Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

The State of Mapping APIs, 5 Years On

timothy posted more than 4 years ago | from the where-are-you-now dept.

Earth 100

macslocum writes "Map APIs took off in 2005, and during the ensuing years the whole notion of maps has changed. Where once they were slick add-ons, map functionality is now a necessary — and expected — tool. In this piece, Adam DuVander looks at the current state of mapping and he explains how mobile devices, third-party services and ease of use are shaping the map development world."

Sorry! There are no comments related to the filter you selected.

Google the first? Not really... (4, Informative)

RobertM1968 (951074) | more than 4 years ago | (#33504318)

"Google had the first mapping API and continues to keep its lead by adding useful new features. "

I suspect the government and the various contractors and outside programmers who worked with them or with their data had the first mapping API. It was used for the TigerLINE data that all the rest of this data is originally based off of. It wasnt some internal govt only project, but something anyone could either download (free) or purchase and use. There are a variety of such tools (various with their own APIs), numerous ancient (Win95/98 era) which far predates Google's tools and APIs.

Re:Google the first? Not really... (5, Informative)

Anonymous Coward | more than 4 years ago | (#33504368)

Obligatory: I work at ESRI and find it hilarious that we're not mentioned in that article apart from the related video.

http://www.esri.com/software/mapping_for_everyone/api/index.html

http://en.wikipedia.org/wiki/Esri

Re:Google the first? Not really... (3, Informative)

Anonymous Coward | more than 4 years ago | (#33504648)

Yes, ESRI is the leading force in the cartographic industry. But outside of the industry 99% of people have never heard of ArcMap or anything else you make. They might see it in the end result if the local government uses your JavaScript/Flash/Silverlight API to cook up something, but Google's the big name in the room as they put something user-friendly on the web before you guys. Yeah, you still had the professional-grade software suite back then, but only the professionals used it then and only they use it now.

By the way, the dev summit back in March was fun (nice weather out there), though I wish I had thought to ask a few more questions/go to specific seminars then. Turns out I missed everything related to a project that came up two months later. Yeah, all of the videos are online but they still miss most of the Q&A afterwards.

Re:Google the first? Not really... (0)

Anonymous Coward | more than 4 years ago | (#33505586)

But outside of the industry 99% of people have never heard of ArcMap or anything else you make

He's the Silent Cartographer!

Re:Google the first? Not really... (0)

Anonymous Coward | more than 4 years ago | (#33504664)

Obligatory: I work at ESRI and find it hilarious that we're not mentioned in that article apart from the related video.

http://www.esri.com/software/mapping_for_everyone/api/index.html

http://en.wikipedia.org/wiki/Esri

Oh I know who this is...

Re:Google the first? Not really... (1)

batkiwi (137781) | more than 4 years ago | (#33504666)

Have you ever tried to you ESRI mapping APIs though? A few THOUSAND poorly documented COM+ components (and 1:1 wrappers if you want to use dotnet) does not an API make.

Re:Google the first? Not really... (0)

Anonymous Coward | more than 4 years ago | (#33505242)

Yes, ArcObjects is completely obtuse and the documentation isn't great. I'm sorry you haven't looked at the Esri software suite for the last three years and seen all the newer APIs.

For the rest of the Esri ecosystem that's moved to the various Web APIs like Javascript, Flex, Silverlight things are quite a bit better.

(not the original AC..)

Re:Google the first? Not really... (1)

batkiwi (137781) | more than 4 years ago | (#33505546)

You're 100% right, the last I looked at esri was almost exactly 3 years ago. At the time we moved away because our local ESRI prof svcs guy essentially told us web apis/soap/etc were not in the pipeline.

Re:Google the first? Not really... (1)

GuidedByVoices (1421045) | more than 4 years ago | (#33507056)

Sadly; some of us are stuck using ArcEngine 9.2 for gov't projects; we don't have the benefit of the recent developments and actually intelligible documentation. Oddly, I work a mere stones throw from the ESRI offices in Ottawa; I should raid their offices for free documentation.

Re:Google the first? Not really... (1)

DarthVain (724186) | more than 4 years ago | (#33507994)

Also stuck with 9.2. Try importing a 9.3 toolbox for fun or even just a 9.3 Python script. Next step, claw eyes out and bash head against keyboard for 2 or 3 days.

Re:Google the first? Not really... (1)

GuidedByVoices (1421045) | more than 4 years ago | (#33510044)

Try uninstalling 9.2 and then installing 9.3 :P 9.2 breaks itself somewhere along the line, making it impossible to entirely uninstall it. Scouring the registry for entries doesn't help either. Im sure there is a workaround, but so far it's easier to just reimage your Windows box :/

Re:Google the first? Not really... (1)

DarthVain (724186) | more than 4 years ago | (#33512412)

Ha, I almost tried to get an upgrade to 9.3 maybe I am glad I didn't now. Perhaps just wait until I get a new work computer and do it then fresh, sounds safer that way.

Re:Google the first? Not really... (0)

Anonymous Coward | more than 4 years ago | (#33505274)

As a programmer who used ESRI MapObjects 2.0a with ESRI IMS (Internet Map Server) 2.0, yes. I wrote them in VB6 in 1998 and they're still ticking today. 12 years ago I had street centerlines, cadastral data, and 1995 DOQQ images working pretty well. The documentation was more than adequate. My only gripe is that the products don't work under Vista or Windows 7. Fortuantely, MapWinGIS has open source controls that are (mostly) capable of the same thing.

The only thing I can credit google for is the use of javascript to allow panning and zooming with seamless refreshes in the background. That and they have gobs of storage space to house imagery for the whole world.

