it from scratch in about three months. Three more to achieve deployment quality.  Generally a new developer can become productive at some narrow specialty area in a few weeks.  I'd like to hear from any of our developers what their personal experience has been.

2. The current status of the Graphene documentation/tutorials and CNX plans in this regard for the next 1-3 months.

We will support whatever community leaders like yourself are able to develop as a viable funded plan.

3. To what extent is the documentation embedded in the source code useful? Are there any examples there?

BM's Blog Article states:  "Anyone can create new smart contracts for BitShares today. There is ample documentation and examples within the code and a community of developers ready to answer any questions those developers may have. A simple tutorial could be drafted on how to write contracts for BitShares."

4. If I hire a C++ developer and give them the task of becoming a fully-fledged Graphene developer able to create smart contracts and deploy them - where should s/he start? What is the path s/he should follow to learn these skills?

Bytemaster would be happy to personally mentor such a person to get them an initial vector and help get them unstuck from time to time.

5. What subjects will the documentation by xeroc cover as opposed to the documentation aimed at developers?

Let's let Xeroc answer that.  He is free to focus where he feels he can do the most good.

[kenCode]

- thoroughly learn Graphene and its API,
- document it by expanding the existing documentation,
- create friendly tutorials for C++ developers wanting to work with Graphene,
- launch a test network,
- and for the next 2-3 months afterwards play the role of an "ask me anything" expert at our full disposal

Except for
  "- create working examples of "hello world" smart contracts and operations,"
that is exactly what I am doing already for quite some time

Exactly. Without xeroc's docs (hey, that rhymes!) my devs would not be able to understand BitShares for a very long time. Xeroc saved the day.
https://docs.bitshares.eu

[xeroc]

- thoroughly learn Graphene and its API,
- document it by expanding the existing documentation,
- create friendly tutorials for C++ developers wanting to work with Graphene,
- launch a test network,
- and for the next 2-3 months afterwards play the role of an "ask me anything" expert at our full disposal

Except for
  "- create working examples of "hello world" smart contracts and operations,"
that is exactly what I am doing already for quite some time

Anyone willing to help out with the documentations is free to do so. Sources are
open and pull requests are welcome. Read the "Meta" page in the docs.

Required skills:
- solid expertise in C++
- proficiency in spoken and written English

Preferred experience:
- familiarity with the blockchain concept and/or BitShares
- already worked as a CNX contractor

I wish I knew C++ as well as it is required. At least, I can read the code
However, I think I may be able to help out any newcomer by answering most basic
questions about the code (not the specifics though)

Besides that: Well done, let's get some talented coders motivated to work for
BitShares!

[jakub]

@bytemaster @Stan
As you can see, we are attempting to make a first step aimed at organizing ourselves in the area of code development (thus relieving the pressure from CNX).
But paradoxically, during this process of becoming more independent from CNX, we are going to be quite dependent on CNX.

So please help us by answering the questions listed in OP.
A blog post would be the perfect form for this but if that is too demanding, posting your input here will do.

[BitShares News]

Nowadays nobody outsider knows what's happening in Bitshares because everything is in here on the forum. If somebody doesn't know what to search for, it's kinda hard to get onboard with development.

FWIW we've been posting the "big topics" on the forum to Twitter, Reddit, Google+ and on BitSharesNews.info.

Hangouts are nice of course, but I have found it difficult to get anybody outsider listen to an hour of talking.

Send them to BitSharesNews.info to read the transcripts.

[jakub]

It looks to me that ken's approach is valuable but not for this particular task.

For this task we need someone really experienced, someone who knows what type of guidelines are needed the most for less experienced developers.
Someone who has made a lot of mistakes in the past and has learned from them.
Someone who will feel confident enough to ask stupid questions and to bother CNX as long as it takes to extract the answers from them.
And good English writing skills is a must.
So we need someone who has both the technical experience and a talent for explaining things. Maybe this means a team of two people, who will share their pay.

As for ken's access to cheaper (yet more dedicated) developers outside the US and Western Europe - I think this would be ideal in the next stage, once the terrain is mapped out and the quality of the work can be validated.
I guess we won't want our first project done without CNX to be a flop. It's better to overpay on the first job than to tarnish our reputation on the very first step of our independence.

[Samupaha]

I'd like to see more posts on the blog. Nowadays nobody outsider knows what's happening in Bitshares because everything is in here on the forum. If somebody doesn't know what to search for, it's kinda hard to get onboard with development. Hangouts are nice of course, but I have found it difficult to get anybody outsider listen to an hour of talking.

Also the technology page on the website needs continuous updating to reflect what's been done and what's unfinished and when it might get finished.

