Upayavira, Please review the Flex archives where Bertrand advises that "consensus" in a discussion like this only has to be "general consensus". Otherwise, you allow rule by a single individual. In all 90+ emails on the original thread and this one, I don't recall any other individual writing in favor of Carlos's change. And that does make it appear that we are repeating our history. For some reason, we've had other individuals decide to hold their ground and call other people names despite clear lack of support for a code or other change. We had a long thread with Carlos earlier about ClassList. It was resolved technically. I am holding out hope this issue will be resolved technically as well. Then we won't be repeating our history.
We have new contributors, one of which is now a new committer. How do we want to model for these new contributors about how to conduct themselves if there is a disagreement? In every social group I'm in, the group generally does what general consensus is (yes, there are exceptions). Individuals are expected to see that there is lack of support and either gather support or defer to others. Otherwise, who is the dictator? Those who cannot be wrong are not good teammates. Nobody is perfect. I've made my share of questionable or incorrect decisions. If I'm the only one with a particular opinion or people point out problems with something I did, I take my shot at convincing them otherwise, but I would like to think that after that I generally defer to the majority. I dearly hope we do not want to model that ignoring general consensus is the way to act. If you read the words in my last post carefully, you will see that I did not actually attack Carlos. However, I see that I should have chosen my words more carefully because I can see how he could see it as an attack. So my apologies for being careless. Instead I was trying to point out to Carlos that he could be seen as a problem personality by not continuing with the technical discussion and not being open to not having his proposed changes accepted. I was hoping that would open his eyes to being more open about his proposal. Quite frankly, I'm still open to the possibility that Carlos is right and we aren't understanding him or even that we are all actually in agreement but just aren't understanding each other, which is why I'm willing to continue the technical discussion and am going to ignore all of the personal attacks. I am hopeful that Carlos will be open to the possibility that his changes are not correct and that that outcome does not make any judgement on Carlos the person. That's key to me: a flaw in my technical offerings does not mean there is some major flaw in me. I try not to take a dislike for one of my a code changes as a personal attack. I will try to summarize my understanding of the technical discussion on the original thread and continue the technical discussion there. My 2 cents, -Alex On 5/11/18, 12:30 PM, "Upayavira" <[email protected]> wrote: Alex, Just to note a dangerous comment below. You rightly note that consensus is important at Apache. Then you say that "consensus is" because four of you agree. But Carlos doesn't, therefore it isn't consensus, unless Carlos is a non-person at Apache - and I don't think that's somewhere you want to go!! Upayavira On Fri, 11 May 2018, at 5:38 PM, Alex Harui wrote: > Carlos, > > Fundamentally, the misunderstanding is that you think you have provided > technical reasons and nobody else thinks so. It appears to me and > others that you are using a bug to justify a non-scalable coding pattern > that is definitely not DRY. > > I'm not sure I want to discuss the technical aspects of your changes in > this thread. I left what I think is a concise question for you in the > original thread. I suggest that we continue the discussion in that > thread. > > IMO, consensus is an important value at Apache. The consensus is that > the approach you are taking is not correct. And socially, the response > to consensus saying that the approach you are taking is not correct > should be "oops, let me fix it quickly" or "oops, I'll revert and put it > in a branch". No one individual is being a dictator here. There are > four of us who don't think your commit was the right thing, and nobody > besides you who thinks it is right. If you are not open to the > possibility that your approach is not correct, then we definitely have a > problem on our hands. Individuals who cannot build consensus but still > think they are right are having trouble communicating, or are problem > personalities. > > My 2 cents, > -Alex > > On 5/11/18, 6:55 AM, "[email protected] on behalf of Carlos > Rovira" <[email protected] on behalf of [email protected]> > wrote: > > Thanks Upayavira, > > for you thought, I think very well brought. I'm fact think that most of the > things you described was happening here. I tried to do my best to explain > lots of time with different words the reasons, maybe much people here are > very busy an some explanations was read "from above". I'm for explaining > more times if is needed, but as well I think we all need in the point that > no body is in posession of truth and that we can get a consensus with a > mixture of all the points. I'm more on this. I think this is not a "one > thing or the other", in this conflict, like in the rest we can have, I > always target for a mixture that makes all people happy, at least in this > conctrete problem we can get to it. > > Thanks > > > 2018-05-11 15:45 GMT+02:00 Upayavira <[email protected]>: > > > All, > > > > When conflict starts, it pretty much *always* starts with a > > misunderstanding. > > > > So what is interesting here is not so much the technical merit, or > > otherwise, within this discussion, but the routes by which > misunderstanding > > occurred and how it led to conflict. > > > > What can we learn from this particular occasion? What could any of > us have > > done differently in our communication? What incorrect assumptions > did any > > of us make? (Usually, in conflict, both sides are making incorrect > > assumptions). > > > > If this does have the same flavour as previous conflicts, what can > we > > learn about collective communication? What (probably small) > adjustments can > > we make? > > > > Often one small adjustment is to watch out for one's own belief in > your > > rightness. I'm certainly often guilty of this. It closes me down > to what > > others have to say. How can we spend more time reading/listening > before we > > respond? > > > > Just some thoughts. > > > > Upayavira > > > > On Fri, 11 May 2018, at 1:14 PM, Carlos Rovira wrote: > > > Hi Harbs, > > > > > > really appreciate the new tone of your words. I think in that's > the way > > to > > > reach consensus on things. I was feel attacked, and the final > point was > > > Alex, in a position using words that should "make me understand > the right > > > position". I sincerely know here nobody here is in the > possession of > > truth, > > > but when people try to convince others in this way, I think > there's no > > > dialog possible. > > > > > > I think all the problem is that before discussion Basic was just > the > > basic > > > implementation or most "raw" way to do things. As you said, the > basic > > > building blocks. After this discussion Basic sundenly becomes a > piece in > > > the framework that we all must to use and can't avoid. I think > that's > > wrong > > > by design. I think Core is what symbolizes that piece. and Basic > is just > > > one more UI set, at least regarding on how is designed. Basic is > not an > > UI > > > set library that has it's own CSS that Links lots of beads and > styles to > > > the application that uses it. And Jewel is in itself a non > dependent UI > > set > > > and is how I'm designing it with lots of hours invested in make > it the > > most > > > simple, reusable solution that enforces PAYG, DRY and > composition. > > There's > > > no point in make me embrace Basic for Jewel since there's no > need to use > > > it. And that should be understood. I give lots of technical > points as you > > > all requested, and hope all that previous emails will be > sufficient for > > you > > > to reconsider the position of making all libraries in Royale > obligated to > > > link Basic. I'll never would want Jewel make that, since I hope > people > > > could get rid of useless dependencies and choose the pieces of > code that > > > server their needs. > > > > > > About the Alex emails, although he wants to agree to the mantra, > I think > > I > > > responded with sufficient valid points to expose that while I > agree > > mostly > > > in the way he envisions royale. This concrete point of making > Basic a > > Core > > > piece is not valid for me. > > > > > > So, I don't mind if finaly some pieces of Basic goes to Core, or > if Basic > > > is splitter in two libraries. But the rest of the project should > not link > > > obligatorily a library that enforces some CSS and beads if > there's no > > need > > > of it. Maybe Core + some more part of Basic (lets's call > BasicPrime), is > > > really the Core