http://ticket.bitzi.com/rdf/urn:ed2k:0123456789ABCDEF0123456789ABCDEF

thank you.

Some ideas:

(1) Identify your requests with a user-agent and version wherever possible, so that if your software does go haywire or overtaxes our service, we can contact you and adapt.

(2) Only perform hits when they will clearly add displayed value to the user. A hit when a user consciously requests it (via a click or menupick) is always OK, dozens or hundreds of hits when they haven't requested such info are not. (If you want to do mass background lookups, please contact us so we can work out a way to enable it with minimal disruption.)

(3) Respect HTTP error codes (like 503-server busy) and the 'retry-after' header for requesting a 'time-out' between requests.

(4) Include attribution to Bitzi when using/displaying Bitzi data, with a clickable link whenever possible.

We'd love to hear of new or planned uses, and see Bitzi-using software as soon as it is publically available. We just have to be careful with our current server resources.

Thanks for your suggestion and interest in Bitzi!

- Gordon @ Bitzi

Here's another extension idea that would come in very useful for me.

http://bitzi.com/lookup/UESBCFZEGEAFIZKW7ZMXNR3XBJ6EJG3M

has audio sha1 value of 

PODBREYKXRSQKHWZHPDB7HT45L56JQXC

but 

http://bitzi.com/lookup/PODBREYKXRSQKHWZHPDB7HT45L56JQXC

Says:

"Bitprint Not Found"

I would like to do a lookup via the audio sha1 hash.

I wish that your website code was opensource so I could send you a patch.

Barring that for the time being, is there any syntax people are using to specify a "naked" file? I (probably mis-)understand that at least Shareaza has an option to strip metadata and/or query for files with duplicate audio content regardless of embedded metadata. audiosha1:{base32sha1}? Suggestions?

I also wish the website were open source (or maybe not open source in the legal sense, but with anon cvs access) as well. I don't understand why there are almost no websites that completely open up their source -- seems like it would be a big win, especially for sites with a tech-savvy community. I can't promise anything for this site, but I strongly endorse the idea.

gojomo, I will use this service responsibly.

> (If you want to do mass background lookups, please contact us so > we can work out a way to enable it with minimal disruption.)

I plan to lookup about 30000 tickets by ed2k hash, please give me a elegance way instead of sending HTTP request to your server.

Thanks for your help.

Eventually we'll have a new full database snapshot available for download; when we do, we'll offer both HTTP and P2P downloads. Then you could perform your lookups offline.

In the meantime, HTTP hits are the best way. To be kind to our server, I'd suggest:

(1) Only one request at a time

(2) Pause between requests. A nice pause time would be the greater of either 2 seconds or however long the last request took. (Then, if the server is bogging down and your requests are taking many seconds, you'd wait many seconds before trying again.)

(3) If you get back any 5xx errors, slow down. Respect any 'retry-after' header that comes back asking for a delay before you retry.

(4) If you just need to scrape it once, and doesn't matter to you the time of day, try to do it during our slow hours -- 12am to 5am pacific time.

Feel free to suggest other interfaces that'd be helpful, and let us know more about the volume and freshness requirements for your application, so we can better plan future offerings.

Thanks!

- Gordon @ Bitzi

I need the data since Aug.2001.

So I'll use your xml lookup service an I'll obey your rules.

I'll send nearlly 30,000 lookups. In my estimation, 25% of them will return a ticket.

Thanks for your great service.

http://ticket.bitzi.com/rdf/0123456789ABCDEF0123456789ABCDEF

or

http://ticket.bitzi.com/rdf/urn:sha1:0123456789ABCDEF0123456789ABCDEF

Think of it as a stepping stone to getting "naked" file support.

I primarily want to use this to share info about corrupt or cookoo egg audio files.