I've been planning to get some referral income, but I want authoritative texts to link to. I can get people somewhat interested to Bitshares but I can't be the only source of information. It's awkward to say "there is not much info on this right now, try to find something on the forum or listen to this hangout where it was discussed a little bit..."

Let's take one example: Fee Backed Assets. Great tool to fund development and possibility for investors to make money.

Is it mentioned on the website? No, not anywhere.

Is there a mention on the blog that we have this kind of feature coming? No. Actually it hasn't been updated in a long time. There is not even a mention that 2.0 has come out.

If the website and official blog aren't updated, for most people it is sign that development is freezed and nothing is happening so they lose their interest.

[kenCode]

I & think most would support an initiative to bring in new Latvian/other devs as individuals or even ultimately as a team under your management, with your position being paid more generously of course, if that's something you'd consider.

I can tell you one thing, I'll get this stuff built a hell of a lot faster with a little fukn support. need coffee now hmmf

[Empirical1.2]

I have been outsourcing since 1993.
I did have a large office with in-house devs and all that overhead for over 16 years but to stay in business you HAVE to outsource and learn how to write detailed project descriptions with milestones, know who the buzzword "devs" are, know who the high value devs are, learn to simplify your english so that you can communicate with anyone in any country, always crosstrain everyone, know what to pay them (don't pay them too little and certainly do not pay them too much (for numerous reasons)).. the list goes on and on. Fuzzy and I chatted about this for a bit, maybe we can chat about this more during the mumble on friday.
 
Call them slaves if you like, but the fact is, $900 goes a looooong way in some countries and those guys drive nice cars, they certainly do not feel like slaves. They feel very successful. They work their ass off too, stay with me for years, never complain about vacation time, or their work hours, or need a fukn gym membership or health benefits from me either. I could go on and on why I only outsource with freelancers. Do you know how much stuff I could build with $45,000? JEEZ
 
This friday (tomorrow) fuzzy and I will be talking about OPENPOS (https://bitsharestalk.org/index.php/topic,20762.0/all.html) and what that 19K project has built and is still building. Those little 2BTS donations (http://i.imgur.com/JyNtgOU.jpg) at the cash registers and mobile wallets will go a long way for us and provide a hell of a lot of instant liquidity and cashflow so that I can build us even more products and take BitShares to the moon.

Shareholders would go crazy for a second team of developers that worked on what they wanted and for a fraction of CNX. That could add immediate substantial value to BTS in their eyes from a cost, direction and decentralisation perspective. I would probably buy into BTS again.

It would also free up BM & CNX to work on whatever they were passionate about and had funding for, independent of BTS shareholders.

I & think most would support an initiative to bring in new Latvian/other devs as individuals or even ultimately as a team under your management, with your position being paid more generously of course, if that's something you'd consider.

Will be very interested in what you and others can bring to the table in this regard.

[kenCode]

I have been outsourcing since 1993.
I did have a large office with in-house devs and all that overhead for over 16 years but to stay in business you HAVE to outsource and learn how to write detailed project descriptions with milestones, know who the buzzword "devs" are, know who the high value devs are, learn to simplify your english so that you can communicate with anyone in any country, always crosstrain everyone, know what to pay them (don't pay them too little and certainly do not pay them too much (for numerous reasons)).. the list goes on and on. Fuzzy and I chatted about this for a bit, maybe we can chat about this more during the mumble on friday.
 
Call them slaves if you like, but the fact is, $900 goes a looooong way in some countries and those guys drive nice cars, they certainly do not feel like slaves. They feel very successful. They work their ass off too, stay with me for years, never complain about vacation time, or their work hours, or need a fukn gym membership or health benefits from me either. I could go on and on why I only outsource with freelancers. Do you know how much stuff I could build with $45,000? JEEZ
 
This friday (tomorrow) fuzzy and I will be talking about OPENPOS (https://bitsharestalk.org/index.php/topic,20762.0/all.html) and what that 19K project has built and is still building. Those little 2BTS donations (http://i.imgur.com/JyNtgOU.jpg) at the cash registers and mobile wallets will go a long way for us and provide a hell of a lot of instant liquidity and cashflow so that I can build us even more products and take BitShares to the moon.

[cube]

I think with a good pay, it is possible to attract good experienced developers to perform this documentation task. FYI, our community have capable people (eg xeroc, pc and roadscape) who can do the job if they are available and like to take up the role. They can form a documentation team or become part of it.  Having quality documentation for bts as it is now is not an easy task.  It would require the team to go deep into the codes in order to map out the logics.  To ease the work, they would need to consult CNX from time to time as Thom suggested.