Re:Google the first? Not really... (1)

Dr.Dubious DDQ (11968) | more than 4 years ago | (#33509542)

I'm not personally heavily into GIS except as a (very) casual hobby, but I get the impression from my Geologist Significant Other that ESRI started stagnating pretty badly when they went "Windows Only".

Re:Google the first? Not really... (3, Informative)

humblecoder (472099) | more than 4 years ago | (#33504758)

Seems like the article focuses on the interface layer (hence the name "The State of Mapping API's). However, I think the hard work is the collection of the underlying map data. One of the more interesting projects is the OpenStreetMap project where map data is crowdsourced.

Re:Google the first? Not really... (1)

DarthVain (724186) | more than 4 years ago | (#33508006)

Considering Google Maps/Google Earth got 99% of their data from someone else using the afore mentioned software...

Re:Google the first? Not really... (5, Insightful)

symbolset (646467) | more than 4 years ago | (#33504764)

You guys make some neat stuff, but the license keeps it away from us.

Re:Google the first? Not really... (4, Interesting)

Anonymous Coward | more than 4 years ago | (#33504778)

Wow. You actually admit that wou work for ESRI? Like that is something to be proud of?! Oh wait, you did that anonymously. I certainly wouldn't want to be associated with that steaming pile of "software".

The only reason you guys are anywhere is due to the way that you've muscled your way into government/state use and thereby forcing anyone wanting to do contract work to use your buggy, ridiculously over priced applications. Seriously, $7,000 for GIS editing software??? Then tens of thousands more for shitty, outdated ArcSDE?

Yeah, I think there is a reason why ESRI wasn't mentioned and that is because your poorly documented APIs, lousy commitments / involvement in OGC (way to be an industry leader) and lack of any real innovation--nothing good has come out of Redlands in more than 5 years.

Re:Google the first? Not really... (1)

Animal Farm Pig (1600047) | more than 4 years ago | (#33504934)

You think Arc is "steaming pile of 'software'"? I've tried a few of the free alternatives (QGIS, MapWindow, Thuban), and, while ambitious projects, they don't come anywhere close to Arc. Can you suggest something better?

Re:Google the first? Not really... (3, Insightful)

grcumb (781340) | more than 4 years ago | (#33505080)

You think Arc is "steaming pile of 'software'"? I've tried a few of the free alternatives (QGIS, MapWindow, Thuban), and, while ambitious projects, they don't come anywhere close to Arc. Can you suggest something better?

MS Word is better and more feature-ful than the competition, but that doesn't mean it's not a steaming pile [imagicity.com] . The two categories are not mutually exclusive.

Re:Google the first? Not really... (1)

DarthVain (724186) | more than 4 years ago | (#33507758)

Agreed. It IS a steaming pile of crap. It just happens to be the BEST steaming pile of crap currently available. I work with it, and have for 10 years. It is amazing the stuff that is wrong with it that never gets fixed. Or the stuff than breaks when they do actually fix or change something.

For instance just last week my pet project / obsession was trying to permanently sort (ie so as soon as you close the attribute table your working on it doesn't revert back to the original sort order) a simple table. After doing some searching in ArcScripts and google, there are scripts going back to freaking ArcView to do it, yet the functionality does not exist in ArcMap. After finding two python scripts, I try to use these. However they are compatible with 9.3 and not 9.2, so not only will it not import and I have to re-build the entire data window and map parameters to variables within a script I didn't write, but once I do it still doesn't work because the actual python code compatibility has changed from 9.2 to 9.3, so I would have to actually re-write someone else's code using new looping structures, etc...

To tie this all in to the article I was trying to use ESRI's 3D Analyst fuction to Export to KML (for Google Earth), in which you need to heavily edit your data first, as the exporter just spits out what it sees and will not let you customize on the fly (and the need to do this whole procedure if your data changes). Hence the need to order it, unless you want it listed in Google Earth in some silly random way, making it impossible for the user to look up a value in the list.

In the end I thought about upgrading to 9.3 simply to save myself the trouble as it was getting ridiculous for the simple thing I was trying to do. Finally I found a independently made extension made by someone else that I could import and use. Is that really the functionality fix for ESRI? Seriously?

I'll not even get into the symbology problems (where you have to rebuild it every time your data changes), where the ESRI tech simply gave up and said, "Sorry it can't do that...".

Re:Google the first? Not really... (1)

DarthVain (724186) | more than 4 years ago | (#33507776)

Oh and they still use DBF table structures. Wah?

Re:Google the first? Not really... (1, Informative)

Anonymous Coward | more than 4 years ago | (#33515162)

Clueless users (or organizations with massive legacy problems) who still use shapefiles, a data format defined almost 20 years ago, still use DBF table structures. Esri has been trying to kill off shapefiles for quite a while now, but they're still heavily used as a data interchange format and simply by users who don't know any better.

Re:Google the first? Not really... (1)

tehcyder (746570) | more than 4 years ago | (#33509232)

If you want to use emacs or Wordperfect 5, then why not just do so? No-one is forcing you to use Word or Write or whatever piece of software. I don't see what the problem is.

And anyway, for most users, a WYSIWYG word processor is a lot easier to get to grips with than a command/text-based system. Sorry, but it's true.

Re:Google the first? Not really... (1)

Greyfox (87712) | more than 4 years ago | (#33505606)

Gaia seems OK but I don't know if it does everything Arc does either. And it looks like you need a license for advanced functionality.

Re:Google the first? Not really... (1)

Mendy (468439) | more than 4 years ago | (#33505950)

uDIG seems the best of the free ones, if a bit slow at times.

