Coming “Soon”

These are features that Mb2 could support “soon”. I have researched each feature and believe that almost every one could be in a proof-of-concept alpha state with two weeks of me working full-time on that feature, although I don’t have much time to program when I’m chasing funding or doing other business related tasks.

I expect a strategic investor or a potential customer will dictate the priority for each feature. In the meantime, native apps and then ring games will be my focus.

At any given time, you can see what I’m working on by looking at the list of high priority issues and see what I’m planning on working on by looking at the list of medium priority issues. In addition to those two labels, some features have their own labels, which are included in the table below.

I update issues on GitHub multiple times a day. I only update this page when I think about it, which means it’s often out of date.

Native apps

I’ve made—but not yet released—native Mb2 apps using Tauri.

Currently, the UI is just the web UI, which is squished and awkward on a phone’s aspect ratio, but is much nicer on a tablet. Additionally, there is not yet code to integrate the login password with the iOS key-chain, so credentials must be supplied if the session expires.

The macOS and Windows native Mb2 clients use a separate window per table. The iOS and Android Mb2 clients have little controls that allow the player to switch from any table to the lobby and from the lobby to any table. These native versions are clunky, but have been what I’ve been playing the daily tournaments on, which has allowed me to find and fix many bugs. I’m now (February 11th) confident that Tauri is solid enough to use in production.

Ring games

Internally, Mb2 already has limited support for ring games. Some of what is lacking is simply functionality related to players joining, changing seats, missing blinds, etc. Those will be easier for me to implement once I can block out some time with someone who knows the ins-and-outs.

Now that native apps are working well enough, I can resume work on ring games once I know how I’ll pay my bills.

Teaching

Craft Poker Co. has a client who commissioned some modifications to help them to teach poker to their new traders. The two major abilities added were:

• pausing a game at the end of a hand so that the teacher and students can discuss the hand.
• allowing the teacher to reveal hole cards at the end of the hand

Additional teaching functionality can easily be added, but will not be done speculatively. Anyone who is interested in using Mb2 to teach poker, is encouraged to contact me so we can work together.

WSOPS compatibility and structures

Each year, starting in 2022, craftpoker.com has hosted a series of WSOP-Style tournaments, where the games and structures are patterned after actual World Series of Poker events that are being held on the same day.

When the official World Series of Poker structure sheets are released for 2026, I’ll generate structures and the WSOP-Style series. Each year I have created tools or systems to allow me to do this quicker.

Customization

It should be possible for anyone to easily “skin” mb2, and give it a custom look-and-feel. It should be possible to share skins either through the server or “behind the server’s back”, where the customization only happens on the client side. Client-side only customization is a privacy feature.

I have been experimenting with client-side customization and have exposed primitive functionality that people can use today to customize the dealer button, the background and the styles of a table. Eventually, rather than customizing each of those aspects individually, it’ll be possible to create a single “asset pack” that customizes those aspects and more (e.g., a custom deck).

I would prefer to work on further customization when I can pay a contractor to work with me. I have both concrete and abstract ideas of how to proceed.

More games

Mb2 is not a poker-engine, but it’s not too far removed from one. In other words, although mb2 doesn’t allow other programmers to create new poker games, doing so isn’t particularly hard for anyone with access to the mb2 source code (which is currently just myself). As an example, adding the Crayfish variant took two days of work, neither of which were particularly close to full time.

Although there are no specific games I plan to implement in the future, typically I add some in anticipation of the WYWAE Series (for 2026, I added Dramaha Zero, Dramaha 21 and Dealer’s Die) and sometimes for the WSOP-Style Series (e.g., Double Board Bomb Pot in 2024).

Membership

Internally, mb2 has tags that can be attached to events (tournaments and ring games) and to players. Tags can be used to track series standings and/or to limit the visibility of events to members with specific tags. There is currently no user interface for adjusting tags, instead tags are rarely used and are created and manipulated in the database.

Adding a user interface to the current tag-based membership functionality and expanding that functionality are both straightforward things that I or almost any programmer can do.

Statistics / Alternate scoring systems

Eventually, craftpoker.com will offer a premium service were people pay a fixed amount per month (e.g., $5) and get functionality unavailable to people playing for free. Statistics are easy to calculate and track and are a logical part of premium membership.

