[XMR] Monero - A secure, private, untraceable cryptocurrency

[AlexGR]

Combining the two together, there were 22 927 lines of code over the first 8 months, which is 18 342 hours of work, so $1 375 620 worth of effort."

Just 1.25 lines of code per hour of work? Isn't that exceptionally slow?

You obviously know nothing about programming. Especially when it pertains to inherited and uncommented code.

It's true I haven't done much coding since mid-90s or something... I played around a bit with mining software and stuff around '13-'14 for scrypt and x11, but other than that, no. Anyway 1.25 lines of code per hour seems totally absurd even with inherited code. After all, there are coins out there that build their anonymity systems from scratch. Their programming work should be valued in the tens of millions per this type of accounting

[GTO911]

The troll is ignored  

Let us all choose not to reply to obvious trolls ever. When in doubt, check the posters post history. If we all do this, their trolling will become useless and the puppet masters will have to find new ways to cause fud

[smooth]

Combining the two together, there were 22 927 lines of code over the first 8 months, which is 18 342 hours of work, so $1 375 620 worth of effort."

Just 1.25 lines of code per hour of work? Isn't that exceptionally slow?

You obviously know nothing about programming. Especially when it pertains to inherited and uncommented code.

It's true I haven't done much coding since mid-90s or something... I played around a bit with mining software and stuff around '13-'14 for scrypt and x11, but other than that, no. Anyway 1.25 lines of code per hour seems totally absurd even with inherited code. After all, there are coins out there that build their anonymity systems from scratch. Their programming work should be valued in the tens of millions per this type of accounting

I think you're confused about the effort of implementing a particular feature vs reworking/improving a large number of things across an entire system. For example, darksend is around 2000 lines of code (didn't bother to strip out blank lines and comments which aren't normally counted), which makes it roughly 10% of the above numbers. https://github.com/darkcoin/darkcoin/blob/master/src/darksend.cpp

This is not meant to denigrate the amount of work on darkcoin/dash, as there has been a lot done on that coin besides just darksend, and lines of code is a rough measure anyway, but you can't say that "build their own anonymity systems from scratch" is necessary more work than, say, bandwidth controls on the p2p (without looking I would guess those commits were of similar size, and more work involved with reverse engineering).

EDIT: regarding 1.25 lines/hour, aminorex says 20 lines/day so that would be a bit higher, but a lot does indeed depend on the nature of the work. There is a some decent research that breaks down programming productivity into a variety of factors and things like quality (within which I would include commenting or lack thereof) and familiarity with the existing code are certainly significant. Also, anything that involves working on an existing system that you can't break/redesign at will becomes much harder (another factor).

IIRC devs generally cost about 600/diem, and produce about 20 lines in a day over the long-run.

[GTO911]

Let me have the honor of nominating Monero's first official troll list, may this list's growth know no bounds


Ignorelist:
BlockaFett
MasterMined710
ElTomeko27
ahoenk

Instructions
Accounts above are considered trolls, that is, their goal is not to be constructive, just to FUD. Since the best way to fight a troll is to not feed him, you are encourage to put every one of them on ignore.

How to ignore someone:
1. On the left panel, look for the word "ignore"
2. Click, it's done. If you changed you mind, just click again, this time on "unignore"
You can access your personal ignore list by going to your profile and click "Ignore user options". Notice ignore is forum-wide, not simply thread-wide, so you won't see any of his post, even on other threads. When pasting this text, remember to update it with new account (or maybe delete some)

[smooth]

^This user DASH troll is currently ignored.

I'm not sure if you were referring to AlexGR but I don't think he's really a troll, just expressing an opposing point of view, which i certainly welcome. No need for an echo chamber of cheerleaders.

There are certainly people who post useless snarky comments, blatantly hostile attacks, etc. but he's not doing that.

[dewdeded]

"AlexGR" is ok.

Add "primer-" & "TaunSew" to the list.

[cAPSLOCK]

^This user DASH troll is currently ignored.

I'm not sure if you were referring to AlexGR but I don't think he's really a troll, just expressing an opposing point of view, which i certainly welcome. No need for an echo chamber of cheerleaders.

