Two issues I've noticed in Feature Suggestions

[Ghastly]

This has nothing to do with Graf, or anything like that, before anyone starts. This is also not to lobby for these particular feature suggestions to get in, I'm just picking out examples. This is aimed at the community, not the devs. I've just been seeing these two particular issues in the Feature Suggestions forum a surprising amount, and this thread illustrates both of them quite nicely: http://forum.zdoom.org/viewtopic.php?f= ... 29#p632474 It's a good suggestion; good thoughts on how it could be used, pretty complete idea but could be built upon further later on. That's not what I have an issue with. It's subsequent posts that seem to fit these formats.

1): You can already do this with *insert thing that has nothing to do with this here*
I don't know if it's just people not understanding what these functions do, or if they aren't reading the whole post or what. In the above example, Neural's core suggestion is a flag for A_AlertMonsters to have alerted monsters target the calling actor. This is because in A_AlertMonsters called by a projectile will have any alerted monsters target the player who owns the projectile. Additionally, A_AlertMonsters uses Doom's "sound" travel code for alerting monsters, which traverses sectors and isn't tied to radius around the calling actor. Thing_Hate won't work at all because there's no way of sector traversal like A_AlertMonsters does; it's a "tagged monster targets tagged player in a specific sort of way." A_RadiusGive won't work at all because that's (get this!) limited to radius! Neither of these suggestions will work in the place of the original post. Please, before giving a post like that, check the wiki and make sure you understand the functions and concepts involved before suggesting them.

A subset of this is "You can already do this with *insert workaround that may have problems in place of another idea that is specifically designed for the issue being worked on*." These kinds of "hacks" are pretty unwelcome. People like clean implementations of things. Graf himself prefers these cleaner implementations in source over gross hacks with stuff we already have in Decorate and ACS (sources: here, here, though this has nothing to do with source, and here). A_Explode, damage type specific pain states and A_GiveInventory would be a hack compared to A_RadiusGive. It has several problems; it requires you to damage the affected actor (which is the major problem), it's not reliable, and it's not extendable. In contrast, A_RadiusGive lets you give to monsters, to players, to corpses and it doesn't require the actor to take damage. Could you imagine someone saying "You can already do this with A_Explode and pain states!" in a feature suggestion thread for A_RadiusGive?


2): Another idea is *insert a much more complicated idea that would never get implemented in a million years*
Admittedly, the example of this in the aforelinked thread is mild, and most of the times I've seen this the ideas given would be pretty welcome. However, they are often horribly complicated to the point where they most likely wouldn't get implemented, and very occasionally aren't good for what the original posted actually suggested it for. It's also more of a problem when all discussion becomes centered on this new idea that won't actually get off the ground. Remember, the simpler the better, in most cases. I wouldn't suggest any massive changes like these unless I knew very well how the program itself worked.


So, this has been my community commentary for this year.

[Enjay]

The thing about #1 is that, often, "you can already do this with" posts are correct. Yes, I fully agree that sometimes the alternatives suggested are hacks or miss the point of the request but sometimes they don't. When it comes down to it, accurate or not, I like to think that posts like that are at least trying to help and made in good faith.

I suppose what people (and I'll include myself in this) really need to ask themselves "have I really understood the request and do I know what I'm talking about" before hitting the submit button.

#2 I broadly agree. Very often, a reasonably straight forward request ends up with a pages-long debate about massively extended options that merely serve to lose the original idea and introduce a level of complexity that dooms the suggestion. However, many features have been implemented in a better or more feature-filled way because someone posted "if you are doing X then it might be an idea to do Y at the same time" or "if this is happening, what about parameters for X, Y and Z too".

The devs are pretty damned good. I can't count the number of times I've suggested something and found it implemented with additional features or far neater ways of doing it than I had ever thought about, and they do this without prompting. However, I do feel that there is a place for discussions about implementation and additional features for a particular request. The question is how to ensure that the discussion of this does not deteriorate into a rambling mess.

[Ghastly]

The thing about #1 is that, often, "you can already do this with" posts are correct. Yes, I fully agree that sometimes the alternatives suggested are hacks or miss the point of the request but sometimes they don't.

However, many features have been implemented in a better or more feature-filled way because someone posted "if you are doing X then it might be an idea to do Y at the same time" or "if this is happening, what about parameters for X, Y and Z too".

Yeah, and I think the key here is the simplicity of the suggestion. More often than not, "you can already do this with X" is going to work when "you can already do this with W, X, Y, Z and R" is a gross hack, and "if this is happening, what about parameters for X, Y and Z too" would get in while "how about a system for doing X, Y, Z so you can do T, U, V and W like this and this" won't.

But yeah, I had a really difficult time writing that first part because of the really fluid and subjective definition of "hack." Every other definition I came up with, "you can already get information on target monsters with SetActivatorToTarget and GetActorProperty" fell under too.

[Ed the Bat]

If it's alright for me to pitch in, I'd like to submit for consideration another issue plaguing the Feature Suggestions forum...

