Pydap to Google Earth
Just saw this on Planet Python. Interesting.
2006 (old posts, page 17)
The REST Dialogues
The REST Dialogues looks like it will be good stuff, although The S stands for Simple is more fun. If I were more clever, I would have taken this approach to discussing W*S.
How to Explain Metadata
No, no, no. As as cook and programmer, I must object. The only analogy that works is ingredients == data. In this case metadata is the well known extra information on the packaging: serving size, nutritional guidance, organic or not, country of origin, year of production, appellation, etc.
Comments
Re: How to Explain Metadata
Author: Jason Birch
Hmm. You must have created a /. effect on that site... A recipe is metadata, documenting the data sources and methods which were used to derive the product :) JasonRe: How to Explain Metadata
Author: Sean
You're right, Jason. One man's data ...Re: How to Explain Metadata
Author: Jason Birch
...is another man's gnocchi?Java Freed
Java is free. Open source GIS folks should read and think about Bray's -- and he's not a Free Software zealot -- common sense arguments for the GPL.
Comments
Re: Java Freed
Author: Steven Citron-Pousty
Hey Sean: glad to see that you will now be using Java - I mean now that it is GPL'ed what complaint could be left ;)primagis.buildout
Kai and I attended Chris McDonough's talk on buildouts -- in general, not specifically about zc.buildout -- at the Plone Conference. I left motivated to start thinking about buildouts. Kai left motivated to do something about common environments for developing PrimaGIS. His solution is primagis.buildout: a script for making complete, isolated PrimaGIS sandboxes. So far, it's only tested on OS X, Debian, and Ubuntu, but already has several new users up and running. For windows, I imagine we could use more binaries, and try to tap into Howard Butler's build kit for PCL-GDAL and PCL-MapServer.
Tests for Goodness' Sake
You know what would impress more than a 1.0 tag? Tests. If it's worth building, it's worth testing. Testing Python is easy, even fun. Ask me how.
Best regards,
The Test Nut
Update: TileCache's HACKING file does serve as a doctest file.
Comments
Re: Tests for Goodness' Sake
Author: Christopher Schmidt
cd tilecache-1.0 python >>> import doctest >>> doctest.testfile("HACKING") UserWarning() (0,20) >>> TileCache has tests, they're just not obvious. I don't think they need to be (although I believe Schuyler disagrees): So long as I consider it my personal responsibility to maintain the TileCache code, it's my responsibility to make sure it does what it says on the tin, which means the tests are important to me, and not to you. I'd love to hear why you feel otherwise.Re: Tests for Goodness' Sake
Author: Sean
I simply didn't find your tests. You've got a well-earned reputation for making cool software, but I think you'll find that adding obvious tests will get you even more attention and contributions from savvy users.Re: Tests for Goodness' Sake
Author: Christopher Schmidt
But you still haven't answered the question: "Why do you need my tests?" OpenLayers has public tests: 905+ of them. They're runnable on the web (http://openlayers.org/dev/tests/run-tests.html), they're using an Open Source testing framework, and they're required for inclusion of new features into the codebase. OpenLayers is an Open Development project. TileCache is an *open source* project, but not an Open Development one. Given a good patch, it would likely be integrated, but (unless I'm sorely mistaken as to the niche this fills), it will never be an Open Development project. There is no public SVN repository. There are no non-MC committers. I can't imagine there ever being a reason for there to be: it's 500 lines of code. There's very little reason to bother committing a patch back, because there's so little to patch that wouldn't mean adding a layer of complexity or four. Tests for open development projects are absolutely essential. Tests for Open Source projects which are not open development don't neccesarily help anyone. If I'm going to check your patch, then it's my responsibility to make sure it doesn't break things, not yours. If I want you to make sure it works, then I've got to think about switching models to be something other than Open Source closed Development. Can you point to some change you might think about making where tests are helpful in TileCache?Re: Tests for Goodness' Sake
Author: Sean
Why do *I* need your tests? Tests are a sign (to me) that the code could be worth my time to evaluate. That's all. OpenLayers' tests were a very positive sign.Re: Tests for Goodness' Sake
Author: Schuyler Erle
Christopher, speaking as your collaborator and admirer, I'll reiterate why I insisted we come up with some kind of unit tests for TileCache: They're useful to *me*. If I break something, I want to know about it. ;-)Web Mapping and Accessibility
Web mappers: what are you doing about accessibility?
Me? I'm using Plone for a solid start. Pleiades is using OpenLayers and Google Earth to visually display georeferenced content; neither of these are fully accessible. It's a hard problem, for sure. One of our technical observers is a CompSci specialist in accessible UIs. He sees the problem as rich and chewy, which means it won't be solved any time soon. Last time we met he had some interesting ideas, and I hope to be able to write about realizations of them next year. Pleiades' feed-based, "viewer"-independent, architecture should help.
Python Cartographic Library 0.11
This release has been a long time coming, and I appreciate the patience I've been shown. At least the wait was less than 18 months. Here are the source release downloads: PCL-0.11.0.tar.gz [MD5, SHA1].
PCL 0.11 includes contributions from Kai Lautaportti, Michael Kerrin, Con Hennessy, Howard Butler, Ludwig Brinckman, Josh Livni, Sally Kleinfeldt, Aleksi Korvenranta, and Chris Calloway.
Comments
Re: Python Cartographic Library 0.11
Author: James Fee
Hey don't kill the messenger.Re: Python Cartographic Library 0.11
Author: Sean
Sorry, James. I've changed the link to the original entry.IronPython Goodies
Responding to my previous .NET post, Seo Sanghyeon wrote to tell me that ElementTree is already ported to -- if I understand correctly -- the IronPython Community Edition, and work on PIL is underway. As I said before, these have always been two of my top 10 favorite modules, standard or not. I guess working on .NET won't be so bad after all.
Rick Santorum (R-AccuWeather) Defeated
Peter Suber points out that users of rich NWS data feeds won big last night. The sponsor of S.786 has been sent home. Suber says it was a good night overall for open access to research data. FRPAA co-sponsor Joe Lieberman was re-elected and pro-OA Sherrod Brown replaces Mike DeWine. Are there any seasoned industry watchers who want to make an assessment of the impact on the geospatial business?