There are certainly people who post useless snarky comments, blatantly hostile attacks, etc. but he's not doing that.

This is very very very important as a general rule.

[fluffypony]

Combining the two together, there were 22 927 lines of code over the first 8 months, which is 18 342 hours of work, so $1 375 620 worth of effort."

Just 1.25 lines of code per hour of work? Isn't that exceptionally slow?

You obviously know nothing about programming. Especially when it pertains to inherited and uncommented code.

It's true I haven't done much coding since mid-90s or something... I played around a bit with mining software and stuff around '13-'14 for scrypt and x11, but other than that, no. Anyway 1.25 lines of code per hour seems totally absurd even with inherited code. After all, there are coins out there that build their anonymity systems from scratch. Their programming work should be valued in the tens of millions per this type of accounting

That figure specifically excludes inherited code. If you read the blog post it's from we go into great detail to explain how we excluded libraries, license changes, and all other irrelevant changes from the count.

The hours / cost estimation is based on The Mythical Man Month - I'm surprised that you didn't read it in the mid-90s when you were "doing much coding", as it was first published in 1975 and then republished with four extra chapters in 1995.

As to why it isn't developed faster, there's an excellent comment Peter Todd made yesterday -

A huge part of the purpose of Bitcoin is to prevent things from happening, things like double-spending, inflation, stealing coins, etc. Additionally, it must prevent those things from happening in the face of active attack.

For instance, in regular distributed systems engineering you can assume modulo failures the components of the system act in predefined ways; when those components do fail the failures happen randomly. I can easily shard a database for a bank across ten servers, and handle validation of the data by random spot-checking to get a sufficiently certain probability of detecting a failure. But Bitcoin is decentralized systems engineering, where security matters. If I'm running one of those shards holding part of the blockchain, I can't assume the others running other parts of the blockchain are honest - it can easily be in their incentive to be dishonest and deliberately not follow the protocol I'm expecting them too. I can't just test my way to ensuring the system works with high probability; I have to imagine how an attacker would attack it.

An analogy: imagine someone told you they had extensively tested a new cryptographic signature implementation by creating running billions of automatically generated unittests, all of which passed. Is this enough to know it works? Hell no! The attacker can do the math to find the one-in-a-trillion-trillion-trillion case that causes your code to fail, something that would simply never happen due to a random failure.

Of course we could develop much faster, it would just require us to abandon all decentralised systems design and instead assume that nodes are trustworthy. Clearly that's not an option.

[AlexGR]

Combining the two together, there were 22 927 lines of code over the first 8 months, which is 18 342 hours of work, so $1 375 620 worth of effort."

Just 1.25 lines of code per hour of work? Isn't that exceptionally slow?

You obviously know nothing about programming. Especially when it pertains to inherited and uncommented code.

It's true I haven't done much coding since mid-90s or something... I played around a bit with mining software and stuff around '13-'14 for scrypt and x11, but other than that, no. Anyway 1.25 lines of code per hour seems totally absurd even with inherited code. After all, there are coins out there that build their anonymity systems from scratch. Their programming work should be valued in the tens of millions per this type of accounting

I think you're confused about the effort of implementing a particular feature vs reworking/improving a large number of things across an entire system. For example, darksend is around 2000 lines of code (didn't bother to strip out blank lines and comments which aren't normally counted), which makes it roughly 10% of the above numbers. https://github.com/darkcoin/darkcoin/blob/master/src/darksend.cpp

This is not meant to denigrate the amount of work on darkcoin/dash, as there has been a lot done on that coin besides just darksend, and lines of code is a rough measure anyway, but you can't say that "build their own anonymity systems from scratch" is necessary more work than, say, bandwidth controls on the p2p (without looking I would guess those commits were of similar size, and more work involved with reverse engineering).

But even a new feature is not something static and, it too, requires maintenance and further work or re-implementation. DarkSend operation, or Masternode operation, has been radically altered and tweaked a lot. DarkSend must have been rewritten as to the way it works at least 3-4 times since Feb '14, to the point that it's no longer DarkSend but rather a ...premix of coins that are sent normally. And it continues to evolve (blinding).

