News from Bookeen about firmware upgrade

[cmbs]

Originally Posted by HarryT

Sorry, but both the Gen3 and the PC version have exactly the same page numbering bugs.

Try this:

Create a book from an HTML file and go to page "100" in it. Make a note of where that page starts. Now add a hyperlink to a point near the bottom of page "100".

Now, if you jump to that hyperlink the point which was previously near the bottom of the page will be right at the top, but the PC version of the Mobi reader will still tell you that you're looking at page 100. Which is the "real" page 100 - the one that you get to by paging through the book, or the one that you get to by jumping to that hyperlink? Different pages are being displayed, so they can't both be the same page, and yet Mobi reader will tell you that both are "page 100".

All versions of the Mobi reader do this. No Mobi device - free or otherwise - does real pagination. Page numbers are an estimates, whether they tell you this or not. This is obvious if you play around with hyperlinks in books.

That's the way that the Mobi readers have always worked. Page numbers are nothing more than estimates based on file position.

Ok. Mobipocket Reader was free. Cybook cost $350. I don't much mind page number bugs in free software. I do mind it in $350 devices.

Personally, I find it useless in it's present state; and the fact they have it leads one to believe it would be easy to add page numbers to the books since the numbers are already being calculated.

A specific page begins with a specific line and ends with a specific line. The fact that you can place a link somewhere in the middle of the page and put that part at the top doesn't change the concept that it's the middle of the page, and further down a new page begins. A link to a specific page number (as in go to page number feature) should go to the top of the page. The number of fixed pages can (theoretically) be calculated, it has nothing to do with the fact that portions of two pages can be displayed at one time. This has never confused me, I don't see the problem.

[delphidb96]

Originally Posted by cmbs

Ok. Mobipocket Reader was free. Cybook cost $350. I don't much mind page number bugs in free software. I do mind it in $350 devices.

Personally, I find it useless in it's present state; and the fact they have it leads one to believe it would be easy to add page numbers to the books since the numbers are already being calculated.

That's your opinion and you have the right to it, but I think most typical readers would find it to be invalid.

A specific page begins with a specific line and ends with a specific line. The fact that you can place a link somewhere in the middle of the page and put that part at the top doesn't change the concept that it's the middle of the page, and further down a new page begins. A link to a specific page number (as in go to page number feature) should go to the top of the page. The number of fixed pages can (theoretically) be calculated, it has nothing to do with the fact that portions of two pages can be displayed at one time. This has never confused me, I don't see the problem.

What I think is happening is that Bookeen started developing their Boo Reader and, as they had a licensing arrangement with Mobipocket, asked for help on how Mobi software does things. And Mobi supplied the algorithm used for determining page positioning, etc. I'm pretty sure Bookeen implemented that way in order to be as compliant to Mobi's internal standard as they could. Well, in much of their implementation, doing so produced quality features. But in this instance - page determination - following Mobi's approach turned out to be, shall we say, 'flawed'?

Yet, here you are, on the one hand demanding they be as close to MobiReader as possible, and on the other hand, bitching and moaning because they are doing it exactly the same way - when doing so doesn't meet your ideals. Boo-hoo-hoo. Cry me a river.

Derek

[HarryT]

Originally Posted by tompe

I would if I used Windows regularly. I will not reboot my laptop just to check this. My observation was done with _1632_ from Baen. The procudure was: jump to page 50, page forward until you see some text you recognize like "chapter 4" and count the number of pages (17 in my case). Jump to page 67 and notice that you are 5 pages off.

I've just tried the experiment on the PC version - it's just as bad.

Here's what I did.

I loaded one of my books (Jane Austen's "Northanger Abbey") and paged through it until I got to the start of chapter 7. The status bar showed page 52. I went back to the start of the book, and then did a "goto page 52". Did this take me to the start of chapter 7? No - it actually took me almost to chapter 8. I had to page back 14 pages - to what was now shown as page 38- in order to find the start of chapter 7, so that's a discrepency of 14 pages in 52.

Page numbers are virtually meaningless in ALL versions of the Mobi reader. I rest my case!

