TO Mass Game Creating Engine

[Robespierre__]

Concise description:

• Please create a means for mass game creation for TO's

Specifics/Details:

• It would be great if we could upload a .csv file to generate all the games we want to create in one blink of an eye. That would allow us to create formulas in Microsoft Excel or Google Sheets. This is especially useful for tournaments that feature participants selecting a home map. Any TO who has run such a tournament will attest to the insane amount of time that adds to his/her life. If the process is a little technical, you could require a training period where the file is first submitted to an experienced volunteer who would make sure that the TO in question knows what he/she is doing. After a training period, the TO could be untethered from supervision and not overburden the volunteer (or non-volunteer) staff. The key to this suggestion is to allow the creation of 100+ games at one time, all with different settings.If you could allow the games to be created with invitations automatically sent out ... wow

How this will benefit the site and/or other comments:

• I guarantee that the number of tournaments that start and do not finish will decrease.
• I guarantee that the number of tournaments with a variety of maps in them will increase.
• I guarantee that allowing automatic invitation sending will multiply the effect of #1 and #2 by a factor of at least 5.
• I guarantee that this suggestion poses a series of real challenges to he who codes it in terms of preventing corrupt/incorrect data from doing harm to the site.
• I guarantee that DaveH will give whoever codes and implements this a foot massage.Every Day. For a year.

Gratefully,

Robespierre__



MOD EDIT: Check this please - James K

[DoomYoshi]

Could this be done with a third-party addon?

[Robespierre__]

The tournament has required the creation of 5,469 unique games thus far (just a few more to go) as you can see by the results google sheet.

https://docs.google.com/spreadsheet/ccc?key=0AtVtXGQIZiUddElzYkZhUEt3anozdGY5NUJFNjJ4TWc#gid=3

Do you know how much time it took to create all the results automation for that sheet? How about when you add in the time for these sheets:

https://docs.google.com/spreadsheet/ccc?key=0AtVtXGQIZiUddFA1RXVfVWIySHBldWRmdldHRGtpb3c#gid=0
https://docs.google.com/spreadsheet/ccc?key=0AtVtXGQIZiUddENfOGQwRGladWdpdG00NTRrSTJkU3c#gid=0

Notice how every how the winning team has their score highlighted in color? I had to teach myself to code that using Google Sheet's scripting language. I have other ideas of what I would like to do, but I honestly can't implement them because of the amount of time that game creation takes unless I want my company to go out of business or to be the worst father that I possibly can be.

I edited this (only cut some paragraphs out) to make it less of an argument from the suggestion phase and more so an example that will help make it a tool. -James K

[patrickaa317]

Though it seems all in charge (and greenoaks) oppose this feature, I just wanted to post in support of it.

I get wanting the human element in the tournament creation but the UI is also not currently all that friendly to doing this. It's getting better in being able to make 5 of the same games at a time but all that did was kept us from hitting F5 four times after the first creation.

The thing that got me initially hooked on this site was Viper's massive 1v1 tournament. If that didn't exist, I probably wouldn't have gotten hooked, once I realized there were tournaments I started signing up for all of them, it was great until I realized how fast 4 freemium spots were filled... Then I purchased premium and have been stuck here since.

No offense Greenoaks as you have put work into your series but most of what we have for tournaments now is a lot of 16 player, standard bracket tournaments. When I see someone with some tournament trophies, I just assume most of them come as a result of winning 3 or 4 games in a row. When I see someone has hosted a shit ton of tournaments and have a ton of medals, I also assume they are the run of the mill, standard bracket, 1 loss and you are out tournament. I've ran a few of those myself (with multiple games per round).

Tournaments and clans are really the two main things to promote cc and assisting your volunteers to run more fun tournaments, like a basketball league or a big round robin, only seems to make sense.

And whoever made the comment about not giving a TO a medal for running one of these, that's fine but in that same regard I think someone who runs 1 simple tournament shouldn't get the same exact medal as someone who successfully completes a more large complex tournament. Now I really don't believe that but right now the minimum work to get a medal is creating a forum thread, creating/watching 18 games, sending 36 invites and maintaining the thread throughout. To me, a TO that creates a medium-large tourney, formats a CSV or XML file, imports it, verifies all inputs are correct, is about the same amount of work.