The Masternode system and the way the nodes get paid has also changed quite a bit due to all the problematic forks and payment cheating due to the semi-voluntary way that is currently in use. Although what I had in mind when posting "from scratch", was Bytecoin itself and Shadow (with marketcaps that wouldn't even justify the work that has been done if we take the accounting provided a bit earlier at face value), and of course the developers also have to maintain the code for problems or bugs.

In any case, I don't think that it's easy to quantify (in terms of value) the coding task and then throw a price like that. I don't have the right formula to provide the precise value of a coding work, but my mind can't accept 6 and 7-digit values for what, again, my mind perceives as much less than that (in terms of work).

We had the same discussion a few days earlier on the 200k USD payment processor and I asked what am I missing... I saw the answers but I still couldn't get my mind to accept them. I finally concluded that in my part of the world, these numbers are simply astronomical for coders while in other parts they may be considered "normal". With 200k USD you'd probably hire at least 10 full-time coders for a year around here.

But then again I was thinking it's not only a difference in salaries between Greece and the US. For example, if XMR's work is indeed 1.4mn USD, then how can that be reconciled with what I read in a linked post to the VIA thread the other day? According to the posts there (explaining what happened to their ICO money) Peter Todd is working for them for 4k per month... So for the alleged 1.4mn for XMR's work, you'd get more than 25 Peter Todds for a year (?). Again something doesn't compute.

[dewdeded]

I live in Germany, good coders here earn 4000 Euro / month. So your $200.000 get 3,7 coders / year (not 10!!!).

And everybody doing Crypto Currency-development or who is intelligent & skilled enough to do so, would be considered a "good coder". Last but not least, you compete with >1000 Bitcoin startups in Europe, USA and Asia for the few skilled people, who can really develope a Crypto Currency.

[Joshuar]

Yea, it really depends on the country and type of work being done. In the USA, expert coders can earn upwards of $100 an hour.

Even the new forum software for bitcointalk.org is costing over 1million USD.

[othe]

I live in Germany, good coders here earn 4000 Euro / month. So your $200.000 get 3,7 coders / year (not 10!!!).

And everybody doing Crypto Currency-development or who is intelligent & skilled enough to do so, would be considered a "good coder". Last but not least, you compete with >1000 Bitcoin startups in Europe, USA and Asia for the few skilled people, who can really develope a Crypto Currency.

But thats the typical german way - better earn less but have a secure job ;-)

Thats really lower end, a normal payment per hour is 75EUR for a self employed programmer - that's the average.

Source: http://www.golem.de/1109/86156.html

It's from 2011 tho, i guess its more today.

Yea, it really depends on the country and type of work being done. In the USA, expert coders can earn upwards of $100 an hour.

Thats low end, in NY or SF etc. you won't find a lot of people who would work for that - granted that a 60square metres appartment there costs easily a few grand per month its obvious they have to charge more.

But even a new feature is not something static and, it too, requires maintenance and further work or re-implementation. DarkSend operation, or Masternode operation, has been radically altered and tweaked a lot. DarkSend must have been rewritten as to the way it works at least 3-4 times since Feb '14, to the point that it's no longer DarkSend but rather a ...premix of coins that are sent normally. And it continues to evolve (blinding).

Well, thats a typical bad design, and you guys don't even have to care about other stuff nor is there serious research published.
Rewriting things 3-4 times clearly shows that design errors have been made from the start.
Designing a correct system often costs more time than implementing it.

We had the same discussion a few days earlier on the 200k USD payment processor and I asked what am I missing... I saw the answers but I still couldn't get my mind to accept them. I finally concluded that in my part of the world, these numbers are simply astronomical for coders while in other parts they may be considered "normal". With 200k USD you'd probably hire at least 10 full-time coders for a year around here.

Thats out of context - that wasn't asked for programming costs - it was a sell of shares in a company; Whoever would have invested gets paid out dividends etc - you know, like apple shares
And it wasn't 200k USD - it was coins, who happened to be worth <1usd/coin at the time, and a lot of people already had them so they wouldn't even need to use their Fiat; and they would have got paid back in coins if they wanted also.

