Bug hunters topic

[stassy]

No problem for me, though if a bug go past the 15 days threshold it's up to the players to report it again, as Orion79 do.

[stassy]

After a quick glance, maybe it's better to choose another icon because the white star outline is not enough to separate the confirmed threads from the others, it should be another bright color but all we currently have is yellow or pink, the flame is unfortunately automatically added if the thread goes past a specific number of post...

[Aswan jaguar]

I agree Stassy I thought that,too.I went for the asterisk as in meaning it could mean "chosen" .The flames icon is different from the one that appears after a number of posts if you believe it's more distinguishable we can go for that.

[stassy]

Well, by applying the same to fixed thread to heart icon it's a bit more visible, let's see how far it will go for a bit.

[stassy]

Ok, seems to be some others icons are useful so I am starting to tag unconfirmed or needing more test with a "?" tag, and unfixable/not fixable with a blushing emoticon.

[Aswan jaguar]

Stassy can you place on top of the Announcements list on Bug Reports forum the -READ THIS BEFORE YOU CREATE A NEW TOPIC-topic so it's the first that people see when visit this forum.Then maybe we will have better chances for better reports,savegames and people start to search if a bug already has been reported.

[stassy]

Unfortunately bumping the topic doesn't put it on top of announcement list
Will have to ask Huggy if it's possible.

[stassy]

On holidays until the 2nd of june, only have a crappy tablet with me so cant bugcheck until then. . .

[stassy]

Due to time restraint and in order to organize my to-debug-list, here is what I am currently planning in order of priority :

Currently checking:
- PM2.dck > PDMN1.dck > PDMN2.dck
- Submitted bugs

Next:
- Gameplay balance in Challenge Mode with the new 95' Challenge reward


Doable but not 100% sure:
- AI base value numbering, estimated one week per block (only have 3 days per week in fact to check around 600 cards per block)

Middle to low priority:
- Re-check confirmed-but-not-fixed-submitted-bug-thread prior to PtM patch
- "Official" ML3 Playdeck decks pool

[stassy]

Pfff PM1.dck finally done, on to PM2.dck then PDMN test decks
AI base value should be a bit faster now as I enter an already tagged area and original game cards (Urza and down)

[Aswan jaguar]

Stassy,Gargaroz what do you think about making a topic (still on the active zone not archive)and put in all those reports that are waiting for an answer,being investigated,checked,etc,if a week passes without any news on the topic would it help if they are all put together in a place?
Making this way also more room(and cleaner path) to the bugs that have not been answered,to?old and news?

[Korath]

I'll put my two cents in here and then never raise the point again:

The "unfixed/unsolveable" archive is worse than useless; a fair number of the posts in them have in fact been solved since being merged there, and it's impossible to find all but the simplest unfixed issues (those that only got their initial post) in it. Hunting through the archive for newly-solveable bugs is in many cases more work than actually fixing the bugs.

The fixed archive isn't a whole lot better. If a fix turns out to be nonfunctional (how many versions has Cryptic Command gone through now?), merging into an unreadable dumping ground like these archives interweaved with a dozen other contemporaneous bugs is exactly the same as starting from the beginning every time. It wastes everybody's time to have to re-collect all the information for a nontrivial bug that seems fixed but regresses a couple versions later, or that is claimed as fixed but actually turns out to be unfixed or only partially fixed on release, instead of just reopening the original bug.

It's especially bad since they get archived on Gargaroz's word without anybody ever having independently tested them. It's considered a worst practice in industry for a developer to close a bug that he worked on himself as "fixed"; for a developer to close a bug he worked on as "fixed" when it hasn't been retested by anyone else and it hasn't been deployed in a release is unheard of - and that's essentially what's happening when you archive a bug that Gargaroz posts "Fixed." on (or, for that matter, that I post a patch for and Gargaroz says he's incorporated it). Usually it's the reporter's responsibility to retest, though that obviously won't work when testing is done by end-users instead of a QA team.

It's understandable that we don't use a real bugtracker like jira or bugzilla or mantis - again, end-users instead of QA - but the one incorporated here for the Magic Album project seems both user-friendly enough to not scare people out of making reports and powerful enough that it lets us actually get work done. If anything, having to look through many pages of search results to see if you're making a duplicate report, then copy-paste in the "Describe the Bug/Which card did behave improperly/etc" skeleton, is less usable; and the bugtracker will do a lot of the things automatically that the forum moderators here are doing manually, by merging topics, editing titles, and asking followup questions. Forums are designed to hold discussions; bugtrackers are designed to improve software. Let's use the right tool for the right job.

If that's impossible or unacceptable, then at least do away with the archives. Make a subforum for threads to be moved to when they've been fixed (equivalent to querying a bugtracker for status=resolved) and leave the threads in unclosed limbo forever, ready to be moved back into the open-bugs forum if someone posts that the fix regresses or was incomplete. If a bug hasn't been fixed, leave the topic open and unarchived; that gives us an idea of how much is left to fix and how to prioritize them. These threads will do us absolutely zero harm sitting on page 5 or 10 or 50, waiting to be fixed in a month or a year or five years when we know more about the engine. Merging threads should only be done for duplicate reports.

If even a subforum is out of the question, at least don't merge threads for fixed bugs until they've been released and retested, and don't merge threads for unfixed bugs at all.

Edit: If, er, this rant seems like it's coming out of nowhere, it's because I thought aww1979's post at the bottom of page 1 was the new one. I hadn't had my morning coffee yet.

[Aswan jaguar]

- Thanks,for sharing your thoughts Korath on how to organise things better.I agree with most of your points.Firstly I want to say that we have made a lot of progress in organising the bugs thanks to Shalbei.You can see what I mean if you see the old bugtopic one of our community begun so people could post bugs in there and not all-over the place.
- I agree about the unfixed/unsolvable forum being rather useless with lot of bugs being already fixed and also that you can't find easily all bug reports about a card.
- For fixed archive I have to agree that reopening the original bug file is better and that we spend a lot of time searching through the archive if a bug returns(also sucks that the reports are archived by date and not by reference)
- I agree with you that it is not practical and effective to mark a bug as fixed only by a developer's word without being tested by a second person at least as a second pair of eyes could help find the problems before a release(e.g for us testers in this forum we test what others report to proceed them to devs and never confirm our own bug reports).
But there are several problems with the above:
1- We are all here doing our hobby and having fun doing so and can't really "force" one another to work in a way that is more efficient-productive(=the last is known to cause people to lose their fun and interest )
2- For about the last 2 years or more we had only Gargaroz dealing with card bugs (besides coding most of the cards)and as you imagine he was the coding team with us trying to make things as convenient as we could to help him.
3- us testers and me particularly don't have the knowledge how to even test the code you attach to bugs for fixes(I hope you can help me with that: a guide and tools needed to deploy code)if we follow this way or it would mean that every day or so the developer will have to give us the updated fixed files to patch in and test them.

- About the bug trucker in the CCGH forum Goblin Hero asked before if we wanted to take part in that and we could ask him to insert our project there.I don't mind at all if you Developers agree on that and it serves your needs better and is more convenient and effective to you so you better talk about it with Gargaroz and Sonic and reach a decision if you want to use this bug trucker or any otherIt may be difficult to persuade users to use it to post bugs at least in the beginning (I could be wrong though).

If you (devs)don't agree to use the bug trucker I am fine with the changes you suggest we should make with no archives and move back "fixed" bugs from a subforum that relapse or leave the unfixed or confirmed(with no answer yet) in the main forum.We will still have a problem though as people don't like to search for bugs committed or post an answer to a bug topic in 200 page.

[stassy]

I will ask Huggy if it's possible to make a subforum of a subforum, I don't remember why I didn't ask before (I think when we made the last bug threads status quo the question was not raised, so we let it like that).

[stassy]

Well, seems to be creating a subforum in a subforum is possible, so if we decide to do it let me know so I will request one.