Card Database Streamlining, and Tokens for Trades

TappedOut forum

Posted on Nov. 4, 2016, 5:59 p.m. by Femme_Fatale

There are essentially 3 topics here, all sort of interlinked. Basically, I'm looking to streamline the card submission process and minimize potential for users to abuse our system to get free tokens (and free cards) for an idea that might be realized in the future; utilizing tokens to help out in trades.

First up is streamlining the submission process.

As some of you know, (or don't know I guess, many people disappeared in the last 6 months), I have a large hand in managing the card database. Anything that could speed up that process is a boon as it takes 3-5 hours to crop, catalogue and organize all the images and then 10-30 hours (depending on the product) submitting them all.

I currently use gyazo as my go to image cropper, and something caught my eye while using it. This something, is called OCR Scanning. OCR stands for Optical Character Recognition. Basically, this is the ability to scan an image and return any text presented on that image into machine coded text.

You can tell where I'm going with this. While yeago will probably be out of commission for a lengthy period of time getting this thing working (unless he can find an open-source one somewhere), it will basically mean that we could just upload a correctly cropped image and be done with our day. Everything would be automatically done for us. It would save me a good deal 5-25 hours a product, and would prevent the need to redo incorrectly done submissions of others. The Gatherer ID will still need to be manually done though, as it isn't present on the card.

However, this is probably only possible if yeago can get the OCR scan to recognize symbols (). Rarity might also be something that needs to be manually done as making the OCR recognize the rarity colour might be difficult to do due set symbols.

Remember, if yeago agrees to do this, he would be out of commission for a while. And if something around the site breaks ... it'll stay broken, for a long while.

Next up is the future potential to use tokens as a supplement for trades that just aren't equal, or when you need to dump a lot of smaller binder stuff that you couldn't normally do. I don't think anyone would be trading anything $20 or more for tokens, hence why I considered it to make those who are trading lesser valued items at high volume, and for those who don't have the trade fodder to make that $5 difference. Further, many people also do a $25 trade limit, mainly to deal with the price of shipping and handling since those costs would accumulate drastically when doing a lot of $1 trades.

The feature is looking like it is going to be only for upgraded users who have a binder. However, there is still potential for people to abuse the systems we have in place that give tokens as incentives. New card submissions, commenting on newly cycled decks, and potentially in the future card fixes on existing cards; are all incentive systems we have or will have. The latter of them all actually introducing decimal tokens, getting 0.1 feature token for every accepted card fix.

While we have a rather successful sinkhole for tokens in regards to the feature queue, I do want to consider the possibility of system abuse and methods to sinkhole anything generated from that. Ideas are welcome in this regard, but the few that I consider are: regularly hosted T/O tournaments that have small token fees with prize pools. A small trading tax when trading with tokens (like 0.2 to 0.5 tokens or something). New methods of deck advertisement.

Now in terms of restricting system abuse, there are a few potentials, and one almost guaranteed. The almost guaranteed is just me constantly working on the database at the rate I do and putting out most others. I have 750 tokens. Probably soon to be more once I work on C16. I keep these tokens for potential event awards. Anyways. In regards to the rest of the database, the card fix .1 feature token reward can be abused, which is why I want to completely alter how the system works (I suggested this a long while back, so much is a repeat if yeago has this in his books).

Basically, I want to completely forgo the entire card submission/fix form for any user who is not a registered database manager. Those of us who regularly manage the database will still get that form (as it works quite well), but for everyone else, what you'll get to send us is an alert. The alert is pretty simple. You press the "notify of card issue" button, or whatever we'll name it, and a pop-up text box will come up. You'll use that to write a description of what is wrong within a minimum of 10 words. Press the send button and wait till one of us addresses the issue.

How this will look for admins Show

This is really for yeago so he knows how to structure it within the site.

Instead of having the admin page as it usually is the place where something like this might take place, it would be a lot more organized if it would be on the card page. Beside each set (promo set being the exception) there would be yellow triangle with an exclamation mark alert sign if in that set there was an issue with one of the cards. Going into that set we would scroll till we found the card with the issue, go to that card's page, and find a button that says "(# of) alerts!" Click it and a list of alerts displayed in place of the discussion section, though displayed like a comment, would appear. There would also be a discussion button as well, so we wouldn't be stuck on alerts. At the bottom right corner of each alert would be two buttons, a green checkmark and a red x. Both would clear the alert from the list but only the green would give the feature token. We would then go about our duties to try and fix any errors, and notify the user who posted the alert as to why we couldn't fix the issue if we indeed couldn't. In this, we can quality control who gets awarded for their efforts if they are being genuine or trying to spam the system.

With this system in place the feature token reward would probably go back up to 1.0 token since we could quality control a lot better, and, it would be a lot easier to get notified of any issues that occur as this could be done by any user, not just upgraded, as we wouldn't have to worry about the quality of the submission. It would also help clear up the T/O section of the forum of questions about the card database as they could simply direct those concerns to the card page, and if an alert came up multiple times we would put a disclaimer here about it.

As for the first cycle thing, there is the word "sometimes" so I guess that automatically limits it? You can't go anymore limiting than WotC and I bombarding yeago with features to consider and fiddle around with and alterations to the database that need to be done within a week. Every set. Goddammit WotC. First flip morph cards (which we had placed some features thinking that that very thing wouldn't occur) and now partner commanders. Ugh.

So, discus and debate about the points I have presented, and hopefully we'll end up to a place where we are known as THE deckbuilding site. I can only wish. Lots of work to be done to get there.

The_Raven says... #2

What would the requirements for the "database manager" be? How many people would you want to become "managers"?

If there's enough potent people to update the database, would it really be nessecary to have this scanner thing?

November 4, 2016 6:37 p.m.

Femme_Fatale says... #3

Yeago has always wanted to completely streamline the entire system as he wouldn't have to worry about relying on others as others are never reliable. Either not knowing enough or simply not being around.

November 4, 2016 7:10 p.m.

Eiti3 says... #4

I personally believe all three suggestions are superb ideas. Although, as a software engineer myself, I understand that any one of these will take a large amount of time to get working properly without the worry of "too" many bugs arriving after completion.

Doing anywhere in between one to three of these will result in possible "Missing" yeago fliers popping up in the forums. However, I am not saying we should avoid doing anything femme has suggested. Just get a solid grasp on what yeago feels capable of undergoing before giving a straw poll as I'm typing this under the impression that this is femme's ideas to get feedback on before pitching the main course to yeago. However, like many times in my life, I could be incorrect.

So, I'll round out my nonsensical blurb to saying I would think streamlining the card uploading process would be the best feature to implement first, as it best benefits everyone - the sooner we have the cards uploaded on the site, the sooner we can use them, no matter upgraded or not.

The "database managers" would be the second best objective, if in the case that yeago is unable to succeed in the previously mentioned. However, as femme mentioned, relying on people is far too risky, especially so when they don't get paid or act as a substantial friend/associate.

The other thing is basically a rainy-day sort of project. Something not necessarily needed ASAP, but worked on when roadblocks are found on more urgent matters, or when one gets bored. That is, however, my opinion.

November 4, 2016 7:48 p.m.