Also, someone said something about more games to be deleted. How is this a concern with the drop games or cancel invite buttons. If a TO screws up, it's up to them to do this.

I can see this being a little lower priority than other things but I think this would benefit more than just a few members. It might actually attract more people to the site that are like me a few years ago.

[Hath]

I don't see how this could take THAT much time to implement. You could have a rigid format, with required fields in a required order, and if they don't match, kill the import. Parse the csv file and pass in to your game creation function (shouldn't be hard). Pass back out a file with the game #'s created in the last column. Seems simple enough to me.

You could make the page only available to TO's who specifically requested it, to prevent overwhelming new TO's (who wouldn't know better), or inform those whose tournament seemed like it might benefit from the functionality. This would limit the exposure, as well as the potential for mass game creation and deletion from mistakes. I think this is of benefit to more than just this one tournament, too. It would encourage more tournaments to be run with choice of home map, 1v1 or otherwise.

I've created a reasonably large percentage of games for the NCAA tournament, and as much as I love clicking through the UI, this would be so much easier. I understand different things have priority though, so if you want to table it, that's fine. But to outright reject it on the notion of too much work involved...that's probably overboard.

[DaveH]

Only a few people could build a proper spreadsheet that could be parsed by the site to create all the correct games. Even if it could be used in every tournament, that doesn't mean the organizer could make it work.

I understand what you are saying, but only a few TO's can manage a tournament as complex as Robes and a few others run. Besides which, a formatted Excel spreadsheet with macros could ensure the format was exactly as required, and could be managed by anyone with basic spreadsheet skills.

The other point is that Robes was specifying the type of tournament with player-set maps and settings - which is why most TO's avoid such interesting tournaments. For these types of tournament the above type of system would open up such tournaments to other TO's who do not have the time nor temperament to manage them under the present system

It would not be required to be used in every type of tournament, so I agree that it would possibly be better as an add-on - if possible.

Cheers

Dave
(If you do have reference to further information, I would be interested in looking into it in more detail.)

[GeneralJestix]

Only a few people could build a proper spreadsheet that could be parsed by the site to create all the correct games. Even if it could be used in every tournament, that doesn't mean the organizer could make it work.

Give requirements for a csv file.

Require headers like: Game Name, Player name, map name, fog, spoils, etc.

Or better yet, create a template for them to fill out that has all required columns listed.

[Night Strike]

Only a few people could build a proper spreadsheet that could be parsed by the site to create all the correct games. Even if it could be used in every tournament, that doesn't mean the organizer could make it work.

Give requirements for a csv file.

Require headers like: Game Name, Player name, map name, fog, spoils, etc.

Or better yet, create a template for them to fill out that has all required columns listed.

That would neglect the benefit that people are wanting this suggestion to achieve because the only way to properly use that spreadsheet would be off a set template.

[Metsfanmax]

Only a few people could build a proper spreadsheet that could be parsed by the site to create all the correct games. Even if it could be used in every tournament, that doesn't mean the organizer could make it work.

Give requirements for a csv file.

Require headers like: Game Name, Player name, map name, fog, spoils, etc.

Or better yet, create a template for them to fill out that has all required columns listed.

I'm having a hard time visualizing a spreadsheet template general enough to account for the variety of tournaments that people host; is anyone willing to work up a simple example?

[Robespierre__]

The "template" as it were is quite easy as the parameters for game creating are EXACTLY the same. The required fields are the same no matter what.

The complexity (to the degree that there is any) comes from the formulas that will fill that data. You need to write formulas that pull data from one sheet based on (say) "Screenname" which will then know what map to select, what spoils setting to select.

If you look at the "Sched" tab for this sheet:

https://docs.google.com/spreadsheet/ccc?key=0AtVtXGQIZiUddHlza1hEYmtiOFdJbmstOGFhcFNncEE#gid=1

You can see that after you enter the team name for columns B and C, the rest of the columns are filled with data (the little icons next to the team name in columns D & E show bonuses earned during the regular season -- that info is pulled from the "List of Players Teams" tab. The map info also pulled from that tab.

