I've worked with PostgreSQL for a while now, but everytime I perform a new install (usually with PostGIS added on soon after the initial install) I am amazed at how simple it is and how smoothly it goes. In the past, I've worked on some pretty complex database installs - especially complex, at least for me, when installing to a Linux system - but it really only takes a few minutes to get PostGIS up and running on a database on any platform.

If you haven't checked them out, you should go download them both now. And if you want an open source desktop interface, there are a number of options that plug directly into PostGIS, including Quantum GIS and uDig. With them you can easily connect to a PostGIS database, and load (check out the Import Shapefiles to PostgreSQL plugin for Quantum GIS) and edit data. On top of this, you can greatly extend Quantum GIS' capabilities by using it as a graphical user interface for GRASS.

I agree with others that the major "gap" in the open source GIS arena is on the client side of things. At this point, there isn't an interface that rivals the ArcGIS Desktop suite. The products that are available, though, do serve a purpose. And if you're willing to put the time into learning and integrating them, they fill a huge niche that is left empty by ESRI.

As for the open source spatially-aware database, however, PostgreSQL and PostGIS offer a viable - albeit not as powerful - alternative to the expensive proprietary solutions. The two combined don't work for every scenario, but they do work for a surprising number of cases.

More to come on all of these solutions later...

DAZ has made their Bryce 5.0 3D modeling and rendering software available for free download through September 6th (their newest version is 5.5). This software package allows you to create 3D environments, including landscapes and animations, through a robust and fairly intuitive interface. You do have to register for a serial number, but, there is no charge for the software!

What's with all these 3D products? (Thanks, Roland, for the info.)

[Open Source Series: Part 1]

Lately, there has been a lot of talk about open source (including a great podcast by Scott Hanselman* about what open source really means). I don't know if it's the sudden overwhelming popularity of Ubuntu (which seems to be gaining more followers every day) and other open source software that's pushing this latest series of discussions, or just that people are finally starting to realize that there is truly an alternative to spending loads of money on software. Either way, I'm happy to see it. I've been a user of open source for sometime now, although I am basically locked into a Microsoft/ESRI world at work. I have a lot of fun experimenting with different applications (and since open source equals free for me to play with, I can experiment all I want), and find that "learning" how to use a new open source application or operating system - even though challenging at times - actually makes me a better overall user in the end. I know, I know, there is great community support available for most open source projects. Most of the time, though, I refuse to use the support. It's actually the same for me and proprietary software; if I can't figure out the solution to a problem on my own, the software doesn't work.

I've heard some say that today's interfaces don't require the end user to really know what (s)he is doing, and I, for the most part, agree. I don't want to hang on to the way things used to be - and I really can't (I'm not old enough ;-)), but I do think that users of technology, in general, are being "dumbed down" by the interfaces. Don't get me wrong, though - for most users this is an okay thing. It means that more advanced technologies are accessible to more people today than ever before. And it should be the goal of every developer to design applications that are accessible to as many people as possible. For those who want to work in technology fields, however, doesn't it make sense to really understand what's going on underneath the hood?

With all that out, I plan on writing a series of entries about my experiences with open source software. Check back for more...

*Note that the link from Scott's blog to his podcast points you to the wrong show. Follow this link for the correct one...

Introduction

Today, more than ever, the line that separates GIS dweebs from IT dorks is growing dimmer*. Whereas in the past users of GIS have had to be local "power users" who were proficient in command line and obscure software technologies (but not necessarily schooled in still-esoteric concepts and definitely not necessarily able to administer and optimize hardware infrastructure and networks), many of today's GIS users are expected to have a broader range of expertise and skills. It's happening on the other side, as well; spatial data and applications are being used more and more in traditional information technology fields to help streamline workflows and improve the efficiency and operations of organizations of all sizes.

Why the convergence? Well, if you look at the big picture, GIS is really just a specialized subset of IT. Here are the definitions, from Wikipedia (emphasis mine):

Information Technology - a broad subject concerned with technology and other aspects of managing and processing information, especially in large organizations.

