Would be nice to get some feedback and your opinion about the current state after a month or something like that.

Right now it feels like, that there are a lot of bugs but only a few game breaker. Even RVR seems working so far, since there are no major bug reports. But yeah, at least the major encounters (Dragon/Legion) and a lot of RA's are still untested and it will take way more time to do that without some beta boosts.

I don't think they will offer any level boosts as long as they are trying to sort out the proper xp distribution in groups, especially when it comes down to big level differences.

We had a perfectly fine working herald, what's wrong with the old one?
Feels like you are coding everything from scratch, no comment

Celteen wrote:We had a perfectly fine working herald, what's wrong with the old one?
Feels like you are coding everything from scratch, no comment

so? insta 50 beta phase in early 2018 is perfectly fine!

btw. i wish the questions in the poll would have had the option "i don't care about this feature at all"

My issue was solved yesterday so i'm good lol mostly about the aggro haha

TheKrokodil wrote:@Priorities:

Right now, we actually don't have a priority system for issues. Mostly because priorities always change depending on what issues are discovered, and also on how much time we have. If we have 1h of time on an evening after work, we're not going to spend it doing hours of research, but we'll use it to fix an obvious, easy-to-fix instead. The cost of context-switching is very real.

The only priority we have so far is the number of votes for an issue. When a GM/Developer/Staff member finishes their current issue and becomes free, they look at the top confirmed issues, and pick one they can solve. Given that higher voted issues appear higher on the list, they're more likely to be picked.

Adding priorities to the tracker would be possible, but I feel like it will end up in a maintenance hassle that won't really give us much of a benefit. Maybe a per-developer bug/task queue would be more appropriate, but I'm not even sure how well that would work.

Sounds to me like you need a QA+project manager that exactly knows the capabilities of the backend developers in order to prioritize their workload while keeping the big picture in view and taking that burden from development in order for them to focus on fixing, while QA groups up similar issues so developers can focus on them and also QA is able to test more efficiently the upcoming builds that contain their fixes.
That is usually how the professional development in the companies I worked for is done.

What I ment with priority are things that affect all players are surely higher in their severity, while crashes/exploits should always have the highest severity. Just because an issue has 20-30 upvotes does not mean it has a high severity. That is why voting system should be more targetted for "nice to have fixed/implemented" issues being prioritized, but not overuling severity.

ascanet wrote:

Raifs wrote:Scroll down a bit and your question will be answered in the issue report itself.

It wasn't a bug, it's working as intented.

Not sure what you mean by working as intended. It is NOT working as they described it works.

0241: Poison status ailment not highlighting name green in party.

"The group window and mini-group window now show the additional following states by changing the affected player's name to an associated color. This functionality is present in Classic, Isles, and Trials of Atlantis clients. These states will be prioritized and displayed in the following order with the associated color in parenthesis: linkdead (blue), dead (red), mezzed (cyan), diseased (orange), poisoned (green), out of zone (grey), nominal (white) From patch 1.67."

The state poisoned (green) is not being displayed

Asca

The server is aiming for 1.65, this is a 1.67 change. Also look at https://uthgard.org/tracker/issue/82/

riad wrote:

Celteen wrote:We had a perfectly fine working herald, what's wrong with the old one?
Feels like you are coding everything from scratch, no comment

so? insta 50 beta phase in early 2018 is perfectly fine!

btw. i wish the questions in the poll would have had the option "i don't care about this feature at all"

The old herald used MySQL to directly access the server's database, which is bad for a number of reasons (security, too slow, can cause lags). And yes -- this did happen. Before it had a robots.txt, whenever the google search bot would crawl the entire player database, the entire server would start lagging. I also personally just don't like the old herald anymore -- I think it has shown age.

If you haven't noticed, I'm leaving the herald development to the community as an experiment, and I don't consider it blocking for the relaunch, so I think the argument of it delaying relaunch is not that valid.

---

@jimsen: Good, enthusiastic people are always welcome to apply!
We are ofcourse implicitly prioritizing issues. Yes, if the server crashes, we drop everything else and fix it ASAP, and the same goes for important db breaking issues. We don't always communicate that, but we think it's also rather obvious that we start looking into what causes the crashes as soon as we're aware of them.

QA is really nice, I agree, but completely infeasible in a project where you don't have enough people. We have a test-server that we use occasionally for making sure things seem right, but we cannot test every possible combination of behavior, and DOL's code is so tightly coupled that unit tests are a nightmare.

TheKrokodil wrote:

riad wrote:

Celteen wrote:We had a perfectly fine working herald, what's wrong with the old one?
Feels like you are coding everything from scratch, no comment

so? insta 50 beta phase in early 2018 is perfectly fine!

btw. i wish the questions in the poll would have had the option "i don't care about this feature at all"

The old herald used MySQL to directly access the server's database, which is bad for a number of reasons (security, too slow, can cause lags). And yes -- this did happen. Before it had a robots.txt, whenever the google search bot would crawl the entire player database, the entire server would start lagging. I also personally just don't like the old herald anymore -- I think it has shown age.

If you haven't noticed, I'm leaving the herald development to the community as an experiment, and I don't consider it blocking for the relaunch, so I think the argument of it delaying relaunch is not that valid.

---

@jimsen: Good, enthusiastic people are always welcome to apply!
We are ofcourse implicitly prioritizing issues. Yes, if the server crashes, we drop everything else and fix it ASAP, and the same goes for important db breaking issues. We don't always communicate that, but we think it's also rather obvious that we start looking into what causes the crashes as soon as we're aware of them.

QA is really nice, I agree, but completely infeasible in a project where you don't have enough people. We have a test-server that we use occasionally for making sure things seem right, but we cannot test every possible combination of behavior, and DOL's code is so tightly coupled that unit tests are a nightmare.

how do you view player's information w/out directly accessing DB (live)?

vadox wrote:how do you view player's information w/out directly accessing DB (live)?

You would periodically dump the live DB to a secondary DB, ideally one on a different server. The dump should be limited to just the information the Herald would use. This is likely what's already being done seeing as the level of my character displayed on the Uthgard page is always lagging behind in levels of my character in game.

Then you would have the Herald query against that DB, or better yet, create a cache of pages up until the next dump.

GreyWyvern wrote:

vadox wrote:how do you view player's information w/out directly accessing DB (live)?

You would periodically dump the live DB to a secondary DB, ideally one on a different server. The dump should be limited to just the information the Herald would use. This is likely what's already being done seeing as the level of my character displayed on the Uthgard page is always lagging behind in levels of my character in game.

Then you would have the Herald query against that DB, or better yet, create a cache of pages up until the next dump.

it's not truly live unless dump happens every minute which is impossible.

TheKrokodil wrote:If you haven't noticed, I'm leaving the herald development to the community as an experiment, and I don't consider it blocking for the relaunch, so I think the argument of it delaying relaunch is not that valid.

Will you crowdsource the API development as well?

edit, also do you have a design vision for this?

I'm working on a design for the new herald player page.

I will run it through TheKrokodil tonight, and when all the API endpoints have been discussed along with design approval, It can be shared for those who would like to help with the Angular 2.0 implementation.

TheKrokodil Also has some architecture guidelines that people should know about before beginning to submit pull requests.

/Arzu

Could we get a sticky of good vetted resources to troubleshoot bug reports?

I'd love to help but I'm new to DAOC in general. I don't have any recollection I can use to question things.

I can research current reports but I'm not always sure what's relevant for our patch level. Some links/pdfs to vetted sources would be amazing. Then I can compare what I find in game to these sources and at least note discrepancies.

Cold Blooded wrote:Could we get a sticky of good vetted resources to troubleshoot bug reports?

I'd love to help but I'm new to DAOC in general. I don't have any recollection I can use to question things.

I can research current reports but I'm not always sure what's relevant for our patch level. Some links/pdfs to vetted sources would be amazing. Then I can compare what I find in game to these sources and at least note discrepancies.

agree, a sticky would be great

meanwhile, Capteo put together a library of sources/info > viewtopic.php?f=9&t=37085, certainly very helpful
also Bubbler is preparing some kind of document consolidating DAoC info

cheers