November 28, 2005
Contents
Creating a Foundation
The Code Base and Product Vision
Backstory
DM Solutions Group's Take
Autodesk's Take
Frank Warmerdam's Take
My Take
Creating a Foundation
Members of the open source MapServer community including MapServer Technical Steering Committee Members, the University of Minnesota and DM Solutions Group along with software industry giant Autodesk, signed their names to an open letter announcing the creation of the MapServer Foundation in recent weeks.Today (11/28/05), that letter, and the first set of goals for the new organization were made public.In short, the group will support and promote "open source web mapping," be broadly based and community driven.The lack of detail in this initial announcement underlies the goal of involving the broadest community as early as possible.
Autodesk has stepped up to the plate with funding and code.The successor to its MapGuide software, built bottom up as open source code, will be turned over to the Foundation.It will be called MapServer Enterprise.An early version of that code is available on the MapServer Foundation website now.Other signatories to the letter include DM Solutions of Ottawa, Canada (the largest company developing on MapServer and host of several sites with open source tools and organizer of three open source/MapServer conferences), the University of Minnesota (home of the MapServer Project), and members of the MapServer Technical Steering Committee.
The Foundation is expected to provide a stable infrastructure for the now extended MapServer family's code base and its growing community.In particular, those involved with the group suggest it will provide for more rigorous testing and management oversight than was previously available.Further, the Foundation will provide legal protection for the code and its programmers.In these times of SCO style lawsuits (where organizations allege copyrighted code has made its way into open source offerings) such backing can be vital to the health and growth of the project and the community.Finally, the Foundation will, in time, become the host for the community's conference(s) [2005, 2006].
The Foundation will serve as the home for contributions to the code base and hopefully remove any concerns over the longevity and nature of MapServer.Some suggest that MapServer's history at the University of Minnesota has tagged it as "just" an academic project.(NASA providing initial funding via the ForNet project, a cooperative effort with the Minnesota Department of Natural Resources.NASA has continued to support MapServer via TerraSIP.)
The Code Base and Product Vision
The existing MapServer code base, currently managed by Steve Lime at the University of Minnesota, will be renamed MapServer Cheetah (to distinguish it from MapServer Enterprise) and will be part of the Foundation's library.Along with the MapServer Enterprise code, Autodesk will also contribute a dozen of its data access tools, called feature data objects (FDOs).Autodesk will retain rights to a few of these including those used to access Oracle and SQL Server.
The two complementary, but in some ways overlapping open source mapping and GIS offerings, MapServer Cheetah and MapServer Enterprise, will offer programmers and end users a choice of open source offerings.MapServer Enterprise includes GIS analysis and related high end functionality and support for ASP, C++, PHP, SWIG (.NET, Java) for programmers.MapServer Cheetah will continue to offer its map publishing framework with support for PHP, Perl, Python, and Java to access the MapServer C API.MapScript, an internal scripting language, allows for the inclusion of disparate data.
While the open source code will be available for both projects via the Foundation website, it is fully expected commercial companies will create their own products based on them.Autodesk plans to release its own version of MapServer Enterprise, Autodesk MapServer Enterprise.Companies in Japan and elsewhere already repackage and sell their own versions of MapServer.
DM Solution Group's Take
Dave McIlhagga, President of DM Solutions Group (DM), uses the term "growing up" to describe what's happening to MapServer and its community.He's first to acknowledge that both have come quite far from the project's start in 1996, but admits both code and community need this sort of organization to reach the next level.
While DM had spent quite a bit of its resources managing, enhancing and offering code, the Foundation, McIlhagga feels, will help it and other companies to focus more on services and less on the core code issues, opening new opportunities and further growing the business.DM, he points out, was in growth mode before this announcement, having recently hired a Chief Operating Officer.
This Foundation, McIlhagga observes, is quite different from other open source foundations that came before.LINUX and APACHE were grass roots efforts that built up code bases and formed communities and foundations.Sun and IBM are taking existing commercial code such as ECLIPSE and Open Office and turning them open source.The MapServer Foundation may be the first, he suggests, to have a bit of both traditions.MapServer has a grass roots history, while Autodesk's contribution of MapServer Enterprise points to a commercial one.