[AlexGR]

EDIT: regarding 1.25 lines/hour, aminorex says 20 lines/day so that would be a bit higher, but a lot does indeed depend on the nature of the work. There is a some decent research that breaks down programming productivity into a variety of factors and things like quality (within which I would include commenting or lack thereof) and familiarity with the existing code are certainly significant. Also, anything that involves working on an existing system that you can't break/redesign at will becomes much harder (another factor).

IIRC devs generally cost about 600/diem, and produce about 20 lines in a day over the long-run.

Aha, ok. Now we have the source.

That figure specifically excludes inherited code. If you read the blog post it's from we go into great detail to explain how we excluded libraries, license changes, and all other irrelevant changes from the count.

Yes, I would expect that to be the case.

The hours / cost estimation is based on The Mythical Man Month - I'm surprised that you didn't read it in the mid-90s when you were "doing much coding", as it was first published in 1975 and then republished with four extra chapters in 1995.

I was relatively (too) young back then to care about the economics of coding.

As to why it isn't developed faster

I have no criticism in this regard. I was just wondering about the figure of lines/code that my mind can't accept as natural.

I live in Germany, good coders here earn 4000 Euro / month. So your $200.000 get 3,7 coders / year (not 10!!!).

Yes, Germany has a relatively high wage for this type of work because there are many companies that want good coders. Wages are a function of demand and supply. There are no major software firms around here, or manufacturing companies that want software written for their equipment. So if you want to work as a coder (here), you'll get a much lower wage. Many coders / few seats / employers have the upper hand in negotiations. Coders get the upper hand in time, as companies become dependent upon them.

Yea, it really depends on the country and type of work being done. In the USA, expert coders can earn upwards of $100 an hour.

Even the new forum software for bitcointalk.org is costing over 1million USD.

That's a lot of money for a forum. I wouldn't pay 1/10th of that. But then again, that's me, not living in the US, seeing "costs" in a different way.

[megges]

To the trolls and wannabe programers in here, after you didn't know it better you should perhaps take a look at:
http://en.wikipedia.org/wiki/Source_lines_of_code

I'm not a XMR fan, i'm more DASH, but in the end i am a software engineer - and i can tell you, lines of code is not a good measurement of anything...

So before you use something like:
"they just did 1 lines of code per hour"
or
"they did 1mio lines of code in just 1 month"
as an argument, you should at least read the wikipedia article, from my experience LOC is not a good measure of development, and i think most programmers would agree with that!

wikipedia:

=== Disadvantages ===
# Lack of Accountability: Lines of code measure suffers from some fundamental problems. Some think it isn't useful to measure the productivity of a project using only results from the coding phase, which usually accounts for only 30% to 35% of the overall effort.
# Lack of Cohesion with Functionality: Though experiments have repeatedly confirmed that while effort is highly correlated with LOC, functionality is less well correlated with LOC. That is, skilled developers may be able to develop the same functionality with far less code, so one program with less LOC may exhibit more functionality than another similar program. In particular, LOC is a poor productivity measure of individuals, because a developer who develops only a few lines may still be more productive than a developer creating more lines of code - even more: some good refactoring like "extract method" to get rid of [[redundant code]] and keep it clean will mostly reduce the lines of code.
# Adverse Impact on Estimation: Because of the fact presented under point #1, estimates based on lines of code can adversely go wrong, in all possibility.
# Developer’s Experience: Implementation of a specific logic differs based on the level of experience of the developer. Hence, number of lines of code differs from person to person. An experienced developer may implement certain functionality in fewer lines of code than another developer of relatively less experience does, though they use the same language.
# Difference in Languages: Consider two applications that provide the same functionality (screens, reports, databases). One of the applications is written in C++ and the other application written in a language like COBOL. The number of function points would be exactly the same, but aspects of the application would be different. The lines of code needed to develop the application would certainly not be the same. As a consequence, the amount of effort required to develop the application would be different (hours per function point). Unlike Lines of Code, the number of Function Points will remain constant.
# Advent of [[Graphical user interface|GUI]] Tools: With the advent of GUI-based programming languages and tools such as [[Visual Basic]], programmers can write relatively little code and achieve high levels of functionality. For example, instead of writing a program to create a window and draw a button, a user with a GUI tool can use drag-and-drop and other mouse operations to place components on a workspace. Code that is automatically generated by a GUI tool is not usually taken into consideration when using LOC methods of measurement. This results in variation between languages; the same task that can be done in a single line of code (or no code at all) in one language may require several lines of code in another.
# Problems with Multiple Languages: In today’s software scenario, software is often developed in more than one language. Very often, a number of languages are employed depending on the complexity and requirements. Tracking and reporting of productivity and defect rates poses a serious problem in this case since defects cannot be attributed to a particular language subsequent to integration of the system. Function Point stands out to be the best measure of size in this case.
# Lack of Counting Standards: There is no standard definition of what a line of code is. Do comments count? Are data declarations included? What happens if a statement extends over several lines? – These are the questions that often arise. Though organizations like SEI and IEEE have published some guidelines in an attempt to standardize counting, it is difficult to put these into practice especially in the face of newer and newer languages being introduced every year.
# Psychology: A programmer whose productivity is being measured in lines of code will have an incentive to write unnecessarily verbose code. The more management is focusing on lines of code, the more incentive the programmer has to expand his code with unneeded complexity. This is undesirable since increased complexity can lead to increased cost of maintenance and increased effort required for bug fixing.

