We need a “cash for clunkers” program to rid us of old busted web GIS
September 22, 2009 33 Comments
A day doesn’t go by were I don’t run into it. You know what I’m talking about, some busted old ArcIMS or MapGuide website that is the only thing between me and getting what I need out of data. Clearly new technology isn’t going to just help these websites get upgraded on their own. No, clearly we need some stimulus money to help foster some new open, standards based, http goodness in our lives.
What I would like the Obama administration to do is start “Cash for ArcIMS” and a “Cash for MapGuide” programs to rid us of these abominations. It is our patriotic duty to replace these old antiquated web technologies with some fine ArcGIS Server, GeoServer, MapGuide Open Source or many of the other great web GIS products out there.
It is time to stand up and take charge. Sing along men!
About James Fee
I mostly agree with you…but not on this. Some real good ArcIMS sites are out there; no need to replace them until the maintenance is too high or something else changes.
I think the “If it’s not broken, don’t fix it” mantra works well here.
I’m all for ArcGIS Server and even the open source items you mention…what about cash for municipalities that have nothing, those are the worst.
There is no good ArcIMS site out there. They are all slow, antiquated and confusing.
Nick, James is proposing cash to solve your ArcIMS problem so no worries mate….
There are good ArcIMS sites out there. You just need the right front end and good attention to detail. There are tons of crappy ones too, I agree. We use Geocortex from Latitude Geographics and then we make things work nicely. We have a few Internet flavors and almost 10 intranet versions. We’re doing ArcGIS Server too; and now using the Flex APIs…but they’re not there yet will the full suite of features and functions (although we’ll probably be there next year) and ArcIMS is more mature and still works.
Here’s an ArcIMS site for a planning department (mine); you tell me.
http://planning.lacounty.gov/gis/interactive
This is no clunker…we’d always like cash to get us farther into the newer technology…but don’t throw these out until you can move over safely.
This is a good idea, but you don’t even need cash , the open-source web servers will do the job just fine. Maybe they need to be given vouchers that buy some ‘james fee consulting time’ to get these installed
I agree with you for the most part, but I think that the problem isn’t always a matter of which map server technology is being used on the back end. There are plenty of good sites around that are still using legacy technology like ArcIMS as their back end map server. The sites that you describe sound more like the ones with a poorly designed front end. The ArcIMS HTML Viewer Template comes to mind, but if we’re going to look at it from this angle, wouldn’t the ArcGIS Server Web ADF be a candidate for criticism, too?
How about this: let’s get a cash for clunkers program going for sites developed using these lousy one-size-fits-all templates!
What user interface to ArcIMS is working? You mentioned the HTML Viewer as crap, the Web ADF suck and every home grown ArcIMS front end is just as bad. Poor cartography, no caching and the data is locked up behind a proprietary web service.
FAIL
Geocortex has a good front end to ArcIMS.
Poor cartography is not solved by ArcGIS Server or the APIs or open source. Caching works well for speed and performance but you can get that other ways too; and not lose the interactive nature of dynamic layers.
Oh man, I’d love my tax dollars to go toward opening up web services. Yes someone needs to pay the lifting on this, but I’d rather see my money go to opening data here at home and not for missile defense in eastern Europe.
http://www.epa.gov/webservices/web/index.htm
Two points:
1) I agree with you.
2) I believe you are referring to a private instance of a more general phenomenon, namely abandoned websites.
If I am correct on #2 (and you know I am), then the issue is not so much the technology, but the need for website (GIS or otherwise) owners to understand that what they own must be maintained, and occasionally upgraded. You can’t just put it up and forget it, which is what many of them do.
I agree with #2.
All to often, there are developed web services scoped and budgeted for a project, but updates and maintenance costs are too abstract for higher management to consider in these budgets and are afraid the client will balk and go with someone cheaper. Fine then, but in this economy upper management does not always have that option. So they either cut it out of the budget or “we’ll figure that out later” syndrome comes into play.
It’s important to explain and educate that if these types of wonderful services being offered to clients, there needs to be some sort of responsibility written into the project scope.
IF you are serious…this isn’t, nor should it be a function of government at any level.
Otherwise I agree…get rid of all the crap out there.
I agree with Mike O. There are just to many with bad front end designs. Unfortunately the templates that were created to teach you how to build a site ended up being the sites.
Most GIS website sites do not need to scale to 100s and 1,000s of users. If you need this then just generate some tiles with a script and use Open Layers. The lack of tiling & caching in ArcIMS is really not that big a deal. You can easily squeeze some nice performance out of ArcIMS by streaming the image and optimizing the cartography and image format.
