1

Many promises that the Coordinator code would be open sourced.

It was stated that when the IOTA Foundation was formed, that the COO would become open sourced.

enter image description here

This was also confirmed by David on reddit: "I will confirm."

And in other places too:

enter image description here

enter image description here

This has yet to be done.

Assuming that the Coordinator works (which it seems to), then what possible reason could there be to not open up that code?

  • 2
    There are several reasons for information to be kept private (google for why data classification is important if you need them). It's hard to make public information private again, so before releasing anything to general public some special procedures need to be done. There is already a public version of the Coordinator, if it's the version that is being run on the mainnet - that's probably a secret. Or not. – Come-from-Beyond Aug 12 '18 at 18:20
  • 2
    This question and its answers are speculative by nature, unless a member of IOTA Foundation, who actually knows the answer, chooses to give a response. – Bayequentist Aug 21 '18 at 22:20
2

I suppose you ask about a full coordinator source, including the Merkle hash tree (or whatever is used) data which is used to sign the milestones.

Not releasing this kind of information is currently crucial to IOTA because everybody else could issue milestones, then.

If you just ask for the source without any private data (like private information to issue milestones, etc.) then reasons could be the fear of security holes in a bad implementation, etc.

Some attacker could gather informations due to analyzing source and possibly find unknown attack vectors.

  • 3
    You asked for possible reasons to not open up the code. You didn't ask for possible reasons to open it up. It's furthermore likely that somebody else wants to know exactly why the "full source" (e.g. all files which are currently used to run the coordinator) shouldn't opened up. – Matthias Aug 22 '18 at 20:01
  • I understand this. But those questions are also for all the other ppl searching the web for answers. And again: I answered your question in a correct manner, you didn't ask for reasons to open it up. So a downvote because of this is somehow rude. – Matthias Aug 23 '18 at 9:04
  • It's not about thinking that other ppl are stupid. Maybe there are just some ppl not knowing that the COO may contain such crucial information. The question you stated will also be interesting for total newbies and have already be asked by ppl on Discord or earlier on Slack. Additionaly the reason why you downvoted me was your claim that I didn't answer the question, which I clearly did. You can continue finding reasons for this behaviour or you can just stop at this point. I won't care about any future excuses or statements regarding this situation from you. – Matthias Aug 25 '18 at 14:04
  • "because everybody else could issue milestones, then." Unless they acquire ownership of the Github account of IOTA Foundation, I don't see any way for anyone to pull that off. – Makan Tayebi Oct 11 '18 at 9:15
1

It is business reason , if they publish coo somebody can copy whole work easily , so once there will be bigger adoption they will make it public and it will be harder for duplicate project to take same adoption. If you are investor in IOTA you aim to not see COO code until big adoption is in place, otherwise it can devalue your investment by many duplicate projet.

  • I think this is the answer to the question. $$$ – Makan Tayebi Oct 11 '18 at 9:18

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy