Help Us Help You Help Us All - Part 2: Five Years Later

[Sqrlmonger]

Five years ago I posted this thread (https://forums.ddo.com/forums/showthread.php/490358-Help-Us-Help-You-Help-Us-All-It-s-Time-for-Logging) requesting the ability to save text that passes through our client-side chat boxes into a log file.

Note: The request was that this logging would be OFF BY DEFAULT.

I feel its been more than long enough to renew the call for this feature.

This is one of the most fundamental MMO features still missing from the game and it could be added so easily.


It has no impact on people who don't want to use it since it defaults off.
It's an extremely easy code change that could be given to even a semi-competent intern (perhaps redundant).
It is likely to provide benefits to everyone who plays with or works on the game.
It allows devs to enlist players in reproducing bugs.
It allows players to collect and utilizes data to assist in tuning characters (e.g. In my last 50 deaths, what is bigger cause? Failed Reflex? Or am I just getting clapped by a Champion with Magic Missile and 5 stacks of DoT Love?)


So how about it Devs? Can we PLEASE get this one knocked out? Preferably before I'm back here in 2027 asking again?

Help Us.
Help You.
Help Us All.

[droid327]

No you can't has a dps meter...

[Sqrlmonger]

No you can't has a dps meter...

I'm sure a lot of people would want it for that purpose, and I might use it if someone made one.

But my personal intentions are more along the lines of tracking daily dice roll results, xp/min, and a few other spicy ideas I've had for a long time but are infuriatingly blocked by this pretty minor hurdle.

[SoulDustar]

Ctrl P works remarkably well it takes a screen shot all you need to do is set your text level to 20 and a color of your choosing and you will have your daily rolls, loot, xp/min etc... all saved and ready to go

[J1NG]

But my personal intentions are more along the lines of tracking daily dice roll results, xp/min, and a few other spicy ideas I've had for a long time but are infuriatingly blocked by this pretty minor hurdle.

Hmmm, forgive me, but aren't all those listed items, not even related to the Combat Log? Which is the only thing that flies by so fast (during combat) that having a an extraction log might be helpful?

Daily Dice can be easily tracked as it happens in the main general chat window in most cases. And most players aren't in a rush during Daily Dice and thus can be tracked easily.
XP/Min is from a player checking their compeltion times and xp earned. Again, usually players interested have plenty of time to note it down to create a data point and don't really need an extraction log of this.

Looking at your previous thread linked, it appears you are under the belief that having the Combat Log would permit a player to be able to test and gain insights into things that may or may not be working correctly. Yet this does not take into account that the information needs to be accurate in the first place that the Combat Log calls from as well as certain testing doesn't reveal everything the players believe it does.

Just recently, we had the Concealment funniness and players believed that Concealment was not working, without realising that it was actually showing that Incorporeal was bypassed and not Concealment. Because of the way the Avoidance defence works; Dodge, Incorporeal and then Concealment. But the Combat Log only showed the Concealment bypass for efficiency (not spamming lots of info that's redundant) but ends up showing inaccurate information (Concealment was bypassed, when it was only Incoporeal that was bypassed). In another incident, I tried to help test for the issue with Lailat in Against the Demon Queen, but the theory I proposed was supported by testing that did not eliminate other variables, giving an inaccurate assessment, despite it fulfilling the criteria that proved my theory of the cause to be accurate.

So without the knowledge of how DDO works already, or painstaking testing that eliminates variables so that the Combat Log only shows things that is accurate to your query, I don't believe having the Combat Log be extractable and placed into a text file for later use by the player to be anything more than giving players nothing of real value at the cost of implementing it.

If you are hoping SSG has the Combat Log also provide other information, then that is unlikely to be granted either. Since that info has to be specifically set to be sent to the chat window itself. So we'd only be bugging SSG to add more to it every time that we want to check on, not a one and done situation if it did get implemented. So again, likely won't be implemented as it offers little value for the time put in to it.

J1NG

[Hafeal]

I will say, as someone who enjoys the story, I have had to do one entry at a time on a copy and paste to share story lines with my daughter. It is a PIA. Not game breaking, but the ability to more easily copy that text and share it would be a welcome change.

