Morten Nielsen read Refractions Research’s “The State of Open Source GIS” white paper and wondered why Paul Ramsey categorized open source GIS as either C or Java. I think Paul only wrote about the open source GIS clique that revolves around Refractions and OSGeo rather than the state of open source GIS as a whole.
About
James Fee - jfee@weogeo.com
James works for WeoGeo helping people organize, share and monetize their geo-content.
Connect
Subscribe
11 Comments
Clique indeed. The paper does recognize mapnik, GMT, and TerraLib as token outsiders, but that’s it. Are C# applications ignored? Yes.
On the other hand, it’s hard to take some of the C# projects seriously (as open source) when they are so plainly oriented towards a proprietary platform: Microsoft .NET.
Sean, saying you can’t take C# projects seriously is a slipperly slope when Java has a similar problem with their semi-proprietary platform. We are just getting dragged down with semantics for sure, but Java itself is not open source.
I’m comfortable on that slope
May be posting early since I havent read the article, but I know that I am not alone in wishing to use the OpenSource efforts along with .NET technologies. I’ve had pretty good luck (including typical OpenSource community support) using csharp_mapscript.
Also, it seems that while .NET is certainly MS targeted there is no reason the source of a solution cannot be open. I think there must be some happy medium for using proprietary products and still developing OpenSource code. ArcScripts seems to be an example of this.
James & Sean:
Actually, I had a different reaction when reading the document, namely about how difficult it is to write a “summary” report about Open Source GIS given the plethora of applications and the different languages and platforms. An interesting conundrum now is that when people go looking for info because they have a vague hankering to find out about “this open-source stuff”, there’s a large amount of information to digest–much of it technical. An unlike a proprietary software vendor, which can ostensibly train a sales force and keep them on the company message, whoever takes it upon themselves to spread the word about the multi-headed hydra that is open source will always be vulnerable to the favoritism criticism.
In short, I’m not sure who of us has the resources (time, for one) to craft the definitive document about “the state of open source GIS as a whole”…and that’s not necessarily a bad thing.
BT
I don’t disagree with your assessment Brian at all, the issue might be with the title of the whitepaper and the totalility of the writing. You’d think this was it for open source GIS.
Personally I enjoyed the report by Paul (I’m a sucker for this kind of stuff).
It is kinda of strange calling .NET proprietary, Mono is no longer a toy. It is a very serious Open Source version of the CLR runtime and the .NET Framework that runs on Linux, OSX and Solaris, desktop computers to Nokia PDA’s.
It is possible to build client side GUI’s or web apps. They track the ECMA spec pretty thoroughly and the community eats its own dogfood with 2 different IDE’s and plenty of active projects on the infrastructure side.
That doesnt sound proprietary to me.
As James also mentions, it is the “totality of the title” that upset me. If this document is made for guiding Refractions clients, this “gray paper” it is rather poor and narrow-minded. Not what I would expect from a thorough advisor. And Mono is completely disregarded in Sean’s comment, even though all the applications I mentioned actually runs on Mono as well.
If you want the totality of open source GIS, go to http://www.freegis.org, fill your boots. There needs to be a culling down to what is “important”, “useful” or “popular”, which is what I attempted. At this point it is fair to say the document is “poor” in that .Net is completely excluded. Since that exclusion is not by design, it is not fair to intimate the author is “narrow minded” or that the document is a “grey paper”. http://geotips.blogspot.com/
Hi Paul. I read your new post and I think I got your point, and I’m ok with that. Only thing is that this isn’t mentioned in the document, and as James Fee commented in your blog, people take what you write seriously (which is kudos to you). Of course .NET is a (kinda) new techonology and of course many of the projects isn’t that mature yet, but many of are already in use in many production environments.
Sean Gillies: > Are C# applications ignored? Yes.
I agree, .NET platform seems to be ignored. In my opinion, one of the most important reason is not that .NET is a “priorietary” platform, but that .NET is NOT portable as C/C++ or Java solutions.
I know, we have Mono but it doesn’t guarantee a portability by design, but by thirdparty hacks.
Java, C, C++ means portability by design / language standards…
One of the huge advantage of Open Source solutions IS portability. As long as Microsoft’s world of .NET is quite far from Open implementations, I can’t see any serious and multiplaform solution based on .NET.
Now, my understanding is that Microsoft develops .NET platform without looking at Open Source communities -> Mono. And, Mono is trying to catch up the Microsoft’s solution and provide some degree of portability. Without any collaboration between Microsoft and Mono, I can’t imagine it will work…
2 Trackbacks
DotNet,OpenSource,Microsoft,Java,GIS…
Java vs .NET: OpenSource GIS…