Here, here. I agree as a now somewhat initiated n00bie. However, Mint has it's own quirks as well. Either one. Avoid Unity though (the new default interface for Ubuntu 11.04). Also, somebody else suggested Edubuntu which is novel, I suppose, but there is indeed a different sort of learning curve for it as well, as the interface definitely does not behave like a traditional OS interface. So if you want to experience something that is somewhat easy, but really different -- Edubuntu.
If you want just plain easy -- Ubuntu 11.04 with the classic interface or the latest Linux Mint.
So yea, I was initiated to the internet via IO.com. And I too used OpenVMS Vax at school before I even knew what the "Open" part even meant. I just wish I had stayed CLI like the rest of you geeks. I struggle to even configure a.profile anymore and regular expressions are not regular to me at all. One thing I do remember that seemed very much a part of the secret-code-ring-key exciting mystery thing of it at the time was the "dot plan" [.pln ] Remeber those ? Everybody had one in the root of thier public account and it was where the coolest and most obscure and frightening ASCII art was paraded. As I remember, Bob Dobbs and Barney the Dinosaur were by far the most popular frameworks for the stram of conssiousness manifestos contained in any groover worth their salt with a "plan".
Does anybody do those anymore ? I wish I still had a copy of mine to be sure.
Oh, and what about that FNORD guy ? Is he really a would-be amature "art" photographer or is he actually, you know, OKOP ?...just wondering...
Wow. Pretty good response, eh? Here is my suggestion:
1) Consider focusing on C and Python as well as or instead of C++
2) Go crazy at Sugarlabs as they are always looking for help.
Let me offer the disclaimer that I know nothing about programming or computers or git or subversion. However, I am kind of kinky and like to watch people. As such, I have noticed that a number of people that seem inappropriately comfortable with mathematics also seem to tend or bend toward C and Python. As for Sugarlabs, it would seem that they are already comfortable with the idea that everything is everything and that would include math, so you might feel at home....or not. Whatever. Anyway, I couldn't help myself.
Good stitching software doesn't distort the image, it actually un-distorts the image. As long as the camera is parallel with the maps, the only distortion in the photographs is actually created by the lens in the act of taking the photograph, lens distortion aka barrel distortion. You have have a circular convex lens producing a rectangular picture. In order to do this, everything not at the center of focus is increasingly distorted as you move farther from that center of focus.
Good stitching software actually undoes this distortion creating a linear image again before doing the stitching. So what you may deem as the software distorting the image, is actually the software correcting the image. Bang for buck, a high res camera taking as many photos as you can stand to take of a single map then stitching them, is the best way to go.
wow, that is like really getting down to brass tacks with the consideration of nadir convolution and so on. dude, you rock! hey, perhaps giving the ASPRS a ring and getting this done right is a good idea. It could be a nice vehicle for an article on process, eh ?
yea, what keosak said. avoid any advice that includes the term "geocoding" as that is not what is going on here at all./* ROTFLO! */
Basically the scan can be accomplished entirely separately from the georeferencing process. You could even do that at, say, Kinko's or something and then bring the.tif into some GIS software and get busy doing what amounts to a pin, stretch, and refine process where your goal is to minimize the RMS for the most important areas or just overall.
Like this guy says, there are strong advantages to putting it in the JPG 2000 format. You might also consider porting it to the new GeoPDF thingy that USGS is now using as a method for topo distribution. If you really wanted to go out on a limb, you could reproduce them and embed anoto in the copies and do fun stuff in situ there. In any case, until you get them reasonably georeferenced, you won't need to worry about the original projections at all for any reason, period. That simply doesn't apply here for a number of reasons that include but are not limited to: a) scalar accuracy of the originals b) datum and/or projection of base reference c) allowable RMS error in an affine transformation will certainly allow for any wack in projection hocus-pocus. That said, it might be interesting to actually know it such that you could get a quantitative handle on the accuracy/ precision of the originals.
If you do get around to it, please let us know on SlashGeo as lots of us folks like old maps......geocoding.....lol
I am the n00b here and I agree with this 100%. Only I will struggle with Ubuntu for now, maybe Mandriva soon, because when GOOgle does do it, I want to be ready to get a job doing something besides GIS (read: shoeshine boy of the IT community). I want a chance of payback to all those people that have helped me to understand that GIS is mostly a vapor career.
If I may, please let me weigh in with another kind of reaction. First, I had the same question you posed, only it was for me and not somebody else. My wife and I share a PC so I couldn't touch that, but I was ready to wipe Bill of my Dell D620 and get into a new hobby and learning curve, right. But I wanted to start out with something for folks like me, rather than say, Slackware, right?
First, let me speak to the idea of "Not to be too mean, but what's so hard about getting linux to run properly on a PC?" As a n00b with some recent experience, I am compelled to speak frankly to this:
1) Getting Linux installed on a PC is definitely easier than a laptop to be sure.
