Someday we have to figure out this app-death thing. There has to be a better way.
If anything, this highlights a problem with the model of SAAS in general: whether it be 3 years out, 10 years out, or 25 years away, these companies and the tools we might begin to depend on are not reliably sustainable.
Is there a better model? What if we all had our own private cloud servers, running apps that we purchased that communicated with each other independent of a central entity? What if they were as easy to set up as the current SaaS signup we know so well? What if it were even easier?
Personally I would love a personally-curated library of tools, on which I could control the end-of-life story. But this also brings up many new problems. New and interesting problems.
There is: Release a piece of software that runs locally. There's no reason document-organizing software needs to be in "the cloud".
One major reason SaaS is so popular is people can't figure out how to write programs worth paying for. If you wrap it up in some vague notion of a "service", customers assume they're getting some kind of ongoing value.
Doo was free. It's great in that it indexes and OCRs everything wherever it is located: you don't have to move the documents into the database like Evernote requires.
Here's the problem: the only revenue source was people who want to store their docs in the cloud. Keep in mind, that would be ALL your documents. That's a tough sell for me.
The thing is, obsolescence is an issue even for desktop software. The company where I work recently went through a fairly painful upgrade process for some software we were using that was no longer supported. We could have kept using it, but it was becoming more and more troublesome as the software failed to keep up with the times.
No matter what software you choose to use and no matter what model the company employs to provide the software, it's really important that the party behind the software is in it for the long haul, because otherwise you're going to have to deal with that.
Basically, look at the business model for the company. Does it look like one that will be around in a few years? It's usually not that hard to tell (in particular, things being given away for free like Doo tend to be transient).
FWIW I think this is the killer issue of this particular point in time. So far the only idea I can see all the pieces for is a new kind of appliance which is your personal "cloud." This would be analagous to an air conditioner or refrigerator in your house, but built with massively open sort of infrastructure (much like existing appliances use standardized fittings and filters and what not). Then into this home 'cloud' a platform for hosting "apps" which you buy and install, and basically run forever until you de-install them or replace them with something better. A server appliance for the rest of the world as it were.
Embrace creative destruction. Apps die and it gives new ones a chance to fill the space. Keeps the ecosystem responsive and gives everybody a chance to innovate.
Want apps to stick around? Enjoy Excel, Photoshop, etc. :)
Right about the time doo started we applied to YC with the similar idea and a working prototype. We were not accepted with the comment that they don't believe that dealing with documents is such a big problem for most people. We continued with the app, but eventually realized that was true. This kind of app would be nice for organizations dealing with tons of documentation - for example legal services, hospitals. But those organizations face their unique challenges and general purpose app will not work for them. Figuring out what would work for them was outside of our expertise and interest, so we shut down out project. Luckily we spent just half a year on it, not 3 years, like doo.
Incidentally, I checked out doo on OS X just few days ago, thinking it would help me with my OCR problem. It was a huge install and in the end it could not connect to my scanner. Not sure it was their problem per se, could be some peculiarity of my OS X setup, but I figured their main purpose was not what I wanted.
Here is what I want and it would be interesting to know how other people on HN solve that. I have no problem managing my digital documents, I have that covered. However paper (carbon-copy, dead-tree) - now this gives me a lot of headache. Even though I subscribed to electronic everything I could, I still many documents coming in paper: user manuals, medical test results, credit card agreements, some bills etc etc.
I hate managing and storing paper. I tried to setup a system that would scan, clean up, OCR documents, store searchable PDFs and (the key part) stored PDFs into folders (in my online backup provider's folder of choice) according to keywords or even performed some kind of machine learning classification (rent, medical, receipts, etc).
I had even created a concoction on Linux from SANE command line, ImageMagick, bunch of Python scripts that did that more or less, but once I moved to OS X that stopped working and I did not have time to port it yet.
On OS X, I use DEVONthink Pro Office with a ScanSnap iX500 scanner, and does pretty much what you're describing. OCR is really good.
While it manages the directory structure, all documents are stored "bare" in its filesystem folder. That is, you can create a Foo.pdf in DEVONthink and then access it from ~/Documents/whatever/PDFS/Foo.pdf (or however it names the directories).
Machine learning classifies incoming documents and can suggest relationships between them.
I'm not related to them, I promise. :-) I'd bought DT a while back but hadn't used it much until I got Doo as part of a software bundle. I played with it a while, realized I already had DT which was much more powerful, and started using it more.
Have you read "Paperless" by David Sparks? It addresses a lot of what you're asking for and has several suggestions for workflows that could achieve your goal.