[HarryT]

Originally Posted by delphidb96

What I think is happening is that Bookeen started developing their Boo Reader and, as they had a licensing arrangement with Mobipocket, asked for help on how Mobi software does things. And Mobi supplied the algorithm used for determining page positioning, etc. I'm pretty sure Bookeen implemented that way in order to be as compliant to Mobi's internal standard as they could. Well, in much of their implementation, doing so produced quality features. But in this instance - page determination - following Mobi's approach turned out to be, shall we say, 'flawed'?

Derek - I think that is exactly what's happened. I completely agree with you. Mobi supplied the algorithms - perhaps even the code - which is flawed.

[cmbs]

Originally Posted by delphidb96

Mobipocket also didn't develop the reader app installed on the Cybook. That app was developed, under license from Mobipocket if I understand correctly, by Bookeen. And if you think you have a flawed product, just remember that *MY* first Cybook Gen3 was an 'engineering sample' and didn't even come with Version 1.0 build 1 of the reader app. I got Version 0.91!!!

Yes, Boo Reader is designed to implement the feature-set that Mobipocket Reader offers. But it IS a work-in-progress! Clearly it supports just a subset of MobiReader features. However, the subset offered in V1.0b538 is far and away SUPERIOR to what I first got to use!

I had NO working dictionary-lookup, NO Secure Mobipocket support, NO Go To, No ability to turn page-flashing on/off, and the list goes on and on. Bookeen has made tremendous strides in what features Boo Reader offers.

I fail to see how this is any defense for them. In fact, it's just proving my point. Why in the heck did they sell such an inadequate product? Why didn't they wait until they had something of higher quality before putting these things on the market? Doesn't this just prove that they are interested in sales over quality? And why are you so satisfied with having paid so much for such a poor product? Because it's better than the competition does not make it good. There is a concept I think a lot of you fail to grasp.

Is it as complete as MobiReader 6.x? No. Is it moving in that direction? Yes. Is it worth the $350-$450 we paid? Well, no. But then, Boo Reader is NOT the entirety of the Cybook. Nosirree, Bob! The vast majority of what we paid for is wrapped up in the hardware and the underlying OS! Believe you me, even if you go for a 'generic' e-ink device nearly exactly the same as the Cybook - but without Boo Reader - you're going to pay about $280-$325 (including shipping). And at that you'll have a device which can read TXT, HTML and *some* PDF. I've seen the "library" functions these offer and they're not even as good as what Bookeen offers in the Cybook.

Granted, buying one of these means we can get a cheaper price, but then anyone who buys one is going to have to figure out how to: 1) "upgrade" FBReader to allow reading of Secure Mobipocket and, 2) "port" FBReader to install and run on the device - or - 3) write from scratch a custom Secure Mobipocket reader for the device. This can be done. But it's going to take some time. Gosh durn! By shelling out $350-$450 a customer can get 85% of what Mobipocket Reader supplies on that same basic hardware framework.

Given that most customers are NOT going to want to have to become C++/Python/Java application and systems programmers, I think you might want to cut Bookeen some slack. But that's just my humble opinion.

Derek

I wouldn't be so vocal about the flaws if so many people weren't on here singing the praises of how wonderful and perfect it is. I can't believe some of what some people are saying, because I know I'm not lying, and I know I treat my cybook well. So how can it be that some people's product works perfectly when mine doesn't? I want prospective buyers to have a complete picture of what they are getting into so they can make an informed decision that is right for them.

[tompe]

Originally Posted by HarryT

I've just tried the experiment on the PC version - it's just as bad.

Here's what I did.

I loaded one of my books (Jane Austen's "Northanger Abbey") and paged through it until I got to the start of chapter 7. The status bar showed page 52. I went back to the start of the book, and then did a "goto page 52". Did this take me to the start of chapter 7? No - it actually took me almost to chapter 8. I had to page back 14 pages - to what was now shown as page 38- in order to find the start of chapter 7, so that's a discrepency of 14 pages in 52.

Page numbers are virtually meaningless in ALL versions of the Mobi reader. I rest my case!

I tested with FBReader and noticed that one page forward not always gave me one new page number. I did not find a jump to page but I suspect that it could work better with FBReader.

I am suprised because it seems simple to implement something that works better. There is no reason why page jump and paging forward should en up on different pages if you define page as what the page number says.

[delphidb96]

Originally Posted by HarryT

Derek - I think that is exactly what's happened. I completely agree with you. Mobi supplied the algorithms - perhaps even the code - which is flawed.

