Andrea Rosso has posted some amazing details about further SVG development on ArcWeb Services. Due to their REST API, ArcWeb Services already offer some basic SVG output.
Apparently ESRI is just about to enhance ArcWeb’s SVG capabilities and release sort of a SVG ArcWeb Explorer, similar to ArcWeb Explorer based on Flash. Unlike the existing Flash version, you’ll be able to download the whole viewer including all sources and customize it in order to fit your needs. Since JavaScript plays an important role when it comes to interactivity and SVG, he points out that it’ll be quite similarly to a JavaScript API. So it should be fairly straightforward to embed ArcWeb SVG into your applications or create your own mashups.
Another highly welcome detail is that he mentions Firefox’s built-in SVG support in his post. As well the screenshot on his site shows Firefox demonstrating ArcWeb SVG Viewer. If ArcWeb SVG works well in Firefox I assume that it will be truly platform independent, something one could hardly achieve while Adobe’s SVG Viewer was the only serious viewer for SVG (cf. DOM implementations on Win/Mac/Linux browsers for instance).
If you got interested by now in what’s going on with ArcWeb and SVG you should keep your eyes open for ArcWeb Labs (to be opened soon). According to Andrea ArcWeb Labs should give some insights and information on unreleased applications.
ArcWeb Services providing an advanced and easy to use SVG interface means a significant progress for the acceptance of SVG as webmapping technology. You may look forward how this will work out.
Update #1:
Head over to Kerry Coffin’s post to get some more information about ArcWeb’s SVG Viewer (e.g. available widgets, perfomance in Firefox vs. Adobe SVG Plugin) and see a few very promising screenshots!
Silently Google added Google Mars to its mapping services. Apart from providing 3 imagery types they offer plenty of information about the surface and Mars missions.
Brian Timothy pointed in his comment on my “Flash vs. SVG”-post to the website of the US Department of Agriculture (USDA) containing maps done completely in SVG. Apart from the “political matter” he mentioned I would like to stress the potential of SVG for mapping statistics.
The USDA maps let the user change and apply various styles immediately without reloading the whole map. Following the same method you could allow users to change thresholds too for instance. Since every single attribute is stored within or linked to the according map object it’s easy to modify the map presentation by JavaScript and XML/SVG/DOM on the client side. As you can read in this short review by Jeff Thurston the application appears fast and with a snappy user interface.
If you’re interested in webmapping using SVG you’ll find below a few well done examples:
In addition you can find my first SVG mapping experiment here. I stopped development 2 years ago. Neither will it work in Firefox 1.5 (compelled to optimize it for Adobes plugin then, *the* showstopper for SVG) nor is it fully functional because as backend I used a PostgreSQL/PostGIS database which I don’t have available online.
The latest Google Maps API update (to version 2.41) allows to display an overview map to the main Google Map by using the GOverviewMapControl. Read the Google Maps API Blog post or head directly over to the Google-Groups discussion for more information on that update.
Reading the word “overview map” I rembered a discussion we had lately about that issue. Almost automatically, without having a second thought, we put an overview map onto our webmapping application where it is now for more than a year now. There is a big chance that you’ll find in 80% of all webmapping tools those tiny overview maps. I simply was wondering if those stamp-sized pictures mean such an indispensable benefit to users. Background of the entire discussion was to find ways to optimize the user interface, to free some more space on the screen for additional information or enlarging the main map.
An overview map does make sense when a large scale map is presented as entrance point, to give an idea where the viewed area is located. Starting with a small scale map, as we do in our application, and showing the whole explorable area, an overview map isn’t an essential element at this point.
As well I consider overview maps as usefull if the time span between user interaction and map delivery is getting to long (due to server perfomances, bandwith bottlenecks, etc.), in particular while zooming and panning. Disorientation of the user could be the result. In times of AJAX and vector graphic driven user interfaces it shouldn’t be a problem to keep that time span short in order to develop a snappy application and in the end achieve a great user experience. Almost every new mapping product of 2005 showed excellent navigational-usablity. Enhancing the overview map like Yahoo! did, interaction on the overview map pans the main map simultaneously, makes navigation through a large area extremely convenient. Once the small overview map is not any longer needed and would only cover parts the main map you can close it. Since screen dimensions are limited and the focus should be put on the map, I would consider the possibility to hide unneeded elements as an essential feature.
