Scan Tailor: A Fussy Yet Effective Scan Scrubber

[Lazy_Kent]

http://www.linuxinsider.com/rsstory/71670.html

[daniel_reetz]

Good to see ST getting some publicity, but I have to disagree with some of the criticism -- especially the one about scanner acquisition. ST does not need to hook up to a scanner driver to be a killer app. That would add so many layers of bloat to a lean, mean app.

The author also makes it sound like the "few filters and tools" in ST are a bad thing - in fact, they are quite sufficient to do everything ST is designed to do. What do they want, a solarize filter? An inscrutable, eye-gouging interface like GIMP?

However, in any case, I am glad to see wider adoption and awareness of Tulon's excellent software.

[emmerick]

Good to see ST getting some publicity, but I have to disagree with some of the criticism -- especially the one about scanner acquisition. ST does not need to hook up to a scanner driver to be a killer app. That would add so many layers of bloat to a lean, mean app.

The author also makes it sound like the "few filters and tools" in ST are a bad thing - in fact, they are quite sufficient to do everything ST is designed to do. What do they want, a solarize filter? An inscrutable, eye-gouging interface like GIMP?

However, in any case, I am glad to see wider adoption and awareness of Tulon's excellent software.

I fully agree, the ST is the best program of post processing I've ever seen. The time you have the perfect running dewarping automatic then there is unbeatable

[rob]

Can you believe that the article author actually thinks that expanding Scan Tailor to be able to acquire images from the scanner and to do OCR would be a good thing? I thought the whole idea behind Linux and UNIX in general was to have tools that you could chain together. ST neatly fits the missing space between image acquisition and digital edition publishing. Why the heck would ST expand to fill the area that plenty of other software already fills?

[Anonymous1]

I find that having a toolkit is less bloat. I don't use OCR much, as my documents are in Slavonic, but I find ST a joy to work with. It would seem a bit overkill for it to scan, tailor, and bind all at once.

When I have to edit individual pages, I do it with GIMP (it's not too bad), not ST. When I want to make an epic book digital, I use ST. It's quite simple, and I don't think it's something you can criticize ST for.

And besides, Bindery binds DjVu files. Why re-invent the wheel yet again?

The thing that bugs me the most is that the author is portraying GIMP as having more features than ST. I don't see a "Split Page" or "Auto-Dewarp" script anywhere, let alone OCR. Some of the criticisms are completely irrelevant to both applications, like OCR.

The review is mediocre at best, and a bit outdated (Auto-dewarping is almost rock-solid. I haven't had a crash in a week!). But at least it gives ST some publicity

[jimboh]

ST is a fantastic program. Try using Scankromsator for comparison! <barf>

[paulica]

I started using ST for research purposes. While the article author may have his points, he has to understand that these kind of softwares are not designed for the masses. From my point of view, Scan Tailor benefits from feedback from a lot of users, while applications built by industry specialists have minimal feedback, since their customer list is generally very small. Not to mention the secrecy surrounding the book scanning industry.

What impressed me most is the way it recognizes borders and output delivery. This forum is sort of like the reasearch lab for ST, every question about using the software is a plus.