McIlhagga wants to be clear that there is much to discuss and explain with the advent of the Foundation.His greatest fear is that the MapServer community will feel that its product will be replaced by MapServer Enterprise."MapServer Enterprise is not a replacement for MapServer.MapServer will continue on as it has." He's quick to remind those that may not be aware that no entity can kill an open source project, only those who choose to work on it, or with it, can do so.
Autodesk's Take
Autodesk has been working on the successor to MapGuide, called MapServer Enterprise and code-named Tux, for some time.The Infrastructure Solutions Division (ISD) decided, even before discussions of a MapServer Foundation, that it would be built on open source technology.(Tux is the name of the Linux mascot, a penguin.) MapServer Enterprise was built from the ground up on an open source development platform called ACE, ADAPTIVE Communication Environment.
Why open source? A few factors influenced the decision, explained Gary Lang, Autodesk ISD's Vice President of Engineering.
"˘ End users and developers were requesting such things as WMS, Web Map Service (an OpenGIS specification) support and that Autodesk was taking far too long to turn these around."With this kind of product," Lang explained, "it's possible to implement and test quickly.There's no need for long waits for functionality." By "this kind of product" he refers to server based products in contrast to desktop ones such as AutoCAD.It's relatively simple to create test suites for servers since there are limited possible requests.For desktop software, it's far more complex.The speed at which open source communities can add these types of changes was appealing, too.
"˘ Third party developers were requesting different ways to build MapGuide apps.In current and previous MapGuide releases, they used JavaScript to build applications on the client side.That meant literally sending code to all the users of the application each time an update was made.These developers were anxious to do their work on the server side, which offered a perhaps simpler way.Autodesk selected open source tools like PHP and SWIG (which enables .NET and Java development) were selected as a way to provide server side programming.
"˘ Autodesk has in recent years aimed to move away from products to applications and services, "moving upstream" as Lang puts it.He points as evidence to the growth of Autodesk Professional Services and ISD's recent acquisition of Topobase.Moving toward open source would free up resources to focus more on applications and less on core code.
"˘ There was full support from Autodesk CEO Carol Bartz.Lang pointed out she sat on the board of VA Linux for some time, and understands how open source works.
Building on open source tools is one thing, but funding and contributing that code to a new foundation is another.Lang has only positive things to say about his interactions with the MapServer community over the past few months."They are not GIS geeks - but people trying to solve a problem.They ground us," he says, noting that only two Autodeskers are involved actively in the Foundation.He's genuinely proud to have his company involved with this group for a few reasons."More alternatives for GIS are better than fewer," he points out."They [the MapServer community members] have done good work that should be continued and the number of [MapServer] users has increased exponentially over the past two years!"
And, while he hopes to grow MapServer Enterprise users, the real goal is "banding together as geospatial professionals." In that context he notes that one of Autodesk's competitors outspends it 10-1 in marketing of Web mapping products.Joining forces with the open source community is one way to offer a clear alternative.
Lang is anxious to see expanding adoption of Autodesk's version of MapServer Enterprise.Increased adoption will mean more FDOs, effective clean up of the code, and new functionality.Those enhancements will come both from Autodesk and members of the MapServer community.
The company will package and support MapServer Enterprise in its upcoming product, Autodesk MapServer Enterprise 2007, just as Red Hat packages and support its distribution of Linux or IBM packages its version of the Apache server.What will Autodesk MapServer Enterprise 2007 look like? It will have a server portion MapServer Enterprise and a new authoring package called MapServer Enterprise Studio.Studio, a "rich client" will be build on the Autodesk Map DLL, so it can read DWG, among other things, and allow the development of map services (deciding what data to publish, how to render it, etc.) as well as data preparation and editing.
Autodesk's commercial version of MapServer Enterprise's native data format will be a new version of SDF built on SQL Lite, an open source database.Other accessible data sources will include (via SDO) ArcSDE, Web Feature Service (WFS), Web Map Service (WMS), Shape files, MySQL, Oracle.FDOs for Oracle and SQL Server will be available only in Autodesk's product, but not in the open source version.Said another way, those two FDOs will not be turned over to the Foundation; they will remain the intellectual property of Autodesk.However, it is entirely possible another developer will build new FDO providers and chose to add them to the MapServer Enterprise code base for all to use.