here, so for me if all libraries and > applications must > > use > > > it, lets put all what's needed in Core. If you like more to > split in > > Core + > > > BasicPrime, for me is ok as well although I think is the same as > one > > Core, > > > and only will make things a bit more complex with two library > core > > projects. > > > > > > And the end of this, and as you can see I think there's room to > make more > > > things here. I only want one concrete thing and is to avoid the > > obligation > > > to link a library of a different UI Set that brings lots of > things to a > > > Jewel App that should not be there due mainly to CSS and things > that > > comes > > > with it. > > > > > > To resolve this, is far easy to move from the current point, > that revert > > to > > > make this. And that can be done if we all work as a team. > > > > > > I'll be glad to help in all that I can, since I'm fully working > all day > > on > > > this. But to do so we must to think that we all want things done > in a > > > concrete way, and my point is that really we are not overlapping > what we > > > need and all what all need have room in this project, since in > the > > essence, > > > we agree in PAYG, composition and the overall structure. > > > > > > Hope this helps to calm things. Again, appreciate your tone and > makes me > > > think all have solution here. Thanks > > > > > > > > > Carlos > > > > > > > > > > > > > > > > > > 2018-05-11 13:13 GMT+02:00 Harbs <[email protected]>: > > > > > > > Hi Carlos, > > > > > > > > I just took a step back and re-read the thread. I’m not sure > exactly > > where > > > > things went off the rails. The discussion seems to be mostly > technical > > to > > > > me. I’m sorry if you feel attacked, and I can tell you that I > did not > > mean > > > > anything like that. I definitely appreciate all the work you > have been > > > > doing. > > > > > > > > When I mentioned that my app was broken I was just explaining > my > > > > frustration. My broken app is not a reason in itself to revert > things > > and > > > > like I’ve mentioned quite a few times already, I did not mean > that I > > would > > > > revert commits without consensus. I’m sorry if it was taken > that way. > > If a > > > > refactor is the right thing to do and others agree, I am fine > to go > > along > > > > with it. I’m not trying to push my own agenda if it’s wrong. > > > > > > > > As hard as it might be to accept, I truly don’t understand the > > technical > > > > drive to do the refactor. If others do understand it, I would > be > > willing to > > > > go along with it even if I don’t. > > > > > > > > However, I’m not seeing that there is a consensus that the > refactor is > > > > correct. I might have missed some other discussion before the > > refactor, but > > > > it was not clear to me before the fact that a refactor was > happening. > > > > > > > > No-one is trying to destroy others’ work and I don’t think > there are > > any > > > > personal attacks going on here. We simply need to come to a > consensus > > on > > > > whether the refactor is the right way to go or not. This > should not be > > a > > > > personal disagreement at all. I think we all generally work > well > > together. > > > > If you feel strongly that it’s the right thing to do, then > please try > > to > > > > find the words to convince others of that. Alex asked some > good > > targeted > > > > questions. That seems like the right place to figure this out. > > > > > > > > I hope we resolve this soon, > > > > Harbs > > > > > > > > > On May 11, 2018, at 12:53 PM, Carlos Rovira > <[email protected] > > > > > > > wrote: > > > > > > > > > > Hi, > > > > > > > > > > in writing this to summarize all the problems, far beyond > code, that > > we > > > > are > > > > > living in the "Container Change" thread. > > > > > > > > > > We're repeating the same pattern with the same problems we > had in > > Apache > > > > > FlexJS project. Now that we created this new project Apache > Royale, I > > > > > though we'll be not living the same, but the problems are > here. And > > are > > > > > exactly the same problems. The difference is that only a > subset of > > > > > contributors are the same, so maybe the problems at that > time wasn't > > > > > exactly as we thought? I must say that at that moment I was > not part > > of > > > > the > > > > > discussions, so not part of that subset. You all can get > your own > > > > > conclusions from this new discussion. > > > > > > > > > > For me the fact that a PMC member state that he's "going to > revert" > > > > > something that is working only based on how this affects at > its own > > > > > personal application code is simply not acceptable at all. > This > > already > > > > > happen in FlexJS, and with the same person. This is not the > project > > of > > > > > anyone here. Is an Apache OS project and for that reason we > cann't > > make > > > > our > > > > > changes thinking in a single one application. > > > > > My changes are not motivated to match a concrete > application, but to > > > > serve > > > > > the general purpose of this technology. > > > > > > > > > > In that thread, people asked me about "technical > motivations". I > > > > expressed > > > > > various times a significant populated list of points based > on the > > core > > > > > points that are part of our nature (PAYG, DRY, composition, > and > > more...) > > > > > and even some significant improvements that we get after the > > refactor: > > > > > reduced size in *all* example applications that use Jewel > about a > > 40%, > > > > > avoid collisions with other UI sets that do the same, and > many others > > > > that > > > > > I don't want to repeat here. > > > > > > > > > > Then I asked for a list of the same "technical motivations" > to have > > Basic > > > > > present in all applications made with Royale. And the only > > motivation was > > > > > not technical was: "since we did in this way before and we > don't > > want to > > > > > change this". So completely philosophical. > > > > > "Why I must to link or depend from Basic? response: Since we > did > > always > > > > in > > > > > the same way..." ¿¿?? > > > > > > > > > > The thread was asking me the same things although I was all > the time > > > > > responding to the same in different ways to make people > understand > > all > > > > the > > > > > implications, and provide my support to make changes and > help with > > any > > > > > issue. But the response continues to be: "No. My application > is > > broken > > > > and > > > > > I want to revert your changes" (a.k.a, my application is > more > > important > > > > > that the Apache Royale project and I think I can dictate how > we > > should > > > > > proceed) > > > > > > > > > > For me this is some kind of dictatorial way of doing things, > and not > > the > > > > > apache way. > > > > > > > > > > This makes me change my way of doing things in this list > though the > > > > > thread..., since people here is "dictating" certain things, > I think, > > I > > > > have > > > > > the same rights than them, so my commit was done in the > right form. > > Since > > > > > as the rest of members in this project, I created a branch, > and made > > two > > > > > commits, and then merged. Could be this discussed more? Yes, > but the > > same > > > > > as with any other commit here. I used to discuss previously, > and I > > used > > > > to > > > > > ask for inputs, acceptance, and more (Something that I don't > > remember in > > > > > all committers) . This time, I'm so sure that the change was > the > > right > > > > one, > > > > > that I committed in the same way you all do when you think > the > > change is > > > > ok. > > > > > > > > > > Now that people is using dictatorial ways, I think I'm > opposed to > > discuss > > > > > my commit. Or if we want to discuss it, I first would want > to discuss > > > > many > > > > > others commits that I'm not conform, but I left in my > philosophy of > > "live > > > > > and let die". I think when people dictates things the way to > proceed > > > > should > > > > > be the same. > > > > > > > > > > I feel attacked, and don't know why. I think I already make > merits in > > > > this > > > > > project so people could trust my work: > > > > > > > > > > * MDL: was a library I started and develop in a huge part > with the > > help > > > > of > > > > > Piotr. And I started and finished until the end, while other > UI sets > > are > > > > > still or incomplete or are only experiments > > > > > * AMF: I worked as well in this part to make sure we had > something > > that > > > > > people on the list has expressed they need. > > > > > * JEWEL: I'm creating alone a complete UI set that looks > polished to > > be > > > > > used in production. > > > > > * Compiler: I finaly are knowing how to do things there and > I finaly > > > > fixed > > > > > many things specially on CSS. > > > > > * Maven: As well I'm more familiar and solved many problems > in > > building, > > > > > some arise thank to this refactor. > > > > > * Website: I develop all the website alone > > > > > * Blog Examples: I'm creating