[cAPSLOCK]

My OT two moneros.

Making this big a deal about ignoring trolls is harmful to our image and makes us look intolerant.  People can use the ignore function independently.

Fighting with trolls is often not worth it.  But keeping a blacklist seems a little heavy handed.

Ignore trolls.  Or not... Freedom of speech is good.  As well as freedom to ignore.
  

[Cryptock]

why is this secure, private, untraceable cryptocurrency still in the title 

[dewdeded]

why is this secure, private, untraceable cryptocurrency still in the title 

Because Evan Duffield said so!

[smooth]

I'm not sure if you were referring to AlexGR but I don't think he's really a troll, just expressing an opposing point of view, which i certainly welcome. No need for an echo chamber of cheerleaders.

too bad because he surely looks like one.

"let me imply how this coin development is exceptionally slow (as opposite of the coin in my sign)"

It's his opinion. He may be right he may be wrong.

In part I don't think this project has done a tremendously good job of communicating all the things we've done. A lot is below the surface, rather than highly visible items like a GUI or a nameplate feature. Yes we have the year in review document and fluffypony posts line counts and such, but you really can't expect people who aren't following the project like a hawk to appreciate all of that, nor the context of it.

Also a lot of the work is on different repos due to the decentralized nature of the project (many developers working independently) and doesn't get merged to the main until it is completely done (recent example of this is the p2p resource limits, which have been under development for >6 months), or maybe not at all if we decide to go a different direction. The latter doesn't mean the work wasn't worthwhile either; sometimes in order to decide not to pursue something you have to at least start implementing it first, then reconsider. But it all makes things less visible if you aren't paying that close attention.

In short I think some of the criticism of "exceptionally slow" progress is uninformed and misguided but not necessarily malicious (though some is).

[Joshuar]

So with the recent news of the bitcoin foundation etc, I thought, why not make a "decentralized" bounty fund for developing Monero. Companies/Individuals would put in $ to the fund periodically, say every 6 months. There would need to be a minimum threshold, so maybe $(X) amount every 6 months, and developers can claim (X) amount of $ as their paycheck with the consensus of the community or miners, etc. The companies or individuals that put in $ towards the fund would have their names in a record book, like a book dedicated towards those that helped monero succeed via donations to pay for development. Maybe even a few perks like they get a claim to physical moneroj every so often.

[smooth]

So with the recent news of the bitcoin foundation etc, I thought, why not make a "decentralized" bounty fund for developing Monero. Companies/Individuals would put in $ to the fund periodically, say every 6 months. There would need to be a minimum threshold, so maybe $(X) amount every 6 months, and developers can claim (X) amount of $ as their paycheck with the consensus of the community or miners, etc. The companies or individuals that put in $ towards the fund would have their names in a record book, like a book dedicated towards those that helped monero succeed via donations to pay for development.

Fluffypony posted something about the upcoming crowdfunding function of the forum recently, which is very close to this (you can not only sponsor specific features/projects, but also propose your own). I don't have a link to it but if someone has that handy go ahead and reply.