As an aside I'm curious to know if anyone has used Manifold [manifold.net] , the website does seem to claim it does everything you might possibly want for one low-low price :)

Re:Google the first? Not really... (1)

deniable (76198) | more than 4 years ago | (#33506192)

And you get a demo by buying the software and asking for a refund within 30 days. We need to see it in action before we jump on board.

Re:Google the first? Not really... (0)

Anonymous Coward | more than 4 years ago | (#33532320)

manifold is OK. Mileage will vary. For example, using it to serve raster data via wms and tiled does not work very well. If the data set that you are working with cannot fit into memory you will be sol. This pretty much precluded us from doing anything useful with it since we use open layers extensively with satellite imagery. For lines and polygons and other shape based data it is better. There are a lot of features. I find the general attitude of the organization to be a little neurotic and condescending to it's customers. The esri vs. Manifold debate has reached a nearly religious level.

Re:Google the first? Not really... (1)

mjwx (966435) | more than 4 years ago | (#33506230)

You think Arc is "steaming pile of 'software'"? I've tried a few of the free alternatives (QGIS, MapWindow, Thuban), and, while ambitious projects, they don't come anywhere close to Arc. Can you suggest something better?

The AC is right, ArcMAP, SDE, SERVER is a steaming pile of crap. Most GIS people would rather use MapInfo or ERDAS Imagine but the edict to use ESRI is handed down from on high. I used to work for a direct competitor to ESRI and our dinky little product outperformed ESRI's IMS and WMS up to twice as fast and reliable at one point but ESRI are incredibly hard to unseat from the minds of government/big mining decision makers.

Re:Google the first? Not really... (0)

Anonymous Coward | more than 4 years ago | (#33507028)

As a past user of ESRI products, I concur that they provide not much more than a steaming pile of "software". In my previous position I worked with an aerospace and defense company which made use of ESRI's stuff for mission critical mapping. Put more than 100 moving targets on a map and watch it choke. When asking ESRI for support they basically stated "reduce the number of targets on the map." Sure buddy, the mission critical surveillance system we were putting together should just ditch targets so you mapping library wont shit itself. We'll just ignore all of the targets we are picking up via AIS and RADAR.

As as aside (and this is outside the realm of normal GIS based application development), a couple of high horsepower GIS APIs can be found in the form of Gallium's InterMAPhics or Envitia's MapLink Pro. These things will do ridiculous amounts of moving targets on a moving map display and ask for more. They'll also do everything the ESRI Arc stuff will do, but do it waaaaaaaaaaay better. The only problem here is these guys want on the order of 100's of grands from you to license their product (which is merely a drop in the bucket for defense contracts that are going for hundreds of millions of dollars). Gallium and Envitia gladly provided working trial versions of their libraries and were more than happy to help with initial R&D based work we did by providing fully documented APIs. ESRI wouldn't even provide us with a trial version of their library. Their line was "It's only $7000, just buy it!"

Re:Google the first? Not really... (0)

Anonymous Coward | more than 4 years ago | (#33510870)

Note to military equipment makers: Be able to launch more than 100 objects and/or decoys, so you overwhelm the opponent's tracking software. Better is the ability to launch more than 256 objects.

Re:Google the first? Not really... (1)

GuidedByVoices (1421045) | more than 4 years ago | (#33507098)

Just 2 cents: Depending on the application, you can get away with quite a few alternatives. Tatouk isn't too bad, it's been in used for nearly a decade in one of our GIS applications. I've investigated SharpMap, which (if you're looking for a barebones mapping app in C#) is great for simple mapping applications/components. A few of our new applications have been using the Google API, which has gone over well for their purposes. I personally don't feel you can suggest a software which is "better" without qualifying what one is desiring to do. Arc does literally everything; though I'd likely guess most developers don't really need this. My issues with ArcGIS would likely be that it's proprietary; during my investigation of SharpMap, when I needed full colour Geotiff and raster support, I built it myself. There was quite a bit of user supplied documentation, and the freedom to open up what I needed should I run into my own road block. It may be rather infantile compared to Arc, but as a developer I much rather open source alternatives. Still, can't fault ESRI for dominating the professional market; no one else was really doing anything with GIS for a good portion of time.

Re:Google the first? Not really... (2, Informative)

garcia (6573) | more than 4 years ago | (#33505030)

I have mod points and I considered moderating your post but I couldn't decide to mark it down or up (it could easily work both ways) so I decided to respond instead:

Yes, ESRI shapefiles are prevalent in the public sector and the software is expensive and difficult to use. Thankfully you can now easily convert SHP files into KML and display a lot of the work done in the public sector for use everywhere else.

I have a small archive of stuff I have converted from SHP to KML in Minnesota available here [lazylightning.org] which are used mainly for geocachers wanting to know what areas are off-limits but as you can see there are MN county boundaries in there as well. Why the county boundaries? Because the company I worked at prior to my current place for employment (a publicly funded educational institution) couldn't afford actual GIS software so I had to use what was freely available to do our analysis of marketing ROI.

So, yeah, ESRI sucks and they need to die but at least there are options now available which are affordable (read: "free") for the general public to get involved and do their own work with what has been made available from their tax dollars being hard at work.

Re:Google the first? Not really... (1)

DarthVain (724186) | more than 4 years ago | (#33508028)

I have used some of the alternatives. They are not good. ESRI has problems I will agree. With how technology is going, if they don't get off their laurels and hike up their socks, they are going to be in real trouble over the next ten years I predict.

