TOPIC: Token API - interest and prior work?

Wade Schwendemann wrote:

Druegar wrote: You really ought to check with Jeff before beginning such a project.

QFT

Assuming you've done this already.

Of course! Jeff has given me his blessing.

I grok the desire for a TDb API. In general, I'm cool with folks being able to use the info locally. One thing that concerns me is people getting upset that the data they're looking at isn't current. Let me try to explain.

One of the values of the TDb is how easily it can be updated. The printed Token Guide is a wonderful resource and beautiful to peruse, but it's extremely difficult to update. As a consequence, it's not always current. Sometimes folks get miffed when they make a decision--particularly a purchase decision--based on out-of-date information. We see this every year with the printed QTR. "I spent a lot of money on a token and now I learn it doesn't work the way I thought it worked!" {queue rage quit}

It is my strong hope that any user of TDb-sourced data not fall into the stale data trap. At the very least, I hope the folks using TDb data provide an obvious disclaimer stating that the live version of the TDb is the ultimate authority on token mechanics and data collected from it may not be up to date.

I agree. On every product page on my site I link directly to Tokendb. I think that's important.

Ideally, speaking as a developer, not as a Druegar who has to keep Tokendb up to date, Tokendb itself would run off of a real database which provides api access to the actual data.

I don't know if it's helpful at all which is why I didn't post earlier. But a couple years ago I started creating an program to scrape TDB and put the results into a proper SQL database. I abandoned that due to lack of time. Getting the database structure right is really tricky though with the amount of different attributes tokens can have. My main "tokens" table had almost 300 columns. Which is probably not ideal. Also I was certainly not developing with an API in mind.

And ultimately, Kirk is right, it would be best if TDB itself is redeveloped. Anything else would be an unofficial hack (not to say a third-party solution wouldn't be useful).