I have no objection to people using it to calculate their dps. I think more tools, more interest is better for the game and community. :)

[Sqrlmonger]

Ctrl P works remarkably well it takes a screen shot all you need to do is set your text level to 20 and a color of your choosing and you will have your daily rolls, loot, xp/min etc... all saved and ready to go

Or I could just roll it and its in my log forever and I can parse it out when I need it without having to do anything.



Hmmm, forgive me, but aren't all those listed items, not even related to the Combat Log? Which is the only thing that flies by so fast (during combat) that having a an extraction log might be helpful?

Daily Dice can be easily tracked as it happens in the main general chat window in most cases. And most players aren't in a rush during Daily Dice and thus can be tracked easily.
XP/Min is from a player checking their compeltion times and xp earned. Again, usually players interested have plenty of time to note it down to create a data point and don't really need an extraction log of this.

Looking at your previous thread linked, it appears you are under the belief that having the Combat Log would permit a player to be able to test and gain insights into things that may or may not be working correctly. Yet this does not take into account that the information needs to be accurate in the first place that the Combat Log calls from as well as certain testing doesn't reveal everything the players believe it does.

Just recently, we had the Concealment funniness and players believed that Concealment was not working, without realising that it was actually showing that Incorporeal was bypassed and not Concealment. Because of the way the Avoidance defence works; Dodge, Incorporeal and then Concealment. But the Combat Log only showed the Concealment bypass for efficiency (not spamming lots of info that's redundant) but ends up showing inaccurate information (Concealment was bypassed, when it was only Incoporeal that was bypassed). In another incident, I tried to help test for the issue with Lailat in Against the Demon Queen, but the theory I proposed was supported by testing that did not eliminate other variables, giving an inaccurate assessment, despite it fulfilling the criteria that proved my theory of the cause to be accurate.

So without the knowledge of how DDO works already, or painstaking testing that eliminates variables so that the Combat Log only shows things that is accurate to your query, I don't believe having the Combat Log be extractable and placed into a text file for later use by the player to be anything more than giving players nothing of real value at the cost of implementing it.

If you are hoping SSG has the Combat Log also provide other information, then that is unlikely to be granted either. Since that info has to be specifically set to be sent to the chat window itself. So we'd only be bugging SSG to add more to it every time that we want to check on, not a one and done situation if it did get implemented. So again, likely won't be implemented as it offers little value for the time put in to it.

J1NG

I'm not looking specifically for the combat log, I think everything that could be filtered to a chat box is beneficial.

As for daily dice, see the above reply.

As for XP/Min, I'm not talking about time in quest alone. Just for a start I want to get to XP/played minute. This has a lot of intricacies people don't think about. For instance, do you actually track all the XP you get from MM deeds? Would it surprise you to learn about 5% of the XP you get from a TR life is Monster Manual deeds? What about optional objectives? Stop to add it all up? Now the rate changes because I've stopped...See the problem?

As for accurate information, it absolutely 100% does take into account that the information needs to be accurate. But for starters, unless you're claiming that literally every piece of information in the log is wrong, this line of reasoning is specious. Inaccurate info in the combat log is itself a bug, and one that can be fixed more easily with logging like this. Aside from that, barring those fixes, there is TONS of accurate info in there that already can be used. The fact that some things are not accurate doesn't change that.

Your example on the concealment is a great example of how this logging would be beneficial. Looking at your overall log with potentially months of data you would have seen the totals for concealment, Incorp, and Dodge individually were wrong but summed together were coherent. You would have been more likely to get to the right answer that there was a display issue as a result with that added information. Not less likely. Logging would have actually helped, not hurt you, if you chose to use it.

The rest of your post is truly silly, you try to argue that 2 examples of how not having logging means having logging would not be useful. I don't even feel the need to reply to that.

[Enoach]

I agree that being able to dump the chat logs could be used in various ways that can be beneficial.

Besides the ill fated MyDDO, DDO has generally steered clear of features that would push data from within the game out. I believe they even have statements in the ULA that outline how they feel about anything that could potentially reverse engineer their product.

