Tangential, but this is such a great illustration of how Blizzard as a company has succeeded time and time again in taking nearly impenetrable games and genres and turning them into something that new players can enjoy, while still retaining enough depth to make the games interesting for advanced players.<p>From MtG and its everchanging sets, significant expense, multiple formats and awful online experience/UI came Hearthstone. A card game that tens of millions have gotten into. It looks fantastic, takes no initial investment, and costs a fraction of what a MtGO deck costs to go from nothing to a top tier competitive deck.<p>Other examples:<p>- Roguelikes -> Diablo<p>- RTS -> Warcraft & Starcraft<p>- MMO -> World of Warcraft<p>- DotA/2 -> Heroes of the Storm<p>- Team Fortress/2 -> Overwatch (not yet released)<p>It's an interesting case study on a company that's succeeded in taking genres usually reserved for a niche audience and bringing them to the masses. It's not a company without flaws and missteps, but it's an interesting lesson in bringing existing markets to a wider audience. I wonder if there's something more general that can be gleaned from their strategy to products outside of games.
I've been playing MTGO since the end of v2. Someone should write a book about how badly managed this game is. The guy in charge of the thing "Worth" has o real technical background whatsoever, and you can see how that affects the vision and goals of the product.<p>It's a disgrace. And should be a software case study.<p>PS: been playing MtG since 1995
Very good recollection of the history of this game. I've been playing since 2002. I really wanted to keep up with this game but WotC just made it difficult to with each successive release of the game which tended to be worse than the previous one.<p>I have no idea what is going on over there. People had to have been throwing money hand over fist to allow them to fail so badly and keep going.<p>It frustrates me to no end that we don't have anything playable on the web or all over mobile. DotP and Duels Origins don't count. I feel that this type of thing should have been figured out.<p>I also wonder what internal conversations were going on about the Hearthstone ascendant. I keep hearing that they're not trying to do the same thing and that they're not truly competitors but I see a whole generation of kids and adults that have moved on to that. I like the deep play of MtG but Hearthstone just feels holistically better.
I used MtGO version 1 as a filthy causal player and had fun, but the whole experience was very frustrating. For a game that involved occasionally moving a few rectangles around the screen, MtGO had huge (and ever increasing) system requirements for the time and never really ran well even on my fairly well-spec'ed machine. As I recall it also had some weird limitations on screen modes which led to stupidly crowded playing surfaces.<p>I hear the recent versions on MtGO are better, but it still isn't cross-platform or system friendly. It's a shame, because the basic game shone through the cruddy front-end and the servers actually enforced the rules well.
Magic the Gathering Online was basically the entirety of online gaming in my teenage years (the rest was World of Warcraft and Guild Wars, ahem). The server problems and managers mentioned in the article triggered a lot of nostalgia, as it taught me back than that online servers are not made of magic and ponies, and QA is actually something companies should care about. I did actually get invited to a few beta tests for new MTGO sets, though. :)<p>I gave up in college after I learned that Magic the Gathering is actually kinda expensive.
For another take on MTGO, see:<p><a href="http://thedailywtf.com/articles/Do-You-Believe-In-Magic-Online" rel="nofollow">http://thedailywtf.com/articles/Do-You-Believe-In-Magic-Onli...</a>
My friend and I spent about 5 years making an online collectible card game. We started around 2003, around the same time that MTG was reporting scaling issues. Their scale issues inspired me to architect a game server that was horizontally scalable-- theoretically infinitely.<p>Our game, from the start, was built to: 1) playable from the web with no client (1mb flash, I suppose) 2) bi-directional socket based to avoid polling and be snappy 3) server is scalable horizontally 4) adding complex cards and rules will not require a client update and would be easy.<p>We managed to complete all these goals and produce what I still consider to be our magnum opus of software development.<p>The scaling worked by having socket servers provide the end-point to the client, and then behind the socket servers were chat services, game services, trading services, card building services and deck services. So, lets say we had 4 socket servers running: you and I might end up on different socket servers, but we could still play against each other because all the socket servers are doing is relaying your commands to a game server that is running our game. In order to scale, we just add more servers. (Also, if we get disconnected you can reconnect easily. Also, spectating is easy.)<p>My friend came up with the idea of making the Flash front-end use generic commands. So instead of the server telling the front-end "Ship A attacks Ship B for 5 damage" it says "Draw a red arrow from Card #123 to Card #456 and display a red number 5". This allowed us to make cards that did all kinds of crazy things without having to do anything to the client. To implement a card that does damage to all of a player's ships you just have to update the server to send the command "Draw a red arrow from Card #123 to Card #456, Card #789, etc...".<p>Finally, the core game engine on the server was an event loop. Thus, making new cards and rules was super easy because all you had to do in code was say "I want this new card to subscribe to the Player Draw Card event, and in that event code: "if drawer is this.owner, draw twice the amount as regular and then this.sendtograve." I was continually blown away by what crazy cards we could make up and how little code it took to implement.<p>As we were creating this game our day jobs became more and more serious and when we finally had the final version complete we both agreed to burn the code and resources to DVD and put it on the shelve. The idea of starting a new journey in marketing and building a company was overwhelming. (Just three months ago I got it all running in the cloud with very little effort for nostalgias sake).<p>I still think about pinging WOTC and asking if they'd like to see what we have and maybe make use of it. I knew WOTC had an online version of MOTG but I had NO idea it had so many rough patches or I would have been down in front of their office with a sign asking them to look at what we have.<p>If anyone from WOTC or other would like to see or talk about the game, feel free to message me.
I've had a glance at MtG and my first thought was that this thing must be a nightmare to code with all the cards that change rules of the game. Anybody care to comment whether that's the case and how well did they cope with the complexity?
I remember arguing in the Magic Usenet group as to whether such a thing was even possible. I was on the "Yes" side, touting the virtues of object-oriented programming.