Geographic Information Systems - a system for creating, storing, analyzing and managing spatial data and associated attributes.

This is not rocket science; while IT is considered to be a broader, all-encompassing field that uses technology to manage and process information (or data), GIS fits into a smaller (and much more specialized) niche. They are, though, at their most basic level one and the same. GIS departments have traditionally remained separate from IT departments, but I see this beginning to change.

Following are some examples of the skills/technologies that fall into the area of overlap, or "intersection" (as it is called in a Venn diagram). And you'll notice when reading the list below that this area of "intersection" is huge - and growing larger all the time.

• Database - this is the most obvious of them all. GIS data are happier in databases. File-based structures, albeit in some ways easier to use and administer, place severe restrictions on what can be done with the data. When stored in a database, though, the only restrictions imposed on data are those brought on by the database administrator's abilities - or lack thereof. A good GIS user needs to not only understand how data are stored in a Relational Database Management Systems (RDBMS), but should also be able to design and optimize a database using accepted standards and techniques. More and more data are being moved from file-based formats into databases, and those who are able to efficiently and effectively use these data will have a significant advantage in leveraging the power of GIS for their own needs.
• Hardware - centralized GIS allows for more efficient allocation of resources and simpler overall maintenance and administration. However, centralization has its difficulties as well. It can put quite a load on finite expensive and mission-critical hardware resources, including memory, processing capacity, etc. A GIS user who understands the relationship between all of these different variables and is able to get the most out of existing resources can get more out of a GIS than one who only understands the software. My prediction: in the coming years, basic working knowledge of IT hardware will be a prerequisite for employment in many fields of GIS.
• Network - with the recent proliferation of network and web-based mapping/spatial technologies, today's GIS user needs to know more about how networks work than a user did even just a couple of years ago. Optimizing applications and services can allow a GIS user to get more out of their network, and can dramatically improve the end-user's experience (in the case of data served over a network or through a web application or service).
• Programming - in the GIS profession, scripts have long been used to extend the functionality of software, and Structured Query Language (SQL) has been used (often in a much simplified manner) to query out data since the inception of GIS. The new allowances of today's technologies, however, make it possible for a GIS user to make his/her data available to a much larger audience - if they can point the data in the right direction and give it a little nudge. Although today's development tools are advanced and user-friendly, there is still a need to know how to write basic scripts to extend interfaces and assist in repetitive tasks. One biggie that all GIS users need to catch up on is XML. With it one can do it all - from documenting their data to connecting two disparate databases that sit on two different platforms.
• Web Design - basic, working knowledge of web design principles and techniques (especially basic CSS and HTML) are helpful to anyone who needs to make any information or services available on the web. And, as I talked about in the "network" section above, the web is the future of GIS.

Conclusion

What does this mean for users of GIS? Well, in my opinion it means that we have an exciting future ahead. A GIS user who understands the fundamental technologies and services that drive IT - in addition to having a passion for spatial technologies - can expect to have only the sky as the limit.

*Note: I use the term "dweeb" lovingly, as I am a proud dweeb myself.

Listening to: Iron & Wine - Our Endless Numbered Days...

As some of you may know, some antivirus programs recognize the RegObj.dll file (C:\Windows\System32\RegObj.dll on most systems) as malicious. This file is required by ArcGIS and, if deleted, ArcGIS' self-repairing feature is triggered, causing a very annoying delay while the install completes. ESRI's solution for the problem is to create a file exclusion in your antivirus software for the RegObj.dll. We use Symantec antivirus, and I personally set up the exclusion on our antivirus server. This hasn't, though, solved the problem. Yes, the repair install doesn't occur as often as it did, but it still happens.