2) Many folks take it as a given (perhaps even Mr. Hudson, I don't know) that one is installing Linux on a PC or laptop that has a reasonable connection to the web (e.g. something other than dial-up). Meanwhile, the fact is that any sort of fast(er) connection than dial-up is simply not available to some of us. Let me put that in perspective. I live in the city limits of Auburn, CA at 112 Porter Lane and commute to the capitol of California, Sacramento, every day. 30 miles. Where I live there is nothing available, period. And I am not alone. This is true even though the Obama administration and the Schwarzeniggar administration go on and on about broadband and so forth. Now, I have found that most Linux types assume that the first thing you do when you install and get logged in is connect and update your packages, right? Well, guess what? Not gonna happen. Not with dial-up, my friend!
3) Though the "in-crowd" may understand that you need to create a distro from a torrent, because the official download is likely to get corrupted and won't load properly when you create the.iso on the DVD from the download, this isn't obviously publicized. This, regardless, of PC or laptop installation.
4) There are some "notorious" WiFi devices that "have issues" and even continue to stump the "experts"; my Broadcom 4311 comes to mind.....
That being said, my laptop is now amongst you guys. And it even happened with the native Broadcom 4311, despite what a kajillion blogs and "support" sites contend. The big trick was getting a cat5 cable connected to it that had some fresh informational juice to pump into it. That was the hardest part to be sure. I mean, there is no place to "plug in" at Starbucks, right?
Anyway, back to the original question. After doing some research on the web, I found that Ubuntu/Gnome and Mandriva/KDE appeared to be two strong candidates. This, with all the talk and strong community "support". I had decided to go with Mandriva because I thought the graphics looked cooler, there was not so much "cult hype" as with Ubuntu, they were based in Paris/Rio, which seems cool, and I was still pissed at some misleading advice offered by "experts" on their site.
Well, it turns out I wound up going with Ubuntu anyway because there is a guy I found nearby me who is a real Linux guru and a nice guy and uses Ubuntu [ www.computer-shoppe.net ] so I figured I would just get what he has to make my configuration questions all that much easier.
Bottom Line from where I sit/stand:
1) Biggest factor for which I chose was about where I could get the best *real* support.
2) You are kidding yourself if you think that any flavor of Linux is ready for the Windows consuming public. It is still either a hobby or a hands-on *supported* tool.
3) Before you count it out, go take a look at the new KDE Mandriva ----cool!
4) Isn't OpenSUSE Novell? Isn't that sort of anti the whole point of being a hobby rebel?
Slashdot Mirror
Here, here. I agree as a now somewhat initiated n00bie. However, Mint has it's own quirks as well. Either one. Avoid Unity though (the new default interface for Ubuntu 11.04). Also, somebody else suggested Edubuntu which is novel, I suppose, but there is indeed a different sort of learning curve for it as well, as the interface definitely does not behave like a traditional OS interface. So if you want to experience something that is somewhat easy, but really different -- Edubuntu. If you want just plain easy -- Ubuntu 11.04 with the classic interface or the latest Linux Mint.
So yea, I was initiated to the internet via IO.com. And I too used OpenVMS Vax at school before I even knew what the "Open" part even meant. I just wish I had stayed CLI like the rest of you geeks. I struggle to even configure a .profile anymore and regular expressions are not regular to me at all. One thing I do remember that seemed very much a part of the secret-code-ring-key exciting mystery thing of it at the time was the "dot plan" [.pln ] Remeber those ? Everybody had one in the root of thier public account and it was where the coolest and most obscure and frightening ASCII art was paraded. As I remember, Bob Dobbs and Barney the Dinosaur were by far the most popular frameworks for the stram of conssiousness manifestos contained in any groover worth their salt with a "plan".
Does anybody do those anymore ? I wish I still had a copy of mine to be sure.
Oh, and what about that FNORD guy ? Is he really a would-be amature "art" photographer or is he actually, you know, OKOP ? ...just wondering...
Wow. Pretty good response, eh? Here is my suggestion: 1) Consider focusing on C and Python as well as or instead of C++ 2) Go crazy at Sugarlabs as they are always looking for help. Let me offer the disclaimer that I know nothing about programming or computers or git or subversion. However, I am kind of kinky and like to watch people. As such, I have noticed that a number of people that seem inappropriately comfortable with mathematics also seem to tend or bend toward C and Python. As for Sugarlabs, it would seem that they are already comfortable with the idea that everything is everything and that would include math, so you might feel at home....or not. Whatever. Anyway, I couldn't help myself.