Re:Google the first? Not really... (2, Funny)

MrNonchalant (767683) | more than 4 years ago | (#33505172)

I work at ESRI and find it hilarious that we're not mentioned in that article

The rest of us also find this hilarious, but for different reasons.

Re:Google the first? Not really... (0)

Anonymous Coward | more than 4 years ago | (#33505792)

Obligatory: I work at ESRI and find it hilarious that we're not mentioned

This comment is insightful I guess but it also explains why you weren't mentioned. You were so far behind the web map revolution that the people involved in it didn't even realize you exist -- and somehow you guys either didn't see that as a problem or had no idea what was happening. ...and you know what? The world is probably a better place because of it.

Re:Google the first? Not really... (1)

mjwx (966435) | more than 4 years ago | (#33506204)

Obligatory: I work at ESRI and find it hilarious that we're not mentioned in that article apart from the related video.

Just to clue everyone in, ESRI is the Microsoft of GIS except MS products typically have less bugs (yes I am trying to deploy ArcSDE 10 at the moment but for some reason it wont create the damn repository and the SDE (esri_sde) service crashes every time we try to start it).

Grumble, grumble, yes ESRI have been at the whole web mapping (WMS - Web Mapping Service) thing a lot longer then Google.

Re:Google the first? Not really... (0)

Anonymous Coward | more than 4 years ago | (#33506858)

Jack!! Is that really you ?

Re:Google the first? Not really... (1)

DarthVain (724186) | more than 4 years ago | (#33507608)

Yes comparing a free online service to a professional editing tool where a single licence runs you 10k is a fair comparison.

Re:Google the first? Not really... (1)

Chasqui (601659) | more than 4 years ago | (#33554986)

Actually, its an interesting comparison. I use ESRI products and more and more I am shifting to using free (such as Google) web and non-web products to do some display / mapping work. Why? Because, for one, as you said, every ESRI Arc 9.3 seat costs a shitload. Secondly, I can achieve a substantially similar result in half the time while leaving the license open for someone else. Will it come to a point that free/lower cost software solutions will eliminate ESRI? Maybe. Should they feel threatened? Absolutely. Their user base is already being eroded and users disgruntled. I can't seem to find many fans. So is it a fair comparison? On the cost, no. On the features and use - absolutely. They are both complementary and competitors.

Re:Google the first? Not really... (1)

DarthVain (724186) | more than 4 years ago | (#33560438)

"On the features and use - absolutely."

Not sure what world you are living in. One has lots of features, the other has zero pretty much. Google simply display's data, there is really nothing you can do with it.

An apt analogy would be Adobe and Adobe Reader. Adobe Reader may be pretty good, and a copy of it may be installed on every persons computer in the whole world, and it is free.... but it ain't going to be hell of a lot of good without Adobe.

ESRI should be wary, but until Google starts expanding into the realm of professional GIS software (and I wish they would), I don't think they have much to worry about. If anything it will fuel more demand for actual data, to which people will have to use ESRI, and I am sure they will like that.

However if Google makes the logical next step after having Google Maps and Earth for a number of years and starts to develop even a rudimentary GIS editing software, I would be scared witless if I was ESRI. They may have the installed base, but as you mentioned, many are fed up with ESRI, and Google has a wide reach and lots of money by several amounts of magnitude compared to ESRI.

Re:Google the first? Not really... (1)

benj_e (614605) | more than 4 years ago | (#33507840)

I found that pretty funny too. I also find it painful to see heatmap used in a cartographic sense.

I've been developing GIS software since 1996, and I have to tell you that while no one toolset is ideal, I've found ESRI's the easiest to use in a production environment. I've use most of the open source GIS tools, even written some papers on them (that apparently were good enough to be cited by other authors), and yet I keep coming back to ESRI's suite.

Perfect, no. Better than the alternatives? definitely. I also like the developer community around ESRI's products - much more friendly and helpful than those associated with OS products. IMHO of course.

Re:Google the first? Not really... (1)

autocracy (192714) | more than 4 years ago | (#33509408)

There's an ESRI office down the hall from me...

I really wonder why they have a satellite office here and what they do there. It's not listed on the website anywhere.

Hey there, how's it going? (-1, Offtopic)

Anonymous Coward | more than 4 years ago | (#33504322)

Can't think of a good joke today, sorry.

Late night pseudo-intellectual wankery (2, Interesting)

cappp (1822388) | more than 4 years ago | (#33504362)

I've always been fascinated by maps and the curious space they occupy in the construction of personal and societal identities. What we map, how we illustrate it, how it's presented, where it's placed, each reveals an aspect of something culturally significant. I wonder if the shifting nature of mapping, as influenced by the demands of mobile devices and almost limitless storage capacity, will have as profound a cultural effect as other mapping technological shifts.

For those interested in mapping - Strange Maps [bigthink.com] has some awesome examples.

Re:Late night pseudo-intellectual wankery (0)

Anonymous Coward | more than 4 years ago | (#33504996)

I've always been fascinated by maps and the curious space they occupy in the construction of personal and societal identities. What we map, how we illustrate it, how it's presented, where it's placed, each reveals an aspect of something culturally significant. I wonder if the shifting nature of mapping, as influenced by the demands of mobile devices and almost limitless storage capacity, will have as profound a cultural effect as other mapping technological shifts. For those interested in mapping - Strange Maps [bigthink.com] has some awesome examples.

Fuck the map. The territory is infinitely more interesting. Being so fascinated by the map is like thinking a condom is more interesting than sex.

