Get the Doxie scanner http://getdoxie.com/ ! It is fast and small, and great for scanning a few documents every now and then. It comes with software that is nicely integrated with the most common cloud services etc.
For me, my "paper load" was not big enough to warrant a bigger purchase of a multiple-page scanner, so if you are still in the paper less minor leagues like me, it is a great choice.
As far as I know there's no Linux support for the Doxie, so you're out of luck there. The ScanSnaps, while being quite a bit more expensive and bigger do have a decent support.
If we're only talking about the odd invoice, any normal scanner would do, too. If you get one of those printer/scanner combinations, you'll also save some space.
Yes, the app situation for Linux isn't that good. The Evernote web app alone isn't as comfortable as a native application, and I haven't seen anything decent out of the open source sector.
I'd recommend going a bit more low-tech. Create a script that invokes the "scanimage" command with the desired parameters, then moves it to your "cloud" directory (Dropbox, Ubuntu One etc.). There, run OCR software over it, dumping the result in a file. If nothing major has changed, tesseract would be the free OCR software of choice.
For simplicity's sake, I'd recommend that the script automatically chooses the file name, you just add some "tags", e.g. if your script is called "scan_stuff", you'd invoke it as "scan_stuff amazon invoice", and the result would end up as "~/Dropbox/scans/2011-09-13-14-46-amazon-invoice.pdf", alongside a "2011-09-13-14-46-amazon-invoice.txt" file generated from that. This makes it easy enough to search by date or by a rather generic type.
The SANE scanimage command supports batch mode, even if you don't have an automatic feeder (--batch-prompt).
This might not be as automatic as some of the specialized solutions for Windows or Mac, but it does handle special cases a bit better. You can just have a parameter or separate script for that and don't have to mess with Preferences each time something slightly different comes in (e.g. color, higher dpi, different target directory…).
I have a whole bunch of invitations - however, I will be leaving town for the weekend and will probably be without internet connection. So if anyone is late on the train but still interested, leave a comment with an email here and I will try to get them to you on monday as I will check the thread as soon as I get back!
The idea of that site is great, thanks! I just started using it. I plan on writing a few sentences as well as a #hashtag containing my general mood for the day. After a few months, I plan on plotting this and seeing what moods were more common when and why.
Yes – programming is fun to hop into, but just a heads up: the most difficult process to learn and master is the marketing and promotion part of releasing an app.
That said, I would recommend Corona - http://www.anscamobile.com/corona/ - for anyone wanting to give iPhone app development a shot. Much easier and fun to jump into than objective-c, especially if you want to make games, and still pretty damn powerful!
It should be noted that the Corona SDK uses Lua on top of Objective-C and is geared for Game apps. If you already have some Objective-C or C knowledge I'd recommend Cocos2d-iphone, which is nearly pure Objective-C, with a little pure C in the rough parts.
Going the Cocos2d route will also help you learn some of the peculiarities of the platform, like memory management and object lifecycle, while giving you enough skills to jump right into pure CocoaTouch.
Nothing against Corona, it has it's place. Bubble Ball[1] was developed with Corona for example.
Really not sure I agree with that. Coding is tough, and in my experience developing iOS apps well (and Cocoa apps in general to a lesser extent) is one of the most difficult forms of programming.
And as for marketing and promotion, I'd say even more challenging is coming up with an idea which doesn't need marketing or promotion :)
Yes, fair enough – I basically agree. Just posted that comment as some food for thought. I think this link was a good read, but perhaps should mention the marketing part of app production as well, if not just coding for coding's sake is the purpose. The reality is not really "you build it and they will come", which is a common mindset that might get people disappointed in the long run.
And yes – the biggest challenge is probably coming up with something original!
Amen to that! It's creating a killer app, by coding, which makes an app worthwhile enough where marketing and promotion almost become moot. I'm currently trying to update my own app after outsourcing most of the code. It's good, but I need other things that people want to talk about. That's what separates apps that sell, and those who don't.
For me, my "paper load" was not big enough to warrant a bigger purchase of a multiple-page scanner, so if you are still in the paper less minor leagues like me, it is a great choice.