Autodesk MapServer Enterprise 2007 will support two vector publishing formats: Autodesk DWF (Design Web Format) and one called eMap, which encodes georeferencing information and can carry attribute information.Clients using those will need to download and install a plug-in their browser, much like previous MapGuide implementations.The server will also support the sending of raster data to a browser, with no plug-in required.That implementation will use tiles and be based on AJAX (Asynchronous JavaScript and XML)."It'll act like Google Maps," says Lang.
Autodesk expects to see a mature code tree for MapServer Enterprise on the MapServer Foundation website by February.That, Lang suggests, will mean an Autodesk release in the second or third quarter of 2006.
Frank Warmerdam's Take
Frank Warmerdam is one of the key players in the open source geospatial space.He's keeper of GDAL (raster libraries) and OGR (vector libraries), two open source libraries used in many open source and commercial software packages.He's a MapServer programmer and a voice of reason in the community.He also was awarded the first Sol Katz award by the open source geospatial community.In the context of this discussion, it's important to know that he's a member of the MapServer Technical Steering Committee and a signatory on the Foundation letter.
As a long time member of the community, Warmerdam is quick to point out the credibility that a company like Autodesk brings to the table.Still, he balances that with skepticism around the company's commitment to MapServer Cheetah, pointing out that the company is at least for now, most interested in the MapServer Enterprise, the code it's written and contributed.
Warmerdam noted his initial intimidation after looking at MapServer Enterprise.It's written, he explains, "by a large group of programmers, with experience and a good deal of oversight.It's heavyweight." That said, after some exploration and fiddling, he was pleased to find he could program FDO providers and transfer his skills.He has a contract to extend FDO and is at work creating FDO providers for GDAL and OGR.Still, he's concerned more casual MapServer Cheetah users may find MapServer Enterprise overwhelming.
Where is the line between MapServer Cheetah and MapServer Enterprise? For Warmerdam, it's about how one chooses to interact with the code.Those who want to "poke around" may well be happier with MapServer Cheetah.Those who want to get up and running fast and use a slick interface may want Autodesk's MapServer Enterprise 2007.Warmerdam is adamant that choice is a good thing.He admits that some part of him wants to "bury his head in the sand and ignore MapServer Enterprise" but realizes that "that's not the best thing for the user." And, he's thinking of the user community, too as he plans to ask the GDAL and OGR communities if they would like to see those libraries, for which he's been a self described "dictator," put under Foundation oversight.
He illustrates another difference between the two offerings in the communities behind them.He observes the difference between how a commercial company and an open source community "markets and sells" software.A commercial company determines a vision, then sells that vision, he explains.The open source community puts out lots of choices and sees "what flies." MapServer has thus been vetted by the community.MapServer Enterprise does not yet have that pedigree.
I posited that aligning the MapServer community with some parts of an Autodesk community might create a force to compete with market leader ESRI.Warmerdam suggests that Autdoesk may see it that way, but he does not see this as a "wedge" to drive ESRI users to open source."ESRI's strength is in the desktop.Its success in online mapping comes from its strength in the desktop.It's 'just easier' for long time users to stay within the family." An open source solution, even one backed by Autodesk is not likely to change that loyalty, he predicts.
Asked for his take on the challenge Autodesk may face educating its existing community about open source, Warmerdam describes it as small."Autodesk is selling what it has always sold: the value of the Autodesk relationship.By purchasing an Autodesk branded version of MapServer Enterprise, the vast majority of open source dilemmas are solved." Those dilemmas include service and support, something in which Autodesk has quite a bit of experience.
It's clear though that Warmerdam and other members of the community have some concerns about sharing a foundation with the likes of Autodesk.For one, offering "two versions" of MapServer may split what until now has been a tightly knit worldwide community.Warmerdam hopes to help keep the two groups linked by working toward shared code such as support for GDAL and OGR and other functions and libraries so that developers and users can move easily between what are now two completely unrelated code bases.There are also concerns that Autodesk, and not the user community of MapServer Enterprise, will drive development in that open source project.Warmerdam hopes leaders in the current MapGuide community will take up roles in the Foundation to combat that possibility.Finally, Warmerdam is sensitive to the fact that even the creation of the Foundation has split the MapServer community."Some were in the know and others were not as the process moved ahead.Some may feel we've sold out to Autodesk." In time, he hopes, the community will rally around the Foundation for the good of the user base.
