Web-services, we've been doing it wrong

Presented at newline: an awesome event organised by Hackerspace Ghent: http://0x20.be/

Transcript

1. Webservices for open data, we’ve been doing it wrong Pieter Colpaert newline - 4th of April 2015 @pietercolpaert
2. some background…
3. More at Open Definition.org A dataset is open if anyone is free to use, reuse, and redistribute it — subject only, at most, to the requirement to attribute and/or share-alike.
4. The gray area of the Web Legal for reuse Damn you, start using open licenses! * not actual size
5. But, confusion arose World of APIs ? World of Open Data
6. Data vs. Services Data dumps Data publishing Data services Algorithms as a service
8. What do you want?
9. What do you want? → and this is true, because his picture is next to the
10. What Open Data people want Raw data → publishing, not a biased service Timely data → the Web is live Maximizing data-reuse → no rate limits!
11. The Web for data crawling machines! Machine experience (MX) matters! If you have to explain an API using external documents, it sucks.
12. Public transit web service of NS Awesome principles! But... Oh come on! You need to request a freaking API key?
13. What does REST+HTTP mean? Glad you ask! App 1 REST webservice Hi server, this is App What can you serve me today?
14. What does REST+HTTP mean? App 1 REST webservice O hi there! I have a limited set of documents which you can discover through these links: http://.../stations http://.../departures …
15. What does REST+HTTP mean? Glad you ask! App 9002 REST webservice Hi server, this is App What can you serve me today?
16. What does REST+HTTP mean? App 9002 REST webservice O hi there! I have a limited set of documents which you can discover through these links: http://.../stations http://.../departures … HTTP cache \o/! Over 9000 others asked the same thing right this minute!
17. What Open Data people want Raw data → publishing, not a biased service Timely data → the Web is live Maximizing data-reuse → no rate limits!
18. Data vs. Services Data dumps Smart servers Data publishing Data services Algorithms as a service Dataset split in smaller pieces accessible over HTTP Smart reusers
19. Data, services and apps Open Data publishing: e.g., • data dump (problem: download entire file on each update) • split your dataset in smaller documents and create links through them (=REST) web-services e.g., • a SaaS route planner: from → to • the closest station to your current location? apps
20. Thanks Pieter Colpaert newline - 4th of April 2015 @pietercolpaert CC BY SA