Where the "template" would be really useful would be in CC supplying a spreadsheet that had all of the map names etc. EXACTLY as the game engine required (or a map code number to further reduce syntax errors).

I really do not think this is that hard of a project. I am guessing 16 hours of developer time. Of course ... those 16 hours are stolen from other projects.

I 100% agree that only certain TO's would be allowed to use it. I am also certain that the kinds of tournaments those TOs would put together would be really great.

Robes__

[DoomYoshi]

That would neglect the benefit that people are wanting this suggestion to achieve because the only way to properly use that spreadsheet would be off a set template.

We already use a restrictive template, it's just not in .csv format.

[patrickaa317]

Done .... apologies.

I see your spreadsheet but think it should be a lot more streamlined/simplified. I can throw one together later; and even have parameters around functions around certain values to ensure appropriate data is inserted in them.

[Robespierre__]

I am not following what you are saying but will await your example. What I have done is use VLOOKUP to pull data based on the value of Column B or Column C. If we were supplied a spreadsheet by CC containing player screennames or #'s/map names or #'s / and then the text or number that they want for all the other settings (spoils, fortification settings, game type, etc), then all that would be pulled formulaically and be really easy.

[patrickaa317]

I guess I was just looking more at your spreadsheet. Columns A-E are irrelevant. You'd need to add the following:

- number of players
- GameType
- Initial Troops
- Play Order
- Fog & Trench in separate columns.
- Round Limit
- Round Length
- Some indicator on the tourney associated and password
- Plus if you are looking at really using an import function, you'd want to have all the player names listed.

I'm stating the obvious here but want to point out that your spreadsheet as it stands would need to be mocked up in a completely different way.

[agentcom]

Yeah, I think Robe's post was just meant to show that data can be pulled, which is also sort of stating the obvious.

What I was picturing (correct me if I'm wrong) is that you'd have a sheet that has the "game information" that will be uploaded to the site. This would be pulled from other spreadsheets, saved as a csv and uploaded. So, the csv would end up looking something like this:

4,agentcom,patrickaa317,robespierre__,greanoaks,S,Classic,A,S,E,C,F,N,20,pw

Basically,

[#players],[player1],[player2],...,[playern],[gametype],[map name or code],[deploy],[turnorder],[spoils],[forts],[fog],[trench],[roundlimit],[password]

Something like that. The gametype and player names are different enough that the player names could be parsed first to ensure there's N players for an N-player game. The map code idea is a good one just requires a separate Vlookup on a table published by CC. All the variables are distinct enough that you can make them unique (I know because I've done this for spreadsheets that I work on and MapRank and the Game Search already do it, too).

Is that the right idea Robes?

If so, the spreadsheets themselves would be fairly easy to design, IMO.

[Robespierre__]

That is precisely what I had in mind. It would be easy to manage a universal template.

I would like to throw out an example of something that we could do given an engine that there is no way on God's Green Earth I would think about doing otherwise.

Clan Baseball.

Each clan puts out 9 players and each plays one singles game for every scheduled game (so you would play 162 singles games over the course of the season). You could have division winners and wild card teams and have a playoff in the end leading to the world series. Perhaps we don't have enough clans to do American and National League, so we just do 16 clans. After a spreadsheet is properly set up, all the TO would have to do is manage the rosters as people go in and out of the 9-person team. All the game creation would be 100% automatic and take 5 minutes tops. If invites are automatically sent, then all missed invites are counted as losses. Tallying results would definitely take some time, but it would not be *that* bad so long as the job were shared amongst 2-3 volunteers. I think 4 people could run this league and no one of them would feel like you hijacked their lives.

This is just one stupid idea showing the kinds of tournaments that you could create which build community that I think would be a real addition to CC.

But I think if I can't convince Blake, Green Oaks, Night Strike, etal, then either (a) they are predisposed to hating my ideas, (b) this idea is not as good as others floating around, or (c) it just is not that good an idea. I doubt it is (a) so that would leave (b) or (c). I am being persistent because I really believe in what I am proposing here.

Robes

[patrickaa317]

Yeah, I think Robe's post was just meant to show that data can be pulled, which is also sort of stating the obvious.