Re:Late night pseudo-intellectual wankery (2, Funny)

cappp (1822388) | more than 4 years ago | (#33505142)

Ah my anonymous friend in some ways condoms are more interesting than sex, at least when considered from a purely academic point of view. Think about all the controversy linked to contraceptives, the evolution of condom technology over the years, the constant shift in identity from disease-control to pregnancy control to disease control, the way in which we sell them, the way in which they're marketed, the politics of advertising, the problem of accurate representation, notions of propriety and sexual manner - there's a wealth of really fascinating topics buried in that there joke matey.

Re:Late night pseudo-intellectual wankery (0)

Anonymous Coward | more than 4 years ago | (#33540970)

Fuck the map. The territory is infinitely more interesting. Being so fascinated by the map is like thinking a condom is more interesting than sex.

Bad idea. The analogy is lost in most slashdotters...

Re:Late night pseudo-intellectual wankery (1)

dj e-rock (700351) | more than 4 years ago | (#33505324)

I had a professor in an anthropology class that I took during my undergrad at UCSD who had the class draw a map of San Diego. It was an interesting exercise since it wasn't prefaced with anything at all. The wide range of maps produced was staggering, with different things out of scale, etc. Mine, I recall, had an emphasis skewed particularly towards freeways that defined my usual routes of travel, combined with major thoroughfares that I would take . . . Genesee, LJV Dr., Nobel, etc.

Given that culture is, more or less, just a set of shared expectations, it's interesting concept.

If I felt more strongly about it, or had more energy, I'd run some statistical or regression analysis to ascertain similarity between different cultures.

Re:Late night pseudo-intellectual wankery (0)

Anonymous Coward | more than 4 years ago | (#33506088)

For those interested in producing Google Maps of images, Piritiles (www.piritiles.com) is an App Engine based tool I wrote during weekends in a couple months to create & publish & embed instant Google Map of any image.

The old Guard from my perspective (5, Insightful)

stimpleton (732392) | more than 4 years ago | (#33504450)

I did a post grad degree in GIS finishing 2003 haveing done IS degree before then. We used professional map tools for analytical work and presentation.

Two things stood out in the culture of GIS:

- A non importance of solid data handling and storage. Flat files were the order of the day.
- Antialiasing was not prevalent. While not required for anayltical work, in presentation it was, but many big name tools did not make the jump. 8 bit was common.
- Presentation was done by govt depts and were fairly snazzy for the day, in 8bit alisaed glory

Now we see were we are today, and its all to do with the fact non mapping companies have got involved without the hangups of the old GIS attitudes from govt depts, universities, and the big name tool vendors influencing them. Companies like MS, Google, have presented maps and GIS so superior to the traditional industry, that even Depts Lands, Mapping and Survey(or whatever called in your country) are resorting to Google maps.

We now have depts of GIS professionals along with proprietry vendor tools being trounced by private enterprise.

Re:The old Guard from my perspective (2, Interesting)

cosm (1072588) | more than 4 years ago | (#33504670)

In response to the old guard comments, some good supporting evidence of the newcomers understanding the importance of getting away from flat layer/shape/whatever files is SQL Spatial [microsoft.com] support in SQL Server 2008. If you haven't played with it, well, it is pretty impressive for Microsoft (at a hefty cost, of course).

I also heard rumoring of upcoming Spatial support in Azure (might be already here), so if that goes well, companies dealing with alot of customer map data will have a sweet way to move it all to the cloud, and serve it out via web services, and with many of the different mapping providers, this will provide excellent venues for serving out customer map-data from the cloud, allieviating them them on-site storage head-aches, and makiing things much more scalable and modular.

I know tools like ESRI's [esri.com] line of products have also come a long way in terms of Web 2.0-ish kind of stuff, as well, and they are worth checking out if you want to get into professional GIS work.

Re:The old Guard from my perspective (2, Informative)

cosm (1072588) | more than 4 years ago | (#33504672)

Meant to post ESRI Web Mapping API [esri.com] link.

Re:The old Guard from my perspective (0)

Anonymous Coward | more than 4 years ago | (#33505130)

Meant to post ESRI Web Mapping API [esri.com] link.

it's okay. you get twice the karma by "forgetting" to post the link.

Re:The old Guard from my perspective (2, Interesting)

Zak3056 (69287) | more than 4 years ago | (#33504794)

. If you haven't played with it, well, it is pretty impressive for Microsoft (at a hefty cost, of course).

Can you amplify on this a bit? Looking at the editions comparison page, it appears that the spatial features are included even in the free version of SQL Server (SQL Server Express). Am I missing something? (like it's feature complete, and just doesn't include some proprietary dataset, or dev tools, or...?)

Re:The old Guard from my perspective (1, Interesting)

Anonymous Coward | more than 4 years ago | (#33504836)

If you want free, go for PostGIS.

Re:The old Guard from my perspective (0)

Anonymous Coward | more than 4 years ago | (#33505364)

Exactly. I use PostGIS every day. The benefits of no lock-in, better price and a superior system to boot. It's amazing that there are still folks who think proprietary first.

Just about everything is a commodity now in the open-source ecosphere, from operating systems to RDBMSs. Postgresql+GIS is solid as a rock and mature to the point now where there's no need for proprietary solutions for this stuff.

Re:The old Guard from my perspective (0)

Anonymous Coward | more than 4 years ago | (#33511276)

And better. Microsoft supports a way out of date OGC version, has little of the SQL spatial toolset, relys on CLR and is rather slow. PostGIS on the other hand outperforms ESRI's ArcSDE and can do a whole lot in the database itself.

