[HN Gopher] Notes on the History of the Map Tile
___________________________________________________________________
Notes on the History of the Map Tile
Author : altilunium
Score : 42 points
Date : 2025-06-15 09:14 UTC (13 hours ago)
(HTM) web link (placing.technology)
(TXT) w3m dump (placing.technology)
| masfuerte wrote:
| I don't understand. There were loads of online maps before Google
| maps and they all used tiles. How else would you do it? What
| Google added was smooth panning between tiles, pretty much as
| soon as native browser technology was up to the job. If they
| hadn't someone else would have.
| 0110101001 wrote:
| > How else would you do it?
|
| Render a viewport, given an API like
| mining/maxing/minlat/maxlat.
| masfuerte wrote:
| Fair enough, but these were solutions that worked without js,
| and they weren't dynamically rendering maps on the front or
| back end. They were just showing squares of pre-rendered
| bitmap, and the square boundaries were fixed. If your point
| of interest was near an edge it could be quite annoying, like
| trying to navigate somewhere in the gutter of a paper atlas.
|
| Even if they'd had an API that took a viewport, the result
| would have been stitched together from bitmap tiles because
| that's what they had.
|
| It seems like the "invention" of tiles for maps must have
| happened as soon as anyone starting using a computer to
| render maps to bitmaps. The Ordnance Survey wouldn't at any
| point have rendered the entire UK to a single bitmap (at
| least not a map with any detail). It would have always been
| tiled.
|
| Edited to add: Actually, the invention was much earlier than
| that. Paper maps were tiled before computers were a thing.
| And this would naturally have carried over to computer-
| rendered maps.
| thrance wrote:
| Yes, to me it's the canonical way to represent maps on a
| computer, that anyone could come up with after spending a bit
| of time pondering the question. And it looks rather
| straightforward to implement, probably a bit less so with
| ancient browser tech.
| JKCalhoun wrote:
| Some of these sure look a lot like mipmaps.
| jbuzbee wrote:
| I worked on a system at Martin Marietta in the late 80's and
| early 90's where we created tiled maps for use by the US Army. We
| had a large scanner we'd use to scan their maps, then we'd
| georectify the scan and slice the result up into tiles of 128x128
| pixels which would be compressed before storing to a whopping 360
| Meg hard drive. I participated in a number of Army field
| exercises in the US and Europe where we'd show the digital maps
| and graphic overlays off to troops who were using paper maps with
| little paper icons they move around to reflect the current
| situation. Our capability never went anywhere because Management
| wasn't really into map-maping and the Defense Mapping Agency
| started doing it themselves, distributing their maps on CD.
| croisillon wrote:
| i thought there would be a line or two about Terravision
| https://en.wikipedia.org/wiki/Terravision_(computer_program)
| zeckalpha wrote:
| Another term for theses is "raster pyramids". Here's an example
| from 1993: https://www.usgs.gov/publications/pyramid-system-
| multiscale-...
| wduquette wrote:
| I worked on supercomputer algorithms to render planetary terrain
| data (image plus digital elevation) using tiling back in the
| early 90's. I'm not sure where my co-worker got the idea, but it
| seemed like an obvious thing to do.
___________________________________________________________________
(page generated 2025-06-15 23:01 UTC)