Kobo future firmware feature request thread

[Francois_C]

I like your post, winsomnia, and I agree with it, except for this assertion

Originally Posted by winsomnia

Car makers don't try to emulate real horse rides by making vehicles drive as bumpy as possible.

This 1894 car is just like our primitive ereaders.

There is a sort of nostalgia about old paper books, which I often hear from people who seldom or never read, about the touch and smell of paper and such nonsense. Why not make a book-smelling paper-touch reader?

PS: About margins, my opinion is also the same as yours: first (with a black Kobo) I thought the wisest was to ue the full screen surface, then I noticed that letters close to the border were less legible, and I adopted a thin margin. However, in Amanuensis, which I mostly use to make or rebuild books, I set margins to 0, as margins are adjustable on Kobo, and a minimum margin is just a limitation to the reader's freedom.

[speakingtohe]

Originally Posted by Francois_C

I like your post, winsomnia, and I agree with it, except for this assertion

This 1894 car is just like our primitive ereaders.

There is a sort of nostalgia about old paper books, which I often hear from people who seldom or never read, about the touch and smell of paper and such nonsense. Why not make a book-smelling paper-touch reader?

PS: About margins, my opinion is also the same as yours: first (with a black Kobo) I thought the wisest was to ue the full screen surface, then I noticed that letters close to the border were less legible, and I adopted a thin margin. However, in Amanuensis, which I mostly use to make or rebuild books, I set margins to 0, as margins are adjustable on Kobo, and a minimum margin is just a limitation to the reader's freedom.

I think we are pretty free overall.

[thymaster]

A couple of simple feature request that is present on my 3 years old Sony eReader and not yet on the Kobo.

1) Lock screen with PIN unlock
2) Customize my own personal screen saver (slideshow option would be nice too)

[dmapr]

Originally Posted by DNSB

Given that epubs don't have fixed page sizes unless you have one of the few fixed layout/non-reflowable ebooks (possible under the EPUB3 standard but the few samples I've seen are not very usable and have a very poor user experience much like trying to use pdf files on an small eInk screen), showing page numbers in the table of contents would be very, very messy to implement. Even using the Adobe pseudo page number algorithm would be messy since that would require scanning the entire file and then somehow adding the page numbers to the table of contents. Just to add to the complexity, are we talking about an embedded table of contents or the toc.ncx file? You can use either with a Kobo ereader.

Regards,
David

I was talking about the toc.ncx one, and I'm happy with the pseudo page numbers that it shows in the footer. It has nothing to do with accuracy and has everything to do with my reading habits. I don't like to stop mid-chapter. I know my reading speed so even with the pseudo-numbers I'll know how long it'll take me to get through to the next chapter if I know how many "pages" it is. Since I often read in the situations where I'll need to stop reading in X number of minutes, I currently use next/prev chapter jump to see the page numbers to see if I should start the next chapter or not. Seeing them in the TOC would be more convenient.

[lsmod]

I want to say Thank You for your work at the new Firmware 2.6.1 !
You implemented most of my wishes like a QWERTY keyboard and it works almost good.

My last really needed wish is that the browser can work in landscape mode.
It is nearly not usable showing webpages only upright.

Another fine feature would be to have an additional filter for the books.
There is enough space for about 1000 books in the internal flash and you got crazy to search one when you have more than a hundred installed.
A filter to show only the bools that are not applied to an bookshelf would help.
Or you give the information of the stored path in the details and filter list.
Then it would be possible to use a path structure.

My dream would be that you give the sources to implement a display driver that allow to implement other applications easy, but i think it is not in the interest of the corporation(s) you are working for.

There are some new bugs / problems i want to report here:

1. In some epubs i can't modify the line spacing
2. Pages are not shown completely in epubs.

I attached a sample technical document where you can see this problems.
It is generated by saving the page with the Firefox of this URL: http://www.mikrocontroller.net/artic...R-GCC-Tutorial.
Then i converted the local HTML with Calibre to EPUB.
So you have the same problems with other documents converted with Calibre.
It is really fine that also tables are shown on the Kobo correct, but it shows only the beginning of the page and then jumps to the beginning of the next page when you skim.
Intresting is that it leaps to the correct internal page number.

[davidfor]

Originally Posted by lsmod

There are some new bugs / problems i want to report here:

1. In some epubs i can't modify the line spacing
2. Pages are not shown completely in epubs.

I attached a sample technical document where you can see this problems.
It is generated by saving the page with the Firefox of this URL: http://www.mikrocontroller.net/artic...R-GCC-Tutorial.
Then i converted the local HTML with Calibre to EPUB.
So you have the same problems with other documents converted with Calibre.
It is really fine that also tables are shown on the Kobo correct, but it shows only the beginning of the page and then jumps to the beginning of the next page when you skim.
Intresting is that it leaps to the correct internal page number.