Re:The old Guard from my perspective (1)

cosm (1072588) | more than 4 years ago | (#33505052)

From a spatial standpoint, there almost identical in terms of datatype support. For tinker development, Express will work just fine. Its just the limitations of express will put a hamper on larger projects (10 GB limitation, 1 CPU, Partitioning). But then again, if your operation is encroaching on those limitations, hopefully the full Server will be within your reach. Something like a web app that has routing and centerline data for the entire country, and with you perhaps offering an web API for accessing that data, well, Express just wont cut it.

Re:The old Guard from my perspective (0)

Anonymous Coward | more than 4 years ago | (#33505398)

Thank you for the feedback. The extremely high cost of a Microsoft offering is what I feared (and expected...). I'll try the open source alternative instead (postgis?) that someone mentioned earlier in the thread...

Re:The old Guard from my perspective (1)

gbjbaanb (229885) | more than 4 years ago | (#33506058)

From what I know, the Microsoft offering is not feature complete, a bit like their free-text search indexing in SQL server, its always a step behind Oracle's. I know we use Oracle for its spatial capabilities because the SQL server doesn't quite cut it for the really cool stuff.

For free versions - MySQL has spatial [mysql.com] in it for some time. That's still free, and hopefully Oracle will improve the capabilities if only to annoy MS and capture the market that you represent - the 'we want it free until we have to buy something, and then that'll probably be Oracle' segment.

Re:The old Guard from my perspective (0)

Anonymous Coward | more than 4 years ago | (#33507632)

If you haven't played with it, well, it is pretty impressive for Microsoft (at a hefty cost, of course).

Can you amplify on this a bit?

Ok.

IF YOU HAVEN'T PLAYED WITH IT, WELL, IT IS PRETTY IMPRESSIVE FOR MICROSOFT (AT A HEFTY COST, OF COURSE).

Re:The old Guard from my perspective (5, Informative)

karavelov (863935) | more than 4 years ago | (#33504822)

There is a module for PostgreSQL called PostGIS with quite impressive feature set... It's there for a while, it is actively developed and have support from PostgreSQL core hackers.

Re:The old Guard from my perspective (1)

sr180 (700526) | more than 4 years ago | (#33505590)

Oracle have supported GIS data for quite some time. In the mapping industry its quite common.

Anyway, WMS was left out of the article. Its been around since 1999. http://en.wikipedia.org/wiki/Web_Map_Service [wikipedia.org]

Can you count to 2?? (1)

syousef (465911) | more than 4 years ago | (#33505774)

Two things stood out in the culture of GIS:

- A non importance of solid data handling and storage. Flat files were the order of the day.

- Antialiasing was not prevalent. While not required for anayltical work, in presentation it was, but many big name tools did not make the jump. 8 bit was common.

- Presentation was done by govt depts and were fairly snazzy for the day, in 8bit alisaed glory
 

If that is how you count to 2, I hope to the spaghetti monster that I never have to use one of your maps.

Re:Can you count to 2?? (1)

micheas (231635) | more than 4 years ago | (#33512824)

Two things stood out in the culture of GIS:

- A non importance of solid data handling and storage. Flat files were the order of the day.

- Antialiasing was not prevalent. While not required for anayltical work, in presentation it was, but many big name tools did not make the jump. 8 bit was common.

- Presentation was done by govt depts and were fairly snazzy for the day, in 8bit alisaed glory

If that is how you count to 2, I hope to the spaghetti monster that I never have to use one of your maps.

How a programmer counts to two: 0, 1, 2.

Re:The old Guard from my perspective (1)

DarthVain (724186) | more than 4 years ago | (#33507934)

Yes and no.

1) Where do you think Google Maps/Google Earth got its data from? The afore mentioned dinosaurs.
2) While easy on the eyes, you can't do analytical work with them. The tools and the data just isn't there, as it has all be simplified for public consumption. If we want a pretty map, in the past we would call a cartographer and build us a pretty map. GIS was and is for analytical work, for use in making decisions.

While serving it up to the public is nice, and sometimes snazzy it hasn't been the primary driver. I am MORE than happy to allow Google etc... take over that business, enjoy! I will happily make XML files all day for Google Maps, or KML for Google Earth. That said, I dare you to try and make either using a 500MB GIS file, or 100MB, or even 50MB. You have to seriously dumb down the data, or use very small datasets, unless you like your user to sit and wait ages for either Google Maps API or Google Earth to load.

Though I will agree with you that over the years, GIS went from bleeding edge in graphics, to no edge. Used to be that a GIS used every new thing the graphics card industry put in graphics cards. Their high end stuff was designed for GIS/CAD. However in most uses you are crunching numbers/tables to get a result, looking pretty is secondary. In fact I would say a faster refresh is better for getting actual work done than slick looking graphics (though large rez/multi monitor). But yeah I would agree GIS has fallen behind.

I remember when GPS was pretty special. We had 'em and like boats did, and they cost 5000$ for a decent one. Now my phone has one? I can buy a stand alone a a local store for like 300$ probably. Hell even Car GPS is passe now! Technology changes...

Licenses? (3, Insightful)

SnapShot (171582) | more than 4 years ago | (#33504918)

What I want is a blog post that actually explains all the various mapping licenses. Preferably in a simple table format. I don't like to read.

Re:Licenses? (0)

Anonymous Coward | more than 4 years ago | (#33505464)

I don't like to read.

That would definitely explain a few things.

Re:Licenses? (1)

dna_(c)(tm)(r) (618003) | more than 4 years ago | (#33505866)

I don't like to read.

That would definitely explain a few things.

You on the other hand, don't like to login, do you?

But seriously, there is so much noise surrounding the information on the internet that I too like that information as usable as possible. No prose, no fancy opinions but usable facts.

