Mark and Jonathan, I would agree with your frustration. I too would very much like to be able to make more actual progress.
While I would agree that email is not a great medium for many of these discussions and still believe that we need some sort of better threaded discussion platform I think the primary reason we are not making more progress is that we continue to lack the
ability to focus on one issue at a time. We are suffering death by a thousand cuts and are bleeding out.
The “agile” process you describe below is basically the process we have all agreed is the one we are supposed to be following now with the exception of hard time boxing in #2. We certainly want to shoot for a quick time box but a deadline cannot force
consensus. We need to work together and compromise where possible to make progress where consensus is not obvious.
We have also published our roadmap of topics that we plan to tackle in the next couple of months. It is
here .
The one issue we are supposed to be focusing on right now is Sightings.
On the last STIX call we went in thinking we had consensus but discovered it was not as solid as we had hoped. The plan was to pull together an overview of the consensus we thought we had and a list of the open issues for us all to continue the discussion
in a focused way. The next morning the MTI ballot was brought up suddenly distracting the priority for several people including myself away from the sightings issue and onto facets of the MTI issue requiring attention. Around the same time and over the next
couple of days additional topics were brought up on the lists for versioning, timestamps, assets, etc. as well as continuing thread discussions on over a half dozen other topics. All of these topics are completely valid topics for the SC to discuss but we
simply cannot try to discuss them all at once or we will be unable to nail any of them down definitively. As Jonathan points out, none of us have the cycles to follow and contribute to all of these topics at once in anything more than a chatroom mode. The
reality is that all of these topics require more deliberative and analytic thinking than that to get right. Getting done fast does no good if it does not get done right.
So, I would like to reiterate the request and encouragement I have given before,
can we please all agree to show discipline and focus on one issue at a time and work our way through our roadmap?
For other topics people are interested in let us please wait until their assigned time in the roadmap to dive into them. If you have new issues to discuss, please make sure to add them as issues to the tracker and if you think the priority order of our
roadmap should change as we work our way through it, please bring that up and we can reach consensus on potentially adjusting it.
I make no claims that this is a magic bullet that will slay all of our problems but I do believe it is the single most important thing we can do to improve our chances of making progress.
For context that may help address some of Mark’s questions on process the intended lifecycle for an issue would follow along the lines of:
Capture issue in the tracker
Anyone can add comments and thoughts whenever they would like within the tracker
Identify priority of the issue and order it within the roadmap Discuss the issue on the list when its appropriate time comes in the roadmap
Identify and discuss facets of the issue Identify and discuss potential proposed solutions to the issue
Provide updated model snippets and examples as much as possible Potentially create wiki writeups providing details for proposed solutions
Discuss the benefits and risk of various proposed solutions Refine and narrow down solution options Achieve and agree to consensus on solution to pursue
If consensus cannot be reached in an appropriate amount of time consider rescheduling continuing discussion for later in the roadmap. This may not be feasible for issues with a high architectural significance.
Capture key elements of discussion as well as details of proposed solution within the tracker Capture proposed to-be state of the model as a branch in the github tracker Move on to next issue Revisit and refine issue and solution as appropriate due to discussions and decisions on other issues In preparation for Release
Review issue solution and integrate to-be model changes from branch into the overall model. Capture final resolution comments in tracker and close issue
Can we all agree to focus on one issue at a time and discipline ourselves not to initiate discussions on other issues until they are our official focus?
I am hopeful that our current frustrations may offer us the motivation to do better at this going forward and allow us to make better progress.
Thank you for your contributions and commitment to making STIX a success.
Sean
From: " " <
> on behalf of "Bush, Jonathan" <
>
Date: Tuesday, December 1, 2015 at 8:36 AM
To: 'Mark Davidson' < >, " " <
>
Subject: [cti-stix] RE: Can we just make a little bit of progress?
Thank you Mark. I think more than a few of the members are feeling what you are feeling – exhaustion from the endless debate. I believe that some have already
tuned out, we need to take steps now to prevent losing others.
I would suggest –
Co-chairs : Borrowing from Agile here - Can we:
1.
pick a topic, one topic, (maybe that is JSON right now, great, we can slow down all other debate and all take a breath)
2.
drive it to conclusion with a quick vote (like a week perhaps),
3.
implement the resolution
4.
then move on to the next topic?
Can we perhaps do this in some small intervals, like maybe 1 month (instead of a big bang approach)? Can we publish our roadmap of topics that we plan on tackling
in the next N months (6?), knowing that it will most likely change as we go?
I think at some point we also need to address group participation, as it impacts voting quite a bit. Do we maybe have too many people in this group? Makes driving
to a consensus or a majority very difficult.
Also, I would suggest that we are trying to solve far too much over email. Am I the only one that doesn’t have time to sit and read some of these huge email chains
every day? I have to admit, I find myself looking at these emails, with topics that are by no means “light”, and saying “wow, doesn’t anyone have a day job!?”. Huge kudos to Aharon for organizing the face-to-face meeting coming up soon, but I would suggest
that it needs to happen more than once a year. 2 at a minimum. I feel that if we don’t change to an agile, small release format as noted above, the more frequent face-to-face meetings are critical. Each one should end in a release of the format.
From:
[ mailto: ]
On Behalf Of Mark Davidson
Sent: Tuesday, December 01, 2015 8:12 AM
To:
Subject: [cti-stix] Can we just make a little bit of progress?
I'd like to offer a perspective, and this is in response to nobody in particular.
STIX 2.0 was kicked off on October 21, which is about ~5 weeks of work time (subtracting the US Thanksgiving holiday week). Since then, topics under discussion have included from
sightings, the indicator type vocabulary, an MTI discussion (which is currently under a TC-wide vote), relationships, and timestamps. From what I can see, the ONLY topic we seem to be anywhere near reaching a conclusion on is the MTI discussion because it's
resolution was motioned.
I hope I surprise nobody when I say that the outsider perspective of our TC, and this SC in particular, is that we debate topics endlessly and never reach a conclusion
on them. I challenge everyone to change this perception. Can we please drive resolution on some of the topics we've discussed? Can some design artifacts be updated with the current state of the discussion?
By now we should have an updated STIX 2.0 architecture with Relationships and possibly Sightings broken out into their own top level objects. Other bits and pieces should
also be documented by now. We seem to have a rough consensus on how timestamps should work (as Bret called out yesterday) - where can that be documented?
We should have a model that is malleable and gets updated on a regular basis as we discuss topics, decide things, and change previous decisions (yes, we are allowed to
do that!). How can we do that?
I find the endless debate tiring, and it discourages me from participating. I want my contributions to help push us toward resolution of issues, not add to the cacophony
of opinions. If we go back to before the STIX 2.0 kickoff, this SC has been at it for 5+ months - what topics have we resolved?
Can we just make a little bit of progress?
Thank you.
-Mark
DTCC DISCLAIMER: This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error, please notify us immediately and delete the email
and any attachments from your system. The recipient should check this email and any attachments for the presence of viruses. The company accepts no liability for any damage caused by any virus transmitted by this email.
