Community
Participate
Working Groups
Reserved bug for EclipseCon
The Eclipse Foundation staff get very thirsty working hard at EclipseCon. Help fix this bug by giving your favourite Foundation employee a frosty beverage.
What type of drinks are we looking for here?
My favorite foundation employee depends whether or not I need to get something through the IP process this week :)
I see Donald with a drink in his hand - anybody else still thirsty?
Alright. I'm willing to prepare a patch. But I couldn't find the drink tickets project in CVS. Did it move to Git already? Than it might be distributed over the earth already.
(In reply to comment #2) > What type of drinks are we looking for here? Yes, request further details. Staff should post their favorite drinks.
Please specify which version EclipseCon you're referring to along with a reproducible test case.
Here is the interface... public interface FrostyBeverage extends Water { public void fill(); public void drink(); } Anyone wants to implement it?
Is this really a problem? From what I've seen, the Eclipse Foundation employees seem to *always* have a drink(s) in their hand. I'm tempted to close this as WORKSFORME.
Happy to lend a +1 vote and even put some of my personal dollars toward solving it.
do we have a i reproducible test case ? and what about unit tests ? :-))
and I am willing to act as QA contact / tester for this bug :-)))
The scope of this bug is a little narrowly focused; People outside the Eclipse Foundation get thirsty, too.
Is this actually a bug or enhancement request? I couldn't find the API specs of foundation personnel to see what's actually needed. Perhaps the water dispensers could be considered as a temporary workaround? Submitters please clarify your requirements, then I can try lobby at the AC for a solution. I've voted +1 on this bug for visibility.
Please note this can't be considered completely fixed until all blocking bugs are fixed.
This bug is especially hard to solve as it's intermittent. It only manifests twice a year: once in California and once in Germany.
> What type of drinks are we looking for here? Denis likes beer -- bottle, glass, warm, cold, it doesn't matter, really. (In reply to comment #15) > Please note this can't be considered completely fixed until all blocking bugs > are fixed. Blocking bugs need to be < 300500. Suggest DUPLICATE. (In reply to comment #14) > I couldn't find the API specs of foundation personnel My API is private and it will stay that way. :-)
The main issue has been addressed tonight. Please reopen as necessary.
I am sad to report that even though after Gunner and Oisin reported successful patches for this bug, I awoke to another case that repeats the problem. Not sure what the say, it is what it is. This needs further examination.
Beware that the suggested sollution may taint your IP. ;)
Yep ... NOT_QUITE_FIXED_YET :)
The current solutions seem to have a number of adverse affects. This bug has caused "some" foundation staff members to miss out on their exercise contributions. I think we need some direction from the PMC on this issue before we release another fix. I'll assume Mike is on the PMC for the "Community Project".
(In reply to comment #22) > The current solutions seem to have a number of adverse affects. That should have been adverse effects.
(In reply to comment #6) > (In reply to comment #2) > > What type of drinks are we looking for here? > > Yes, request further details. Staff should post their favorite drinks. Maker's Mark Manhattan, up!
We are working hard on adding the drink specification in Java EE 7, so all can benefit, not only the Eclipse Foundation staff. Reference Implementation JCP fight between Napa Valley and Bordeaux...President Sarkosy is now involved.
I worry that bugs get more attention at EclipseCon when we all see each other face to face and have the potential to languish once everyone gets back to their busy daily lives. Some kind of automated testing, perhaps nightly unit testing seems in order, so we can know within several hours if there is a regression in this area.
Walter provided a H patch yesterday to Denis. It looks like Denis reviewed the patch and accepted it. So this might have temporarily fixed part of the equation.
With EclipseCon about to end today, I think that we need to make another effort at digging to the root cause of this issue such that it can be fixed. As per comment #16, this has been reported to be intermittent by time of the year, so we need to take our chances at a reproducable test case NOW to fully understand the issue! I'm thus increasing priority to a P1. I feel like we need to join forces in an organized Community effort, not only patching the symptoms but digging to the root causes. I have already alerted the full Eclipse Architecture Council since we are fully aware of the severity of this issue. In fact we're going to meet this morning at 8am in the Magnolia Room (Registration Area) to brainstorm on possible solutions before the viral nature of this affects larger parts of the Community. I invite all interested parties (submitters, contributors, committers) to join us this morning for finding a real solution to avoid this creeping up again in the years to come. One interesting observation that I have made while monitoring availability of drinks at EclipseCon was that throughout the day there's free drinks available whereas in the evenings it gets harder to find a drink. Can the submitters confirm that this issue only manifests in the evenings? If so, what are the acceptance criteria for a real fix?
(In reply to comment #26) > I worry that bugs get more attention at EclipseCon when we all see each other > face to face and have the potential to languish once everyone gets back to > their busy daily lives. Some kind of automated testing, perhaps nightly unit > testing seems in order, so we can know within several hours if there is a > regression in this area. They all need a subscription to either Beer of the Month or a Wine of the month, to keep the symptoms that cause the problem from re-occuring too often. I do agree an Smoke Test would be ideal, preferably some place close to their offices weekly.