Backstory
The creation of the Foundation did not come out of nowhere.There have been signs for some time that MapServer was ready for a "next step." I've heard many in the open source community make a statement on the order of, "to get any further, we need to get our act together." I got that sense, and the sense of excitement at what was possible, at the second open source GIS and MapServer conference in Ottawa in 2004.There was even more buzz at the 2005 event, with a bit of wonder and head scratching about the attendance of an ESRI staffer and several very high level Autodesk and Intergraph staffers.
The attendance of Gary Lang and Geoff Zeiss of Autodesk and Ignacio Guerrero of Intergraph made many at this year's event nervous.To be fair, these are all very well respected, knowledgeable and professional individuals who were honest and open about why they were there."We can't ignore open source anymore; we need to understand it," one of them told me.I did see other odd pairings of commercial GIS vendors with open source people at Directions Media's Location Intelligence conference, as well.
The formation of the MapServer Technical Steering Committee in July was another indication that some sort of more formal structure was developing.Later this year, blogs (including All Points Blog) pointed to a proposal and discussion on a MapServer Foundation on Tyler Mitchell's "next generation" MapServer site.(The documents in question are no longer online.) Sean Gilles, a member of the Technical Steering Committee, raised the issue of "secrecy" in his post at Zcologia.
My TakeI agree with one of the members of the MapServer community who suggested that this pairing and the formation of the Foundation are perhaps the most important events in the geospatial arena in 20 years.Why? This is not a merger or acquisition.This is not even a business relationship in the traditional sense.It's the realization at very high levels in the commercial geospatial community that the open source model and process are valid and valuable for both software providers, third party developers, and software users.Moreover, a company like Autodesk must believe that choosing the open source approach will generate substantial revenue and open new doors.
I offer these short lists of pros and cons regarding the creation of the Foundation along with a "wish list" for its future.
Pros
MapGuide/MapServer Enterprise gets a huge re-launch.MapGuide has been a terrific product with little visibility.This "whole new MapGuide story" can only bring it out into the light in many new ways.
MapServer/Cheetah gets a boost.MapServer has been growing in name recognition and use, but the creation of the Foundation, backing by a huge industry player, can only enhance that growth.
Open Source get a boost.While fear, uncertainty and doubt (FUD) about open source continues in IT circles, I hate to admit how much is still running amuck in geospatial circles.Hopefully, the Foundation's efforts will continue to clear the air.
More platforms.Moving to open source is pushing Autodesk from its "Windows only" retreat.
A home for other open source projects.It's too early to say for sure, but the Foundation may help house the growing number of open source geospatial projects.
Cons
Challenges to community abound.The nature of the MapServer community has thrown on its head.Skepticism and fear are to be expected.They are frankly healthy during times of change, though they can ultimately pull groups apart.
Integrating products and people.While this is not a merger of products, finding common ground between MapServer Cheetah and MapServer Enterprise will be a task that may make or break the Foundation.Holding it together is key if other open source GIS projects are to come under its umbrella down the road.
Head to head competition.To date open source consultants only competed (and often cooperated!) with one another.In time, open source programmers tackling open source MapServer Enterprise may well face off against those with experience with Autodesk MapServer Enterprise.(That might be considered a pro by some.)
Wishes
I'd like to see the Foundation, in time, become the home to open source GIS technologies that are ready for this next step.Which ones? GRASS? PostGIS? QGIS?
I'd like to see other commercial companies consider signing on the foundation.Would it make sense for Intergraph to explore using the FDO technology in GeoMedia? Might that company turn some of its projects into open source ones?
I'd like to see the foundation take a firm stand on standards, perhaps even Open Geospatial Consortium standards.MapServer has many OGC specifications implemented (WMS 1.1, WMS 1.0, WMC 1.0, WFS 1.0, SLD 1.0, GML 2.0, Filter 1.0, WMS 1.1.1), but the lack of a "body" to the product certified compliant has been lacking.
