colophon
Thursday, 08 May 2008 08:09 Hrs
seldomlogical.com / articles / colophon
Summary Seldomlogical is built using some simple ideas in mind. Build tools to give me control over my own data. The technology changes, the data does not
Rationale
Reference
[0] I have been thinking and hacking on this problem for a while. You can read more about nextgen here http://flickr.com/photos/bootload/sets/72157600015450188/
[1] You can read more about my Python hacking here http://flickr.com/photos/bootload/tags/python
[2] Using webpy as a development server was a no brainer. Googles appengine is based on webpy and webpy has less moving parts than Django. You can read more about my webpy hacking here http://flickr.com/photos/bootload/tags/webpy
[3] Google appengine has just changed the face of web hosting. Consider the costs. At the moment it is free but shaped. Now if I was to get a host in Australia that could a) have a python account, b) work with webpy it would be in the order of AUD$500 to $1000 dollars per year. Domain name and email hosting another $100.
[4] The Atom 1 feeds are valid. I am also adding meta-data into the feeds. For example geospatial information in the form of geo.lat, geo.lon per Atom.item. The intention is to add useful meta information for general consumption. More so than the marked up pages. The data is more important than the presentation.
[5] The templates are valid XHTML transitional.
[6] maxdesign: CSS Centering - fun for all!, http://www.maxdesign.com.au/presentation/center/
[7] In the article Bret explains how the appengine is based on Webpy.
[8] Teh, the minimal blog is written by Pradeep Gowd using Google appengine technology. The source code can be found at http://github.com/btbytes/teh/tree/master
I have previously described how I named Seldomlogical. I'm going to quickly discuss the technology behind it. The codename for this project is nextgen [0]. It has a few simple aims:
- Data is more important than the technology
- Let me save all that data I enter into the web
- Allow raw data to be exported as feeds
- Give me back control of my own data
- Where possible use my own technology
So while the technology may have changed the data remains in re-usable formats that I control. Enough of the ideas what about the technology?
TechnologyThe latest incarnation of nextgen is built using:
- Python [1]
- Webpy [2]
- Google App engine [3]
- data is stored using Datastore api
- admin authentication using google Users api
- Feeds are valid Atom 1.0 [4]
- Feed syndication is via Feedburner
- Templates are XHTML Transitional [5]
- The valid css is homegrown, based on this article [6]
- Images are from my flickr site
- The sitename and byline are discussed here
- Inspired by Bret Taylors appengine blog [7]
Reference
[0] I have been thinking and hacking on this problem for a while. You can read more about nextgen here http://flickr.com/photos/bootload/sets/72157600015450188/
[1] You can read more about my Python hacking here http://flickr.com/photos/bootload/tags/python
[2] Using webpy as a development server was a no brainer. Googles appengine is based on webpy and webpy has less moving parts than Django. You can read more about my webpy hacking here http://flickr.com/photos/bootload/tags/webpy
[3] Google appengine has just changed the face of web hosting. Consider the costs. At the moment it is free but shaped. Now if I was to get a host in Australia that could a) have a python account, b) work with webpy it would be in the order of AUD$500 to $1000 dollars per year. Domain name and email hosting another $100.
[4] The Atom 1 feeds are valid. I am also adding meta-data into the feeds. For example geospatial information in the form of geo.lat, geo.lon per Atom.item. The intention is to add useful meta information for general consumption. More so than the marked up pages. The data is more important than the presentation.
[5] The templates are valid XHTML transitional.
[6] maxdesign: CSS Centering - fun for all!, http://www.maxdesign.com.au/presentation/center/
[7] In the article Bret explains how the appengine is based on Webpy.
[8] Teh, the minimal blog is written by Pradeep Gowd using Google appengine technology. The source code can be found at http://github.com/btbytes/teh/tree/master