I came across a couple of messy, but doable, workarounds in the ESRI support forum today, and am hoping that these will take care of the problem forever! The first solution is as follows:

• Copy the RegObj.dll file from the \System32 folder to another folder on your harddrive (e.g. C:\Workspace). Symantec only deletes the .dll if it resides in your \System32 folder, so it should be safe here.
• Create a simple batch file that moves the RegObj.dll from the C:\Workspace folder to the \System32 folder.
• To create the batch file, right-click in the directory where you want it to reside and select New>Text Document in the context menu (picture illustration).
• Assuming that Windows is installed to your C:\ partition, open up the new text file and paste in the following line:
• copy "C:\Workspace\RegObj.dll" C:\WINDOWS\system32\RegObj.dll /Y
• Rename the text file to YourName.bat. You will have to confirm that you want to change the file extension by clicking "Yes" (picture illustration).
• Next, point to Start>Run and type in "Tasks". Press enter.
• Double-click on "Add Scheduled Task", and the "Scheduled Task Wizard" dialog will open. Set the scheduled task parameters as follows:
• Click on "Browse..." and browse to the location of the batch file that you just created. When you get to it, double-click on it.
• In the next window, choose to "Perform this task" Daily and click "Next".
• Select the task "Start time" (I chose 12:01 AM). Choose to perform this task "Every Day" and specify the task "Start date" (today's date).
• This is important. Specify a valid username and password in this dialog - the task will run as if started by this user. This can be tricky if you authenticate against Active Directory. If you know the credentials of a local administrator, use them (as you won't have to worry about forgetting to go back in and reset the password when an Active Directory password expires).
• At the next dialog, check the "Open advanced properties..." box and click "Finish".
• When you get into the advanced properties dialog, click on the "Schedule" tab at the top and push the "Advanced..." button.
• In the "Advanced Schedule Options" dialog, check the "Repeat task" box and choose to repeat the task at a set interval (I chose 5 minutes).
• Chose the "Time:" radio button and set the scheduled task end time (I chose 11:59 PM) and click "OK".
• Click "OK" to close out the other open dialog and you should be good!

The second solution uses the same method but doesn't have to run as often, as it unzips and then registers the RegObj.dll file before moving it into the \System32 folder. I'm not going to document this method, as it has already been done. I will, however, point you in the right direction (zip uploaded to the ESRI Support Forum by Tunrayo Alabi, here's a link to the full discussion).

Digital Data Services, Inc. (DDS), a company based in Denver, Colorado, put together a very helpful ArcGIS Reference Card for the 2006 ESRI User's Conference.

It includes common map scales, a UTM diagram of the continental United States, ArcMap editing and troubleshooting tips, and some basic Visual Basic functions and is available for download as a .pdf at the Neer Cartography website. This could be especially useful to those who only occasionally dabble in the ArcGIS Desktop environment.

Microsoft released Windows Live Writer this past week, and I thought I would give it a try (like everyone else in the blogging world, it seems). Here are my initial impressions:

• The interface is more intuitive and less complex than my current blogging tool, w.bloggar. This doesn't mean that it isn't as powerful, though; it seems to have all of the features that I used with w.bloggar - and looks better doing it.
• Connecting to dasBlog to post was almost too easy. Live Writer supports the new Really Simple Discoverability (RSD) standard, which means that for those blogging engines that support the standard, connecting and posting is as simple as pointing the interface your blog's URL. At 1.8, dasBlog doesn't support RSD, but it will in 1.9.
• The availability of plugins and an API should be a huge boon to the Live Writer project. Omar Shahine has already got a couple of plugins up his sleeves that I am excited about, and I plan on trying the "Currently Listening" plugin later this week.
• Another strength that I see in Live Writer is its image and mapping capabilities. You can add images, change the layout and image properties (including adding custom watermarks), and edit images without leaving the blogging window. The image editing capabilities remind me a lot of Picasa. Adding custom maps from the Windows Live Local service is just as easy, and is a real asset to the application.
• The ability to preview your post - CSS and all - from within the Windows Live interface is outstanding. Whereas before I often had to publish the post and then go into the built-in editor to correct formatting errors, now I can just post and go.

Overall, I am - as you can see - very impressed with Live Writer. It has the exact capabilities that I need, and nothing that I don't. I really look forward to using some of the plugins that will undoubtedly come out of the API.