So what needs to happen, over time, is Bookeen has to decide whether to black-box a new implementation which doesn't rely upon *any* of Mobi's 'flawed' algorithms. Has anyone checked out whether FBReader makes the same mistake? Perhaps the algorithms used by it's display modules might be a good source for 'updating' Boo Reader.

Derek

[HarryT]

Originally Posted by cmbs

I fail to see how this is any defense for them. In fact, it's just proving my point. Why in the heck did they sell such an inadequate product?

Because when you licence a product from someone you're generally not allowed to make ad hoc changes to it. When Bookeen licenced the rights to implement the MobiPocket Reader on the Gen3 they probably had to produce a product that performed to the specs and algorithms that Mobi supplied them with - bugs and all.

[cmbs]

Originally Posted by delphidb96

That's your opinion and you have the right to it, but I think most typical readers would find it to be invalid.

Which part?

I think the "typical reader" expects more from an expensive product than they do from a free one, and I think it's completly valid.

I also think the "typical reader" is going to assume, by looking at what's presented, that they are calculating page numbers and offering to send you to a specific page. They would assume that because that's exactly what it looks like.

Originally Posted by delphidb96

What I think is happening is that Bookeen started developing their Boo Reader and, as they had a licensing arrangement with Mobipocket, asked for help on how Mobi software does things. And Mobi supplied the algorithm used for determining page positioning, etc. I'm pretty sure Bookeen implemented that way in order to be as compliant to Mobi's internal standard as they could. Well, in much of their implementation, doing so produced quality features. But in this instance - page determination - following Mobi's approach turned out to be, shall we say, 'flawed'?

Yet, here you are, on the one hand demanding they be as close to MobiReader as possible, and on the other hand, bitching and moaning because they are doing it exactly the same way - when doing so doesn't meet your ideals. Boo-hoo-hoo. Cry me a river.

Derek

I have not demanded that they be as close to mobireader as possible, although it would be an improvement if they were more like it. Nor am I bitching and moaning that they are. Mobipocket Desktop Reader was free AND it works better. Granted the cybook offers features the desktop reader doesn't. Do you always spend your money like it's water? Some of us don't appreciate spending a lot of money on things that don't work correctly (and don't underestimate the importance of poor customer service). In fact, it's possible that the "typical reader" cares about the product they get for the money. Even assuming that most people don't care how they spend their money and what they get for it, some people do, and it's completely valid. Just because a feature is unimportant to you, doesn't make it unimportant.

[delphidb96]

Originally Posted by cmbs

I fail to see how this is any defense for them. In fact, it's just proving my point. Why in the heck did they sell such an inadequate product? Why didn't they wait until they had something of higher quality before putting these things on the market? Doesn't this just prove that they are interested in sales over quality? And why are you so satisfied with having paid so much for such a poor product? Because it's better than the competition does not make it good. There is a concept I think a lot of you fail to grasp.

'Inadequate'... If you feel that strongly against the Cybook, whyinsamhell did you bother to buy one? First, the Cybook was released, what, three months ago? That means it's still first-generation for the software - Version 1.0! Second, all of the currently-available-in-the-US e-ink readers are, at the e-ink display level, 'first-generation'. That means any person buying one is clearly in the 'innovator' or 'early-adopter' category. Yet here you are behaving as if you are a 'mainstream' or 'late-adopter' who demands perfection from the get-go. Sorry, but you can't be an 'early-adopter' of new technology and demand 'late-adopter' perfection from what you're buying.

I wouldn't be so vocal about the flaws if so many people weren't on here singing the praises of how wonderful and perfect it is. I can't believe some of what some people are saying, because I know I'm not lying, and I know I treat my cybook well. So how can it be that some people's product works perfectly when mine doesn't? I want prospective buyers to have a complete picture of what they are getting into so they can make an informed decision that is right for them.

No, you're not lying. But you are ignoring reality to present biased views. The reality is that even the Kindle is a 'first-generation' device (that being the newest gee-whiz e-ink version) and anyone who buys a 'first-generation' of any new technology has to expect bugs, kinks and flaws in the operation of same. We do point out the flaws, but we don't try to paint it as a complete rip-off and abdication of developer responsibility - which is, even if you don't use those words, exactly what you're doing.