Now if you want to talk about map platforms then you really can’t include ArcGIS Server in this. Its just to expensive and with all there crazy licensing. I don’t think its worth it. Especially since MapServer, GeoServer, MapGuide & PostGIS can do the same thing.
James,
I think you’re man overboard on this one.
Most of the (woefully iterated, poorly funded at the outset and operating on less money now) ArcIMS and older MapGuide sites are doing exactly what they were designed to do. They answer the mail and deliver information for citizens…and that they do it in a proprietary and inaccessible manner is probably a selling point for folks that have the heebie jeebies about giving away their data.
The poor schmo in the bowels of some county GIS program who wrapped his laptop box in duct tape to make a new wastebasket would no doubt love to upgrade…but with what budget, on what machine, using what new software and database?
This profile is probably more prevalent than we’d like to admit across the GIS community…and it’s why Shapefiles and DXF refuse to die.
These cobwebs live on because they work, and the change cost is untouchable in this climate of worker furloughs (thanks for the free vacation, pal) and staffing cutbacks.
To quote my colleague on this, “Cache for Clunkers†…
Way to go James! Second time you raise a good point crying for statistics, and if I knew where to get them I’d be the first to post it!
1) what is the proportion of Win2K / XP / Vista desktop clients
2) what is the proportion of ArcIMS / Mapguide / ArcGIS Server / MapGuide OpenSource and other web servers
3) what is the proprotion of HTTP, netbook, phone etc. web clients
Beaching you metaphor, the credit crunch hit foot soldiers hard, from sergeant to colonels we know it, but do the generals know? Such statistics migh give us some ammunition to shore up our troops budgets!
Donny V’s point on ArcGIS Server’s cost and licensing had me wondering something… Is there any other solution that handles distributed versioned editing (e.g. geodata services)?
This is the one AGS feature that I have not been able to replicate through other means.
Sure using this PostGIS extension.
http://geoserver.org/display/GEOSDOC/Versioning+PostGIS+datastore
Does that work disconnected though?
That’s the part that I really need (so that munis can post their changes every couple of months).
A lot of cities and counties are strapped for cash now as we all know and quite possibly have more important issues, like social services, to fund than the newest slickest way to zoom to a parcel.
Better tools to make better decisions saves everyone money Jeff.
But a packet of paper maps with a live person are generally going to be viewed as “better tools” than a web map for the decision makers.
I couldn’t disagree more. Interactive mapping in which users are immersed in the data is by far a better tool than a plotted map.
A plotted map + a live person, not just a plotted map. (But I put “better” in quotes, because it is about perception too, not just some objective measure of “better”.)
The user still has to know how to interpret the data too as well as be immersed in it. While you can add on the fly 4D adaptive kernel densities, component analysis, ANOVAs, and explanations of what all the stats mean to your application; for the time spent creating such an application, decision makers would much rather never see -any- of that and just have an expert say “Here’s what it all means”.
“Are we preventing crime?” is a lot harder question from a map alone than “Where is the crime?”
If the site still works, and delivers the ‘illusion’ of access, oft-times that is the goal of the service.
‘Better’ tools might actually provide a citizen with the method to second-guess or validate the actions of the government entity that sponsored the site in the first place.
Even more prosaically, I wonder how many of these cobwebs live on after the original sponsor and geospatial guru has moved on to greener pastures, and are perpetuated under some thready little support contract by the original vendor or integrator.
I have seen many small geospatial programs with hard-working low-on-the-totem-pole folks rattling away on old technology, suffering through little or no budget or ability to affect change in the system above them…and to them, a web site that sucks is better than no site at all.
Sigh. The reality of many a county-level GIS, I think.
“I wonder how many of these cobwebs live on after the original sponsor and geospatial guru has moved on to greener pastures…” — Most of them, I bet.
There *is* a Stimulus program like this – it’s called broadband mapping and its a $350million national program (with even more in the USDA budget…much more). Savvy grant-hounds are already attached at the hip to the money spigot. Folks like, you know, that sales guy over in Redlands.
Who says ArcIMS can’t cache http://ims0.osiemaps.ie/website/publicviewer/main.aspx?id=&utype=&ecom=S1&user=
Pingback: Map Me the Money « geo-fi
Uh, based on my experience of working for Fat Consultancies on public sector contracts, this sounds like a great recipe: implement a real clunker this year, then apply for your Cash For Clunkers grant next year to re-implement it, and get paid twice for the same job. And most government clients I’ve seen are probably dumb enough to go for it too…
nicely put
Pingback: GIS in XML » Blog Archive » Storm Runoff modelling and MRLC
Pingback: Public sector tender specialists - ways2win