What I was picturing (correct me if I'm wrong) is that you'd have a sheet that has the "game information" that will be uploaded to the site. This would be pulled from other spreadsheets, saved as a csv and uploaded. So, the csv would end up looking something like this:

4,agentcom,patrickaa317,robespierre__,greanoaks,S,Classic,A,S,E,C,F,N,20,pw

Basically,

[#players],[player1],[player2],...,[playern],[gametype],[map name or code],[deploy],[turnorder],[spoils],[forts],[fog],[trench],[roundlimit],[password]

Something like that. The gametype and player names are different enough that the player names could be parsed first to ensure there's N players for an N-player game. The map code idea is a good one just requires a separate Vlookup on a table published by CC. All the variables are distinct enough that you can make them unique (I know because I've done this for spreadsheets that I work on and MapRank and the Game Search already do it, too).

Is that the right idea Robes?

If so, the spreadsheets themselves would be fairly easy to design, IMO.

Love it, except I might suggest putting the players at the very end given they are the only one that can vary in number. You would want to to ensure that there is >= N players for N-Player game; I don't think you'd want to force all to be invited at once. It might be a situation where i want to create the game now, just not send 100% of the invites in case it is dependent on another result.

I am being persistent because I really believe in what I am proposing here.

Completely agree here. I'm mostly quite disappointed this was rejected so fast, especially for a vague reason of 'not enough resources'. If it's something that is beyond the ability of the dev team, then that's fine, just own it. If it truly is the case of not enough resources, let's throw out at least half of the things sitting in the submitted box now. IMO, this would be way easier and quicker than GameLog 2.0, Unlimited Adjacent forts, etc. Plus would have more benefits for TO's to run tournaments like the Clan Baseball league like you suggested which WILL ATTRACT MORE MEMBERS.

[Robespierre__]

Love it, except I might suggest putting the players at the very end given they are the only one that can vary in number. You would want to to ensure that there is >= N players for N-Player game; I don't think you'd want to force all to be invited at once. It might be a situation where i want to create the game now, just not send 100% of the invites in case it is dependent on another result.

The beauty of this kind of format is that none of the games are dependent on any of the others so they can just keep flowing out from the TOs. Only once you reach the playoffs will it have dependent games.

Robes

[patrickaa317]

Love it, except I might suggest putting the players at the very end given they are the only one that can vary in number. You would want to to ensure that there is >= N players for N-Player game; I don't think you'd want to force all to be invited at once. It might be a situation where i want to create the game now, just not send 100% of the invites in case it is dependent on another result.

The beauty of this kind of format is that none of the games are dependent on any of the others so they can just keep flowing out from the TOs. Only once you reach the playoffs will it have dependent games.

Robes

I was referencing more the format of the CSV than your Clan Baseball tournament, though I do think that'd be a fun tourney to play in.

[Robespierre__]

I see what you are saying now. It would be great if you could leave the invite field empty and therefore extend no invites. But if we had to wait to start games only when there are active participants awaiting, I could live with that.

[patrickaa317]

I see what you are saying now. It would be great if you could leave the invite field empty and therefore extend no invites. But if we had to wait to start games only when there are active participants awaiting, I could live with that.

Completely agree with that.

This would also be great for clan wars where clans use a bunch of the same map/settings for each war.

[JamesKer1]

This thread was once a suggestion that was rejected. With some editing, I have eliminated a bunch of talk and summarized it below, but made sure to leave notes and conversations on the technical side of things. This is a copy of its old thread.

-This suggestion had lots of support from the community (actually 100% from non-Team CC members), but it was shot down by blake and NightStrike for using up too much time and not effecting enough people.
-An argument was made that, if created, it would lead to more complex tournaments, fewer abandoned tournaments, and faster tournaments, along with making tournaments more "magical instead of torture."
-Some parts may be a little choppy since chunks of totally irrelevant conversations have been deleted. I tried to fix one major one, but the others I left to try and keep everything original.
-Robes and I have discussed this, and would like to present this as a Tools Suggestion now, hoping to bring it to someone who does have the time to code it, work with it, and benefit TO's for years to come.


If you really want to see all the talk in and not in favor of this, check out rejected suggestions. Any questions, contact Robes.

James K