Seems as though this is an evening issue. With the conference staff blatantly violating the "all lanyards must match" rule at lunch, no shortage of thirst-quenchers were available. As well, a constant stream of beer has been seen on the patio outside the bar; I suspect this bug can be closed, but we may still have to reopen it later.
Whew! Now that it's 3 weeks past EclipseCon, I think we've sufficiently recovered to declare this bug fixed. Thanks to the many contributors for all the patches applied. Until next year... :)
Created attachment 190982 [details] Reopening for EclipseCon 2011 Reopening bug for EclipseCon 2011
REopening
This issue was fixed in the Helios release. Please upgrade to Indigo and open a new bug if you find a regression. Be sure to add your Build ID and collect logs of your thirst such that we can investigate.
Martin is right... In fact, if this has not been properly ported to Indigo then this bug should be 'cloned' (see the bottom of the bug report for the clone action), and the work should happen there. Having said that, I assume you are trying to reuse the ribbons from last year, so cloning this bug might have technical problems. Maybe this should be raised with the Architecture Council ;-).
(In reply to comment #34) > This issue was fixed in the Helios release. EclipseCon isn't part of the release train. Anyway, this looks like one of those housekeeping bugs that occur again and again. ;)
This isn't even a defect, it's an enhancement request. I've changed it accordingly.
Depending on how thirst is quenched there can be severe performance and security issues that needs to be reviewed and understood before any drinks are administered. I propose we elect a committee of responsible community members to define the criteria for releasing *any* hydrating substances, and to define a juste distribution policy. Given the difficulty of the work, they could probably have finished a policy by EclipseCon 2012. I think an exception for "Water" can be accepted for EclipseCon 2011.
I recall submitting a patch to fix this bug about this same time last year. I suppose it's my own fault for not regressing it properly. I guess I'll have to resubmit after we're back from vacation. I would be willing to fund research into a more permanent solution.
I just wanted to document that I ran at least a dozen unit tests on this last evening, but the thirst issue remains. I will be looking for additional unit tests this evening.
Come on you people. This is the best you can do? I'm not even at EclipseCon and I've had more to drink than this bug suggests. And I hardly drink.
Re-opening due to a user's request. Trying to reproduce the issue...
Setting hardware platform to "other" as this only reproduces on carbon rather than silicon.
Oh and OS should of course be "other".
(In reply to comment #40) > I will be looking for additional unit > tests this evening. It seems that that execution has been aborted earlier last night. You should try again today. ;)
(In reply to comment #45) > (In reply to comment #40) > > I will be looking for additional unit > > tests this evening. > > It seems that that execution has been aborted earlier last night. You should > try again today. ;) this bug seems to have some side effect or impact on the eclipse runner group in the morning. Today the group was significantly smaller. Did you make some heavy testing recently ?
(In reply to comment #46) > this bug seems to have some side effect or impact on the eclipse runner group > in the morning. Today the group was significantly smaller. That's just proof that the load balancer is working properly.
The problem is recurrent and widespread on carbon installations, but cocoa seems to improve matters.
I am happy to report that much thirst was quenched last evening. A fine showing by the Eclipse community !
Comment on attachment 190982 [details] Reopening for EclipseCon 2011 I'm glad I didn't meet any star wars characters last night.
Mike initially reported this bug appears fixed, and I can verify that fact. Thank you everyone who helped with this bug!
Removing the 'security' keyword as this bug is not concerned with actual security vulnerabilities.
boooo....
REOPENING. Given that EclipseCon is not in sunny -forget 2011, will you?- California anymore, and that Lynn "Bunny" Gayowski is back in the business this year, I am wondering whether there will be hot chocolate served during receptions. FWIW, chocolate fountains would definitely fall into the "hot chocolate" category!
I can confirm that I am coming to EclipseCon. However, I am thirsty for non-chocolate beverages.
A critical amendment is needed: Eclipse Foundation staff, PAST and present, are thirsty... I would find chocolate-based liqueurs acceptable.
+1! Drinking chocolate martinis with Lynn Gayowski is on my bucket list.
I can attest that the Conservatory Bar makes a good Manhattan, and they have some nice Port choices in the Market Street bar. Both Port and Bourbon go very well with chocolate. I know this to be absolutely true.
Also +1 for drinking chocolate martinis with Lynn :)
Just checking the bug symptoms in the summary. Is chocolate really thirst-quenching?
I'm afraid this bug needs more testing to be sure we've covered all use cases.
(In reply to comment #61) > I'm afraid this bug needs more testing to be sure we've covered all use cases. Alright. See you at the bar in a few minutes!
Posting this link so that we're all aware [1]. It appears that the community feels this bug is particularly important. Looks like we might need to re-evaluate the importance of this bug within the next few months. [1] http://www.ericcloninger.com/?p=517
Just bumping this one again, I hear they are thirsty again.
This problem seems to be a recurring one yet frustratingly there is still no reproducible test case. Every time I try to debug this, I find the foundation staff drinking yet the claim is that they're still thirsty. This time around we must determine the root cause of the problem rather than merely treating symptoms with copious random beverages. I think it's best if we install a thirst monitor that automatically reports the level of thirst when it exceeds acceptable tolerance. The Code Recommenders folks have some experience with such monitoring, though there may be some privacy concerns that need careful consideration. In any case, we need a guinea pig volunteer prone to thirst for said instrumentation...
