That story makes no sense at all. They started developing an exchange for cards in 2007, then somehow modified the system to work with Bitcoin four years later? There's absolutely no similarity between those services. I provided a reference to archive.org for the date I mentioned, and you can verify that no such service ever existed until the domain was repurposed in 2011.
It's ridiculous to even begin to think that there's any similarities between the sites outside of the domain name and it's owner.
You're free to hold that belief, I don't agree. The public story seems very clear to me that the original owner pivoted to become a bitcoin exchanges and then sold to the current owners. It's reasonable to believe it's based on the original code for trading cards.
If you don't agree, then don't agree, I don't care.
Please explain the similarities in the backend to me. Trading cards is basically an online store, nothing like Bitcoin and nothing like a currency exchange. The similarities begin and end in the name.
Are you a programmer? I am, and I damn well know that every programmer reuses code between his own projects and both the original site and the current site were originally written by the same guy, he pivoted. If you don't think he reused any code you're just being foolish.
I am, but there's no code to reuse in this situation. Maybe some scraps of user authentication code, all the rest is completely different. What you're describing is like trying to adapt a VW Beatle to tow cruise ships, sure you could maybe reuse the steering wheel, but why would you want to when you could build one better suited for the job?
Then disagree with me and be done with it and stop f'ing arguing when it's clear what I believe and why.
And I still disagree, plenty of the site could have been re-used because if the new code uses the same core frameworks, a home-brew persistence framework for example, it could have the same core issues; issues that may not have mattered on a trading card site but matter greatly when money is involved, rounding problems, lack of transactions, stupid architectural choices like MongoDb, who knows.
If you don't agree, I don't care. I've been doing this for 15 years, I trust my own expertise far more than some disagreeable person on the internet.
Your argument seems to be "reusing trading-card code for a currency exchange is stupid and would result in a terrible, bug-ridden product." Mt.Gox is terrible and bug-ridden. I'm not clear what you're disagreeing with.
Looking at archive.org, there's nothing but a "contact us / patent pending" text-only web page that exists of the hypothetical MtG trading card site. Also, the Wikipedia assertion appears to have been copied from Wired, potentially a seed for citogenesis[1], as Wired is the one and only source of this claim and we have no supporting evidence.
We cannot know what was reused from what, absent new evidence, as there is no public example of an actual MtG trading card site that I can find anywhere, just a quick note saying that the idea was "patent pending" and that one might someday be found there until a Bitcoin exchange popped up.
I would be quite interested if anyone has new evidence. The record, so far, is a bit thin. Perhaps the Jed McCaleb has a Github or other public repository containing items of interest?
Interestingly, if you go back and look through the Wikipedia page's history, you can see what appears to be exactly that process happening.
Before the Wired article, the Wiki article implies that the site was an operating Magic card exchange, but the citation is a link to the Internet Archive's copy of the stub page, with absolutely nothing there to substantiate the idea that it was actually an online exchange. The Wired article appeared later and is unsourced, but could plausibly have taken the Wiki article's unsupported statement at face value. Then just the other day (February 9) the Wiki page was edited by a user named Agyle to use the Wired article instead of the Wayback Machine reference.
Trading cards are fungible, and it's not hard to build an exchange where current price is based on the last price paid. There's no fundamental reason that trading cards couldn't be used as a currency if enough people agreed about their convertibility, for that matter. Do you know for a fact that the code base was written from scratch, or are you just assuming that it must have been?
Do you have additional insight beyond what the Wired story says?
"But then McCaleb turned [Mt. Gox] into a website where people could exchange cash for bitcoins, a digital currency that had only just found its way onto the internet, and just as the exchange started to take off, he sold it to Karpeles."
Are you saying McCaleb started with a fresh codebase and just hosted it on the same domain? I haven't seen any evidence to support that. But I also don't understand what support for your claims you think that archive.org link provides. Just that a card exchange was hosted there for several years?
There's no evidence to suggest any code reuse, much less 7 years later. The card exchange never existed in a usable public form, it's just a page saying to contact them. What code could possibly be reused between the services had there not been a 4 year gap?
It's ridiculous to even begin to think that there's any similarities between the sites outside of the domain name and it's owner.