I'd like to see the geospatial community open its doors to the Foundation.I'd like to see the Foundation showing off what it does and how it does it at trade shows and on GIS Day.I'd like to see it invited to professional meetings to get users up to speed.
But these are dreams for some time in the future.Today, this week, we are simply adjusting to a space/time warp in the continuum of our geospatial marketplace.Processes, products and relationships are all changing in ways with which none of us have experience.
|
Your Comments Post a comment All comments provided in this section are those of the individual who has created the post. These are not the opinions of Directions Media, its editors, staff or owners unless otherwise noted. Directions Media retains the right to edit or delete any comments posted herein.
|
|
||||||
| This is great news, with wins for all involved (and potential challenges for those not involved). First, Autodesk can gain market share for its vertical solutions and, indirectly for its DWF format. It can also differentiate on its commercial product with support, stability, a commercial-quality autoring tool, and supported enterprise database FDO providers. Second, the entire online mapping community benefits from a code base that has been professionally developed from the ground up to provide a rich mapping interface with fairly minimal technical knowledge required to implement. I don't feel competition between the two groups of developers is going to pan out. My feeling on this is that all Mapserver Enterprise developers will develop to the code, and the platform that their clients adopt will depend on what works best for that client. There will no longer be a distinction between open source developers and commercial developers. There will just be developers. On the topic of the two offerings. When we migrate from MapGuide 6.5 to Mapserver Enterprise, I fully expect to continue licensing the Autodesk commercial offering, but changing our strategy to use the free version in perhaps our public internet site and possibly for load balancing if Autodesk's commercial EULA permits combining the two. Jason |
||||||
|
||||||
| My foundation ideas/discussion documents that were mentioned are available once again. Due to a glitch in a server upgrade, they needed to be re-loaded. You can find them at: http://ms.gis.umn.edu/community/foundation Tyler |
||||||
|
||||||
| While the Google Earth/Google Maps launches were certainly the watershed developments of 2005, this announcement sets the stage for the MapServer/MapGuide combo to be the major development in the industry in 2006. As part of a start-up in the first Internet boom that tried (and failed) to sell the superior performance of MapGuide to ESRI-entrenched clients, I fully appreciate where Frank Warmerdam is coming from when noting that ArcIMS/ArcServer will always have a captive market of sorts. But the sea-change now is that a big chunk of the new demand for easy-to-use geospatial interfaces and data storage is coming from non-GIS constituencies who have few allegiances. They just need something that works and is cost-effective. And from an independent consultant's perspective, I can assure you that having the Autodesk name to throw around rather than the University of Minnesota will help alleviate much of the vague skepticism clients may have about MapServer being open-source. I'm not saying that bias is fair--it's just that perceptions matter a lot in corporate environments. So yes, very good news indeed. Brian |
||||||
|
||||||
| While internally at Autodesk this may have been known as the "next MapGuide", what is now known as MapServer Enterprise shares no code with MapGuide. It is a completely new architecture with an entirely web based API with 100% consistency between the Java, PHP and ASP.NET API's. As a long-time MapGuide developer and having been using this new software in beta and pre-release for a number of months, I can say that it is quite different (and better) than MapGuide. When people start to utilize this software, they need to evaluate it for what it is: a completely new product that happens to be made by Autodesk that is now being contributed as an open source project. |
||||||
|
||||||
| With Earth Resource Mapping funding the development of GDAL and Autodesk now involved with Mapserver Enterprize, the grass roots, results-oriented-days that Minn Mapserver is known for could well be gone forever. Just as the case with ESRI Desktop and Server software products, some Mapserver users will be overwhelmed with fuzzy product line distinctions ( cheetah vs enterprize). Instead of innovating solutions and providing them free to the GIS community ( ala dmsolutions and Refractions Research ), Autodesk's move to not put FDO database connectivity into their 'enterprize' opensource seems like a last ditch effort to get out of the software business they never dominated anyway. Instead of GIS users sending money to Autodesk in future, they should provide to Refractions Research for PostGIS development...we, as a community, would all be alot better off. |
||||||
|
||||||