The line height problem is because of the style in the book. Each page is actually a table. The actual text is within a td tag that has style "calibre19". This is defined with "line-height: 1.45". The reader app does not override this. Simply removing this from the style definition will allow the line heights to be set. But, it might mess up the layout.

The incomplete pages looks to be a layout issue. There are levels of div's and tables around the actual text. When I stripped those out for a couple of chapters, the pages flowed correctly.

[lsmod]

Originally Posted by davidfor

The line height problem is because of the style in the book. Each page is actually a table. The actual text is within a td tag that has style "calibre19". This is defined with "line-height: 1.45". The reader app does not override this. Simply removing this from the style definition will allow the line heights to be set. But, it might mess up the layout.

Hmmm - i understand.
It's really complicate to understand and find out the correct options to convert documents in calibre.

For documents from the internet and other direct sources it's a good alternative to use the epub plugin for open office.

Originally Posted by davidfor

The incomplete pages looks to be a layout issue. There are levels of div's and tables around the actual text. When I stripped those out for a couple of chapters, the pages flowed correctly.

But the beginning is showed correctly.
It's not correct to jump over the rest of the page.

How do you analyze this?
I did not work with epub tools before.

[davidfor]

Originally Posted by lsmod

Hmmm - i understand.
It's really complicate to understand and find out the correct options to convert documents in calibre.

Yes it can be. Starting from a web site like this make it hard.

For documents from the internet and other direct sources it's a good alternative to use the epub plugin for open office.

But the beginning is showed correctly.
It's not correct to jump over the rest of the page.

It think that because each chapter is in a table, the reader app is trying to render that as a single page. But, there might be something in one of the styles defined that is forcing this.

I had a look at the original page. If you edit the HTML and delete the code for the site index, it will probably fix this when you do the conversion. That is deleting everything before "<h1 class="firstHeading">AVR-GCC-Tutorial</h1>" and then deleting the matching table ends at the end of the document.

How do you analyze this?
I did not work with epub tools before.

I opened the epub with Sigil and had a look at the code. Sigil is a very good epub editor. There is a forum for it here.

[lsmod]

Thank you for the advice.
I will have a look at Sigil.

There seems no general easy way to convert documents to epub that will fit for the most circumstances.
But it's a pity that calibre generates output formats that makes such problems.
Maybe it helps to try out the possibilites with the profiles ...

[Francois_C]

Sigil is great for polishing ebooks and correcting errors, removing unused styles and images, creating a table of contents when there is none, removing useless portions of code, correcting misspellings. (Thanks to ebooks, now, when we find a misspelling in a book, we are no longer condemned to read it again and again till the end - which is bad for my blood pressure, as I'm a former French teacher - we can fix it
Calibre is the most essential tool for e-books, but it is not the best epup creating tool. Amanuensis and Writer2epub, and even the writer2xhtml addon for LibreOffice create cleaner output files.
But maybe my poor opinion about Calibre for making ebooks can also be explained because it is mostly used by noobs who convert books without understanding what they are doing.
Every time I made a book from an ODT with Calibre, it proved to be good, except for some too complicated stylesheets. But if you avoid direct formatting and use predefined word processor stylesheets for everything except bold and italic within a paragraph, Calibre, Amanuensis, writer2epub and Atlantis make excellent and light epubs.

[1drey]

I want very much:

1. opportunity to export all highlighted and commented text from the book;
2. archive and restore highlighted and commented text
3. sync notes, bookmarks and notes across all my reading devices where Kobo reader is used (including android devices).

[Rev. Bob]

Heck, I'd just like the ability to update my Goodreads status from my Glo...but with the Amazon deal, that's probably not in the cards. Still, even being able to connect my Twitter account (highlight and tweet, for instance) would be a plus.

[OverHaze]

store.kobobooks.com no longer shows the DRM used in the book you are about to buy. Thats an important piece of info, I'd like to know before purchasing if I'll be dealing with Adobe DRM or not.

So less of a feature request and more of an appeal to return some basic functionality to the website.

[davidfor]

Originally Posted by 1drey

.
2. archive and restore highlighted and commented text
3. sync notes, bookmarks and notes across all my reading devices where Kobo reader is used (including android devices).

If you buy books from Kobo and read the kepub version, this already happens. It isn't going to happen for other formats as there are privacy concerns.

[Mad Beer Oil]

1. Option to use custom dictionaries/More dictionaries offered (English <-> Russian would be nice)
2. Dictionary function for PDF files