There may be reasons they decide this, it is also possible they will change their mind on this as well. I would suggest, don't just ask for a feature, outline the Use Case. Be specific - much like your nearly 5 year ago post where you outlined what you were after. But then actually outline how that would be beneficial to SSG - They are the ones that would need to dedicate resources and implement the feature. This request is asking them to spend those resources but you haven't outlined an RoI.

Possible reason they have not added this feature
1. Avoid Third Parties from profiting off their intellectual property
2. Lessons learned during the life cycle of MyDDO
3. Concerns of having a logging system that may flood users systems or unknowingly export sensitive data

[J1NG]

As for XP/Min, I'm not talking about time in quest alone. Just for a start I want to get to XP/played minute.

Ah. That is something different then.


This has a lot of intricacies people don't think about. For instance, do you actually track all the XP you get from MM deeds? Would it surprise you to learn about 5% of the XP you get from a TR life is Monster Manual deeds? What about optional objectives? Stop to add it all up? Now the rate changes because I've stopped...See the problem?

Yes I can agree this can be helpful. However, would not a quick copy of your log in and log out XP + a quick glance at the clock each time not suffice in this instance? Unless if you are aiming to use it for maximised real life minute for xp in game through planning ahead? Which wouldn't really be ideal, since it wouldn't be able to take into account RNG actions on stuff like Rares and Rare encounters offering XP or sudden world crashes that stop your originally expected 100k xp return for 4 minutes of game played. So usually it's a reactive (examine at the end after it's obtained) type of analysis. But I can see the appeal for those who like to spend time planning everything out for what they do.


As for accurate information, it absolutely 100% does take into account that the information needs to be accurate. But for starters, unless you're claiming that literally every piece of information in the log is wrong, this line of reasoning is specious.

I am claiming the one with the problem is often the one sitting at the computer, not the computer itself. Or, the player analysing data or information is the one usually with the problem (because of x, y and z; one such example as I put myself in, was a theory that appears to be proven true to be the cause of the problem raised but turns out isn't really in the end despite the test output data proving I was right), not the data from the Combat Log itself necessarily being inaccurate or wrong.


Inaccurate info in the combat log is itself a bug, and one that can be fixed more easily with logging like this. Aside from that, barring those fixes, there is TONS of accurate info in there that already can be used. The fact that some things are not accurate doesn't change that.

This is true. However, the part above covers this. It's not the Combat Log that's the problem in many cases, but the players seeing something which the log is not actually showing but have been interpreted to be such by the player. There are many instances of players who did minor testing or had observations and used the Combat Log to aid their case, only for it to be checked properly and found to be inaccurate in their assessment and often these are Combat Logs that are 100% correct; it was the player who interpreted them incorrectly.

:: edit :: Whether through insufficient knowledge of the game, or just misunderstanding the what the Combat Log is showing entirely through lack of controlled use of the Combat Log with variables such as in combat.


Your example on the concealment is a great example of how this logging would be beneficial. Looking at your overall log with potentially months of data you would have seen the totals for concealment, Incorp, and Dodge individually were wrong but summed together were coherent. You would have been more likely to get to the right answer that there was a display issue as a result with that added information. Not less likely. Logging would have actually helped, not hurt you, if you chose to use it.

It depends. Logging alone would not have solved the Concealment issue that I was talking about.

The Combat Log was actually 100% accurate, so a player just looking at the log will assume there's a bug there, or something is wrong with Concealment. When there was nothing wrong at all. Concealment worked as it was intended (especially if Concealment checks were done in isolation as well, which Combat Logs showing Concealment checks in isolation would not reveal anything strange only that there was something strange *some times*).

The mystery was solved through use of another client intead and comparing both Combat Logs at the same time to uncover what was going on. Whilst this could have been done with two logs that were extracted (one from each client), this doesn't really help tracking this mystery/bug down any quicker, because it's not a "many times to show the current average doesn't tally with the expected average" type of issue that can be tracked down in that fashion. But an understanding of how the combat sequence actually goes and then working it back into analysis of the two Combat Logs and specifically working test inputs into the testing as it happened and then comparing it to expected outputs. This didn't need much from the Combat Log itself, but accurate testing methods and application.