In terms of zoom and pan work-flow I always appreciated back and forth buttons. If I get lost I follow my trace back until I know where I am. Just like hitting ctrl-z, the most used shortcut ever on my keyboard.
Jacktracker looks like a fun travel guide for 24 fans. It tracks all movements of Jack Bauer and his followers around LA in a map.
Recently I came across the mapping tool called Social Explorer (via Cartography). It’s based on Flash technology and offers exploration of U.S. census data.
Since more than a year now we are working on a similar application for Austria. It’ll let users explore Austrian and European census data along with other information relevant to spatial development. Within the last year I could observe the focus of development shifting from technology driven discussions (e.g. offering advanced GIS functionality) to cartographic *correct* geodata and finally emphasising on user-needs. After the event people often have the wonderful gift of hindsight, so now I know that we should have put more weight on initial user-requirements discussions. Technology is a fascinating thing and we all were caught by visions of tools we would like to integrate.
A noticeable aspect of the Social Explorer is its simplicity. The first screen of our application contains more or less the same elements. We start with pre-defined maps, the user can switch between various topics and load different maps, we have of course zooming & panning, something similiar to that slide-show tool and the possibility to create reports containing tables and diagrams. But somehow the Social Explorer interface appears more clearly and easier. At a glance you know what you can do and what this tool has to offer, in our application it’s still more trial and error. A colleague hit the nail on the head by mentioning “It’s the American approach, keep it simple. We got the European approach, to make it more complicated than it actually is.”.
During the next months we’ll concentrate more on interface design, clean it up while keeping the same functionality to achieve a higher deegree of usability. In the end, usabality (beside marketing) is crucial for a popular application.
If you’re interested in interface design and mapping/cartographic applications you should definitely have a look at the Atlas of Switzerland, an amazing cartographic product.
Every time thinking about usability, interface design, etc. I recall an interview about Apple’s one button mouse, which they used to produce until last year. Asked why they still put only one button on their mice while other PC-mice offer I don’t know how many buttons and wheels the Apple employee answered: “Considering that Apple users have only one button available, Apple developers are forced to keep the interface of their applications simple in order that users can operate them with only one button.”. A way to guarantee Apple’s hallmark, the ergonomically designed user interface.
Since when is Ask.com providing maps? Did I miss something? Not only they are catching up to Google, Microsoft, Yahoo!, ESRI, etc. but also they offer some detailed street maps of Europe. Quite exciting for us here in Europe!
The whole application is very similar to other (familiar) mapping products. Based on image tiles (no Flash so far), with zooming, panning and search options as usual.
Basically the gazetteer seems to work for european places. Well, basically. I wasn’t able to locate Vienna (“Wien”), the austrian capital, instead it suggested me almost every european one-horse-town containing the word “Wien”. But, after all, I achieved to find out the walking distance from “Waidhofen an der Ybbs” to “Sankt Pölten”. Switching between walking and driving gives you different routes, as it should be.
What’s missing, or at least it was hiding from me, is an API or a documentation respectively. I’m not sure if Ask.com is working on an API or if they simply don’t want you to take advanage of their maps and embed them within other websites.
However, good to see a more detailed Europe appearing in those mapping products. Since our INSPIRE will probably end up as NMA-coopartion to avoid publicly available geodata we will partly depend on the goodwill of those companies.
Update #1:
The Cartography weblog and All Points Blog already reported yesterday about it. Check those sites for some more information on Ask.com maps.
The german IT news site Heise reported that Google is about to release a Version of their AdSense program that refers ads to coordinates instead of keywords. Thus map featured web sites could place ads related to the location the user is currently viewing.
Some years ago SVG was considered as the next generation of rich media and animation in the internet, as a sort of Flash killer. Nothing happened. Flash technology enhanced and is still widely spread. SVG enhanced too and even went mobile but it’s still a good kept secret.
Now, ESRI & Yahoo! are implementing their webmapping services based on Flash technology. Last week Google silently enabled SVG in their latest API update (API v2.37) for Google Local.