As time passes, items get lost to the ages and long buried. Now, I'm not saying anyone's necessarily at fault. It's just a thing that happens. Sometimes a suggestion gets missed or glossed over, and nothing ever comes of it. There are instances of suggestions from literally years ago for features that now exist in the engine, but the sands have buried the suggestion threads and they remain unclosed. These things create a very real issue, in that new features are being brought into a suggestion box that's overfilled with unnecessary padding, which can only be a hinderance for the process. Once in a while, I wonder if it would be fruitful to suggest an effort (perhaps on the part of the community, as I know the devs always have their plate full with important matters) to pick out these loose threads (pun intended) and help clear out the clutter. It always saddens me when incoming suggestions don't receive enough attention (again, not blaming anyone in particular for the phenomenon), and I think it would really be helpful for everyone if the issue could be fixed.

[Graf Zahl]

The main problem why stuff gets lost is that randi doesn't do much there. If I'd no everything I don't want to implement myself the forum would be much smaller. But that's no solution.

[Ed the Bat]

The main problem why stuff gets lost is that randi doesn't do much there. If I'd no everything I don't want to implement myself the forum would be much smaller. But that's no solution.

Understandable. Maybe if another set of free hands were tasked with sifting through it on behalf of the busy devs, at least a few things would come out as 'Hey boss, this is already in. We should close this', or if things seem like a lot of work to implement but aren't altogether bad ideas, they'd go 'What do you think of this, boss? Should we add it to the to-do list, or just chuck it out now?', which would at least kill the suspense of having to wait and see what the final answer is. Personally, I argue with myself a lot on suggestions I make that don't seem to get attention, because I'm divided between 'Maybe I should make a little bump after a while, in case it got missed/forgotten', and 'I shouldn't be a pest about this; it might be rude to keep bringing it back up every few months.'

Don't get me wrong, none of these are complaints about how anyone's operating. I'm just a huge fan of process improvement and streamlining, whenever it might be possible.

[Nash]

If it's alright for me to pitch in, I'd like to submit for consideration another issue plaguing the Feature Suggestions forum...

As time passes, items get lost to the ages and long buried. Now, I'm not saying anyone's necessarily at fault. It's just a thing that happens. Sometimes a suggestion gets missed or glossed over, and nothing ever comes of it. There are instances of suggestions from literally years ago for features that now exist in the engine, but the sands have buried the suggestion threads and they remain unclosed. These things create a very real issue, in that new features are being brought into a suggestion box that's overfilled with unnecessary padding, which can only be a hinderance for the process. Once in a while, I wonder if it would be fruitful to suggest an effort (perhaps on the part of the community, as I know the devs always have their plate full with important matters) to pick out these loose threads (pun intended) and help clear out the clutter. It always saddens me when incoming suggestions don't receive enough attention (again, not blaming anyone in particular for the phenomenon), and I think it would really be helpful for everyone if the issue could be fixed.

Every now and then I actually DO go through the several threads and report the suggestions that have already been implemented. This started out as me just looking for my own suggestions that have been added recently but as I went on, I managed to pick out a few threads from others as well.

(Moderators will have no doubt seen me doing this XD)

Although I don't know if only the devs have access to moving/close threads in that section or general mods have access, too.

[Ghastly]

I think the community takes care of that third issue. Nash, as he said, has been picking through and saying "this has been added already!" for example, and there's the code submissions forum so users can implement and submit things of their own (to illustrate this, would any coders like to take a look at my HealRange property suggestion? [I'm joking] ). The only hard fix to this, however, is paying the developers so they can pick through suggestions for a living (something tells me they'd burn out fast, though ).

[Ed the Bat]

Go ahead and look at the very oldest page in the suggestions forum. We have proposed suggestions that are over five years old for things that already exist. And this is just the final page; I can't imagine how many more such languishing threads there are in the rest of the catalog.

[wildweasel]

Every now and then I actually DO go through the several threads and report the suggestions that have already been implemented. This started out as me just looking for my own suggestions that have been added recently but as I went on, I managed to pick out a few threads from others as well.

(Moderators will have no doubt seen me doing this XD)

Although I don't know if only the devs have access to moving/close threads in that section or general mods have access, too.

Moderators unfortunately don't have the permissions to send threads to Closed; since I'm unsure if the Devs (i.e. not randi) have access to the Moderator Control Panel, I'm not sure that anybody is actually seeing the reports. I leave them there in the report queue in case someone does, but the last time this happened, the one report on a feature that had already been implemented was actually still in the queue upwards of two months later.

I almost wonder if it would be a wiser idea for you to simply bump the threads in question with a link to the bug or suggestion that is related and implemented.

[Blzut3]

I don't see anything in the moderation queue so I guess not. I'd definitely be fine if threads were bumped if the feature has been added.

[Nash]

I could do that (I see several threads that are bump worthy already) but

A) I could come of as spammy
B) What if they still remain unattended... :P

[Blzut3]

I guess if it makes you feel better, send me a PM with links and rational. For the most part I think one of us will notice. We do read the suggestions even if we don't reply to them you know.

[Graf Zahl]

I could do that (I see several threads that are bump worthy already) but

A) I could come of as spammy
B) What if they still remain unattended...

If you bump a thread wit a request to close, that's not spamming. Nobody would ever complain if you did.

[Ghastly]

If you bump a thread wit a request to close, that's not spamming. Nobody would ever complain if you did.

Well, nobody would complain if the feature was already in, anyway.