Right, got me some power (until I get chased off by the organisers) and 3G dongle… back to the conference.
I missed a chunk of this session getting the previous post out, so here is what is left of this one…
A nice talk about Google Maps, and how you can use it with your site. The Sopranos, in the final series, had a site with a map of real New York with pins. If you clicked on the pin you would be shown an video clip with what happened in that episode at that place.
Googlebot understands KML links, which are used in Google Maps. If you upload a KML file your result will be shown in the search results prepended with [kml] and you can click there and be taken to the Google Maps placeholder.
Tiago Robert Santos – Building your next business with Google Technologies
(Using Google App Engine and Google Web Toolkit)
The web is an application development platform. You can write once, run it in multiple platforms from PC, Mac, Linux, Smartphones et al.
Open Source Options give a low entry barrier.
The problem space for the development is on the client and server sides.
He is showing us a nice timeline of what the history of web development looks like, from HTML in 1990 and 1993 the Mosaic browser and the Apache webserver in 1994 over to Gmail in 2004 and Google App Engine in 2008.
Google App engine is:
- A cloud Computing Platform
- Run your Web Apps on Googles infrastructure
- Local development, easy deployment
- No Machines to configure or manage
- On-Demand capacity
- enterprise class reliability: 99.9%SLA
- Provides the COntainer and Services (PaaS)
- Hardware, connectivity
- operating system
App Engine supports both Java and Python.
The system functions a lot like a Linux system, you can set up something similar to a cron job to automate certain tasks.
The datastore is Scalable, built on top of bigtable. It is distributed, transactional, hierarchical, schema-less. It is not a relational database or an SQL engine.
Memcache is persistend, distributed in-memory cache. You can speed up dynamic database-driven websites by caching data and objects in RAM. This is Primarily intended for fast access to caced results of datastore queries.
Tons of other APIs available. Gmail, Gadget API, Google Accounts, Tasks Queue etc.
Pricing, you can buy as you go – scaling your outlay as you expand. The pricing is rather cheap – I wonder if the prices are in Rands or Dollahs…
Another infographic on screen about the life of a Google App Engine Request. Looks complicated.
Requests for Static Content is routed to the nearest Google Datacenter. Travels over Googles Network, Uses the same infrastructure other Google Products use.
Requests for Dynamic Content checkes for Cached runtime – if it existsm no initialization, if not – executes requests.
Summary Morning Session Two
Nice informative sessions. A bit hectic typing away. More after lunch. If you are here come say hi!