| I'm not quite sure what you're really talking about. GDAL development, AFAIK, has always been funded by both proprietary and non-proprietary (gov't agencies, universities, other open source-based companies, etc.) entities. MapServer will continue to be MapServer, shaped by its community. As for the GDAL/MapServer demise, only time will tell although I think you're wrong. When put in the context of customized web mapping applications, the MapServer/GDAL/PostGIS combo is still hard to beat. |
||||||
|
||||||
| Re: GDAL I'm surprised at this take on the funding of enhancements to the GDAL libraries. Having a commercial entity pay for the development of a portion of code that will benefit the entire community is not a bad thing. Frank's got to live too. Other commercial entities such as Safe Software have paid for enhancements to GDAL/OGR that have benefited us all as well. Re: FDO I think that you have misinterpreted something here. Autodesk has open sourced FDO, and provided it in its software grant to the Mapserver Foundation. All it has apparently held back from this version of the code are the source for some specific FDO providers such as Oracle and MSSQL. There is nothing stopping the developer community from creating their own versions of these providers or, more likely, sticking with open source databases like MySQL and PostGIS. Other: I agree that it is important that our existing open source champions like Refractions Research are supported--though I'm not clear how you would see this happening, as third party funding of open source projects is apparently suspect--but I also see the need to encourage commercial entities to view open source as a viable business model. I hope the Mapserver folks don't take this the wrong way, but what I have seen so far of the Open Source MSE product goes way beyond much of what Cheetah can currently do, and having this available to the entire community is a huge benefit. Companies like DM Solutions and Refractions Research are primarily service-based. The MSE offering allows them to offer more competitive services using open source software. If you have not already done so, I would highly suggest downloading and evaluating the capabilities of MSE before discounting the importance of this grant to the open source community by Autodesk. In my opinion, this is a Win-Win solution. Jason (Opinions expressed are my own, not my employer's) |
||||||
|
||||||
| Autodesk has taken the Mapserver branding without really providing anything of long term value in return ( good PR move) and DMSolutions now can bow gracifully out of software development and make money with MSE implementation ( could business move ). Is the open source community really getting serviced here? No, Just two companies trying to capitalize on the open source branding and the mapserver branding. The foundation is a great idea...the top secret rollout of this one though makes the motives of Autodesk and DMSolutions appear unauthentic. |
||||||
|
||||||
| I understand why some people may have concerns about motives of Autodesk or DM Solutions, but fundamentally having the Foundation is good for all the MapServer users. For users and developers, let's give them the benefit of the doubt. If you question the intent of every foundation established, it is mainly FUD that drives it. For example, would you question Mr. Gates' intent in establish a foundation to fight against malaria and AIDS? Availability of more web-mapping technologies and having more choices represent a good thing. It is a gutsy move for all parties involved in the formation of the MapServer Foundation. This may be considered a hat-trick. At least in what we can see in the short term. It is a win for open source web mapping users and developers, a win for the MapServer Foundation, and a win for Autodesk and DM. Now the question is how do you measure success in the long term? For the Foundation, one of the most important criteria may be the number of new customers/users and developers joining the MapServer/MapServer Enterprise camp. For Autodesk or DM and others running similar businesses for profit, then it ultimately comes down to how much money they will be able to bring in at end of the month, or end of the year. How many new opportunities will it generate? Create a metric, start counting. If the Foundation becomes a playground where ESRI, Intergraph, MapInfo and the likes come to play, then the formation of this Foundation would be considered the foundation of the new web-mapping/Internet-GIS revolution. However, if no other gorillas come to play, then what happens here may be considered a market disturbance for the 800-pound gorilla. Question is, will they come to play? One of the questions posted in the "wishes" list is: "Would it make sense for Intergraph to explore using the FDO technology in GeoMedia?" It's funny to see such a question. Maybe FDO in GeoMedia would be the first step of a Autodesk/Intergraph union? Truth is that Intergraph is unlikely to take over ESRI as the number one player in GIS any time soon, nor can Autodesk do that. However, if there's any intelligence, then the AI union of these two CAD-GIS businesses would represent the most serious threat to ESRI. Fundamentally, the formation of the Foundation is good for its community. The spirit of the community and its people are reasons for hope of success for the Foundation. At the same time one should wonder if GIS folks are only looking at a portion of a bigger picture... Have we missed the cruise-liner that Google, Microsoft and Yahoo left on for web-mapping? True that they are serving different markets at this point. But with critical mass they are building, who is to say the 8-ft MapServer dingy or the 50-ft ESRI yacht is safe from the big waves of a cruise-liner? (dingy and yacht used for shake of discussion) The Internet Services Disruption email by Msft CTO is a good read. A copy available at http://www.scripting.com/disruption/ozzie/TheInternetServicesDisruptio.htm. Sam |
