DEVONThink is great for getting electronic content into a single database. Love it. However, it needs the ability to scan traditional paper documents into the database with one click, like Kip (see thekip.com). This would make the application truly awesome.
Yes, I looked at Kip and thought its functionality was great but I really don’t want to have two different repositories for my resources. It would be great if DTP could duplicate this.
I love working with internal betas. One button, with OCR. But OCR isn’t free of course, as there’s more code in the “plugin” than in DT Pro.
Actually, with a ScanSnap scanner, one can already do 1-button scanning and transfer of image-only PDFs into DT Pro.
Personally, I don’t care about OCR. All I want to do is get scanned PDFs into my database.
May I ask why you don’t care about the textual contents?
My understanding is that DT indexs the content of PDFs any way. I don’t need to reuse the text. All I need is the PDF for reference.
I think I need to clear a misconception here. The PDF files created by the ScanSnap Manager consist of JPEG or TIFF files encapsulated in PDF. Therefore there is no text at all to be found in these files. Furthermore, interesting meta data such as the title, creation date, author, etc. are not present.
The new functionality that we will provide addresses all of these issues. All the OCR does is include an invisible layer of text over the original scan that can be indexed. The document still looks like the original scanned one.
That’s very cool and would be a welcome addition. Is this available now or in a future release. (Sorry if I missed this.)
This will be in an upcoming release. Details to be announced…
And I know you guys are working hard on the integrated-OCR, and I appreciate that you don’t want to pre-announce.
I have a time window in which I can upgrade Read-IRIS from version 7 to version 11 for $59. Soon, I’ll have to pay over $150 for the privilege. Without preannouncing anything… do you think the upgrade would be a good idea?
Stephen, I probably can’t give you a completely satisfying reply.
I’ve got ReadIRIS 11 myself, and haven’t used it since beginning beta testing on the version of DT Pro capable of one-button scan/OCR/save to DEVONthink database.
The OCR version will be a special edition of DT Pro that will cost more than the regular version of DT Pro, because of the additional development costs and license fees for the OCR engine.
Not every user of DT Pro may need the OCR feature.
It works beautifully with the Fujitsu ScanSnap scanner, for up to 50 1-sided or 25 2-sided pages in the feed hopper. It may not work as well with other brands of sheet feeder scanners in the fully automatic mode in which it works with the ScanSnap, as there are substantial variations among scanner brands. But the special edition will be capable of OCR/save to database from images produced by other scanners.
Can’t give you a definite date of release or upgrade cost for the special edition.
But if I were you (and you already have a ScanSnap) I would wait.
I was just about to buy a scanner online, when I remembered something about DT + ScanSnap.
In anticipation of this development, and considering I was going to buy a scanner anyhow, should I purchase the Fujitsu ScanSnap…or should I wait? Is there more than one make of it, a particular one which DT will only work with?
As far as I know there is only one model ScanSnap for the Macintosh. This is the one we will support. And from personal experience: it’s a great machine for document scanning.
I have the PC FI-5110 EOX2 version of the ScanSnap and use the latest Japanese driver to make it work on my MacIntel.
Aside from the inconvenience of having the menus all in Japanese (sigh!) it works great–though the bare driver doesn’t do OCR, obviously. Is the Japanese driver likely to work with the forthcoming OCR solution from Devon-Tech?
Update: now have English menus with the Japanese driver, thanks to this post:
Since the application can send its output to a third-party application, I see no reason why not.
Yes, the Japanese software should work as it is in most parts identical to the international version. The forthcoming ScanSnap S500 will use a localized driver that is similar to the former Japanese driver. Confusing? Yes But it should work.
It is a VERY nice machine. Well worth the dough, if you do much scanning.
I’ve been using a Canon LiDE 500F and it works great. Press one button on the scanner and it scans a pdf+text and sends it to the program of your choice. This file format means that you see the jpg image, but the OCR text is also included for searching and indexing. You can even select the OCR text by dragging the cursor over the appropriate words in the scanned image.
There is no document feeder add-on so it’s not for huge stacks of documents. Othere than that it’s about as convenient as they come. (I still miss my old PaperPort. Sigh.)
I consider Abbyy Finereader technology to be the best in the OCR field – though currently unavailable to MacIntel.
I note that Abbyy does have an OEM SDK with their latest technology for Mac OS-X, even though they don’t have an end-user application for Mac that uses it.
Here’s hoping we get that Abbyy technology in the new Devon-Tech product.
Honestly, we’ll get ReadIRIS technology as we have already an agreement with them because the Abbyy SDK was not available at that time. Technically, this doesn’t stop us from changing the OCR plugin at any time we think it makes sense, though.