Now I'm off arranging my "get off" signs in a sensible pattern on my lawn.

Re:Licenses? (1)

mjwx (966435) | more than 4 years ago | (#33506256)

What I want is a blog post that actually explains all the various mapping licenses. Preferably in a simple table format. I don't like to read.

I second that and I do like to read but reading license agreements is like torture.

Slow... (0, Flamebait)

Schnoogs (1087081) | more than 4 years ago | (#33504970)

...news day I guess.

Tomorrow it will be "the state of sorting APIs"

More Important than the Maps (4, Interesting)

geoffrobinson (109879) | more than 4 years ago | (#33505050)

I think the thing that was most impressive, at the time, was how they got JavaScript to do all that in a browser. Now, that was impressive.

Is it my impression or did AJAX really take off after people saw Google Maps?

Re:More Important than the Maps (2, Interesting)

PCM2 (4486) | more than 4 years ago | (#33505634)

Is it my impression or did AJAX really take off after people saw Google Maps?

I kind of remember it that way, too, but there were already AJAX (or AJAX-like) toolkits in the works by the time Google Maps was announced in 2005. Tibco General Interface certainly predates it, and so does Dojo Toolkit. Prototype came out around the same time. I think the truth is that a lot of people got the same ideas around the same time, but Google was among the first to market with a cool (and visually impressive) use for those ideas.

Re:More Important than the Maps (1)

maxume (22995) | more than 4 years ago | (#33512160)

Yeah, after a whole bunch of people noticed that Outlook web access wasn't messing around with iframes.

ok seriously (3, Interesting)

JeanBaptiste (537955) | more than 4 years ago | (#33505222)

why cant they take all this map data and have racing or GTA type games where you can drive in places you're familiar with?

Re:ok seriously (-1, Troll)

Anonymous Coward | more than 4 years ago | (#33505452)

Seriously Jean Baptiste, you're an asshole.

Re:ok seriously (0)

Anonymous Coward | more than 4 years ago | (#33506008)

only if we get to run over people we know

Re:ok seriously (1)

BradleyUffner (103496) | more than 4 years ago | (#33507720)

I'd be happy with a flight simulator that had an option to pull its ground textures from these sites. yeah, i know about the flight sim in google earth... I'm talking about something a more complex., like x-plane, ms flight sim, or even orbiter.

I was looking forward to read an article on ORMs (0)

Anonymous Coward | more than 4 years ago | (#33505474)

Who cares about real, geographical maps?

UMN Mapserver (2, Informative)

tuxicle (996538) | more than 4 years ago | (#33505608)

No mention of the University of Minnesota Mapserver [mapserver.org] program. It can render GIS data with antialiasing, auto-placing labels, etc. With some work on a configuration file, you can have it produce images that are nearly identical to Google Maps, but can be arbitrarily scaled (ie, not based on tilesets). Support for WMS makes images from Mapserver available to external programs via a simple HTTP-based RPC. You can also query for features using WFS. I use it to generate maps for use with a weather radar display.

GeoServer (0)

Anonymous Coward | more than 4 years ago | (#33511050)

Let's not forget GeoServer [geoserver.org] .

Satellite images (1)

dargaud (518470) | more than 4 years ago | (#33505902)

Closely related to maps are satellite images, and they are in serious need of a cleanup as well. I was recently asked to find some images for a specific area in Antarctica [gdargaud.net] and do specific processing on them. I'll pass on how hard it was to find images (for free or fee), the USGS was the best source but with other commercial satellites it was simply impossible to get samples. Not, the problem was in the format: basically every generation of satellite has its own data format and/or associated geolocation description file, and even 'standard' formats such as GeoTIFF (a TIFF file with embedded geolocation information) were hardly ever read by GIS softwares. I must have tried 10 packages and it's a nightmare, particularly if you want to do processing in addition to visualisation.
Photoshop/Gimp can open and enhance GeoTIFF files. Google Earth Pro can overlay them on the globe. But call one after the other and the geoloc info is lost. It's all like this. Last time I had to do this I ended up writing my own image processor in C [gdargaud.net] , that was almost a decade ago. Nothing has changed. Of course big vendors tell you that their 10k$ software can do it, except that from my trials I have no faith in that, that's too expensive for processing 10 files, and it takes 3 months to get a handle on the software.

Re:Satellite images (1)

cryokinetic2 (1888240) | more than 4 years ago | (#33509586)

If you want a software package comparable to ENVI or ERDAS Imagine, check out Opticks. http://opticks.org/confluence/display/opticks/Welcome+To+Opticks [opticks.org] It's open source and runs on a variety of systems. GDAL is also worth a look. It's good for simple, quick jobs, large, compex scripts etc. http://www.gdal.org/ [gdal.org]

Re:Satellite images (1)

penguinchris (1020961) | more than 4 years ago | (#33514500)

I have had a similar experience with satellite imagery and data, but I have to say, despite the ESRI bashing going on earlier in the comments (much of which I agree with) ArcInfo was ultimately able to handle everything I threw at it without too much fuss. They certainly don't make it easy to figure out how to avoid the "fuss", though.

Re:Satellite images (1)

penguinchris (1020961) | more than 4 years ago | (#33514566)

I was distracted and forgot the main points I wanted to make, regarding getting at the data in the first place and processing it. Getting the data also involves a lot of "fuss" but once you figure out where the thing you want comes from (NASA, USGS, etc. all have their own sites and systems for accessing the data) it's fairly trivial. Again, they don't make it easy to figure out how to avoid the fussing around, so I definitely agree with you there.

