Silly Patents

If you word things right, even pre-existing concepts can be re-packaged as bright original ideas. Google has filed a patent on an indexing method that is nothing but a specialization of a quad-tree, and a packing of level/row/column information into a 64-bit address space.

The specialization of quad-tree is to always use powers-of-two: cut your parents into four identical children; ensure the children are at a scale exactly half that of the parents. This yields nice behavior in the row/column, you can traverse to the parent row/column of a cell just be dividing the current row/column by two.

The packing of the information into a cell id is not completely clearly explained, because there is some talk of compacting and stop bits, to fit 31 levels, but even without compacting a 64-bit space can hold 29 levels quite easily (5 bits of level information, 29 bits of row address, 29 bits of column address).

I am a bit incredulous at the implied assertion that no one thought of chopping up the map of the world in descending powers-of-two before. The specific claims about the packing method might indeed be “original” but in such a trivial way as to be unimportant.

The whole process of software patenting reminds me of the historical acts of enclosure in England.

They hang the man, and flog the woman,
That steals the goose from off the common;
But let the greater villain loose,
That steals the common from the goose.

– Oliver Goldsmith

Meeting of the Tribes

Wow, what a great synthesis of larger FOSS4G philosophies and culture! Danny de Vries absolutely nails it, on a whole bunch of topics, including interoperability:

What makes open-source so different from a corporate system like ESRI is its fundamental interest in building software according to universal standards. This in contrast to the strategic interest of any closed, corporate system to somehow make users reliant on their system alone.

Melting Snow

Note: In the spirit of my own criticism, I won’t delete this post, but apparently the problem at Directions is technical (lost comments) not political (deleted comments) so I am completely out-to-lunch and offer apologies to Directions! Fool == Me. Ignore the below.*

The Directions Magazine news item on which the owner of Blue Marble Geographics was caught posing as a Blue Marble customer in the comments section has been purged of its comments. The internet is such a fickle place… down the memory hole we go!

FOSS4G: Was it all a Dream?

Just a week ago, I was addressing the opening plenary session at FOSS4G! It all feels like a blur now, and it is very hard to remember anything distinctly.

Ironically, for all the talking and speaking I did, the things I remember most clearly are the few talks I had time to get into, and listen quietly, on pgRouting, and the new spatial ETL tool from Camptocamp. Perhaps because these are the two talks where I learned something new and interesting. The rest was just reciting things I already knew, which is interesting perhaps (I hope) for those on the receiving end, but less stimulating for me.

Level Two

I walked through the conference centre on the way home from work yesterday… it was shut tight, sadly quiet, and geek-free.

New Tenant

Meanwhile, the second geospatial conference in as many weeks ran today. Make of it what you will. :)

Snow Job in September

In response to Adena Schutzberg’s update on the acquisition and release of the Mentor Software CS-MAP reprojection library as open source, Ted Florence, of Avenza Systems writes:

While we (Avenza) did for a time use Safe’s FME Objects (which uses CS-Maps) in one of our products we have recently pulled it completely in favour of the Blue Marble file format and co-ordinate conversion products (GeoTranslate and GeoCalc). The major reason for this was our dis-satisfaction with the CS-Maps co-ordinate engine which underlies FME.

In investigating and researching our needs, we found that opposite to your comments, most organizations are actually using Blue Marble and not Mentor and that more and more have been leaving Mentor over the last few years. Almost every major oil company and defense contractor I can think of uses Blue Marble.

What he does not write is that Avenza Systems is not his only affiliation. The corporate register of the State of Maine indicates that Florence is also the President and sole Director of Blue Marble Geographics. What an interesting coincidence! Perhaps like Victor Kiam, he liked Blue Marble’s reprojection library so much be bought the company! Regardless, he should declare his interest when making comments like the one he put on Adena’s article.