Well, I'm not foundation staff but I am a tester. To ensure our tests are covering the correct target groups and risks I'd suggest having more than one persona - though I am willing to be one of them. In terms of test strategy, I think I'd go for a manual rather than automated approach on this one.
I can reproduce the problem, too! Happy to help with any drink^Htesting.
Seeing this issue come up every year ... we clearly have a regression. The timespan of delivery is way to long. Even trying to patch the issue with Democamps does not seem to fix this from coming up again. I would suggest to start thinking about continuos delivery!
I have not seen a clear "condition of satisfaction" for this enhancement request. Therefore the development team ist guessing. Is seems to me that there is an implicit assumption that the solution to 'thirsty' implies beer or other alcoholic beverages. What we need is a testable condition of satisfaction. We could make some minimal assumptions and we could test if plain water would fulfill the implicit requirement of quenching the thirst of the Eclipse Foundation staff at EclipsCon. There are studies that show that drinking beverages with alcohol, sugar or caffeine actually increases the thirst [http://www.livestrong.com/article/309477-the-effects-of-alcohol-caffeine-on-dehydration/]. From other projects I have learned that giving customers what the think they want may actually not actually the best solution for them. In goal directed design, we ask what is the goal behind the requirement. Is quenching thirst really the goal of the user or are there are some other implicit goals. We should do some qualitative and quantitative studies. It may be that those studies reveal that the goals are - socializing - saving money - getting drunk - having fun - getting attention - enhancing the relationship between the community and the foundation - becoming more likable (studies show that asking for a favor makes you more likable (http://www.awaionline.com/2012/04/three-little-known-ways-to-make-people-like-you/)
(In reply to Ed Merks from comment #65) > I think it's best if we install a thirst monitor that automatically reports > the level of thirst when it exceeds acceptable tolerance. The Code > Recommenders folks have some experience with such monitoring, though there > may be some privacy concerns that need careful consideration. In any case, > we need a guinea pig volunteer prone to thirst for said instrumentation... I will volunteer but only because I value the community contributions.
(In reply to Ed Merks from comment #65) > This time around we must determine the root cause of the problem rather than > merely treating symptoms with copious random beverages. > FWIW, I disagree strongly here.
(In reply to Michael Scharf from comment #69) > I have not seen a clear "condition of satisfaction" for this enhancement > request. > > Therefore the development team ist guessing. Is seems to me that there is an > implicit assumption that the solution to 'thirsty' implies beer or other > alcoholic beverages. > > What we need is a testable condition of satisfaction. > > We could make some minimal assumptions and we could test if plain water > would fulfill the implicit requirement of quenching the thirst of the > Eclipse Foundation staff at EclipsCon. There are studies that show that > drinking beverages with alcohol, sugar or caffeine actually increases the > thirst > [http://www.livestrong.com/article/309477-the-effects-of-alcohol-caffeine-on- > dehydration/]. > > From other projects I have learned that giving customers what the think they > want may actually not actually the best solution for them. > > In goal directed design, we ask what is the goal behind the requirement. Is > quenching thirst really the goal of the user or are there are some other > implicit goals. We should do some qualitative and quantitative studies. It > may be that those studies reveal that the goals are > - socializing > - saving money > - getting drunk > - having fun > - getting attention > - enhancing the relationship between the community and the foundation > - becoming more likable (studies show that asking for a favor makes you more > likable > (http://www.awaionline.com/2012/04/three-little-known-ways-to-make-people- > like-you/) To your list of goals, I would add - becoming more attractive as companions become more intoxicated (http://io9.com/5830663/just-why-do-people-get-more-attractive-when-youre-drunk)
Might be easier if we open specific bugs and tackled them one at a time. I propose we start with Foundation staff named Ian. Of course, if the thirstiness of other Ian's were addressed, that would be fine too.
You mean to deliver a MVP to Ian(s) first, get feedback, improve and then expand the wider community? Sounds like a great idea.
Depending on the turn-around times!
Closing this bug :)
Created attachment 259842 [details] Directions to World of Beer from EclipseCon 2016 If there is any residual thirstyness I would like to point out EclipseCon 2016 is 1000 feet from a "World of Beer" franchise. I highly recommend World of Beer locations.
Based Donald's input this needs to be reopened so we can visit World of Beers.
Fixed bug description. Who cares if Ian is thirsty....he's a lightweight compared to others in the EMO :)
(In reply to Mike Milinkovich from comment #79) > Fixed bug description. Who cares if Ian is thirsty....he's a lightweight > compared to others in the EMO :) Let's be careful how we throw the lightweight title around. Some of us have worked hard to earn titles...
(In reply to Mike Milinkovich from comment #79) > Fixed bug description. Who cares if Ian is thirsty....he's a lightweight > compared to others in the EMO :) +1 :)
Thanks you everyone! EclipseCon Europe was amazing!