||||||
|
||||||
| Jim wrote: "Autodesk has taken the Mapserver branding without really providing anything of long term value in return ( good PR move)..." Have you actually downloaded the preview version of the new MapServer Enterprise software? Wow! I wholeheartedly agree with Jim's assertion that Autodesk is taking advantage of Mapserver branding. But I could not disagree more with his assertion that the company provided nothing of long-term value. The software is very impressive and easy to work with "out of the box". Moreover, somebody else noted that the Mapserver community has suffered from its own branding problem as a loose group of academics with only the semi-official backing of an academic institution. Give credit where it is due; Autodesk has a reputation for professional, quality software. The company's participation in the community gives us more credibility. If larger organizations begin to implement Mapserver "Cheetah" or Mapserver Enterprise because they know that Autodesk is a player and this gives the software more credibility, we will all benefit. Why? Because these organizations will contribute to the community and development of the software. And they will undoubtedly bring far more resources to this effort than even our first few years could. I encourage anybody who hasn't installed the preview version of the software yet to do so. So far, it has been fun for me to explore code that clearly comes from a crack team of engineers. Nothing of value? Look again. |
||||||
|
||||||
| The simplicity and popularity of Google Maps API/Earth has completely overshadowed the impact of Autodesk-DMSolutions-Intergraph VS. ESRI point made above. Even collectively, these over engineered software products do not capture the imagination of the public as Google does. Calling this foundation 'the beginning of a new web mapping/Internet-GIS revolution' is not at all accurate. However, a foundation is a great idea. As far as participating in revolutions, we may be better off reading the latest Che Guevara book or navigating around the globe in Open Source World Wind instead. |
||||||
|
||||||
| I have to agree with Jim on this one. It comes as a surprise to me that most of the posted comments here are in support of the Autodesk "brand". In a profession that is absolutely dominated by industry giants, I would think that the last thing the open source community would want is one of those giants calling alot of the shots. It seems to me that the risks involved far outweigh the potential advantages, at least for the Mapserver name. We have to remember that, bottomline, Autodesk is a company in business to make money. As Jim stated, I fear that they are doing nothing more than trying to capitalize on the open source branding. |
||||||
|
||||||
| I know how Autodesk works. I have to compare the Autodesk Empire to Wal-Mart, where the motto is Always screwed customers Always. AutoDesk along with ESRI are trying to capitalize on the little man and I think that sucks. Fokes its not going to be open, they are going to make it so hard to use you will have no option but to attend their little campus and learn high level programming. You think a major software company is going to give their $30,000 software out for free. I think open source needs to stay open source and not be influence (big $$s) by the giants of Autodesk or ESRI. I defiantly have to agree with Todd and Jim on this one. I hope the media gets wind on this. |
||||||
|
||||||
| John, Which part of open source do you not understand? IBM has done wonders with open source. So has Sun. Why should this be different? Anon |
||||||
|
||||||
| Maybe somebody (independent of AutoDesk) out there can clarify this for me. I just read AutoDesk's Dec. 2005 FAQ on Mapserver Foundation. It appears from the material I have reviewed that Mapguide Open Source will use DWF as the vector format to stream to a client component. Although AutoDesk calls the format open, the only way to access the data is through their API. They give appear to be currently giving the API for free for (for now at least), but the format itself is apparently not published. It is not clear if the client component that interprets the DWF and renders them is open source. For this the whole product to be open source – both the server and client components. Also, AutoDesk says that under the LGPL, you can “freely modify, extend, and improve, the MapGuide Open Source source code” but you can only do this through dynamic linking. Is this an open API or open source. I may have missed something here, anybody clear this up? |
||||||
|
||||||
| The MapServer Foundation is now known as the Open Source Geospatial Foundation: https://www.osgeo.org/ |
||||||