However, I will agree that had the Combat Log showed all stages of the Combat Sequence, and showed that actually, it was Incorporeal that was bypassed, and not Concealment, that would have made it much clearer and wouldn't have caused this confusion to some players (but at the expense of a Combat Log that is bloated on screen, besides that to be found in an extracted file, which hinders testing that doesn't need numerous data points from a Combat Log lasting hours on end - so there are advantages and disadvantages to having and not having extra data/information).


The rest of your post is truly silly, you try to argue that 2 examples of how not having logging means having logging would not be useful. I don't even feel the need to reply to that.

That is not what I said. But you interpret what you want, how you want. :)

J1NG

[SoulDustar]

Or I could just roll it and its in my log forever and I can parse it out when I need it without having to do anything.

Or you could download an install an OCR program and now SSG does not need to waste dev time for something you can already do with with a screenshot or a video as well.

What is the accuracy of OCR to extract the text?

98 to 99 percent accuracy
Obviously, the accuracy of the conversion is important, and most OCR software provides 98 to 99 percent accuracy, measured at the page level. This means that in a page of 1,000 characters, 980 to 990 characters will be accurate. In most cases, this level of accuracy is acceptable.

[Sqrlmonger]

Or you could download an install an OCR program and now SSG does not need to waste dev time for something you can already do with with a screenshot or a video as well.

What is the accuracy of OCR to extract the text?

98 to 99 percent accuracy
Obviously, the accuracy of the conversion is important, and most OCR software provides 98 to 99 percent accuracy, measured at the page level. This means that in a page of 1,000 characters, 980 to 990 characters will be accurate. In most cases, this level of accuracy is acceptable.

Waste dev time? This will save dev time. Full Stop. If you think otherwise you have never tried to debug with and without logging. The only sense in which this will "waste" dev time is that it will cause more bugs to be found and fixed. If that is your definition of "wasting" dev time then I can't help you out of the logic pretzel you are trapped in.

We are talking about something like 10 lines of practically pre-written code. The whole task is instantiating the file stream, writing to the file via the stream, closing the file, and error trapping. This isn't some huge dev time investment, this is quintessential intern stuff. If the intern cannot do this they really truly have no business being a developer anywhere until they get more training.

We aren't talking about some head-scratching mind-bending development puzzle. Also, your entire argument of "well if you can do it yourself, its a waste of their time" is an argument that works on literally any change. I could modify the assembly to add logging as well (though it would likely get me flagged as a cheater, with time even that could be overcome). And the same is true of a LOT of changes on this forum. Your entire argument is essentially "No changes. Ever."

It's not new to me how much people on this forum gate-keep plenty of perfectly reasonable suggestions. But this? This is such an absurdly simple AND profoundly beneficial change. This is some sort of crab buck mentality to a scary degree.

We are talking about a log file. A LOG FILE people!

At this point its clear I'm not changing your mind. So I'm not going to continue arguing, feel free to take the last word.


PS - This whole things reminds me of when I was first starting out as a developer and I was implementing a CR for a banking client on one of their end of day reports. On the report header, the name of the bank was blatantly misspelled (e.g. "Regions Bank" was spelled "Reigons Bank"). So, in addition to adding the new functionality, I fixed the typo. I submitted my changes noting the typo fix, and they flew through testing and QA out to the bank the next day. The following day I'm on a call with the bank and my boss as they are complaining about how they didn't ask for the typo to be fixed. They actually made us do an entirely new version of the report putting the typo back in, we asked if they needed the typo because it helped them distinguish something (they had a lot of reports) or if they were scraping the report? But no, they just wanted the change put back because it wasn't in the CR, and they didn't want to get in trouble with their Boss for not calling out an unrequested change.

This entire thread reminds of that self-defeating mentality where obvious improvements aren't encouraged, they are discouraged. Sorry I don't live in that world, I live in a place where people make the improvements they can from where they are, and if you think someone's idea has a problem, you don't shoot it down entirely, you suggest ways to hone it and make it better.

But I digress.