Also I was quite frustrated with doing analysis. GeoTIFF sounded great until I realized as you did that once you do anything to it outside of GIS software, such as in GIMP or Photoshop, the geolocation is lost. What a pain :) Ultimately I just didn't bother doing much processing that I couldn't do right in ArcInfo, but then I was just using it for a master's thesis research project and not doing anything too serious where it mattered.

No way to add POI on GMaps? (2, Interesting)

bemymonkey (1244086) | more than 4 years ago | (#33505982)

Why is there no way to add/update POI on apps like Google Maps?

So many of the existing POI are out of date, and many are missing... why can't I just edit the information myself directly from my Android phone or Maps on the PC? Of course, the changes would need to be approved before they're actually integrated into Maps, but I feel like they're leaving a lot of potential untapped here.

OSM supports this, of course, but the Android apps are absolute crap... not to mention the maps of Southeast Asia or Eastern Europe. :(

Re:No way to add POI on GMaps? (1)

gbjbaanb (229885) | more than 4 years ago | (#33506072)

most PoI are created from external data sources, not entered by hand. So your beef is with the companies that supply that data. The problem with manually modifying them is that there are a lot of them, and you'd so quickly build up a backlog longer than anyone could manage. Maybe they could get an automated system in place, where enough people who update the same POI 'validate' their own work and have the update accepted would work.

Re:No way to add POI on GMaps? (1)

bemymonkey (1244086) | more than 4 years ago | (#33506202)

Ah, good point. Is there a place where I could check which sources Google uses for their Maps POI? Some of them are clearly marked/stated but others are just there without any hint as to how they got there...

Re:No way to add POI on GMaps? (1)

gbjbaanb (229885) | more than 4 years ago | (#33507790)

no idea, sorry. It will change according to the country anyway. Usually its government datasets, plus 'yellow pages' type data from commercial companies.

Re:No way to add POI on GMaps? (0)

Anonymous Coward | more than 4 years ago | (#33506734)

If by poi you mean points of interest , check out wikimapia , it's got my city charted out and i live in the middle of nowere.
It's a google maps overlay.

Re:No way to add POI on GMaps? (1)

bemymonkey (1244086) | more than 4 years ago | (#33506798)

Interesting... how do I get it to display as an overlay? I can't really seem to select both Google Maps and the WIkiMapia map at once.

Also, is there an Android app? :p

Re:No way to add POI on GMaps? (1)

bemymonkey (1244086) | more than 4 years ago | (#33506814)

Found the app... Doesn't really seem to work though. Just displays a list of stuff that's been mapped out with checkboxes... no way to actually display a map :(

http://www.appbrain.com/app/com.MapiaExplorer [appbrain.com]

Re:No way to add POI on GMaps? (1)

T.E.D. (34228) | more than 4 years ago | (#33513740)

They used to do that back in the early 2000's, but some loser named vpotus@whitehouse.gov kept erasing his own house from the database.

They need reduce also. (0)

TechyImmigrant (175943) | more than 4 years ago | (#33505988)

The 'map' API will remain fairly useless until they also develop the 'reduce' API. Then developers will get the full power of a map-reduce API.

Another way of summarizing the article... (1)

MrCrassic (994046) | more than 4 years ago | (#33507122)

Yeah, remember when people used to print maps from MapQuest to each other because in-car GPS modules were expensive and a bit "beta?"

Now, Garmin's newest GPS doubles as an Android phone, the iPad (still waiting for an Android tablet) is the biggest GPS module available and one can find their way from practically anywhere with internet access just by knowing the destination.

Paper maps aren't dead, but they are needed much less nowadays.

Missing the Point (0)

Anonymous Coward | more than 4 years ago | (#33507432)

Okay some people are really missing the point (surprise surprise /.)

There are two huge components to mapping, data creation and data delivery. Google, M$, Mapquest, etc.. all blow ESRI out of the water with their front end for data delivery. They all SUCK at data creation/maintenance etc. Googles Enterprise client for data maintenance is a disaster. ESRI is the de-facto tool for this. They simply serve different purposes.

disclaimer, I am an SDE admin for county government. I develop apps using the ESRI Javascript API. My app will never be as good or as slick as google. However, my data is WAY better.

3D GIS Modeling with PixelActive CityScape (0)

Anonymous Coward | more than 4 years ago | (#33513422)

Hey Everybody,

I work at PixelActive and we've seen tremendous interest in 3D GIS modeling from all of our clients. Whether it's mapping companies or video game developers who want to make games based in known cities, they are using CityScape to model cities quickly.

When we look at established GIS companies, it appears there is a lot of legacy support that they need to handle, which slows them down. We've been able to move much faster, supporting other commenters' posts about how private companies are finding success here. Our problems is supporting many of the legacy formats that are not well documented or designed for real-time rendering (which CityScape is designed for).

Joseph
www.pixelactive3d.com

where are my free maps? (1)

Finite9 (757961) | more than 4 years ago | (#33518386)

I recently became interested in hiking/orientiering, and i tried to find maps geographical maps online that could be used for this purpose. Couldn't find any apart from the government (Sweden). I will have to pay for the maps.

I was quite surprise to tell the truth. I have no involvment with the mapping industry, but I would have thought by now after Google came out with Google Earth and satellite images became so prevalent that it was simply a matter of time before all mapping was free? I can go onto the local yellow pages equivalent and see photos of my car in the driveway! And there are overlays for sat images to show POIs.

Isn't it just a matter of time before there are mapping solutions for everything, free, online, and won't this basically kill the current mapping industry (from he who knows nothing about this)?

Check for New Comments
Slashdot Login

Need an Account?

Forgot your password?