> stored PDFs into folders (in my online backup provider's folder of choice) according to keywords or even performed some kind of machine learning classification (rent, medical, receipts, etc).
Sorry to piggyback: I have had a similar problem - hating paper, but drowning in it. I'm readying an OSX program to do this, if you are anyone else would be interested in trying it out/beta testing it (once it is ready! It isn't quite yet ready) in exchange for a free copy that would be awesome. So far designed to work for Scansnap or a workflow that OCR's the PDF's first.
Their execution history looks really problematic. Launching on Windows 8, then OSX, then Android, then iOS, then desktop Windows is probably the exact opposite of a successful roll out. If they were constrained to only releasing one platform at a time, they did it entirely the wrong way.
[+] [-] calinet6|12 years ago|reply
If anything, this highlights a problem with the model of SAAS in general: whether it be 3 years out, 10 years out, or 25 years away, these companies and the tools we might begin to depend on are not reliably sustainable.
Is there a better model? What if we all had our own private cloud servers, running apps that we purchased that communicated with each other independent of a central entity? What if they were as easy to set up as the current SaaS signup we know so well? What if it were even easier?
Personally I would love a personally-curated library of tools, on which I could control the end-of-life story. But this also brings up many new problems. New and interesting problems.
[+] [-] ForHackernews|12 years ago|reply
One major reason SaaS is so popular is people can't figure out how to write programs worth paying for. If you wrap it up in some vague notion of a "service", customers assume they're getting some kind of ongoing value.
[+] [-] cjensen|12 years ago|reply
Here's the problem: the only revenue source was people who want to store their docs in the cloud. Keep in mind, that would be ALL your documents. That's a tough sell for me.
[+] [-] cmer|12 years ago|reply
[+] [-] chc|12 years ago|reply
No matter what software you choose to use and no matter what model the company employs to provide the software, it's really important that the party behind the software is in it for the long haul, because otherwise you're going to have to deal with that.
Basically, look at the business model for the company. Does it look like one that will be around in a few years? It's usually not that hard to tell (in particular, things being given away for free like Doo tend to be transient).
[+] [-] ChuckMcM|12 years ago|reply
[+] [-] msutherl|12 years ago|reply
Want apps to stick around? Enjoy Excel, Photoshop, etc. :)
[+] [-] skolos|12 years ago|reply
[+] [-] mynegation|12 years ago|reply
Here is what I want and it would be interesting to know how other people on HN solve that. I have no problem managing my digital documents, I have that covered. However paper (carbon-copy, dead-tree) - now this gives me a lot of headache. Even though I subscribed to electronic everything I could, I still many documents coming in paper: user manuals, medical test results, credit card agreements, some bills etc etc.
I hate managing and storing paper. I tried to setup a system that would scan, clean up, OCR documents, store searchable PDFs and (the key part) stored PDFs into folders (in my online backup provider's folder of choice) according to keywords or even performed some kind of machine learning classification (rent, medical, receipts, etc).
I had even created a concoction on Linux from SANE command line, ImageMagick, bunch of Python scripts that did that more or less, but once I moved to OS X that stopped working and I did not have time to port it yet.
[+] [-] kstrauser|12 years ago|reply
While it manages the directory structure, all documents are stored "bare" in its filesystem folder. That is, you can create a Foo.pdf in DEVONthink and then access it from ~/Documents/whatever/PDFS/Foo.pdf (or however it names the directories).
Machine learning classifies incoming documents and can suggest relationships between them.
I'm not related to them, I promise. :-) I'd bought DT a while back but hadn't used it much until I got Doo as part of a software bundle. I played with it a while, realized I already had DT which was much more powerful, and started using it more.
[+] [-] ciniglio|12 years ago|reply
http://macsparky.com/paperless/
[+] [-] konradb|12 years ago|reply
Sorry to piggyback: I have had a similar problem - hating paper, but drowning in it. I'm readying an OSX program to do this, if you are anyone else would be interested in trying it out/beta testing it (once it is ready! It isn't quite yet ready) in exchange for a free copy that would be awesome. So far designed to work for Scansnap or a workflow that OCR's the PDF's first.
[+] [-] momchenr|12 years ago|reply
[+] [-] dublinben|12 years ago|reply
[+] [-] zAy0LfpBZLC8mAC|12 years ago|reply
[+] [-] dewey|12 years ago|reply
[+] [-] computer|12 years ago|reply
[+] [-] almosnow|12 years ago|reply
[+] [-] almosnow|12 years ago|reply