[EstefanTT]

If the de dev needs to learn before beeing efficient, I think is better to find a motivated person accepting at first a low payrate. Probably in a country were life is cheap.
Once that dev have achieved his first contract, we can re-hire him with a better payrate because he will be more efficient. That way, we don't pay too much for his "formation" and we also keep him motivated for further jobs.


for your proactivity Jakub, well done !

[yvv]

Your crowd is very funny guys, but your business suffers from one fundamental problem: you are all completely detached from reality. You think that buying slaves from Latvia for $900 will solve your problems? You are wrong. Think again. It will work for a while. But then your slaves will refuse to work for you until you give them all benefits which golden billion in EU and US has. And your government will fuck you hard to not allow this. What are you going to do then? Find more slaves in Africa? It is not going to work.

[Empirical1.2]

Marvellous idea jakub. Perhaps kencode can find a Latvian developer with the requisite skills in addition to very good English?

  Great idea

[Ben Mason]

Marvellous idea jakub. Perhaps kencode can find a Latvian developer with the requisite skills in addition to very good English?

[Thom]

I can get expert coders out of Latvia for $900 month who will work on average 10-12 hour per day (because they WANT to) and will produce typically DOUBLE what any of my American coders ever could. Also, Russia, Turkey and Ukraine are about the same rates. I still have an expert python/C++ guy in India too but his rate is close to American salaries these days, India just isn't as competitive anymore.

Think outside the borders. You guys want to get out of this mess? Talk to me.

@kenCode
The thing is, do "expert coders out of Latvia for $900" speak (and write) fluent English so that they can communicate with CNX and produce comprehensive documentation?

Awesome thread Jakub, I hope to see your efforts succeed.

@kenCode, I am not sure it's practical to achieve the goals Jakub outlined here by someone that doesn't reside in Blackburg VA. It takes a lot of very short convos to distill the info required to become proficient in a system as complex as the BitShares backend.

Not saying it can't be done, but it will be much harder to do without being in VA. It also requires CNX's willingness to carve a portion of the dev's time out for consulting. Also, if the dev is goes away after the contract, as opposed to getting hired by CNX / BCL permanently, a good portion of the investment goes away with them (besides the docs and examples).

You can't expect the contractor to do well in the "ask him anything" role until well into and possibly near the end of the contract period. Makes me wonder if the worker proposal needs to have a phase 1 and 2 for docs vs. api helpdesk.

[jakub]

For me (speaking as a shareholder) having such a person at our disposal (plus all the results of his/her educational work) is worth spending at least 3M - 4M BTS (for 3 months).

[jakub]

I can get expert coders out of Latvia for $900 month who will work on average 10-12 hour per day (because they WANT to) and will produce typically DOUBLE what any of my American coders ever could. Also, Russia, Turkey and Ukraine are about the same rates. I still have an expert python/C++ guy in India too but his rate is close to American salaries these days, India just isn't as competitive anymore.

Think outside the borders. You guys want to get out of this mess? Talk to me.

@kenCode
The thing is, do "expert coders out of Latvia for $900" speak (and write) fluent English so that they can communicate with CNX and produce comprehensive documentation?

[jakub]

Listen, I never said CNX should disappear from bts.  I just don't want them to be the dominant player in both bts knowledge and bts stake.  That is centralization.

Why don't we fund a worker proposal for employing for 2-3 months a C++ developer with these goals:
- thoroughly learn Graphene and its API,
- document it by expanding the existing documentation,
- create friendly tutorials for C++ developers wanting to work with Graphene,
- create working examples of "hello world" smart contracts and operations,
- launch a test network,
- and for the next 2-3 months afterwards play the role of an "ask me anything" expert at our full disposal

Required skills:
- solid expertise in C++
- proficiency in spoken and written English

Preferred experience:
- familiarity with the blockchain concept and/or BitShares
- already worked as a CNX contractor

The main goal of this position would be to "get under the hood and completely map out all the terrain" for all future developers.

Basic info we need to establish:
1. How big is the Graphene source code? (in terms of lines of code) How much time do you expect an experienced C++ developer (not familiar with the blockchain concept) would require to become productive with Graphene?
2. The current status of the Graphene documentation/tutorials and CNX plans in this regard for the next 1-3 months.
3. To what extent is the documentation embedded in the source code useful? Are there any examples there?
4. If I hire a C++ developer and give them the task of becoming a fully-fledged Graphene developer able to create smart contracts and deploy them - where should s/he start? What is the path s/he should follow to learn these skills?
5. What subjects will the documentation by xeroc cover as opposed to the documentation aimed at developers?

This is what we already know:

C++ Dev, 80k-ish / year full time salary?
http://www.payscale.com/research/US/Skill=C%2b%2b/Salary#by_Years_Experience