Honestly, it’s quite a time since I did my last Flash development but back then (it was version 5 I guess) I never would have choosen Flash as a tool to be used in a webmapping application. Actually the decision for a vector graphics based webmapping application was then in favour of SVG. Flash’s object handling and ActionScript was too unfunctional for my purpose: I wanted to load geographic objects on demand (e.g. on changing view extents) out of a PostgreSQL/PostGIS database and map them in a web browser, without continuously reloading the whole map of course. SVG was the easiest way to go. As a XML based vector graphic format it allows you to edit, put attributes and styles as you want to every single object within your map. On the other hand vector graphics as SVG required more potential clients compared to raster pictures delievered by mapservers. All objects were handled – requesting & rendering – on the client side (JavaScript was used to access and parse XML, something like “pre AJAX”, to name a buzzword) while the user was interacting with the map. However, after this work I became a big fan of SVG, especially when it comes to webmapping. If there wasn’t this plugin-dependency, which resulted in every single discussion to be the main showstopper for SVG. With native SVG support in a popular browser like Firefox hopefully things start to change.
Back to Google Local: in SVG compatible browsers (such as Firefox 1.5) one can now use SVG rather than PNGs to draw polylines if following parameters are set:
_mSvgEnabled = true/false … to enable/disable SVG in SVG compatible browsers
_mSvgForced = true/false … to force the use of SVG in every browser (even if not capable of displaying SVG correctly) or limit SVG to compatible browsers.
If SVG is not “forced” Internet Explorer 5.5+ will still render polylines with VML (something like a rudimentary SVG…).
This simple comparsion gives an idea what can be done at what perfomance with SVG.
“It draws a random polyline with 100 segments. You can change the
number. There are “Redraw with SVG” and “Redraw without svg” buttons.”
SVG is a well elaborated standard. Maybe Google Local is the push that this technology needs.
Another interesting point on the Flash vs. SVG discussion is the take-over of Macromedia by Adobe. Now we have both technologies under one umbrella. Why not merging them together? Or at least bundle them to one single plugin. Since XML support is already built into Flash, why not extend it to SVG?
A few days ago I discussed with a colleague ways to publish maps, analysis results or to provide analysis tasks within an organisation. We were talking about a quite common corporate geomarketing challenge (I suppose, since we’re no geomarketing experts): a specialised department analyses spatially distributed marketing data and provides other departments with their results. Whereas the presentation of results should be somewhere in between PowerPoint slides and a full functional desktop GIS client. Honestly it was quite hard to seperate various publishing options, identify the optimal tasks and features and combine them to the best individual publishing solution. Along with recent developments on server-side GIS software, powerful new generation geo viewers like the popular Google Earth or the upcoming ArcGIS Explorer offer content providers quite an alternative to well-established map publishing solutions like mapservers do for instance.
So today I tried to sum up a few map publishing solutions I know and figure out their pros and cons, their qualities. Just to clarify: this list is not exhaustive. It’s an overview and the result of a short brainstorming to have a note for further discussions on that topic in my pocket. Comments or suggestions will be highly appreciated!
ArcGIS Publisher & ArcReader
Pros:
- Out of the box viewer
- Local map viewer
- Optional 3D (ArcGlobe)
- “Familiar” GIS application interface
- Advanced map layouts
- Data encryption
- Viewer customisation
Cons:
- Viewer installation
- Local map file
- Limited base maps (e.g. provided by geography network)
- Extension for map authoring needed
- Limited viewer functionality
Google Earth, ArcGIS Explorer (upcoming)
Pros:
- Out of the box viewer
- 3D
- already integrated access to base maps
- OGC Webservice support (upcoming in GE?)
- Local data support (KML, Shp, PGDB)
- Customised tasks (AGX)
- User acceptance (GE)
Cons:
- Viewer installation
- 3D only
- Mainly data viewer
- Limited map layout authoring
- Internet/Intranet access required or recommended
- Ads (upcoming in GE)
Webmapserver e.g. Mapserver, ArcIMS, deegree,…
Pros:
- Out of the box browser interface
- Platform independent
- Customisable user interface and functionality
- OGC Webservice support (embed or provide)
- Access to base maps by third party API (Google local, ArcWeb, Yahoo!, etc.)
Cons:
- Browser interface
- Last & performance
- Development effort
ArcGIS Server, GRASS, GDAL, etc.
Pros:
- Browser interface
- Platform independent
- Customisable user interface and functionality
- Customised advanced tasks
Cons:
- Browser interface
- Last & performance
- Development effort