Derek

[tompe]

Originally Posted by HarryT

Because when you licence a product from someone you're generally not allowed to make ad hoc changes to it. When Bookeen licenced the rights to implement the MobiPocket Reader on the Gen3 they probably had to produce a product that performed to the specs and algorithms that Mobi supplied them with - bugs and all.

You are making excuses. I refuse to believe that they are not allowed to implement a working goto page.

[HarryT]

I think that my experiment with the Windows Mobi Reader shows that this behaviour is "generic" to all versions of the product. I can only assume that there is a rational explanation for this, such as Mobi specifying to a licensee what can and cannot be changed. As you say, it would have been reasonable for Bookeen's programmers to have changed this blindingly obvious flaw had they been able to. Therefore, there must have been a reason that they did not. I don't believe them to be incompetent.

[delphidb96]

Originally Posted by cmbs

Which part?

I think the "typical reader" expects more from an expensive product than they do from a free one, and I think it's completly valid.

I also think the "typical reader" is going to assume, by looking at what's presented, that they are calculating page numbers and offering to send you to a specific page. They would assume that because that's exactly what it looks like.

The typical reader is going to assume that a 'new' technology is going to have bugs, kinks and flaws. Witness Windows 1.0, 2.01, 3.0, 95, 98, ME, NT, XP and Vista. There's a saying - perhaps you've heard it - "Never buy 1.0 of *anything*." There's an addendum to that. "But if you do, expect you'll have to pay more and live with the bugs until they get fixed." Now just what part of that do you have a problem understanding? You CHOSE to buy the Cybook when it was in it's first generation. You didn't have to.

I have not demanded that they be as close to mobireader as possible, although it would be an improvement if they were more like it. Nor am I bitching and moaning that they are. Mobipocket Desktop Reader was free AND it works better. Granted the cybook offers features the desktop reader doesn't. Do you always spend your money like it's water? Some of us don't appreciate spending a lot of money on things that don't work correctly (and don't underestimate the importance of poor customer service). In fact, it's possible that the "typical reader" cares about the product they get for the money. Even assuming that most people don't care how they spend their money and what they get for it, some people do, and it's completely valid. Just because a feature is unimportant to you, doesn't make it unimportant.

But YOU are NOT a typical customer. You chose to buy a Cybook when it was 'first genertion'. You chose to be an 'early-adopter'. You chose, despite knowing all this and making these prior choices, to then bitch and moan over minor inconveniences as if they were the worst possible failures! That is what makes your most recent posts on this nothing more than bitching and moaning while distorting reality.

No, that is not Bookeen's position, nor is it Not Another E-Book's. That opinion is solely mine. And I'm not wasting any further time on what you have to say about this.

Derek

[cmbs]

Originally Posted by delphidb96

'Inadequate'... If you feel that strongly against the Cybook, whyinsamhell did you bother to buy one? First, the Cybook was released, what, three months ago? That means it's still first-generation for the software - Version 1.0! Second, all of the currently-available-in-the-US e-ink readers are, at the e-ink display level, 'first-generation'. That means any person buying one is clearly in the 'innovator' or 'early-adopter' category. Yet here you are behaving as if you are a 'mainstream' or 'late-adopter' who demands perfection from the get-go. Sorry, but you can't be an 'early-adopter' of new technology and demand 'late-adopter' perfection from what you're buying.

No, you're not lying. But you are ignoring reality to present biased views. The reality is that even the Kindle is a 'first-generation' device (that being the newest gee-whiz e-ink version) and anyone who buys a 'first-generation' of any new technology has to expect bugs, kinks and flaws in the operation of same. We do point out the flaws, but we don't try to paint it as a complete rip-off and abdication of developer responsibility - which is, even if you don't use those words, exactly what you're doing.

Derek

That's funny. I'm pretty sure you're the one ignoring reality.

[tompe]

Originally Posted by HarryT

I think that my experiment with the Windows Mobi Reader shows that this behaviour is "generic" to all versions of the product. I can only assume that there is a rational explanation for this, such as Mobi specifying to a licensee what can and cannot be changed. As you say, it would have been reasonable for Bookeen's programmers to have changed this blindingly obvious flaw had they been able to. Therefore, there must have been a reason that they did not. I don't believe them to be incompetent.

You are trying to find explanations to something that I am convinced of is just that they did not have time (or money to hire more people) to do it.
Or they just do the minimal thing they can get away with and still earn the same amount of money.