issue tracker

[tex]

So the thoughts from my Presentation thread have been that we need an issue tracker or some such. I was going to throw down my knowledge, meager as it is, before I set about my own wtactics work.

• hiveminder is in the browser and free. I'm pretty sure it collaborates but I don't know the details. I know somebody who works there.
• Google Code is also in the browser and free. I'm using it for a couple things and the built-in issue tracker works alright.
• One of 37 Signals' browser-based services might be helpful. It was helpful for collaborating over a project I did once anyway.

Back when I was gainfully employed I admin'd a small bugzilla installation and it was simple enough for folks to get work done, but that'd mean installing one's own bugzilla, which is a pain.

[aspidites]

Hive Minder looked kinda sexy (of couse, I only looked at the presentation, and those are generally pretty shiny/sexy anyways)

I'd rather not use google code. From my personal experience, it can be difficult to find projects (or more accurately, find forks/personal branches). I think the current industry standard is github. Sourceforge gets a lot of exposure from the open source world, but some of its interfaces (again, in my own experience) were clunky, and github encourages social collaboration. Arguments might be made as to the primative nature of github's ticketing system, but I find the combination of milestones and tags more than sufficient. For larger projects, or those simply wanting more control, they have plenty of service hooks (including one for 37signals' campfire).

I've not used any 37signals products, but have heard good things about them.

Precedent dictates that proof of concepts are more effective and demonstrative than theoritical discussion around here, so I'm going to create a github project. As part of this evaluation, I'll be looking at usability and storage of binary assets as well. Will report back in probably less than 24 hours. If I don't, call an ambulance.

PS- Snowdrop, I can hear you already: "Don't bother aspidites, this is only taking away from ViCE development time". My answer to this is two-fold:
1. Often, I get tired of coding, so distractions that are still project related are more productive than sleeping, fapping, or any other kind of unsolicited activity
2. ViCE is still being worked on. Since I'm not working and not in school (for the next couple of weeks), I have plenty of extra time to allocate to assorted tasks.

[tex]

Another two I didn't think of earlier:

• The wiki can be used as an issue tracker,
• and so can the forums,

it's just a matter of using it as such and sticking to it. No complex rules need to be applied so much as actually making people stick to using it.

Hmm . . . the more I think about it, the more I think starting a new project just for the issue tracker would be lame, so I withdraw the Google Code suggestion. A general-purpose WTactics issue-tracker needs to be general purpose, which means there's no reason to use one specific to software development.

[aspidites]

I just checked out hiveminder, and it seems to be nothing more than a beefed up todo list/issue tracker. Creating an account just for that single purpose seems pointless, so I'd quickly downvote that suggestion.

I tend to be against forums as issue trackers. Sure, you could sticky a post, but given posts only have one author, it becomes tedious to update them. Not to mention queries of tasks would become a real bother. Same arguments for the wiki, actually [1].

I think I'll be able to finish my evaluation of github (and DVCSs in general) in a couple of hours, rather than in 24 hours, so stay tuned for that.

[1] Disclaimer: I generally hate wikis. I like the ideal of what they aim to be, but in practice have found them unmanagable, unorganised, and overall useless. The only time that I've seen wikis executed effectively was as documentation for a couple of distros (gentoo and arch linux in particular).

[tex]

I just checked out hiveminder, and it seems to be nothing more than a beefed up todo list/issue tracker. Creating an account just for that single purpose seems pointless, so I'd quickly downvote that suggestion.

Are you saying that you think issue tracking isn't important enough to merit its own solution, or are you saying that we need to solve more problems than just that, so we might as well do it all in one go?

I tend to be against forums as issue trackers. Sure, you could sticky a post, but given posts only have one author, it becomes tedious to update them. Not to mention queries of tasks would become a real bother. Same arguments for the wiki, actually.

On further reflection I think I see what you mean for using the forums. It's just too hard to get the flexibility you'd need. And I wouldn't want you to have to use a tool that you say you hate, so the wiki is out too it would seem.

[aspidites]

Are you saying that you think issue tracking isn't important enough to merit its own solution, or are you saying that we need to solve more problems than just that, so we might as well do it all in one go?

I'm saying that there are better tools for the job. Trac, for example, which would offer better integration.

And I wouldn't want you to have to use a tool that you say you hate, so the wiki is out too it would seem.

I think I'm the only one that has such a disdain for wikis, so that needed be the conclusion here, lol. That said, I honestly don't think it's the right tool for the job.

[Q_x]

From infrastructure point of view, WTactics is a bunch of projects, so every leader would be most satisfied to care for bugs on his or her own.

There is a problem though, as there are some media binded to WTactics (artwork, world and history, GDD) - but I guess those won't need any bugtracker at all. ORC, cards - those may need a tool like this, but in most cases it's something like a misspelled word - correcting it takes less time than filling a bug report. Simple shoutbox or an e-mail would suffice, not even forums are needed.

[aspidites]

Spent a couple of hours playing with github. While github (or git in general) is a good idea for keeping certain things under version control, github's wiki feature is far too limiting to be of use. As for issue tracking (back on topic, finally), in the strictest since of reporting issues, I think it defeats the purpose of having a wiki. That is, wiki's are editable by the general public, so if there are "bugs", you should just dive in. From the sake of keeping track of what is working on, I still think a separate issue tracker is the best option here. If nothing else, it allows us to gauge productivity and keep users informed about what we are working on.

Before it gets mentioned (again?), blog posts are good for status updates, but aren't as permanent as milestones. Also, if show-stoppers are found, its much more reasonable to file a ticket declaring so, rather than add a comment to a blog post.

[Knitter]

Are we talking about an issue tracker for the core WTactics project (rule creation)? If so I don't see why we would need one, in fact it could be yet another cumbersome tool to maintain.

We already have a forum, a wiki and each project uses it's own tools. Will we have the will to maintain one more issue tracker that users seldom use, and at the moment we have no users at all to report issues. Any issue that arises in creating rules can just be dealt with in the forum or IRC and I see those issues appearing when playtesting a card, therefor it should be simply enough to just talk in the forum or while playing.

Do the devs that playtest cards really miss an issue tracker?

Sandscape uses trac for issue tracking and it's only used by me to keep an easier track of what I'm doing, if I were to use the forum I would be speaking to myself so the tickets keep me from sounding to lunatic . I think vise uses it's own tools and Cardscape (still being developed?) has it own issue tracker also.

[snowdrop]

http://wtactics.org/wiki/index.php?title=Roadmap shows an old example of how wiki can be used. It is somewhat clumsier than traditional way of doing it but is much more flexible as it allows us to create it however we want. It also uses one of the many sites we already have which is under-used, so no need to register or maintain another service.

Syntax is:

Code: Select all

{{name of tracker template|name of person|title of issue|brief description.}}

Available tracker templates (names):

Todo
AssignedTo
InProgress
Discussed
InProgress
Postponed
Declined

...they can be seen and easily modified at http://wtactics.org/wiki/index.php?titl ... dTemplates

Those interested in using a tracker may use that for anything that is not code related (meaning, sandscape, cardscape and vice have their own trackers or whatever their devs feel like). I have no problems doing it myself if it helps, but won't put down time in searching for alternatives to it.

Most things on the roadmap were probably written by Q_x long time ago and it needs some re-work. Also, when adding tasks, add on a small level.. for example a task called "playtest" is too wide. It should be a title instead and made up of several smaller tasks (initial dev playtest, public playtesting, beta.. etc)

Edit: Upload a 40x40px avatar in jpg format to the wiki with your exact wiki/forum-nickname. That way your avatar will appear "automagically" when using some of the tracker templates.