one or two post to engage the > community > > > > > * Social Networks: We have now Twitter, Facebook, Google+ > and > > LinkedIn > > > > > working and moved continuously only by me, and getting a > good > > traction. > > > > > > > > > > I'm working fully on this project from many months (8-12 > hours day), > > and > > > > > what I get is a person that is frustrated since my change > broke his > > > > app... > > > > > a co-mate working in his app ask about some changes > (although I > > don't see > > > > > real align to one or another line, but assume that if asked > he's in > > the > > > > > same boat that the first one), other people join to the > discussion to > > > > then > > > > > apologize to me privately and personally, and finally other > PMC > > members > > > > > only gave me a reason non technical that thing was "as is" > from the > > > > > beginning. > > > > > > > > > > I'm all for communication, and I'll be spending all the time > needed > > in > > > > this > > > > > point to go forward and fix whatever is not working. But the > current > > way > > > > is > > > > > not the way to go. I can go back from now on and ask about > how people > > > > > thinks about things I want to do. I usually do (The latest > was about > > > > > spending facebook donation, how, when,...), but demand the > same for > > > > others, > > > > > and end this kind of dictatorial ways of managing things, we > are all > > in > > > > the > > > > > same boat, there's no leader here, there's no boss here, > there's > > only an > > > > > Open Source project that at least I want to make it shine > and we are > > > > > loosing our time with the noise generated by a single person > and his > > > > broken > > > > > app, instead of joining the effort to make all working ok > and that > > we all > > > > > get all the requirements we all want, we are trying to > impose things > > > > while > > > > > my changes are done to get more freedom. I'm against that > > dictatorial way > > > > > of doing things. > > > > > > > > > > People that doesn't want how I'm designing Jewel, can simply > not use > > it, > > > > > and can make it's own UI Set in parallel, stick with Basic > or > > whatever > > > > they > > > > > want. As well I'm for "Live and Let die", I expect the same > from the > > > > rest, > > > > > and only help when is required to help the project. I'm not > forcing > > > > > anything in this project to make people need Jewel, while > others > > want we > > > > > depend obligatorily from Basic and that's not needed.Having > the > > > > possibility > > > > > to make people choose what they want and not force anyone to > use what > > > > they > > > > > don't want (in this case Basic) is crucial. I don't plan to > mess > > more in > > > > > Basic, since I want to avoid this situations > > > > > > > > > > So, I'll be one last time asking here to reconsidere how > things are > > > > > managed, how things are asked, avoid unilateral actions that > destroy > > > > > other's work. > > > > > > > > > > I propose to left the discussion to cold this weekend. Hope > people > > return > > > > > on Monday with the motivations restored and want to continue > forward, > > > > left > > > > > the discussion and we talked about how to fix whatever build > is > > broken, > > > > > plan to release, do more work on social netiworks, publish > new blog > > > > > content, follow working on Jewel, MXRoyale, and more. > > > > > > > > > > Let me know, what you prefer. > > > > > > > > > > Thanks > > > > > > > > > > > > > > > > > > > > -- > > > > > Carlos Rovira > > > > > > https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fabout.me%2Fcarlosrovira&data=02%7C01%7Caharui%40adobe.com%7C68a39fc1a1ac4c6e7a7308d5b746cee4%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C636616437110115902&sdata=GBUHO%2BzEdYQ5ydYry1CSTUIBt%2BR1c4AZtjpsemoZ9ns%3D&reserved=0 > > > > > > > > > > > > > > > > > -- > > > Carlos Rovira > > > > https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fabout.me%2Fcarlosrovira&data=02%7C01%7Caharui%40adobe.com%7C68a39fc1a1ac4c6e7a7308d5b746cee4%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C636616437110115902&sdata=GBUHO%2BzEdYQ5ydYry1CSTUIBt%2BR1c4AZtjpsemoZ9ns%3D&reserved=0 > > > > > > -- > Carlos Rovira > > https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fabout.me%2Fcarlosrovira&data=02%7C01%7Caharui%40adobe.com%7C68a39fc1a1ac4c6e7a7308d5b746cee4%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C636616437110115902&sdata=GBUHO%2BzEdYQ5ydYry1CSTUIBt%2BR1c4AZtjpsemoZ9ns%3D&reserved=0 > >