Good stitching software doesn't distort the image, it actually un-distorts the image. As long as the camera is parallel with the maps, the only distortion in the photographs is actually created by the lens in the act of taking the photograph, lens distortion aka barrel distortion. You have have a circular convex lens producing a rectangular picture. In order to do this, everything not at the center of focus is increasingly distorted as you move farther from that center of focus. Good stitching software actually undoes this distortion creating a linear image again before doing the stitching. So what you may deem as the software distorting the image, is actually the software correcting the image. Bang for buck, a high res camera taking as many photos as you can stand to take of a single map then stitching them, is the best way to go.
wow, that is like really getting down to brass tacks with the consideration of nadir convolution and so on. dude, you rock! hey, perhaps giving the ASPRS a ring and getting this done right is a good idea. It could be a nice vehicle for an article on process, eh ?
yea, what keosak said. avoid any advice that includes the term "geocoding" as that is not what is going on here at all. /* ROTFLO! */
Basically the scan can be accomplished entirely separately from the georeferencing process. You could even do that at, say, Kinko's or something and then bring the .tif into some GIS software and get busy doing what amounts to a pin, stretch, and refine process where your goal is to minimize the RMS for the most important areas or just overall.
Like this guy says, there are strong advantages to putting it in the JPG 2000 format. You might also consider porting it to the new GeoPDF thingy that USGS is now using as a method for topo distribution. If you really wanted to go out on a limb, you could reproduce them and embed anoto in the copies and do fun stuff in situ there. In any case, until you get them reasonably georeferenced, you won't need to worry about the original projections at all for any reason, period. That simply doesn't apply here for a number of reasons that include but are not limited to: a) scalar accuracy of the originals b) datum and/or projection of base reference c) allowable RMS error in an affine transformation will certainly allow for any wack in projection hocus-pocus. That said, it might be interesting to actually know it such that you could get a quantitative handle on the accuracy/ precision of the originals.
If you do get around to it, please let us know on SlashGeo as lots of us folks like old maps... ...geocoding.....lol
I am the n00b here and I agree with this 100%. Only I will struggle with Ubuntu for now, maybe Mandriva soon, because when GOOgle does do it, I want to be ready to get a job doing something besides GIS (read: shoeshine boy of the IT community). I want a chance of payback to all those people that have helped me to understand that GIS is mostly a vapor career.
If I may, please let me weigh in with another kind of reaction. First, I had the same question you posed, only it was for me and not somebody else. My wife and I share a PC so I couldn't touch that, but I was ready to wipe Bill of my Dell D620 and get into a new hobby and learning curve, right. But I wanted to start out with something for folks like me, rather than say, Slackware, right? First, let me speak to the idea of "Not to be too mean, but what's so hard about getting linux to run properly on a PC?" As a n00b with some recent experience, I am compelled to speak frankly to this: 1) Getting Linux installed on a PC is definitely easier than a laptop to be sure. 2) Many folks take it as a given (perhaps even Mr. Hudson, I don't know) that one is installing Linux on a PC or laptop that has a reasonable connection to the web (e.g. something other than dial-up). Meanwhile, the fact is that any sort of fast(er) connection than dial-up is simply not available to some of us. Let me put that in perspective. I live in the city limits of Auburn, CA at 112 Porter Lane and commute to the capitol of California, Sacramento, every day. 30 miles. Where I live there is nothing available, period. And I am not alone. This is true even though the Obama administration and the Schwarzeniggar administration go on and on about broadband and so forth. Now, I have found that most Linux types assume that the first thing you do when you install and get logged in is connect and update your packages, right? Well, guess what? Not gonna happen. Not with dial-up, my friend! 3) Though the "in-crowd" may understand that you need to create a distro from a torrent, because the official download is likely to get corrupted and won't load properly when you create the .iso on the DVD from the download, this isn't obviously publicized. This, regardless, of PC or laptop installation.
4) There are some "notorious" WiFi devices that "have issues" and even continue to stump the "experts"; my Broadcom 4311 comes to mind.....
That being said, my laptop is now amongst you guys. And it even happened with the native Broadcom 4311, despite what a kajillion blogs and "support" sites contend. The big trick was getting a cat5 cable connected to it that had some fresh informational juice to pump into it. That was the hardest part to be sure. I mean, there is no place to "plug in" at Starbucks, right?
Anyway, back to the original question. After doing some research on the web, I found that Ubuntu/Gnome and Mandriva/KDE appeared to be two strong candidates. This, with all the talk and strong community "support". I had decided to go with Mandriva because I thought the graphics looked cooler, there was not so much "cult hype" as with Ubuntu, they were based in Paris/Rio, which seems cool, and I was still pissed at some misleading advice offered by "experts" on their site.
Well, it turns out I wound up going with Ubuntu anyway because there is a guy I found nearby me who is a real Linux guru and a nice guy and uses Ubuntu [ www.computer-shoppe.net ] so I figured I would just get what he has to make my configuration questions all that much easier.
Bottom Line from where I sit/stand:
1) Biggest factor for which I chose was about where I could get the best *real* support.
2) You are kidding yourself if you think that any flavor of Linux is ready for the Windows consuming public. It is still either a hobby or a hands-on *supported* tool.
3) Before you count it out, go take a look at the new KDE Mandriva ----cool!
4) Isn't OpenSUSE Novell? Isn't that sort of anti the whole point of being a hobby rebel?