Mb2 uses Fun Money to keep score, which is honest in that there is no buying of extra chips, and an account’s surplus or deficit is tracked for all time and other periods (i.e., year to date, month to date, week to date, since a year ago, since a month ago and since a week ago). Zero sum scoring can be off-putting for someone who plays for entertainment value.

All buy-ins and pay-outs are tracked in a database, and the leaderboard options available are all simple reports generated via SQL. Adding new reports is trivial, as is adding a play rake or bad beat jackpot. People may be interested in simulating the drop at a casino they’re considering visiting in the future, so an even more realistic scoring system that includes a rake may be important to them.

On the other hand, people may want the ability to reset their score (or Fun Money) weekly or change it based on external events (e.g., get more Fun Money for every 10,000 steps you take) or perhaps not use Fun Money at all and have a point system.

I do not plan to work on statistics or alternate scoring systems in the near future, but mb2 already has some rudimentary capabilities that will be extended fairly quickly when it’s appropriate.

White-labeled sites

Craft Poker Co. currently has two instances of mb2 running, the public one at https://craftpoker.com, and one that is leased to a financial firm who uses it to teach their new traders how to play poker. The private site is currently a “one-off”, in that it is updated by hand in parallel to (but not as frequently as) craftpoker.com.

Currently, another instance or two can be brought up fairly easily, but it is an ad hoc process. It doesn’t make sense to automate white-labeled sites until more customization work has been done, although with interest and funding, both could be done in parallel.

Structure and Series editors

Working alone, it would likely take more than two weeks full time work for me to come up with an alpha version of a structure or series editor, because it would be UI intensive work and I’m very slow when doing UI work. Click the triangle for additional technical details.

Impossible to rig deck

People are good at seeing patterns, unfortunately that includes patterns that aren’t actually there. Confirmation bias is a thing and that means that online poker always looks rigged. Online poker deals more hands per hour than are dealt in a casino or home game with physical cards. Consequently improbable things happen more often in online poker than in manual poker and players’ minds will come up with an explanation for why the various improbable things are happening (e.g., the site has a bias toward newbies, to let them win more and hook them). Additionally, it’s obvious that a site could be rigged even if it isn’t actually rigged. An “impossible to rig deck” can mathematically show that certain types of rigging aren’t being done.

Here’s a hand-wavy description:

Each client has a way to send an encrypted value to be used in the generation of a random number for any next event that requires randomness. At that time, everyone in the hand sends their encrypted values. Once they’re all received, all the encrypted values are sent back to everyone in the hand. After everyone receives the encrypted values, everyone sends their key to the server which then uses the keys to recover all the values. It then puts those values into a pre-defined algorithm that yields a random number given set of values. The random number is then used. After the hand is over, all the keys are sent to everyone who sent values. This allows people to verify that everyone’s values were used in exactly the way that was defined before the start of the hand.

That’s the complicated version where it’s clear that everyone’s client participated in the random number choice in a way that can’t favor anyone. A less complicated manner is to simply encrypt the seed to the hash routine and to send that encrypted message to all the clients, then send the encryption key after the end of the hand. The biggest drawback is that since the provenance of the seed isn’t known, there’s no guarantee that the bot isn’t sending better (or worse) starting cards to specific players. A lesser drawback is that if anyone can figure out the key, then the deck is effectively transparent. The more complicated version, above, has multiple keys, each of which could conceivably be generated in a different manner, and to see through the deck, someone would need to recover each of the keys.

Although I’ve done no coding of the Impossible to rig deck, implementing the “happy path” is easy. Dealing with edge cases, is not something I’ve thought much about, nor will I until there’s a reason to.

QR-code based pick-up games

Once the user interface is pleasant to use on mobile devices, especially after ring games are working well, it makes sense to add an easy way to create ad hoc tournaments and games, e.g., by displaying a QR code that can be scanned to join an event.

This is trivial to implement but won’t be particularly useful until the whole user experience is a fun one.

The server running on a phone or tablet

Although I haven’t written any of the relevant code, my research into native apps makes me think it’s very likely that putting the poker server into an app that runs on a phone or tablet is fairly easy to do.

I have no plans for doing this soon and perhaps might never want to do it, but it could be worthwhile from a privacy standpoint. So, if there is a privacy-conscious potential investor or potential customer who is interested in this feature, let me know.